Analysis of endemicity and regionalisation of the freshwater ichthyofauna of Middle America

Amairany Bernal-Portillo; Fabian Pérez-Miranda; Omar Mejía

doi:10.21425/fob.18.164902

Abstract

The first step towards understanding the spatial and temporal patterns of diversity amongs freshwater fishes is through the delimitation of areas of endemism. In this work, we used two of the most popular methodologies for recognising areas of endemism, Parsimony Analysis of Endemicity (PAE) and Endemicity Analysis (EA) to describe and compare the number of endemic regions of freshwater fishes in Middle America using both grid cells (0.5° and 1° resolution). Besides, we built a regionalisation scheme to compare PAE and cluster analysis using different natural basins as study units. The number of areas of endemism recovered varied widely depending on the method used, PAE being a better approach than EA. PAE identified six provinces and 38 sub-provinces for the ichthyofauna of Middle America, most of which were similar to those described previously. We describe an ichthyological province not previously listed and recovered both ocean slopes from southern Mexico to Panamá as a single province instead of two as previously suggested.

Highlights

We used Parsimony Analysis of Endemicity (PAE) and Endemicity Analysis (EA) to assess the patterns of endemicity amongst the freshwater fishes of Middle America;
PAE regionalisation recovered a total of eight provinces and 26 sub-provinces;
Most of the provinces recovered by PAE were congruent with earlier approaches, allowing the support of a general purpose regionalisation scheme.

Keywords

cluster analysis, grid cells, hydrobasins, indicator species, PAE, VNDM

Introduction

The recognition of areas of endemism is a fundamental task for the understanding of spatial and temporal distribution patterns in freshwater fishes. These units, defined as regions of congruent and non-stochastic distributions of two or more taxa (Nelson and Platnick 1981; Morrone 1994; Estrada-Márquez et al. 2021), represent the starting point in evolutionary biogeography that allows tracking a unique evolutionary history (Ferrari et al. 2010; Pinton et al. 2013; Fattorini 2017). Such knowledge makes it possible to analyse the processes that determine species turnover in landscapes in terms of species richness and community composition and structure (Smith and Bermingham 2005). In addition, identifying areas of endemism allows the definition of ichthyological provinces that represent geographic regions of relatively homogeneous fauna (Roberts 1975; Lévêque 1997). Despite this, there is no universally accepted method for identifying such areas (Carine et al. 2009; Escalante et al. 2009; Torres-Miranda et al. 2013; Hoffmeister and Ferrari 2016). At present, multiple approaches (Rosen and Smith 1988; Harold and Mooi 1994; Morrone 1994; Crisp et al. 2001; Linder 2001; Hausdorf 2002; Szumik et al. 2002; Hausdorf and Hennig 2003; Fattorini 2017) have been implemented; within these, we focus on two of the most common. First, Parsimony Analysis of Endemicity (PAE) constructs area cladograms from a presence-absence data matrix, where areas of endemism are recognised by at least two synapomorphic taxa (Morrone 1994, 2014); meanwhile, Endemicity Analysis (EA) uses an optimality criterion for the identification of areas of endemism through the explicit evaluation of congruence amongst species distributions over grid cells, allowing the recognition of nested, overlapping and disjunct areas (Szumik et al. 2002; Szumik and Goloboff 2004; Carine et al. 2009; Casagranda et al. 2009, 2012).

Middle America represents a transition zone between the Nearctic and Neotropical Regions, covering a wide area with a complex topography and climate, resulting in a region with high species richness and endemicity (Darlington 1957; Brown and Gibson 1983; Cox and Moore 1985; Ortega and Arita 1998; Albert and Reis 2011; Winker 2011). As in other transition zones, the boundaries between the Nearctic and Neotropical Regions are not clearly defined; however, in this paper, we define Middle America as the region that extends from México to the border between Panamá and Colombia (sensu Ortega and Arita (1998); Albert and Reis (2011); Winker (2011)). This delimitation excludes 15 species of Neotropical origin that could reach the United States; nevertheless, they represent less than 2% of the total species richness and are unlikely to have a significant impact on the overall pattern.

Historically, there have been several attempts to regionalise this geographic area using chorological, phenetic and cladistic methods (Miller 1966; Bussing 1976; Miller et al. 2005; Smith and Bermingham 2005; Matamoros et al. 2012; Říčan et al. 2013; Matamoros et al. 2015; Říčan et al. 2016; Elías et al. 2020; Rico et al. 2022), but none has included the complete assemblage of freshwater fish species of the region. In the present work, we first analyse and compare the endemicity patterns of freshwater fishes of Middle America through PAE and EA, using different grid cells as area units. After it, we develop and compare a regionalisation scheme for this geographic area using PAE and cluster analysis with different natural basins as geographic units.

Materials and methods

Distribution data

Collection records for 685 species of native freshwater fish species of Middle America were retrieved from the Global Biodiversity Information Facility server (GBIF) (Mejía 2022) using the dismo R library (Hijmans et al. 2017) (Suppl. material 1). We restricted the analysis to species that spend the critical phases of their life cycles in inland freshwater bodies and that eventually penetrate estuarine brackish water (Contreras-MacBeath et al. 2022). Several quality filters were applied to curate the database (Jin and Yang 2020). First, human observations were pruned, leaving only collection records that had a voucher in a biological collection. Data for each species were projected on to the geographic space of Middle America, the records with coordinates located in the sea were deleted and, finally, using checklists for the study area (Lee et al. 1980; Ruiz-Campos et al. 2002; Kullander 2003; Lucinda 2003; Smith et al. 2004; Miller et al. 2005; Valdez-Moreno et al. 2005; Matamoros et al. 2009; Angulo et al. 2013; McMahan et al. 2013; Espinosa-Pérez and Ramírez 2015; Říčan et al. 2016), we removed geographic outliers according to the geographic boundaries of each species and updated the nomenclature of each one according to Eschmeyer’s fish catalogue (Fricke et al. 2022).

Analysis of endemicity

Two different geographic units, 0.5° and 1.0° grid cells, were employed to compare the performances of the two different methodologies (PAE and EA). The grid cells were obtained with the raster R library 3.6–30 (Hijmans 2025). For the PAE analysis, a presence-absence matrix for each of the grids resolution was obtained using the fossil 0.4 R library (Vavrek 2011) (Suppl. material 2). The parsimony analysis was implemented in TNT 1.5 (Goloboff and Catalano 2016) using a heuristic search with a driven search algorithm (Goloboff et al. 2008). The most parsimonious trees were used to construct a strict consensus tree. Species distributions (characters) were optimised in WINCLADA 5.3 (Nixon 2002) and areas of endemism were recognised as the more inclusive nodes supported by at least two synapomorphies. Synendemic species (species with a congruent distribution that supported a node) were removed from the matrix and the PAE procedure was iterated until no more areas of endemism were recovered (Santiago-Alvarado et al. 2022).

The Endemicity Analysis was implemented in the NDM/VNDM software version 3.0 (Szumik and Goloboff 2004) with the following parameters for both grid cell sizes, ten random seeds and 10 repetitions. The radius was set to zero in order not to infer species distributions in neighbouring cells. The minimum number of endemic species was set at 2 for each area, with 50% of the unique species reserved for area maintenance. Finally, a flexible consensus of unique species was obtained at 30% and 50% thresholds. As in the PAE, synendemic species were removed and the procedure was iterated until no more areas were recovered.

Comparison amongst methods

The number and composition of areas of endemism recovered differed amongst the methods; thus, instead of comparing the areas recovered by each method, we opted for a simpler technique to evaluate the level of congruence amongst the different approaches. We compared the level of overlap amongst areas of endemism polygons using the dissolve and overlap tools implemented in QGIS 3.16.12 (QGIS 2021).

Icthyological provinces of freshwater fish of Middle America

For the regionalisation of provinces and sub-provinces, Pfastetter hydrological basins of levels 4 and 5 (Lehner and Grill 2013; available at: https://www.hydrosheds.org/) as well as the basins and sub-basins of the world listed by the Food and Agriculture Organization of the United Nations 2022 (available at: https://data.apps.fao.org/catalog//iso/7707086d-af3c-41cc-8aa5-323d8609b2d1) were used. For the PAE of FAO basins, an exhaustive search was implemented using the implicit enumeration algorithm; meanwhile, for the rest of the geographic units (FAO sub-basins and Pfastetter hydrological basins of level 4 and 5), a driven search was performed as described earlier.

The cluster analysis and the recognition of indicator species were performed according to the method proposed by Rico et al. (2022). Briefly, starting from the presence-absence matrix (PAM), we first estimated the Jaccard dissimilarity index using the betapart R library (Baselga and Orme 2012). Then, we constructed a dendrogram using the UPGMA algorithm and estimated the goodness-of-fit between the original data and the tree matrices using the co-phenetic correlation coefficient (Sokal and Rohlf 1962) in the R stats library (R Core Team 2023). Finally, we estimated the number of geographic clusters using the Kelley-Gardner-Sutcliff (KGS) penalty algorithm (Kelley et al. 1996) implemented in the R library maptree 1.4-8 (White and Gramacy 2022). After estimating the number of geographic clusters, we tested for significant differences using an ANOSIM (Clarke and Warwick 1994). Finally, we used indicator species analysis (ISA) (Dufrêne and Legendre 1997), where species with values above 0.75 were mostly restricted to a geographic province or sub-province and absent from others. The analysis was implemented using the R library indicspecies 1.8.0 (De Cáceres and Legendre 2009).

The proposed provinces, sub-provinces and districts for Middle America were based on the PAE results and the names proposed followed the FAO nomenclature for basins and sub-basins (Food and Agriculture Organization of the United Nations 2022) (see below).

Results

A total of 52131 collection records were recovered after quality filtering. The results for the 0.5° grid cells are available in Suppl. material 3, in the same way, the results of Pfastetter 4 and 5 hydrobasin levels for the comparison of PAE and cluster analysis are available in the Suppl. material 7. In addition, all the input files and raw results of all analyses are available at https://github.com/HOmarMejiaG/Fishes-of-Mexico-and-Central-America/tree/main.

Areas of endemism

The analysis using 1° grid cells with PAE allowed 39% of the total area of Middle America to be regionalised, recovering 31 areas of endemism with three nested areas within area 31 (Fig. 1A) These areas are supported by the distribution of 130 species of 19 families, with Poeciliidae, Cichlidae and Cyprinodontidae being the most represented families (Suppl. material 4). Meanwhile, for the 0.5° grid cells, 35 areas of endemism were recovered and only 9% of the total area was regionalised (Suppl. material 3). For the Endemicity Analysis (EA) of one degree (1°) grid cells, a total of 24 areas of endemism were recovered using the 30% flexible consensus threshold, where the endemicity index (EI) ranged from 0.785 to 21.316 and included a total of 330 endemic species (Fig. 1B, Suppl. material 5). The 50% consensus threshold recovered 45 areas of endemism, where the EI ranged from 0.785 to 21.316 and included a total of 352 endemic species (Fig. 1C, Suppl. material 5). Meanwhile, the half degree (0.5°) grids cells recovered a total of 33 areas of endemism with 30% consensus, where the endemicity index (EI) ranged from 1.000 to 11.413 and included a total of 212 endemic species (Suppl. materials 3, 5). Finally, the 50% consensus threshold recovered 39 areas of endemism, where the EI ranged from 1.000 to 11.413 and included a total of 207 endemic species (Suppl. materials 3, 5).

Figure 1.

Endemism areas recovered for the freshwater icthyofauna of Middle America using 1° grid cells with different methods. A. PAE; B. EA with a 30% consensus threshold; C. EA with a 50% consensus threshold.

Comparison amongst methods

The 1° grid cells analysis using both methods showed higher levels of geographic overlap compared to the 0.5° grid cells. For the 1° grid cells, the comparison between PAE and the consensus areas of EA at 30% showed 44.528% overlap and a slightly lower value (43.363%) for the comparison between PAE and the consensus areas of EA at 50%. For the 0.5° grid cells, the comparison between PAE and the consensus areas of EA at 30% showed 30.273% overlap and a slightly lower value (30.723%) for the comparison between PAE and the consensus areas of EA at 50%. Finally, the comparison between PAE and consensus areas of EA at 30% recovered an overlap of 29.447%.

Regionalisation of Middle America

Six areas of endemism were recovered using the FAO basins; this scheme allowed regionalising of 99% of the complete surface of Middle America (Fig. 2A, B). The areas are supported by 59 species belonging to 20 families, with Poeciliidae being the most represented, with 15 species (Table 1). In contrast, analysis of FAO sub-basins regionalised 47% of the total surface of Middle America, recovering a total of 35 endemism areas, with three smaller areas nested within area 23 (Fig. 2C). These areas are supported by 130 species belonging to 22 families, with Leuciscidae, Cichlidae and Poeciliidae being the most represented (Table 3). For the cluster analysis, a total of six icthyological provinces were recovered for FAO basins (Fig. 3A, B), despite the co-phenetic correlation index being very high (R_cof = 0.92), with a total of 575 indicator species, of which 181 had values ≥ 0.75, implying that species are distributed in most of the basins that comprise the province, the overall ANOSIM results were significant, but the pairwise comparisons amongst areas were not. Thus, under these criteria, none of the six postulated provinces could be considered as valid (Suppl. material 6). The FAO sub-basins recovered eight sub-provinces, although two were not continuous in the geographic space (Fig. 3C). The co-phenetic correlation was high (R_cof = 0.84), with a total of 648 indicator species; however, only nine species had values ≥ 0.75, with Poeciliidae and Cichlidae being the families with the most species; in this case, the ANOSIM results suggest that only areas 2 and 7 were significantly different from the others (Suppl. material 6). Our regionalisation scheme is based on the results obtained using PAE and, due to in the cluster analysis using UPGMA, the ANOSIM failed to recover significant differences. The FAO system recognised 16 basins and 307 sub-basins for Middle America; thus, we named our provinces and sub-provinces based on the names suggested by FAO, considering the name of the basin or sub-basin with the largest area. We added two additional provinces well supported as areas of endemism from Pfafstetter level 4 (areas 1 and 2 in Suppl. material 7), for a total of eight ichthyological provinces (Fig. 2A, Suppl. material 7).

Table 1.

Download as

CSV

XLSX

Provinces recovered for the freshwater icthyofuana of Middle America obtained with PAE.

Number	Name	FAO basins included	Species richness	Supporting species
1	Lower Colorado	Colorado	6	Gila elegans , Ptychocheilus lucius, and Xyrauchen texanus.
2	Lower Santa Cruz-San Pedro	Santa Cruz, San Pedro	9	Agosia chrysogaster , Catostomus insignis, and Pantosteus clarkii
3	Baja California	Baja California	4	Fundulus lima and Fundulus parvipinnis
4	Central Mexican Highlands	Mexico Interior	52	Astyanax argentatus , Cyprinodon inmemoriam, Cyprinodon latifasciatus, Megupsilon aporus, Characodon garmani, and Stypodon signifer
5	Mexican Pacific Coast	Mexico Northwest Coast, Río Lerma, Pacific Central Coast, Río Balsas	186	Mayaheros beani , Ictalurus dugesii, Poecilia butleri, Poeciliopsis latidens, Poeciliopsis presidionis, Poeciliopsis prolifica, and Poeciliopsis viriosa
6	Mexican North Atlantic	Río Grande-Bravo, North Gulf, Río Verde	174	Fundulus grandis , Pylodictis olivaris, Poecilia formosa, and Poecilia latipinna
7	Isthmic-Peninsular	Papaloapan, Isthmus of Tehuantepec, Grijalva-Usumacinta, Yucatán Península	205	Atherinella alvarezi , Atherinella schultzi, Hyphessobrycon compressus, Maskaheros regani, Paraneetroplus bulleri, Thorichthys callolepis, Thorichthys helleri, Trichromis salvini, Dorosoma anale, Gambusia yucatana, Priapella compressa, and Priapella intermedia
8	Central American	Southern Central America, Caribbean Coast, Colombia-Ecuador Pacific Coast	236	Brycon argenteus , Brycon striatulus, Astyanax panamensis, Compsura gorgonae, Gephyrocharax atracaudatus, Pseudocheirodon arnoldi, Roeboides occidentalis, Andinoacara coeruleopunctatus, Panamius panamensis, Ctenolucius beani, Cyphocharax magdalenae, Hoplias malabaricus, Hoplias microlepis, Pimelodella chagresi, Brachyhypopomus occidentalis, Lebiasina panamensis, Ancistrus chagresi, Ancistrus spinosus, Chaetostoma fischeri, Leptoancistrus canensis, Rineloricaria uracantha, Brachyrhaphis episcopi, Hiatirhaphis cascajalensis, Neoheterandria tridentiger, Poecilia gillii, Poeciliopsis elongata, and Trichomycterus striatus

Figure 2.

Endemism areas recovered for the freshwater icthyofauna of Middle America with PAE A. FAO basins; B. Area Cladogram recovered with FAO basins; C. FAO sub-basins.

Lower Colorado province

This area encompasses the southern portion of the Colorado Basin and comprises a total of six species across four families. Leuciscidae is the most abundant. This area is supported by the distribution of three species (Table 1).

Lower Santa Cruz-San Pedro province

This area comprises the lower portions of the Santa Cruz and San Pedro basins. The species richness is low, with only nine species in the Leuciscidae and Catostomidae. This area is supported by the distribution of three species (Table 1).

Baja California province

This area comprises the Baja California basin and includes only four species. Fundulidae is the dominant family. This area is supported by the distribution of two species (Table 1).

Central Mexican Highlands province

This area comprises the inner Mexican basins and includes 52 species, with Leuciscidae being the dominant family. This area is supported by the distribution of six species (Table 1).

Mexican Pacific Coast province

This area comprises the basins of Northwest Mexico, the Pacific central coast and the Lerma and Balsas basins. The area harbours 186 species, where Leuciscidae, Goodeidae, Atherinidae and Poeciliidae are dominant. This area is supported by the distribution of seven species (Table 1).

Mexican North Atlantic province

This area comprises the Río Grande-Bravo, North Gulf of Mexico and Río Verde basins with a richness of 174 species, with Leuciscidae and Poeciliidae the most represented. This area is supported by the geographic distributions of four species (Table 1).

Isthmic-Peninsular province

The area comprises the Papaloapan, Tehuantepec isthmus, the Grijalva-Usumacinta basin and the Yucatán Península. The area harbours 205 species, with Cichlidae and Poeciliidae being the most abundant. This area is supported by the geographic distribution of 12 species (Table 1).

Central American province

This area comprises the southern basins of Central America, as well as the Caribbean and Pacific Coasts, with a total of 236 species. Acestrorhamphidae, Cichlidae and Poeciliidae are the most abundant families. This area is supported by the distributions of 27 species (Table 1).

A total of 38 sub-provinces and three districts within sub-province 26 were recovered supported by two to 13 species (Fig. 1B, Table 2).

Figure 3.

Endemism areas recovered for the freshwater icthyofauna of Middle America with Quantitative phenetic methods A. FAO basins; B. Cluster analysis phenogram obtained with FAO basins; C. FAO sub-basins.

Table 2.

Download as

CSV

XLSX

Sub-provinces and districts for the freshwater icthyofuana of Middle America obtained with PAE.

Number	Name	FAO sub-basins included	Species richness	Supporting species
1	Bajo Colorado	South portion of Colorado basin	6	Gila elegans , Ptychocheilus lucius and Xyrauchen texanus
2	Lower Santa Cruz-San Pedro	Santa Cruz and San Pedro	9	Catostomus insignis and Pantosteus clarkii
3	El Paso–Las Cruces	El paso/Las Cruces	3	Notropis orca and Notropis simus
4	Concepción-Fuerte	Concepción/Arroyo Cocaspera, Sonora, Yaqui, Matepe, Mayo and Fuerte	52	Atherinella elegans , Gila brevicauda, Gila minacae, Meda fulgida, Poeciliopsis jackschultzi, Poeciliopsis occidentalis and Poeciliopsis sonoriensis
5	Santa María	Santa María-Panamá	21	Cyprinodon fontinalis and Cyprinella bocagrande
6	Presa El Granero	Conchos and Presa El Granero	35	Cyprinodon julimes , Cyprinodon pachycephalus, Cyprinella panarcys and Gambusia zarskei
7	Florido-Presa de la Colina	Florido and Conchos/Presa de la Colina	39	Cyprinella panarcys and Gambusia alvarezi
8	La Amistad	Elm/Sycamore and La Amistad	31	Gambusia clarkhubbsi and Gambusia krumholzi
9	Bravo-Sosa	Bravo/Sosa	49	Herichthys minckleyi , Cyprinodon atrorus, Cyprinodon bifasciatus, Lucania interioris, Cyprinella xanthicara, Notropis megalops, Notropis saladonis, Etheostoma lugoi, Etheostoma segrex, Gambusia longispinis, Gambusia marshi, Xiphophorus gordoni and Xiphophorus meyeri
10	Laguna de Mayrán	Laguna de Mayrán/Viesca	9	Cyprinodon latifasciatus , Characodon garmani and Stypodon signifer
11	Río Grande-Bravo	Bravo/ San Juan	43	Moxostoma albidum , Gila modesta, Notropis megalops and Xiphophorus couchianus
12	San Fernando	San Fernando	48	Cyprinodon bobmilleri y Fundulus philpisteri
13	San Pablo	San Pablo	14	Cyprinodon inmemoriam and Megupsilon aporus
14	San Pedro	San Pedro	32	Cyprinodon meeki , Characodon audax, Characodon lateralis, Notropis aulidion and Poeciliopsis presidionis
15	Santiago-Huicicila	Santiago/Aguamilpa and Huicicila/San Blas	19	Atherinella pellosemeion , Algansea avia and Poeciliopsis presidionis
16	Ameca-Atenguillo	Ameca/Atenguillo	34	Allotoca goslinei , Xenotoca doadrioi, Zoogoneticus tequila, Algansea amecae, Notropis amecae and Yuriria amatlana
17	Santiago-Lerma	Santiago Guadalajara, lago Chapala, Lerma/Chapala, lago de Pátzcuaro/Cuitzeo and lago de Yuriria	58	Chirostoma labarcae , Algansea popoche, Yuriria chapalae and Tetrapleurodon spadiceus
18	Coahuayana	Coahuayana	21	Allodontichthys tamazulae , Allodontichthys zonistius and Xenotoca lyonsi
19	Tepalcatepec-Infiernillo	Tepalcatepec/Infiernillo	27	Chirostoma copandaro and Chirostoma zirahuen
20	Tamesí-Moctezuma	Tamesí, Tamuín and Moctezuma	79	Notropis tropicus , Gambusia vittata, Xiphophorus montezumae and Xiphophorus pygmaeus
21	Tecolutla	Tecolutla and Cazones	20	Rheoheros coeruleus and Gambusia senilis
22	Nautla	Nautla	16	Astyanax tamiahua and Herichthys deppii
23	Atoyac A	Atoyac A	23	Poblana alchichica , Poblana letholepis, Poblana squamata and Notropis boucardi
24	Atoyac B	Atoyac B	11	Notropis cumingii and Profundulus oaxacae
25	Ixtapa-Omotepec	Ixtapa, Coyuquilla, Costa de Atoyac, Papagayo, Nexpa and Omotepec	18	Gobiesox mexicanus , Profundulus balsanus and Profundulus mixtlanensis
26	Grijalva-Usumacinta	Jamapa, Papaloapan, Coatzacoalcos, Tonalá/Lagunas del Carmen/Machona, Grijalva/Villa Hermosa, Chixoy, Lacantún, Salinas, Pardón, San Pedro/Guatemala, Laguna de Términos, Champotón, Yucatán, Quintana Roo, Cerradas, Bahía de Chetumal, New, Freshwater Creek/Cowhead Creek and Belize River	171	Thorichthys meeki and Phallichthys fairweatheri
26 A	Papaloapan-Grijalva	Jamapa, Papaloapan, Coatzacoalcos, Tonalá/Lagunas del Carmen/Machona, Grijalva/Villa Hermosa, Chixoy, Lacantún, Salinas, Pardón and San Pedro/Guatemala	146	Maskaheros argenteus , Oscura heterospila and Rheoheros lentiginosus
26 B	Yucatán-Quintana Roo	Yucatán-Quintana Roo	35	Typhlias pearsei and Carlhubbsia kidderi
26 C	Cerradas	Cerradas	42	Cyprinodon beltrani , Cyprinodon labiosus, Cyprinodon maya, Cyprinodon simus, Cyprinodon suavium and Cyprinodon verecundus
27	Tuxtla-Concordia	Grijalva/Tuxtla Gutiérrez and Grijalva/Concordia	50	Chiapaheros grammodes and Vieja breidohri
28	Moho-Lago de Izabal	Grande/Moho, Sarstum, Polochic and lago de Izabal/Dolce	47	Potamarius izabalensis , Cincelichthys bocourti, Carlhubbsia stuarti and Pseudophallus galadrielae
29	Motagua	Motagua	43	Cathorops melanopus and Chuco microphthalmus
30	Choluteca	Choluteca	19	Amphilophus hogaboomorum and Tlaloc portillorum
31	Lagos de Nicaragua	lago superior de Nicaragua and lago Nicaragua	49	Atherinella sardina , Astyanax nasutus and Amphilophus labiatus
32	Tárcoles	Tárcoles/Virilla/Colorado	16	Cynodonichthys glaucus and Cynodonichthys siegfriedi
33	General-Sierpe	General, Terraba and Sierpe	31	Pseudocheirodon terrabae and Imparfinis lineatus
34	Cuervito	Cuervito	20	Astyanax kompi and Cynodonichthys uroflammeus
35	Telire-Calovébora	Estrella, Telire, Sixaola, Teribe/Culubre, Changuinola/Cricamola and Cricamola/Calovébora	46	Amatitlania kanna , Amatitlania myrnae, Cribroheros bussingi, Cribroheros rhytisma and Phallichthys quadripunctatus
36	Coclé del Norte	Coclé del Norte	24	Odontostilbe mitoptera , Roeboides loftini and Cynodonichthys monikae
37	Canal de Panamá - Costa del archipiélago de San Blas	Canal de Panamá and Costa del archipiélago de San Blas	59	Astroblepus mendezi , Astroblepus pirrensis, Roeboides carti, Mesoheros atromaculatus and Cynodonichthys montium
38	Grande Panamá-Tuira	Santa María, Grande-Panamá, Antón/Caimito, Caimito/Juan Díaz/Pacora, Canal de Panamá, Costa de Bahía de Panamá, Tuira and costa del Golfo de Panamá	87	Trachelyopterus amblops , Sturisomatichthys panamensis and Priapichthys darienensis

Discussion

Areas of endemism

The majority of historical biogeographical studies rely on the recognition of areas of endemism, as these represent the basis for primary biogeographical homology (Morrone 2008, 2017; Parenti and Ebach 2009). However, there is no consensus concerning how these areas should be delimited. In the present work, we utilised two of the most popular methods for delimiting areas of endemism for the freshwater ichthyofauna of Middle America. Previous studies have compared the performance of methods, based on three criteria, i.e. presence, diagnosability and contiguity (Carine et al. 2009). Presence refers to the possession of at least two endemic species in the set of grids; diagnosability refers to the exclusive presence of at least two shared species and contiguity implies that grids are not disjunct in the geographic space. Based on the third criterion, we suggest that both methods, PAE and EA, are useful for the delimitation of areas of endemism, as they explicitly incorporate distribution data (Aagesen et al. 2013).

The number of areas of endemism recovered with both methods was similar, although there was less than 50% overlap between the results. Several authors have criticised the use of PAE for delimitation because this method does not take into account the spatial relationships amongst cells, PAE is a hierarchically structured approach and this precludes the recognition of overlapping areas of endemism, leading to recovering fewer areas with lower numbers of grid cells (Szumik and Goloboff 2004; Moline and Linder 2006; Carine et al. 2009; Casagranda et al. 2012; Canales et al. 2016). The results of this study support the aforementioned predictions, as the EA recovered a higher number of areas of endemism that nearly covered the total area of Middle America (Fig. 1B, C). The analysis identified numerous overlapping regions, an expected result in transition zones (Fattorini 2017).

The fundamental aspect that must be evaluated when comparing PAE and EA is the strict overlap or sympatry of endemic species (Escalante 2015). In this work, the number of species that supported the areas derived from PAE ranged from two in the 0.5° grid cells to nine in the 1° grid cells (Suppl. material 4). Meanwhile, the EA results indicated that the number of species recovered in areas of endemism ranged from one to 57 (Suppl. material 5). While these results suggest the better performance of EA, inspection of the consensus areas of EA with at least two species with endemicity indices of 1.0 (i.e. that the species is distributed in each cell of the grid) leads to an extreme reduction in the endemism areas that fulfil this criterion. For the 1° grid cell analysis, only three regions were recovered at 30% consensus and six areas were recovered at 50%, while, for the 0.5° grid cells, four and six endemism areas were recovered, respectively. Thus, based on our results, we found that PAE represents a better approach than EA for delineating areas of endemism in the freshwater fishes of Middle America; nevertheless, this statement cannot be generalised to other biological groups of study areas.

The use of different grid cell sizes enables the recovery of patterns at various geographical scales. Large grid cells allow the recognition of patterns at higher taxonomic ranks, while small grid sizes will enable the examination of lower taxonomic groups that have recently diverged (Crother and Murray 2011). For example, in this study, we recovered from two to six areas of endemism nested within the Grijalva-Usumacinta Basin (Fig. 1, Suppl. material 3) as was previously suggested by Elías et al. (2020), who recovered six areas of endemism. In summary, there is a lack of agreement regarding which method is the best for recovering areas of endemism. There are advantages and disadvantages to each of the cladistic methods used in this study (Fattorini 2017). Therefore, a comparison of both methods is mandatory for recognising areas of endemism, as in some circumstances, they can be complementary and provide additional information (Canales et al. 2016).

Regionalisation of Middle America

Biogeographical regionalisations constitute a reference system for several other areas in biology (Morrone 2018); therefore, multiple regionalisation schemes have been proposed for the freshwater ichthyofauna of Middle America, based on different methodologies, including chorological (Miller 1966; Miller et al. 2005; Lévêque et al. 2007; Abell et al. 2008; Říčan et al. 2013), phenetic (Smith and Bermingham 2005; Matamoros et al. 2012; Elías et al. 2020; Rico et al. 2022) and phylogenetic approaches. The method used in the present study is the first to include all families of freshwater fishes and the complete area using a phylogenetic approach.

In the last years, we have witnessed a resurgence of cluster analysis that group areas based on similarities and differences without recognition of homologies (Ebach and Parenti 2015; Escalante 2017), leading to postulate that, in large-scale biogeographic studies, the use of quantitative or phenetic methods represents the best choice compared to PAE due to differences between an area of endemism and a spatially inclusive regionalisation (Kreft and Jetz 2010). Nevertheless, in this work, the phenetic approach failed to properly recover ichthyological provinces, despite the analysis identifying from six to eight regions not supported by ANOSIM results (Fig. 3A, B) indicating that these areas were not significantly different or were not continuous in geographic space.

A natural region must include an historical relationship of the biota (Ebach and Parenti 2015); thus, a biogeographical province recognised by the presence of a distinct set of endemic taxa allows us to recognise how life on Earth has been moulded by historic and contemporary biotic and abiotic forces (Escalante 2017; Morrone 2018).

The ichthyological provinces recovered in this study using PAE (Fig. 1A) are similar to six of the seven provinces recently described by phenetic methods for North America by Rico et al. (2022) (Table 3). The provinces “Bajo Colorado” and “Bajo Santa Cruz-San Pedro” recovered in this study are nested within the Colorado province (Rico et al. 2022), sharing Ptychocheilus lucius, Xyrauchen texanus and Gila elegans as supporting and indicator species; the “Baja California” provinces in both schemes are similar, sharing Fundulus lima and Fundulus parvipinnis. The “Costa del Pacífico Mexicano” province is similar to the “Sonora-Lerma” province (Rico et al. 2022), sharing Mayaheros beani. However, our region does not extend to cover the North-Central Mexico Highlands as in Rico et al. (2022). Finally, the “Centro-Norte del Atlántico mexicano” province is similar to the “Grande-Pánuco” province (Rico et al. 2022), sharing Pylodictis olivaris. Again, our province is restricted to the lowlands of the Atlantic slopes due to the Sierra Madre Oriental comprising an effective barrier (Fig. 1A, Table 1). Thus, in this study, we recovered the “Central Mexican Highlands province” supported by six endemic species (Astyanax argentatus, Cyprinodon inmemoriam, Cyprinodon latifasciatus, Megupsilon aporus, Characodon garmani and Stypodon signifer), a province that was not previously recovered by Rico et al. (2022).

In previous studies, both the Pacific and Atlantic slopes were recognised as independent provinces: “Chiapas-Nicaragua” province in the Pacific and “Grijalva-Usumacinta” in the Atlantic (Miller 1966; Bussing 1976; Říčan et al. 2013, 2016; Matamoros et al. 2015; Rico et al. 2022). In the present work, we recovered a single province called “Ístmica-Peninsular” that was supported by 13 species (Table 1, Suppl. material 3). Finally, from the south of the Motagua Basin to the southern border of Panamá, a single province called “Centroamericana” was recovered in this study, one that does not match with any of the previous works that have split this area into nine different provinces (Matamoros et al. 2015).

At a finer scale, it is difficult to compare the sub-provinces found in this study to those previously postulated due to the effects of different methods, area sizes and numbers of fish families. In this work, we postulated seven sub-provinces for lower America Media (Costa Rica and Panamá) that match with the proposal of Smith and Bermingham (2005). For the Nuclear Middle America / Chortís Block region, however, Matamoros et al. (2012, 2015) suggested four to ten sub-provinces, but in the present work, we recovered only two (Choluteca and Lagos de Nicaragua). The Pacific slope from the Río Verde and Tehuantepec to the Choluteca River in Honduras has been recovered as a single sub-province in several previous studies (Miller 1966; Bussing 1976; Miller et al. 2005; Abell et al. 2008; Říčan et al. 2013; Matamoros et al. 2015; Říčan et al. 2016), but this was not indicated in the present work.

Table 3.

Download as

CSV

XLSX

Comparison of the regionalisation schemes of Middle America Icthyofauna obtained by Rico et al. (2022) using an UPGMA clustering from a dissimilarity index and the results obtained in this study through the use of a parsimony analysis of endemicity. Name of regions, indicator species (IS), endemic species supporting the area and shared species between two approaches are displayed.

Rico et al. (2022)	Indicator species	PAE this study	Endemic species	Remarks	Shared species
Baja California	Fundulus lima , Fundulus parvipinnis	Baja California	Fundulus lima , Fundulus parvipinnis	Provinces are identical in both regionalisation schemes	Fundulus lima , Fundulus parvipinnis
Colorado	Pytocheilus lucius , Xyrauchen texanus, Catostomus latipinnis, Gila elegans, Gila robusta, Pantosteus discobulus, Gila cypha	Lower Santa Cruz- San Pedro and Lower Colorado	Agosia chrysogaster , Catostomus insignis, Pantosteus clarkii, Gila elegans, Pytocheilus lucius, Xyrauchen texanus	These provinces are part of the Colorado region that extends to Mexico and were recovered as independent units in our study	Pytocheilus lucius , Xyrauchen texanus,Gila elegans
Grande Panuco	Cyprinella lutrensis , Pylodictis olivaris, Gambusia affinis, Ictalurus punctatus	Mexican North Atlantic	Fundulus grandis , Pylodictis olivaris, Poecilia formosa, Poecilia latipinna	In our scheme, this province is restricted to the lowlands of Atlantic Coast due to Sierra Madre Oriental comprising an effective barrier in comparison to the Rico et al. (2022) scheme where this region extends beyond Sierra Madre Oriental	Pylodictis olivaris
NA	NA	Central Mexican Highlands	Astyanax argentatus , Cyprinodon inmemoriam, Cyprinodon latifasciatus, Megupsilon aporus, Characodon garmani, Stypodon signifer	This province is recovered in our scheme, but not in Rico et al. (2022), comprises a series of endorreic basins nestled within both Sierras Madre Oriental and Sierra Madre Occidental	NA
Sonora Lerma	Campostoma ornatum , Mayaheros beani	Mexican Pacific Coast	Mayaheros beani , Ictalurus dugesii, Poecilia butleri, Poeciliopsis latidens, Poeciliopsis presidionis, Poeciliopsis prolífica, Poeciliopsis viriosa	Very similar, nevertheless, in Rico et al. (2022) scheme, this province extends inland and comprises our Central Mexican Highlands province, besides, in our scheme the south limit extends beyond Transmexican Vocanic Belt and includes a section of the Rio Balsas-Nacaome of Rico et al. (2022)	Mayaheros beani
Balsas Nacaome	Poecilia nelsoni , Poeciliopsis turrubarensis, Amphilophus trimaculatum	NA	NA	Partially included in our Pacific Coast province and in our Grijalva Usumacinta province	NA
Grijalva Usumacinta	Pseudoxiphophorus bimaculatus , Trichromis salvini, Poecilia mexicana, Rocio octofasciata, Criboheros robertsoni, Xiphophorus maculatus, Rhamdia guatemalensis, Thorichthys meeki	Isthmic-Peninsular province	Atherinella alvarezi , Atherinella schultzi, Hyphessobrycon compressus, Maskaheros regani, Paraneetroplus bulleri, Thorichthys callolepis, Thorichthys helleri, Trichromis salvini, Dorosoma anale, Gambusia yucatana, Priapella compressa, Priapella intermedia	Our scheme comprises the entire Grijalva-Usumacinta province of Rico et al. (2022), but also the Pacific slope of their Balsas-Nacaome province from Oaxaca to Guatemala due to the fact that we recovered both Oceanic slopes of Mexico in the same province	Trichromis salvini
NA	NA	Central American province	Brycon argenteus , Brycon striatulus, Astyanax panamensis, Compsura gorgonae, Gephyrocharax atracaudatus, Pseudocheirodon arnoldi, Roeboides occidentalis, Andinoacara coeruleopunctatus, Panamius panamensis, Ctenolucius beani, Cyphocharax magdalenae, Hoplias malabaricus, Hoplias microlepis, Pimelodella chagresi, Brachyhypopomus occidentalis, Lebiasina panamensis, Ancistrus chagresi, Ancistrus spinosus, Chaetostoma fischeri, Leptoancistrus canensis, Rineloricaria uracantha,achyrhaphis episcopi, Hiatirhaphis cascajalensis, Neoheterandria tridentiger, Poecilia gillii, Poeciliopsis elongata and Trichomycterus striatus	This province extends from the Motagua-Polochic fault in Honduras to Panama, but also comprises the south limit of the Balsas-Nacaome province of Rico et al. (2022)	NA

Conclusions

The lack of agreement between regionalisation schemes is a consequence of the different scales used, as well as other methodologies and the number of families and species. It is necessary to point out that, despite similarities being focused on species turnovers, cluster analysis could be still considered to deal with natural kinds (Rieppel 2006); so, once the areas has been defined, it is possible to recognise their endemic taxa (Morrone 2018); the above also applies to ecoregions that can be compared with historical regionalisations due to both being moulded by the same processes (Morrone 2025).

There is no universally accepted method for the delimitation of biogeographical regions and, thus, the selection of a proper approach needs to be evaluated by taking into account the nature of the study (Kreft and Jetz 2010). It is necessary to use and compare different methodologies, as well as various geographic scales (Wiens 1989; Mayer and Cameron 2003; Rahbek 2005) in order to identify patterns and choose the best method for a single dataset due to, in a final instance, the general purpose classification being useful for a wide variety of goals (Jensen 2009).

Acknowledgements

We kindly appreciate the comments of Juan J. Morrone, Augusto Ferrari and two anonymous reviewers for their useful comments that allowed improving the manuscript. This work was supported by SIP-IPN multidisciplinary project number 2355.

Author contributions

OM conceived the research and acquired the data. ABP and FPM curated the geographical distributions of each species. ABP developed the analysis and conducted the statistical analyses. OM and FPM supervised the research. ABP wrote the initial manuscript draft, all authors contributing to the final version of the manuscript. This work is part of ABP Master’s degree thesis.

Data accessibility statement

All of the data that support the findings of this study are available in the main text or Supplementary Information.

References

Aagesen L, Szumik C, Goloboff P (2013) Consensus in the search for areas of endemism. Journal of Biogeography 40: 1–14. https://doi.org/10.1111/jbi.12172

Abell R, Thieme ML, Revenga C, Bryer M, Kottelat M, Bogutskaya Coad B, Mandrak N, Contreras-Balderas S, Bussing W, Stiassny MLJ, Skelton P, Allen GR, Unmack P, Naseka A, Ng R, Sindorf N, Robertson J, Armijo E, Higgins JV, Heibel TJ, Wikramanayake E, Olson D, López HL, Reis RE, Lundberg JG, Pérez MHS, Petry P (2008) Freshwater ecoregions of the world: a new map of biogeographic units for freshwater biodiversity conservation. BioScience 58: 403–414. https://doi.org/10.1641/B580507

Albert JS, Reis R (2011) Historical biogeography of Neotropical freshwater fishes. University of California Press, California, 408 pp. https://doi.org/10.1525/9780520948501

Angulo A, Bussing W, Garita-Alvarado C, López M (2013) Annotated checklist of the freshwater fishes of continental and insular Costa Rica: additions and nomenclatural revisions. Check List 9(5): 987–1019. https://doi.org/10.15560/9.5.987

Baselga A, Orme CDL (2012) betapart: an R package for the study of beta diversity. Methods in Ecology and Evolution 3(5): 808–812. https://doi.org/10.1111/j.2041-210X.2012.00224.x

Brown JH, Gibson AC (1983) Biogeography. C. V. Mosby Company, St. Louis, Missouri, 643 pp. https://www.resolutionmineeis.us/documents/brown-gibson-1983

Bussing WA (1976) Geographic distribution of the San Juan ichthyofauna of Central America with remarks on its origin and ecology. Investigations of the ichthyofauna of Nicaraguan lakes. Nebraska, USA, 103 pp. https://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1010&context=ichthynicar

Canales GM, Mayer-Goyenechea IG, Badillo LF, Cerón JMC (2016) Evaluación de AE y PAE para identificar trazos generalizados, a partir de las serpientes del estado de Hidalgo, México. Revista de Biología Tropical 64: 1611–1624. https://doi.org/10.15517/rbt.v64i4.21219

Carine MA, Humphries CJ, Guma IR, Reyes-Betancort JA, Santos Guerra A (2009) Areas and algorithms: evaluating numerical approaches for the delimitation of areas of endemism in the Canary Islands archipelago. Journal of Biogeography 36: 593–611. https://doi.org/10.1111/j.1365-2699.2008.02016.x

Casagranda MD, Roig-Juñet S, Szumik C (2009) Endemismo a diferentes escalas espaciales: un ejemplo con Carabidae (Coleóptera: Insecta) de América del sur austral. Revista Chilena de Historia Natural 82: 17–42. https://doi.org/10.4067/S0716-078X2009000100002

Casagranda MD, Taher L, Szumik CA (2012) Endemicity analysis, parsimony and biotic elements: a formal comparison using hypothetical distributions. Cladistics 28: 645–654. https://doi.org/10.1111/j.1096-0031.2012.00410.x

Clarke KR, Warwick RM (1994) Similarity-based testing for community pattern: the two-way layout with no replication. Marine Biology 118: 167–176. https://doi.org/10.1007/BF00699231

Contreras-MacBeath T, Ardón DA, Quintana Y, Angulo A, Lyons T, Lardizabal C, McMahan C, Elias DJ, Matamoros WA, Barraza JE, González R, Fuentes-Montejo CE, Armbruster JW, Carrasco JC, Brito Rodriguez M (2022) Freshwater fishes of Central America: Distribution, assessment, and major threats. Diversity 14(10): 793. https://doi.org/10.3390/d14100793

Cox CB, Moore PD (1985) Biogeography: an ecological and evolutionary approach. Blackwell Scientific Publications, London, 244 pp. https://mariomairal.com/wp-content/uploads/2020/10/Biogeography-An-Ecological-and-Evolutionary-Approach-Cox-Moore-Ladle-9-Edition.pdf

Crisp MD, Laffan S, Linder HP, Monro A (2001) Endemism in the Australian flora. Journal of Biogeography 28: 183–198. https://doi.org/10.1046/j.1365-2699.2001.00524.x

Crother BI, Murray CM (2011) Ontology of areas of endemism. Journal of Biogeography 38: 1009–1015. https://doi.org/10.1111/j.1365-2699.2011.02483.x

Darlington PJ (1957) Zoogeography: the geographical distribution of animals. John Wiley y Sons, New York, 675 pp. https://archive.org/details/zoogeographygeog0000darl/page/n3/mode/2up

De Cáceres M, Legendre P (2009) Associations between species and groups of sites: indices and statistical inference. Ecology 90: 3566–3574. https://doi.org/10.1890/08-1823.1

Dufrêne M, Legendre P (1997) Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecological Monographs 67: 345–366. https://doi.org/10.1890/0012-9615(1997)067[0345:SAAIST]2.0.CO;2

Ebach MC, Parenti LR (2015) The dichotomy of the modern bioregionalization revival. Journal of Biogeography 42: 1801–1808. https://doi.org/10.1111/jbi.12558

Elías DJ, McMahan CD, Matamoros WA, Gómez-González AE, Piller KR, Chakrabarty P (2020) Scale (s) matter: Deconstructing an area of endemism for Middle American freshwater fishes. Journal of Biogeography 47: 2483–2501. https://doi.org/10.1111/jbi.13941

Escalante T (2015) Parsimony analysis of endemicity and analysis of endemicity: a fair comparison. Systematics and Biodiversity 13: 413–418. https://doi.org/10.1080/14772000.2015.1046966

Escalante T (2017) A natural regionalization of the world based on primary biogeographic homology of terrestrial mammals. Biological Journal of the Linnean Society 120: 349–362. https://doi.org/10.1111/bij.12898

Escalante T, Szumik C, Morrone JJ (2009) Areas of endemism of Mexican mammals: reanalysis applying the optimality criterion. Biological Journal of the Linnean Society 98: 468–478. https://doi.org/10.1111/j.1095-8312.2009.01293.x

Espinosa-Pérez H, Ramírez M (2015) Exotic and invasive fishes in Mexico. Check List 11(3): 1–13. https://doi.org/10.15560/11.3.1627

Estrada-Márquez AS, Morrone JJ, Villaseñor JL (2021) Areas of endemism of two biogeographic provinces in central Mexico based on their endemic Asteraceae: a conservation proposal. Revista Mexicana de Biodiversidad 92: e923470. https://doi.org/10.22201/ib.20078706e.2021.92.3470

Fattorini S (2017) Endemism in historical biogeography and conservation biology: concepts and implications. Biogeographia–The Journal of Integrative Biogeography 32: 47–75. https://doi.org/10.21426/B632136433

Ferrari A, Paladini A, Schwertner CF, Grazia J (2010) Endemism analysis of neotropical pentatomidae (Hemiptera, Heteroptera). Iheringia, Série Zoologia 100: 449–462. https://doi.org/10.1590/S0073-47212010000400018

Food and Agriculture Organization of the United Nations (2022) FAO Land and Water Division. Major hydrological basins of the world. https://data.apps.fao.org/catalog//iso/7707086d-af3c-41cc-8aa5-323d8609b2d1 [Visited on 05/06/2022]

Fricke R, Eschmeyer WN, van der Laan R [Eds] (2022) Eschmeyer’s Catalog Of Fishes: Genera, Species, References. http://researcharchive.calacademy.org/research/ichthyology/catalog/fishcatmain.asp [Electronic version accessed 08 08 2022]

Goloboff PA, Farris JS, Nixon KC (2008) TNT, a free program for phylogenetic analysis. Cladistics 24: 774–786. https://doi.org/10.1111/j.1096-0031.2008.00217.x

Goloboff PA, Catalano SA (2016) TNT version 1.5, including a full implementation of phylogenetic morphometrics. Cladistics 32: 221–230. https://doi.org/10.1111/cla.12160

Harold AS, Mooi RD (1994) Areas of endemism: Definition and recognition criteria. Systematic Biology 43: 261–266. https://doi.org/10.1093/sysbio/43.2.261

Hausdorf B, Hennig C (2003) Biotic element analysis in biogeography. Systematic Biology 52: 717–723. https://doi.org/10.1080/10635150390235584

Hausdorf B (2002) Units in biogeography. Systematic Biology 51: 648–652. https://doi.org/10.1080/10635150290102320

Hijmans RJ, Phillips S, Leathwick J, Elith J, Hijmans MRJ (2017) Package ‘dismo’ 1.3-14. https://cran.r-project.org/web/packages/dismo/index.html

Hijmans R (2025) raster: Geographic Data Analysis and Modeling. R package version 3.6-31. https://github.com/rspatial/raster

Hoffmeister CH, Ferrari A (2016) Areas of endemism of arthropods in the Atlantic Forest (Brazil): an approach based on a metaconsensus criterion using endemicity analysis. Biological Journal of the Linnean Society 119: 126–144. https://doi.org/10.1111/bij.12802

Jensen RJ (2009) Phenetics: revolution, reform or natural consequence? Taxon 58: 50–60. https://doi.org/10.1002/tax.581008

Jin J, Yang J (2020) BDcleaner: A workflow for cleaning taxonomic and geographic errors in occurrence data archived in biodiversity databases. Global Ecology and Conservation 21: e00852. https://doi.org/10.1016/j.gecco.2019.e00852

Kelley LA, Gardner SP, Sutcliffe MJ (1996) An automated approach for clustering an ensemble of NMR-derived protein structures into conformationally related subfamilies. Protein Engineering, Design and Selection 9: 1063–1065. https://doi.org/10.1093/protein/9.11.1063

Kreft H, Jetz W (2010) A framework for delineating biogeographical regions based on species distributions. Journal of Biogeography 37: 2029–2053. https://doi.org/10.1111/j.1365-2699.2010.02375.x

Kullander SO (2003) Family Cichlidae. In: Reis RE, Kullander SO, Ferraris Jr C (Eds) Check list of the freshwater fishes of South and Central America. Porto Alegre, 605–654.

Lee DS, Gilbert CR, Hocutt CH, Jenkins RE, McAllister DE, Stauffer Jr JR (1980) Atlas of North American freshwater fishes (p. 854). North Carolina State Museum of natural history. https://doi.org/10.5962/bhl.title.141711

Lehner B, Grill G (2013) Global river hydrography and network routing: baseline data and new approaches to study the world’s large river systems. Hydrological Processes 27: 2171–2186. https://doi.org/10.1002/hyp.9740

Lévêque C (1997) Biodiversity dynamics and conservation: the freshwater fish of tropical Africa. Cambridge University Press, 438 pp. https://horizon.documentation.ird.fr/exl-doc/pleins_textes/divers20-10/010012957.pdf

Lévêque C, Oberdorff T, Paugy D, Stiassny MLJ, Tedesco PA (2007) Global diversity of fish (Pisces) in freshwater. In: Balian EV, Lévêque C, Segers H, Martens K (Eds) Freshwater animal diversity assessment. Springer, Dordrecht, 545–567. https://doi.org/10.1007/978-1-4020-8259-7_53

Linder HP (2001) On areas of endemism, with an example from the African Restionaceae. Systematic Biology 50: 892–912. https://doi.org/10.1080/106351501753462867

Lucinda PHF (2003) Family Poeciliidae. In: Reis RE, Kullander SO, Ferraris Jr C (Eds) Check list of the freshwater fishes of South and Central America. Porto Alegre, 555–581.

Matamoros WA, Schaefer JF, Kreiser BR (2009) Annotated checklist of the freshwater fishes of continental and insular Honduras. Zootaxa 2307(1): 1–38. https://doi.org/10.11646/zootaxa.2307.1.1

Matamoros WA, Kreiser BR, Schaefer JF (2012) A delineation of Nuclear Middle America biogeographical provinces based on river basin faunistic similarities. Reviews in Fish Biology and Fisheries 22: 351–365. https://doi.org/10.1007/s11160-011-9232-8

Matamoros WA, McMahan CD, Chakrabarty P, Albert JS, Schaefer JF (2015) Derivation of the freshwater fish fauna of Central America revisited: Myers’s hypothesis in the twenty first century. Cladistics 31: 177–188. https://doi.org/10.1111/cla.12081

Mayer AL, Cameron GN (2003) Consideration of grain and extent in landscape studies of terrestrial vertebrate ecology. Landscape Urban Plan 65: 201–217. https://doi.org/10.1016/S0169-2046(03)00057-4

McMahan CD, Matamoros WA, Álvarez-Calderón FS, Henríquez WY, Recinos HM, Chakrabarty P, Barraza E, Herrera N (2013) Checklist of the inland fishes of El Salvador. Zootaxa 3608(6): 440–456. https://doi.org/10.11646/zootaxa.3608.6.2

Mejía O (2022) Fishes of Mexico and Central America. Occurrence download https://doi.org/10.15468/dd.faguz5 [accessed via GBIF.org on 03/03/2022]

Miller RR (1966) Geographical distribution of Central American freshwater fishes. Copeia 4: 773–802. https://doi.org/10.2307/1441406

Miller RR, Minckley WL, Norris SM (2005) Freshwater fishes of Mexico. University of Chicago Press, Chicago, USA, 490 pp.

Moline PM, Linder HP (2006) Input data, analytical methods and biogeography of Elegia (Restionaceae). Journal of Biogeography 33: 47–62. https://doi.org/10.1111/j.1365-2699.2005.01369.x

Morrone JJ (1994) On the identification of areas of endemism. Systematic Biology 43: 438–441. https://doi.org/10.2307/2413679

Morrone JJ (2008) Evolutionary biogeography an integrative approach with case studies. Columbia University Press, New York, 301 pp. https://doi.org/10.1600/036364409789271227

Morrone JJ (2014) Parsimony analysis of endemicity (PAE) revisited. Journal of Biogeography 4: 842–854. https://doi.org/10.1111/jbi.12251

Morrone JJ (2017) Neotropical biogeography regionalization and evolution. CRC Press, Boca Raton, Florida, 312 pp. https://doi.org/10.1201/b21824

Morrone JJ (2018) The spectre of biogeographical regionalization. Journal of Biogeography 45: 282–288. http://dx.doi.org/10.1111/jbi.13135

Morrone JJ (2025) A dual biotic nature: What can history tells us about the biogeographical regionalization of Argentina? Revista de la Sociedad Entomológica Argentina 84: e0201. https://doi.org/10.25085/rsea.840201

Nelson GJ, Platnick NI (1981) Systematics and biogeography: Cladistics and vicariance. Columbia University Press, New York, 582 pp. https://doi.org/10.1093/sysbio/syq050

Nixon KC (2002) WinClada, version 5.30: 08. Published by the author, Ithaca, New York 734: 745.

Ortega J, Arita HT (1998) Neotropical-Nearctic limits in Middle America as determined by distributions of bats. Journal of Mammalogy 79: 772–783. https://doi.org/10.2307/1383088

Parenti LR, Ebach MC (2009) Comparative biogeography: Discovering and classifying biogeographical patterns of a dynamic earth. University of California Press, Berkeley, California, 312 pp. https://doi.org/10.22201/ib.20078706e.2010.002.246

Pinton A, Agnèse JF, Paugy D, Otero O (2013) A large-scale phylogeny of Synodontis (Mochokidae, Siluriformes) reveals the influence of geological events on continental diversity during the Cenozoic. Molecular Phylogenetics and Evolution 66: 1027–1040. https://doi.org/10.1016/j.ympev.2012.12.009

QGIS.org (2021) QGIS Geographic Information System. Open Source Geospatial Foundation Project. http://qgis.org

R Core Team (2023) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.

Rahbek C (2005) The role of spatial scale and the perception of large scale species richness patterns. Ecology Letters 8: 224–239. https://doi.org/10.1111/j.1461-0248.2004.00701.x

Říčan O, Piálek L, Zardoya R, Doadrio I, Zrzavý J (2013) Biogeography of the Mesoamerican Cichlidae (Teleostei: Heroini): colonization through the GAARlandia land bridge and early diversification. Journal of Biogeography 40: 579–593. https://doi.org/10.1111/jbi.12023

Říčan O, Piálek L, Dragová K, Novák J (2016) Diversity and evolution of the Middle American cichlid fishes (Teleostei: Cichlidae) with revised classification. Vertebrate Zoology 66: 1–102. https://doi.org/10.3897/vz.66.e31534

Rico CN, Hoagstrom CW, Elías DJ, McMahan CD, Matamoros WA (2022) Biotic regionalization of freshwater fishes in Northern Middle America highlights high beta diversity created by prominent biogeographic barriers. Frontiers of Biogeography 14: e58095. https://doi.org/10.21425/F5FBG58095

Rieppel O (2006) The merits of similarity reconsidered. Systematics and Biodiversity 4: 137–147. https://doi.org/10.1017/s1477200005001830

Roberts TR (1975) Geographical distribution of African freshwater fishes. Zoological Journal of the Linnean Society 57: 249–319. https://doi.org/10.1111/j.1096-3642.1975.tb01893.x

Rosen BR, Smith AB (1988) Tectonics from fossils? Analysis of reef-coral and sea-urchin distributions from late Cretaceous to Recent, using a new method. Geological Society Special Publication 37: 275–306. https://doi.org/10.1144/GSL.SP.1988.037.01.19

Ruiz-Campos G, Castro-Aguirre JL, Contreras-Balderas S, de Lourdes Lozano-Vilano M, González-Acosta AF, Sánchez-Gonzáles S (2002) An annotated distributional checklist of the freshwater fish from Baja California Sur, Mexico. Reviews in Fish Biology and Fisheries 12: 143–155. https://doi.org/10.1023/A:1025067202218

Santiago-Alvarado M, Luna-Vega I, Rivas G, Espinosa D (2022) Effect of cell size and thresholds in NDM/NVDM methods on recognizing areas of endemism. Zootaxa 5134: 1–33. https://doi.org/10.11646/zootaxa.5134.1.1

Smith SA, Bermingham E (2005) The biogeography of lower Mesoamerican freshwater fishes. Journal of Biogeography 32: 1835–1854. https://doi.org/10.1111/j.1365-2699.2005.01317.x

Sokal RR, Rohlf FJ (1962) The comparison of dendrograms by objective methods. Taxon, 33–40. https://doi.org/10.2307/1217208

Szumik CA, Cuezzo F, Goloboff PA, Chalup AE (2002) An optimality criterion to determine Areas of Endemism. Systematic Biology 51: 806–816. https://doi.org/10.1080/10635150290102483

Szumik CA, Goloboff PA (2004) Areas of endemism: an improved optimality criterion. Systematic Biology 53: 968–977. https://doi.org/10.1080/10635150490888859

Torres-Miranda A, Luna-Vega I, Oyama K (2013) New approaches to the biogeography and areas of endemism of red oaks (Quercus L., Section Lobatae). Systematic Biology 62: 555–573. https://doi.org/10.1093/sysbio/syt021

Vavrek MJ (2011) Fossil: palaeoecological and palaeogeographical analysis tools. Palaeontologia Electronica 14: 16.

Valdez-Moreno ME, Pool-Canul J, Contreras-Balderas S (2005) A checklist of the freshwater ichthyofauna from El Petén and Alta Verapaz, Guatemala, with notes for its conservation and management. Zootaxa 1072(1): 43–60. https://doi.org/10.11646/zootaxa.1072.1.4

White D, Gramacy RB (2022) maptree: Mapping, Pruning, and Graphing Tree Models. R package version 1.4-8.

Wiens JA (1989) Spatial scaling in ecology. Functional Ecology 3: 385–397. https://doi.org/10.2307/2389612

Winker K (2011) Middle America, not Mesoamerica, is the accurate term for biogeography. The Condor 113: 5–6. https://doi.org/10.1525/cond.2011.100093

Supplementary materials

Supplementary material 1

List of the 685 species of freshwater fish of Middle America used in this study (.xlsx)

https://doi.org/10.21425/fob.18.164902.suppl1

Download file (25.46 kb)

Supplementary material 2

PAM matrix of the species of freshwater fishes of Middle America using FAO basins as geographic units (.xlsx)

https://doi.org/10.21425/fob.18.164902.suppl2

Download file (600.03 kb)

Supplementary material 3

Endemism areas recovered with PAE and EA smethodologies using half degree grid cells (.pdf)

https://doi.org/10.21425/fob.18.164902.suppl3

Download file (2.41 MB)

Supplementary material 4

List of species that support the endemism areas recovered with PAE for half degree grid cells (.xlsx)

https://doi.org/10.21425/fob.18.164902.suppl4

Download file (41.90 kb)

Supplementary material 5

List of species that support the endemism areas recovered with EA for one degree grid cells at 30% consensus (.xlsx)

https://doi.org/10.21425/fob.18.164902.suppl5

Download file (51.33 kb)

Supplementary material 6

ANOSIM results test for the pairwise comparisons among areas recovered from half degree grid cells with phenetic approach (.xlsx)

https://doi.org/10.21425/fob.18.164902.suppl6

Download file (180.35 kb)

Supplementary material 7

Endemism areas recovered with PAE and cluster methodologies for Pfastetter hydrobasins levels 4 and 5 (.pdf)

https://doi.org/10.21425/fob.18.164902.suppl7

Download file (5.07 MB)

﻿Abstract

﻿Highlights

Keywords

﻿Introduction

﻿Materials and methods

﻿Distribution data

﻿Analysis of endemicity

﻿Comparison amongst methods

﻿Icthyological provinces of freshwater fish of Middle America

﻿Results

﻿Areas of endemism

﻿Comparison amongst methods

﻿Regionalisation of Middle America

﻿Lower Colorado province

﻿Lower Santa Cruz-San Pedro province

﻿Baja California province

﻿Central Mexican Highlands province

﻿Mexican Pacific Coast province

﻿Mexican North Atlantic province

﻿Isthmic-Peninsular province

﻿Central American province

﻿Discussion

﻿Areas of endemism

﻿Regionalisation of Middle America

﻿Conclusions

﻿Acknowledgements

﻿Author contributions

﻿Data accessibility statement

﻿References

Supplementary materials

Abstract

Highlights

Introduction

Materials and methods

Distribution data

Analysis of endemicity

Comparison amongst methods

Icthyological provinces of freshwater fish of Middle America

Results

Areas of endemism

Comparison amongst methods

Regionalisation of Middle America

Lower Colorado province

Lower Santa Cruz-San Pedro province

Baja California province

Central Mexican Highlands province

Mexican Pacific Coast province

Mexican North Atlantic province

Isthmic-Peninsular province

Central American province

Discussion

Areas of endemism

Regionalisation of Middle America

Conclusions

Acknowledgements

Author contributions

Data accessibility statement

References