The genus Habromys: an untold biogeographic history

Deborah V. Espinosa-Martínez1,3 , César A. Ríos-Muñoz2* , and Joaquín Arroyo-Cabrales3 .

1Posgrado en Ciencias del Mar y Limnología, Universidad Nacional Autónoma de México, Ciudad de México, México. E-mail: dvem@ciencias.unam.mx (DVE-M)

2Facultad de Ciencias, Universidad Nacional Autónoma de México, Ciudad Universitaria, CP 04510, Ciudad de México, México.

3Laboratorio de Arqueozoología, Subdirección de Laboratorios y Apoyo Académico, Instituto Nacional de Antropología e Historia, Moneda 16 Centro, CP 06060, Ciudad de México, México. E-mail: joaquin_arroyo@inah.gob.mx (JA-C)

*Corresponding author: cesar.rios@unam.mx

The genus Habromys includes seven species of rodents restricted to Mesoamerican cloud forests across the mountain systems of central and eastern Mexico and northern Central America. Previous studies have suggested that diversification within the genus largely shaped by vicariant scenarios in which range expansion was followed by isolation driven by habitat fragmentation. However, these hypotheses have not been formally evaluated within a time-calibrated framework using explicit model-based approaches. Here, we reconstructed the biogeographic history of Habromys using a time-calibrated phylogeny based on mitochondrial genes, previously used in published studies, and evaluated alternative parametric biogeographic models implemented in BioGeoBEARS. The DEC+j was the best-supported model explaining geographic range evolution, indicating that founder-event speciation and within-area cladogenesis (narrow and subset sympatry) were the most frequent processes shaping diversification, whereas vicariance and anagenetic dispersal/extinction were inferred as comparatively uncommon. Divergence-time estimates suggest that crown diversification in Habromys began during the Pliocene, whereas most interspecific divergences occurred during the middle Pleistocene. These patterns are consistent with a biogeographic history driven by climatic oscillations that periodically increased connectivity among cloud forest habitats and later promoted isolation in montane regions. Overall, our results support a scenario in which repeated colonization and isolation events associated with Pliocene–Pleistocene climatic fluctuations played a greater role than vicariant processes linked to mountain building.

Keywords: Ancestral reconstruction area, cloud forest, Cricetidae, distribution, Mesoamerica, parametric biogeography.

El género Habromys incluye siete especies de roedores restringidas a los bosques mesófilos de montaña de Mesoamérica, distribuidos en los sistemas montañosos del centro y este de México y el norte de Centroamérica. Estudios previos han sugerido que la diversificación dentro del género estuvo determinada principalmente por escenarios vicariantes, en los cuales una expansión del área de distribución fue seguida por aislamiento impulsado por la fragmentación del hábitat. Sin embargo, estas hipótesis no han sido evaluadas formalmente dentro de un marco temporal calibrado ni mediante enfoques explícitos basados en modelos. En este estudio reconstruimos la historia biogeográfica de Habromys utilizando una filogenia calibrada temporalmente basada en genes mitocondriales previamente empleados en estudios publicados, y evaluamos modelos biogeográficos paramétricos alternativos implementados en BioGeoBEARS. El modelo DEC+j fue el mejor soportado para explicar la evolución del rango geográfico, indicando que la especiación por efecto fundador y la cladogénesis dentro de área (simpatría estrecha y simpatría por subconjunto) fueron los procesos más frecuentes que moldearon su diversificación, mientras que la vicarianza y la dispersión/extinción anagenética fueron inferidas como relativamente poco comunes. Las estimaciones de tiempo de divergencia sugieren que la diversificación en Habromys inició durante el Plioceno, mientras que la mayoría de las divergencias interespecíficas ocurrieron durante el Pleistoceno medio. Estos patrones son consistentes con una historia biogeográfica impulsada por oscilaciones climáticas que incrementaron periódicamente la conectividad entre los bosques mesófilos y posteriormente promovieron el aislamiento en regiones montañosas. En conjunto, nuestros resultados apoyan un escenario en el que eventos repetidos de colonización y aislamiento asociados a las fluctuaciones climáticas del Plioceno–Pleistoceno desempeñaron un papel más importante que procesos vicariantes vinculados al levantamiento de los sistemas montañosos.

Palabras clave: Biogeografía paramétrica, bosques mesófilos de montaña, Cricetidae, distribución, Mesoamérica, reconstrucción de áreas ancestrales.

The genus Habromys Hooper and Musser 1964 (Cricetidae: Neotominae) includes seven species of rodents (H. chinanteco, H. delicatulus, H. ixtlani, H. lepturus, H. lophurus, H. schmidlyi, and H. simulatus), all endemic and ecologically restricted to Mesoamerican cloud forest from central and eastern Mexico to northern Central America (León-Paniagua et al. 2007; Ceballos 2014; Ramírez-Pulido et al. 2014; Contreras-Medina et al. 2024). The species are considered unique due to their arboreal habits and high ecological specialization (Colunga Salas 2014; 2016; Marines-Macías et al. 2018). In some species, populations may occur at low densities (Colunga Salas 2016; Castañeda-Rico et al. 2023) and small home ranges (Marines-Macías et al. 2018), which has contributed to their recognition as threatened at the national level (H. simulatus: subject to special protection, SEMARNAT 2010) and the international level (H. delicatulus: endangered, H. ixtlani: critically endangered, and H. lophurus: near threatened, [Vázquez 2017; Álvarez-Castañeda et al. 2018; Emmons et al. 2019]).

The geographic distribution of Habromys species is restricted or highly fragmented. H. delicatulus and H. schmidlyi occur in the Trans-Mexican Volcanic Belt (Romo-Vázquez et al. 2005); H. simulatus has a disjunct distribution in the Sierra Madre Oriental and eastern Trans-Mexican Volcanic Belt (León-Paniagua et al. 2007; Castañeda-Rico et al. 2011); while H. chinanteco and H. ixtlani are found in the Sierra de Juárez and H. lepturus in Cerro Zempoaltepetl, all part of the Sierra Madre de Oaxaca (León-Paniagua et al. 2007; Contreras-Medina et al. 2024); and H. lophurus is distributed in the mountains east of the Isthmus of Tehuantepec in southeastern Mexico (Chiapas), through southern and central Guatemala (Sierra de las Minas), to El Salvador (Hall 1981; León-Paniagua et al. 2007; Contreras-Medina et al. 2024) (Figure 1).

In addition to formal descriptions of the species (Merriam 1898; Osgood 1904; Goodwin 1964; Robertson and Musser 1976; Carleton et al. 2002; Romo-Vázquez et al. 2005), published works have focused on identifying new distributional records (Briones-Salas et al. 2012), ecological aspects and genetic diversity for some species (Castañeda-Rico et al. 2011; 2023; Colunga Salas 2014; 2016; Marines-Macías et al. 2018), analyzing the effect of Pleistocene climate change (León-Paniagua and Guevara 2019), and understanding phylogenetic relationships within the genus (León-Paniagua et al. 2007; Rogers et al. 2007; Castañeda-Rico et al. 2011; 2023). Nevertheless, studies have emphasized phylogenetic relationships and distributional patterns rather than explicit hypothesis testing of biogeographic processes.

The first phylogenetic hypotheses for Habromys were presented by Rogers et al. (2007), based on the complete mitochondrial cytochrome-b gene (1143 bp), and by León-Paniagua et al. (2007), based on mitochondrial ND3, tRNA-Arg, ND4L, and ND4 sequences (1131 bp). Both studies supported the monophyly of Habromys using broad taxonomic sampling within Neotominae, and recovered geographically structured lineages largely associated with montane regions of Mexico and northern Central America, suggesting that isolation among cloud forest massifs has promoted diversification. Rogers et al. (2007), which included six of the seven recognized species, emphasized the role of the Isthmus of Tehuantepec as a major geographic barrier potentially driving divergence between lineages west and east of the isthmus, and suggested that this vicariant event may predate Pleistocene climatic fluctuations. They also proposed the Cajonos River as a potential vicariant barrier separating H. ixtlani and H. lepturus. León-Paniagua et al. (2007) included all described species of Habromys and additionally documented geographically structured lineages within H. simulatus and H. lophurus, suggesting a history of range expansion followed by diversification associated with fragmentation of montane forests during climatic oscillations, and suggesting that the Isthmus of Tehuantepec is not a major force in the differentiation of the genus. Subsequent work supported the differentiation of two evolutionary units within H. simulatus using microsatellite data (Castañeda-Rico et al. 2011), whereas the phylogeographic structure suggested for H. lophurus remains untested. However, although these studies provided a critical phylogenetic framework, their biogeographic interpretations were not evaluated under explicit model-based approaches, and the relative roles of biogeographic processes in shaping current distributions remain unclear.

Parametric biogeography provides a statistical framework to test alternative evolutionary scenarios by explicitly modeling biogeographic processes (anagenetic: dispersal, extinction, range switching; and cladogenetic: within-area cladogenesis, vicariance, founder-event) through time, involved in the reconstruction of biogeographic history. In addition to considering a temporal framework that allows identifying the congruence of biogeographic events, and relating them to independent geographic information (Sanmartin 2012). Based on a time-calibrated phylogenetic hypothesis and knowledge of species distributions, the geographic evolution of lineages is modeled using a stochastic process that determines the rate of change between areas, allowing the reconstruction of ancestral areas and inference of processes that shaped the distribution of a given taxon (Sanmartin 2012; Matzke 2013b). Despite the availability of a phylogenetic hypothesis and well-documented geographic distributions, Habromys has not been examined using parametric biogeographic methods that explicitly compare alternative evolutionary scenarios.

Given the strong association of Habromys with cloud forests, the high fragmentation of these habitats, and the presumably limited dispersal ability of these rodents, we hypothesize that vicariance played an important role in the diversification of the genus among montane regions during climatic fluctuations of the late Neogene and Pleistocene (León-Paniagua et al. 2007; Rogers et al. 2007; León-Paniagua and Guevara 2019). We evaluated this scenario against alternative processes such as founder-event speciation and within-area cladogenesis. Specifically, we predict that (1) ancestral areas will be reconstructed in eastern Mexican and Central American highlands, (2) diversification events will coincide with major geographic barriers such as the Isthmus of Tehuantepec and the separation of the Trans-Mexican Volcanic Belt, Sierra Madre Oriental, Sierra de Juarez and Cerro Zempoaltepetl, and (3) dispersal events will be limited and mostly restricted to adjacent mountain systems. To test these predictions, we conducted a model-based historical biogeographic analysis using a time-calibrated phylogeny and explicit parametric models. Here we reanalyzed the mitochondrial dataset of León-Paniagua et al. (2007), complemented with an additional sequence for H. simulatus from Castañeda-Rico et al. (2014), to evaluate alternative parametric models of historical biogeography.

Materials and methods

Phylogenetic analysis. We performed a phylogenetic analysis including 31 individuals representing the seven species described for the genus Habromys and 14 outgroup species (one individual per species), following the taxon sampling and outgroup selection by León-Paniagua et al. (2007). Partial mitochondrial sequences of ND3 and ND4 were obtained from GenBank (Supplementary Material 1), following León-Paniagua et al. (2007), with one additional H. simulatus sequence from Castañeda-Rico et al. (2014). The two gene fragments were concatenated for subsequent analyses. Sequences were aligned in MEGA11 (Tamura et al. 2021) with the Multiple Sequence Comparison algorithm based on Log-Expectation (MUSCLE) (Edgar 2004) under default settings.

We inferred a time-calibrated phylogeny using BEAST 2.7.7 (Bouckaert et al. 2019). We selected the substitution model by running Modeltest 3.6 on our alignment, which recovered HKY model (Hasegawa et al. 1985) with invariant sites, four Gamma categories as the best-fit model, consistent with León-Paniagua et al. (2007). To ensure comparability with previous analyses, we implemented this same model in BEAST. We implemented a relaxed log-normal molecular clock, and the Calibrated Yule speciation model (Heled and Drummond 2012). The speciation rate (birthRateY) was modeled using a Gamma distribution with parameters α = 3 and β = 0.09. Node ages were calibrated using four secondary fossil-based constraints taken from León-Paniagua et al. (2007). For each calibrated node, we implemented an MRCA (most recent common ancestor) priors with a lognormal distribution and enforced monophyly (Table 1), these priors constrain node ages while enforcing monophyly. For the relaxed molecular clock, Gamma distributions were used with parameters α = 3 and β = 0.00667 for the mean (ucldMean), and α = 4 and β = 0.05 for the standard deviation (ucldStdev).

Twenty million generations were run, sampling every 1000 generations. We assessed the stationarity of the Markov Chain Monte Carlo (MCMC) in Tracer v.1.7.2 (Rambaut et al. 2018), ensuring that the effective sample sizes (ESS) were > 200 for all priors, indicating adequate sampling of the posterior distribution. The Maximum Clade Credibility Tree (MCCT) and the most probable common ancestor were obtained using TreeAnnotator v.2.7.7, discarding 25% of burn-in. The maximum credibility tree was visualized in FigTree v.1.4.4 (http://tree.bio.ed.ac.uk/software/figtree/).

Ancestral area estimation. The ancestral areas for the genus Habromys were estimated using the BioGeoBEARS package (Matzke 2013a) in R v.4.3.3 (R Core Team 2024). To reduce the number of geographic states, we initially retained one representative per species of the genus was initially used. In the case of H. schmidlyi, two individuals were considered because they do not form a monophyletic group and one of the individuals is a sibling of H. delicatulus. This non-monophyly has been previously reported and may reflect incomplete lineage sorting or taxonomic complexity. On the other hand, for H. simulatus and H. lophurus, León-Paniagua et al. (2007) and Castañeda-Rico et al. (2011) report that there are two well-differentiated populations in two different areas in each species, so we decided to retain one individual from each of the populations in order to reconstruct biogeographic processes involved (Figure 1). Considering that BioGeoBEARS analyses have a computational limit associated with the number of states (number of combinations of relationships between the areas used, Matzke 2013a, b), we pruned most outgroups, leaving only Peromyscus boylii (distributed in the south and southwest of the United States of America and northern Mexico, up to the Central Mexican Plateau [Kalcounis-Rueppell and Spoon 2009]) and P. slevini (distributed on Santa Catalina Island in the Gulf of California [Álvarez-Castañeda and Cortés-Calva 2002]), as both are recovered as a sister lineage to Habromys (León-Paniagua et al. 2007).

To estimate the biogeographic history of the genus, we considered three sets of biogeographic models implemented in BioGeoBEARS: (1) Dispersal-Extinction-Cladogenesis (DEC, Ree et al. 2005; Ree and Smith 2008), (2) the likelihood version of the Dispersal-Vicariance Analysis (DIVALIKE, Matzke 2013a, b), which differs from the original version of DIVA (Ronquist 1997; Ronquist and Sanmartín 2011) in that it is not based on parsimony methods, but rather on a likelihood interpretation, and (3) the likelihood version of the BAYAREA model (BAYAREALIKE, Matzke 2013a, b) which, unlike the original version of BAYAREA (Landis et al. 2013), considers cladogenesis as part of the evolutionary process. Likewise, models that consider speciation because of the founder effect (parameter +j, Matzke 2013b, 2014) were also tested: DEC+j, DIVALIKE+j, and BAYAREALIKE+j. The best model was selected based on the likelihood values (LnL) and the Akaike Information Criterion (AIC) values obtained from model fitting (Table 2).

We performed biogeographic stochastic mapping (BSM) under the best-fitting DEC+j model to estimate the frequency of biogeographic events across the evolutionary history of Habromys. We generated 50 stochastic map replicates and recorded, for each replicate, the number of inferred events corresponding to anagenetic dispersal, within-area cladogenesis (narrow sympatry), subset sympatry, vicariance, and founder-event speciation. The resulting frequency distributions were used to assess the relative contribution of each process.

Results

Phylogenetic analysis. Our phylogenetic hypothesis recovered the same overall topology as León-Paniagua et al. (2007), consistent with the use of the same mitochondrial markers (ND3 and ND4) and substitution model HKY+I+G. Some differences were observed in the relationships among individuals of H. lophurus, H. lepturus, H. delicatulus, H. schmidlyi, and H. simulatus, reflected in moderate posterior probabilities (0.5–0.8) for several internal nodes (Figure 2).

Divergence-time estimates indicate that the neotomine clade containing Habromys diversified during the Miocene, whereas crown diversification within Habromys began in the Pliocene, with most interspecific divergences occurring during the Pleistocene.

Ancestral area estimation. DEC+j was the best-supported model explaining the geographic range evolution of Habromys, with an AIC weight of 0.80, compared to much lower values for the other six models tested (Table 2). Biogeographic stochastic mapping under this model indicated that founder-event speciation (+J) and within-area cladogenesis (narrow and subset sympatry) were inferred as the most frequent processes shaping range evolution and diversification in Habromys in the mountain systems of Mexico and the highlands of northern Central America, whereas anagenetic dispersal/extinction and vicariance were comparatively uncommon (Figs. 3 and 4; Supplemented Material 2).

Crown Habromys was inferred to originate in the Sierra Madre Oriental (SMO) at 4.69 Ma via founder-event speciation from the outgroup range (Baja California + Central Mexican Plateau) (Figure 3; Supplemented Material 2). The split leading to H. simulatus was dated at ~4.1 Ma, and this lineage was inferred to have colonized the eastern Trans-Mexican Volcanic Belt during the Middle Pleistocene (0.77 Ma) through a founder-event (Figure 3, Supplemented Material 2). A subsequent founder-event from the SMO to the southern Trans-Mexican Volcanic Belt (S-TMVB) was inferred at ~4.1 Ma. Within S-TMVB, H. schmidlyi was inferred to originate via within-area cladogenesis (narrow sympatry) at 0.67 Ma, followed by a founder-event into the northern Trans-Mexican Volcanic Belt, where H. delicatulus originated at 0.27 Ma (Figure 3; Supplemented Material 2).

A founder-event from the S-TMVB to the Sierra de Juárez (SJ) in the Sierra Madre de Oaxaca was inferred at 3.47 Ma (Figure 3). Within the SJ, diversification among H. ixtlani, H. lepturus, and H. chinanteco was dated at 2.66 Ma at the end of the Pliocene. A founder-event was inferred at 1.5 Ma from the SJ to Cerro Zempoaltepetl, by the lineage that includes H. lepturus, whereas H. ixtlani remains in SJ. This event was followed by within-area cladogenesis (narrow sympatry) at 0.83 Ma, associated with the divergence of H. lepturus. A subsequent founder-event from Cerro Zempoaltepetl back to the SJ was inferred at 0.37 Ma, associated with the divergence of H. chinanteco (Figure 3; Supplemented Material 2). Finally, a founder-event from the SJ into the Isthmus of Tehuantepec and the highlands of Chiapas and Guatemala was inferred at 2.66 Ma, followed by colonization of Honduras and El Salvador at 0.82 Ma. Diversification within H. lophurus was inferred to occur in these Central American highland areas (Figure 3; Supplemented Material 2).

Discussion

Previous studies on the diversification of the genus Habromys suggest that its presence in the mountain systems of central and eastern Mexico and northern Central America is the result of vicariant scenarios, in which range expansion was followed by isolation associated with habitat fragmentation (León-Paniagua et al. 2007; Rogers et al. 2007). However, these hypotheses were not formally evaluated in a time-calibrated framework that allows explicit comparison among alternative biogeographic processes. Our results indicate that the biogeographic history of Habromys is best explained by repeated founder-event speciation, whereas within-area cladogenesis (narrow sympatry) was inferred only for the nodes involving the lineages leading to H. schmidlyi and H. lepturus (Figure 3; Table 2; Supplemented Material 2).

Founder-event speciation, unlike speciation by vicariance, occurs when a new area is colonized by a small number of individuals through long-distance dispersal or rare dispersal events. Such events may be facilitated when a geographical barrier disappears (due to various factors such as climate changes, continental collisions, the formation of mountain ranges, or changes in sea level). When the barrier reappears or a new one forms, populations on both sides of the barrier begins to separate, reducing gene flow and allowing them to differentiate into different species (geodispersion). Therefore, barriers are not static elements, but rather dynamic throughout history (Lieberman and Eldredge 1996; Matzke 2013c; Wiley and Liberman 2011). In Habromys, the best-fitting model (DEC+j, Table 2) supports a biogeographic scenario dominated by multiple founder-events speciation events from the early Pliocene to the middle Pleistocene (Figure 3, Supplemented Material 2), consistent with the climatic oscillations that periodically expanded and contracted cloud forest habitats, and not with the formation of mountain systems. This differs from the scenario proposed by León-Paniagua et al. (2007), who interpreted the diversification of Habromys and the closely related genera (Podomys, Osgoodomys, and Megadontomys) as the result of an initial range expansion followed by vicariant differentiation driven by climatic fragmentation of montane forests.

The Pliocene has been considered a transitional climatic period characterized by warm climates that changed rhythmically every 40 000 years to large-scale glacial-interglacial periods, which preceded the Pleistocene conditions that occurred with greater frequency and magnitude (Dowsett et al. 2010; Lawrence et al. 2010; Salzmann et al. 2011). In this context, the founding-events occurred during the Pliocene leading to the colonization of the southern Trans-Mexican Volcanic Belt (4.69 Ma), Sierra Madre Oriental (4.1 Ma), Sierra de Juárez (3.47 Ma), and the Isthmus of Tehuantepec, Chiapas, and Guatemala (2.66 Ma), coincide with warm periods of this era (Haywood et al. 2009; Salzmann et al. 2011; Jennings et al. 2025). These have been associated with the dynamics of cloud forests and may have increased their connectivity, which correspond to the habitat of Habromys species, whose processes associated solely with the Pliocene have been overshadowed by Pleistocene climate changes (Sosa et al. 2016).

Although colonization of the areas began during Pliocene, the diversification of the species that make up Habromys within their current ranges took place during the Middle Pleistocene (1.5–0.27 My) (Figs. 2 and 3). This coincides with the Middle Pleistocene Transition (MPT, 1.2–0.5 My), when the frequency of glacial periods changed from approximately every 41 000 years to approximately every 100 000 years, and remained so until the Late Pleistocene (Head and Gibbard 2005; Head et al. 2008; Sun et al. 2019; Herbert 2023). In the case of Habromys, our divergence-time estimates suggest that their presence distribution areas coincide with glacial periods that took place during the MPT (Cohen and Gibbard 2019; 2022), which coincides with León-Paniagua and Guevara (2019). The colonization of the eastern Trans-Mexican Volcanic Belt from the Sierra Madre Oriental by H. simulatus (Figure 3) is temporally consistent with a glacial period that took place 0.77 Ma. Likewise, the colonization southern and northern Trans-Mexican Volcanic Belt, 0.67 and 0.27 Ma respectively (Figure 3), shows that the presence and diversification of H. schmidlyi and H. delicatulus are also the result of glacial periods that occurred during the MPT (Cohen and Gibbard 2019; 2022).

Unlike the findings of Rogers et al. (2007), who argue that the colonization the Sierra de Juárez and Cerro Zempoaltepetl resulted from a single vicariant event occurred in the valley of the Cajonos River, our results suggest a biogeographic scenario were an initial colonization of the Sierra de Juárez by H. ixtlani (1.5 Ma) took place during a glacial period prior to the onset of the MPT (Head and Gibbard 2005; Head et al. 2008; Cohen and Gibbard 2019; Sun et al. 2019; 2022; Herbert 2023); subsequently, the colonization of Cerro Zempoaltepetl (0.83 Ma) and the recolonization of the Sierra de Juárez (0.37 Ma) by H. chinanteco and H. lepturus also coincide with glacial periods that occurred during the MTP (Cohen and Gibbard 2019; 2022). As for the presence of H. lophurus in the Isthmus of Tehuantepec, Chiapas and Guatemala, and Honduras and El Salvador (0.82 Ma) may be associated with the same glacial episode that led to the presence of H. lepturus on Cerro Zempoaltepetl. This differs from other studies that have proposed that the presence of H. lophurus in the mountain systems of northern Central America is the result of vicariance events due to the presence of the Isthmus of Tehuantepec, which acts has acted as a geographical barrier to the dispersal of Habromys and other small mammals (Carleton et al. 2002; Rogers et al. 2007; León-Paniagua and Guevara 2019). However, it is necessary to consider that the cold and dry conditions that occurred during glacial periods may have promoted the contraction of cloud forests, restricting them to possible refuges, as has been suggested for other periods in Mesoamerica (Toledo 1982; Graham 1999).

The evolution of Habromys and its distribution areas is the result of glacial and interglacial events during the Pliocene and Pleistocene epochs. The oldest events allowed the colonization of new areas, while the most recent ones generated isolation and, therefore, the differentiation of each species within its distribution area. This is consistent with the proposal by León-Paniagua and Guevara (2019), who mention that glacial events allowed connections through the formation of corridors between mountainous areas, enabling dispersal events between them, while during warm periods these processes were prevented because valleys and lowlands acted as barriers for species of the genus Habromys. However, it is worth mentioning that the climatic changes associated with the Last Glacial Maximum, although they may have had effects on distribution (León-Paniagua and Guevara 2019), did not coincide with major inferred divergence events in our phylogeny.

Our results suggest that the biogeographic history of Habromys has been strongly shaped by long-term climatic fluctuations, in which colonization and subsequent isolation processes appear to have played a greater role than vicariant scenarios linked to the formation of Mesoamerican mountain systems. Nevertheless, our inferences are based on a phylogeny reconstructed from mitochondrial genes, which represent a single maternally inherited locus. Mitochondrial datasets may not fully reflect the species tree due to processes such as incomplete lineage sorting, introgression, or sex-biased dispersal, potentially affecting both divergence-time estimates and ancestral-area reconstructions. Therefore, although our results support a founder-event dominated biogeographic scenario, future studies incorporating multilocus nuclear data or phylogenomic datasets will be critical to test the robustness of this hypothesis and to evaluate whether alternative processes such as vicariance or anagenetic dispersal gain support under different phylogenetic reconstructions.

Acknowledgments

We would like to thank the editors for inviting us to contribute to this special issue honoring the academic career of Dr. Livia León-Paniagua. We would also like to thank Dr. Livia León-Paniagua for inspiring so many generations and serving as an example of tenacity and dedication.

Declaration of Artificial Intelligence use

We declare that the authors have not used any Artificial Intelligence in the elaboration of this paper.

Author contributions

Conceptualization: Deborah V. Espinosa-Martínez and César A. Ríos-Muñoz; Methodology: Deborah V. Espinosa-Martínez and César A. Ríos-Muñoz; Supervision: Joaquín Arroyo-Cabrales; Drafting: Deborah V. Espinosa-Martínez, César A. Ríos-Muñoz, and Joaquín Arroyo-Cabrales.

Supplementary data

SD1. Genbank accession numbers of specimens used in the construction of the dated phylogenetic reconstruction.

SD2. Cladogenetic biogeographic events are inferred from biogeographical stochastic mapping (BSM) under the DEC+J model. The table shows the divergence time (Ma), process type, ancestral area, descendant areas (1 and 2), and the dispersal areas in founder events. A = X, B =, C =, D =, E =, F =, G =, H =, I = y J =. Geographic areas correspond to those shown in Fig. 1: A = Isthmus of Tehuantepec + Chiapas + Guatemala; B = Honduras + El Salvador; C = Sierra Madre Oriental; D = Eastern Trans-Mexican Volcanic Belt; E = Northern Trans-Mexican Volcanic Belt; F = Southern Trans-Mexican Volcanic Belt; G = Cerro Zempoaltepetl; H = Sierra de Juárez; I = Baja California; J = Central Mexican Plateau.

Literature cited

Álvarez-Castañeda ST, and Cortés-Calva P. 2002. Peromys-cus slevini. Mammalian Species 705:1–2. https://doi.org/10.1644/0.705.1

Álvarez-Castañeda ST, Vázquez E, Castro-Arellano I, and Lacher T. 2018. Habromys ixtlani. The IUCN Red List of Threatened Species 2018. e.T136582A22376638. [Accessed December 3rd, 2025]. https://dx.doi.org/10.2305/IUCN.UK.2018-2.RLTS.T136582A22376638.en

Bouckaert R, Vaughan, TG, Barido-Sottani J, Duchêne S, Fourment M, Gavryushkina A, et al. 2019. BEAST 2.5: an advanced software platform for Bayesian evolutionary analysis. PLoS Computational Biology 15:e1006650. https://doi.org/10.1371/journal.pcbi.1006650

Briones-Salas M, Hernández-Allende A, Coronel MM, and Pérez GG. ٢٠١٢. New records of the endemic Chinanteco deermouse Habromys chinanteco (Rodentia: Cricetidae) in the Sierra Madre de Oaxaca, Mexico. The Southwestern Naturalist 57:221–222. https://doi.org/10.2307/23258057

Carleton MD, Sánchez O, and Urbano Vidales G. 2002. A new species of Habromys (Muroidea: Neotominae) from México with a generic review of species definitions and remarks on diversity patterns among Mesoamerican small mammals restricted to humid montane forests. Proceedings of the Biological Society of Washington 115:488–533.

Castañeda-Rico S, León-Paniagua L, Ruedas LA, and Vázquez-Domínguez E. 2011. High genetic diversity and extreme differentiation in the two remaining populations of Habromys simulatus. Journal of Mammalogy 92:963–973. https://doi.org/10.2307/23259931

Castañeda-Rico S, León-Paniagua L, Vázquez-Domínguez E, and Navarro-Sigüenza AG. 2014. Evolutionary diversification and speciation in rodents of the Mexican lowlands: the Peromyscus melanophrys species group. Molecular Phylogenetics and Evolution 70:454–463. https://doi.org/10.1016/j.ympev.2013.10.004

Castañeda-Rico S, Parker LD, Sánchez E, Rivas-Trasvina S, Hawkins MTR, Edwards CW, and Maldonado JE. 2023. Novel genomic resources contribute to the systematics of threatened arboreal deer mice of the genus Habromys Hooper and Musser, 1964 (Cricetidae, Neotominae) within a neotomine–peromyscine phylogeny. ZooKeys 1179: 157–168. https://doi.org/10.3897/zookeys.1179.108759

Ceballos G, editor. 2014. Mammals of Mexico. Baltimore (EEUU): Johns Hopkins University Press.

Cohen KM, and Gibbard P. 2019. Global chronostratigraphical correlation table for the last 2.7 million years, version 2019 QI-500. Quaternary International 500:20–31. https://doi.org/10.1016/j.quaint.2019.03.009

Cohen KM, and Gibbard P. 2022. Global chronostratigraphical correlation table for the last 2.7 million years v.2019 (poster version). Mendeley Data V5. https://doi.org/10.17632/dtsn3xn3n6.5.

Colunga Salas PF. 2014. Ámbito hogareño y uso de hábitat de Habromys schmidlyi (Rodentia: Cricetidae). [Bachelor’s Thesis]. [Mexico City (MEX)]: Universidad Nacional Autónoma de México.

Colunga Salas PF. ٢٠١٦. Diversidad genética y patrones ecológicos de Habromys schmidlyi. [Master’s Thesis]. [Mexico City (MEX)]: Universidad Nacional Autónoma de México.

Contreras-Medina R, Santiago-Alvarado M, Espinosa D, Rivas G, and Luna-Vega I. ٢٠٢٤. Distributional patterns and conservation of the genus Habromys (Rodentia: Cricetidae) in Mesoamerica. Studies on Neotropical Fauna and Environment 59:175–191. https://doi.org/10.1080/01650521.2022.2085071

Dowsett HJ, and Caballero Gill RP. 2010. Pliocene climate. Stratigraphy 7:105–110.

Edgar RC. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research 32:1792–1797. https://doi.org/10.1093/nar/gkh340

Emmons L, and Vázquez E. 2019. Habromys lophurus. The IUCN Red List of Threatened Species 2019. e.T9610A22376801. International Union for Conservation of Nature; December, 2025. [Accessed December 3rd, 2025]. https://dx.doi.org/10.2305/IUCN.UK.2019-3.RLTS.T9610A22376801.en

Goodwin GG. 1964. A new species and a new subspecies of Peromyscus from Oaxaca, Mexico. American Museum Novitates 2183:1–8.

Graham A. 1999. The tertiary history of the northern temperate element in the northern Latin America biota. American Journal of Botany 86:32–38. https://doi.org/10.2307/2656952

Hall ER. 1981. The mammals of North America. 2ª ed. Vol. 2. New York (EEUU): The Blackburn Press.

Hasegawa M, Kishino H, and Yano T. 1985. Dating of human-ape splitting by a molecular clock of mitochondrial DNA. Journal of Molecular Evolution 22:160–174. https://doi.org/10.1007/BF02101694

Haywood AM, Dowsett HJ, Valdes PJ, Lunt DJ, Francis JE, and Sellwood BW. 2009. Introduction: Pliocene climate, processes and problems. Philosophical Transactions of the Royal Society A 367:3–17. https://doi.org/10.1098/rsta.2008.0205

Head MJ, and Gibbard PL. 2005. Early–Middle Pleistocene transitions: an overview and recommendation for the defining boundary. In: Head MJ, and Gibbard PL, editors. Early–Middle Pleistocene transitions: the land–ocean evidence. London (UK): Geological Society, Special Publications; p. 1–18. https://doi.org/10.1144/GSL.SP.2005.247.01.01

Head MJ, Pillans B, and Farquhar SA. 2008. The Early–Middle Pleistocene Transition: characterization and proposed guide for the defining boundary. Episodes 31:255–259. https://doi.org/10.18814/epiiugs/2008/v31i2/014

Heled J, and Drummond AJ. 2012. Calibrated tree priors for relaxed phylogenetics and divergence time estimation. Systematic Biology 61:138–149.

Herbert TD. 2023. The Mid-Pleistocene Climate Transition. Annual Review of Earth and Planetary Sciences 51:389–418. https://doi.org/10.1146/annurev-earth-032320-104209

Hooper ET, and Musser GG. 1964. Notes on classification on the rodent genus Peromyscus. Occasional Papers of the Museum of Zoology, University of Michigan 635: 1–13.

Jennings CE, Aber JS, Barendregt R, Bierman PR, Mason J, Rovey CW, et al. 2025. Middle Pleistocene glaciations in North America. In: Elias S, editor. Encyclopedia of Quaternary Science, 3ª ed. London (UK): Elsevier: p. 647–655. https://doi.org/10.1016/B978-0-323-99931-1.00123-9

Kalcounis-Rueppell MC, and Spoon TR. 2009. Peromyscus boylii (Rodentia: Cricetidae). Mammalian Species 838:1–14. https://doi.org/10.1644/838.1

Landis MJ, Matzke NJ, Moore BR, and Huelsenbeck JP. 2013. Bayesian analysis of biogeography when the number of areas is large. Systematic Biology 62:789–804. https://doi.org/10.1093/sysbio/syt040

Lawrence KT, Sosdian S, White HE, and Rosenthal Y. 2010. North Atlantic climate evolution through the Plio-Pleistocene climate transitions. Earth and Planetary Science Letters 300:329–342. https://doi.org/10.1016/j.epsl.2010.10.013

León-Paniagua L, and Guevara L. 2019. Quaternary range-shifts of arboreal rodents of the genus Habromys (Cricetidae, Neotominae) in Mesoamerica and their evolutionary consequences. Mammalian Biology 94:4–10. https://doi.org/10.1016/j.mambio.2018.10.005

León-Paniagua L, Navarro-Sigüenza AG, Hernández-Baños BE, and Morales JC. 2007. Diversification of the arboreal mice of the genus Habromys (Rodentia: Cricetidae: Neotominae) in the Mesoamerican highlands. Molecular Phylogenetics and Evolution 42:653–664. https://doi.org/10.1016/j.ympev.2006.08.019

Lieberman BS, and Eldredge N. 1996. Trilobite biogeography in the Middle Devonian: geological processes and analytical methods. Paleobiology 22:66–79. https://doi.org/10.1017/S009483730001602X

Marines-Macías T, Colunga-Salas P, Verde Arregoitia LD, Narajo EJ, and León-Paniagua L. 2018. Space use by two arboreal rodent species in a Neotropical cloud forest. Journal of Natural History 52:1417–1431. https://doi.org/10.1080/00222933.2018.1459921

Matzke NJ. 2013a. BioGeoBEARS: biogeography with Bayesian (and likelihood) evolutionary analysis in R scripts. V 0.2.1. R package. https://rdrr.io/cran/BioGeoBEARS/.

Matzke NJ. 2013b. Probabilistic historical biogeography: new models for founder-event speciation, imperfect detection, and fossils allow improved accuracy and model-testing. Frontiers of Biogeography 5:242–248. https://doi.org/10.21425/F55419694

Matzke NJ. 2013c. Probabilistic historical biogeography: new models for founder-event speciation, imperfect detection, and fossils allow improved accuracy and model-testing. [PhD thesis]. [Berkeley (EEUU)]: University of California.

Matzke NJ. 2014. Model selection in historical biogeography reveals that founder-event speciation is a crucial process in island clades. Systematic Biology 63:951–970. https://doi.org/10.1093/sysbio/syu056

Merriam CH. 1898. Descriptions of twenty new species and a new subgenus of Peromyscus from Mexico and Guatemala. Proceedings of the Biological Society of Washington 12:115–125.

Osgood WH. 1904. Thirty new mice of the genus Peromyscus from Mexico and Guatemala. Proceedings of the Biological Society of Washington 17:55–77.

R Core Team. 2024. R: A language and environment for statistical computing. R Foundation for Statistical Computing. V. 4.4.2. Vienna (Austria). https://www.R-project.org/.

Rambaut A, Drummond AJ, Xie D, Baele G, and Suchard MA. 2018. Posterior Summarization in Bayesian Phylogenetics Using Tracer 1.7. Systematic Biology 67:901–904. https://doi.org/10.1093/sysbio/syy032

Ramírez-Pulido J, González-Ruiz N, Gardner AL, and Arroyo-Cabrales J. 2014. List of recent land mammals of Mexico, 2014. Special Publications, Museum of Texas Tech University 63:1–69.

Ree RH, and Smith SA. 2008. Maximum likelihood inference of geographic range evolution by dispersal, local extinction, and cladogenesis. Systematic Biology 57:4–14. https://doi.org/10.1080/10635150701883881

Ree RH, Moore BR, Webb CO, and Donoghue MJ. 2005. A likelihood framework for inferring the evolution of geographic range on phylogenetic trees. Evolution 59:2299–2311. https://doi.org/10.1111/j.0014-3820.2005.tb00940.x

Robertson PB, and Musser GG. 1976. A new species of Peromyscus (Rodentia: Cricetidae), and a new specimen of P. simulatus from southern Mexico, with comments on their ecology. Occasional Papers of the Museum of Natural History, The University of Kansas 47:1–8.

Rogers, DS, Funk CC, Miller JR, and Engstrom MD. 2007. Molecular phylogenetic relationships among crested-tailed mice (genus Habromys). Journal of Mammalian Evolution 14:37–55. https://doi.org/10.1007/s10914-006-9034-2

Romo-Vázquez E, León-Paniagua L, and Sánchez O. 2005. A new species of Habromys (Rodentia: Neotominae) from Mexico. Proceedings of the Biological Society of Washington 118:605–618. https://doi.org/10.2988/0006-324X(2005)118[605:ANSOHR]2.0.CO;2

Ronquist F. 1997. Dispersal–vicariance analysis: a new approach to the quantification of historical biogeography. Systematic Biology 46:195–203. https://doi.org/10.1093/sysbio/46.1.195

Ronquist F, and Sanmartín I. 2011. Phylogenetic methods in biogeography. Annual Review of Ecology, Evolution, and Systematics 42:441–464. https://doi.org/10.1146/annurev-ecolsys-102209-144710

Salzmann U, Williams M, Haywood AM, Johnson ALA, Kender S, and Zalasiewicz J. 2011. Climate and environment of a Pliocene warm world. Palaeogeography, Palaeoclimatology, Palaeoecology 309:1–8. https://doi.org/10.1016/j.palaeo.2011.05.044

Sanmartín I. 2012. Historical biogeography: evolution in time and space. Evolution: Education and Outreach 5:555–568. https://doi.org/10.1007/s12052-012-0421-2

SEMARNAT. 2010. Norma Oficial Mexicana NOM-059-ECOL-2010. Diario Oficial de la Federación DCLXXXVII (23, 2ª sección). Mexico City (MEX): Secretaría de Medio Ambiente y Recursos Naturales; p. 1–78.

Sosa V, Ornelas JF, Ramírez-Barahona S, and Gándara E. ٢٠١٦. Historical reconstruction of climatic and elevation preferences and the evolution of cloud forest-adapted tree ferns in Mesoamerica. PeerJ 4:e2696. https://doi.org/10.7717/peerj.2696

Sun Y, Yin Q, Crucifix M, Clemens SC, Araya-Melo P, Liu W, et al. 2019. Diverse manifestations of the mid-Pleistocene climate transition. Nature Communications 10:352. https://doi.org/10.1038/s41467-018-08257-9

Tamura K, Stecher G, and Kumar S. 2021. MEGA 11: molecular evolutionary genetics analysis version 11. Molecular Biology and Evolution 38:3022–3027. https://doi.org/10.1093/molbev/msab120

Toledo VM. 1982. Pleistocene changes of vegetation in tropical Mexico. In: Prance GT, editor. Biological diversification in the tropics. New York (EEUU): Columbia University Press; p. 93–111.

Vázquez E. 2017. Habromys delicatulus. The IUCN Red List of Threatened Species 2017. e.T136683A22376548: International Union for Conservation of Nature; December, 2025. [Accessed December 3rd, 2025]. https://dx.doi.org/10.2305/IUCN.UK.2017-2.RLTS.T136683A22376548.en.

Wiley EO, and Lieberman BS. 2011. Phylogenetics: theory and practice of phylogenetic systematics. New Jersey (EEUU): Wiley-Blackwell.

Associated editors: Giovani Hernández Canchola and Pablo Colunga Salas

Submitted: December 30, 2025; Reviewed: February 9, 2026

Accepted: May 8, 2026; Published online: May 29, 2026

Supplementary data

SD1. Genbank accession numbers of specimens used in the construction of the dated phylogenetic reconstruction.

SD2. Cladogenetic biogeographic events are inferred from biogeographical stochastic mapping (BSM) under the DEC+J model. The table shows the divergence time (Ma), process type, ancestral area, descendant areas (1 and 2), and the dispersal areas in founder events. A = X, B =, C =, D =, E =, F =, G =, H =, I = y J =. Geographic areas correspond to those shown in Fig. 1: A = Isthmus of Tehuantepec + Chiapas + Guatemala; B = Honduras + El Salvador; C = Sierra Madre Oriental; D = Eastern Trans-Mexican Volcanic Belt; E = Northern Trans-Mexican Volcanic Belt; F = Southern Trans-Mexican Volcanic Belt; G = Cerro Zempoaltepetl; H = Sierra de Juárez; I = Baja California; J = Central Mexican Plateau.

Species (individual)	Locality	Genbank accession number
Habromys chinanteco (Hchi136584)	21 km S Vista Hermosa, Oaxaca, México.	DQ793118.1
H. chinanteco (Hchi124129)	28.6 km S Vista Hermosa, Oaxaca, México.	DQ793117.1
H. schmidlyi (Hschm6071)	8 km SW Zacualpan, Almoloya,de Alquiciras, Estado de México, México.	DQ793116.1
H. schmidlyi (Hschm6070)	8 km SW Zacualpan, Almoloya,de Alquiciras, Estado de México, México.	DQ793115.1
H. simulatus (Hsim169)	Puerto de la Soledad, Cuicatlán, Oaxaca, México.	DQ793114.1
H. simulatus (Hsim173)	Puerto de la Soledad, Cuicatlán, Oaxaca, México.	DQ793113.1
H. simulatus (Hsim159)	Puerto de la Soledad, Cuicatlán, Oaxaca, México.	DQ793112.1
H. simulatus (Hsim171)	Puerto de la Soledad, Cuicatlán, Oaxaca, México.	DQ793111.1
H. simulatus (Hsim018TD)	Tenango de Doria, Hidalgo, México.	DQ793110.1
H. simulatus (Hsim019TD)	Tenango de Doria, Hidalgo, México.	DQ793109.1
H. simulatus (Hsim017TD)	Tenango de Doria, Hidalgo, México.	DQ793108.1
H. simulatus (Hsim39686)	El Potrero, 10 km SW, Tenango de Doria, Hidalgo, México.	DQ793107.1
H. simulatus (Hsim83263V)	3 km W Zacualpan, Zacualpan, Veracruz, México.	DQ793106.1
H. simulatus (Hsimu)	El Potrero, Tenango de Doria, Hidalgo, México.	KF885886.1
H. lepturus (Hlep33630)	Cerro Zempoaltepetl, 4.5 km N Santa María Yacochi, Tlahuitoltepec,Oaxaca, México.	DQ793105.1
H. lepturus (Hlep35232)	Cerro Zempoaltepetl, 4.5 km N Santa María Yacochi, Tlahuitoltepec,Oaxaca, México.	DQ793104.1
H. lepturus (Hlep35237)	Cerro Zempoaltepetl, 4.5 km N Santa María Yacochi, Tlahuitoltepec,Oaxaca, México.	DQ793103.1
H. lepturus (Hlep35239)	Cerro Zempoaltepetl, 4.5 km N Santa María Yacochi, Tlahuitoltepec,Oaxaca, México.	DQ793102.1
H. ixtlani (Hixt33599O)	27.5 km NNE Llano de las Flores, Oaxaca, México.	DQ793101.1
H. ixtlani (Hixt00173O)	6.5 km ESE, Santa María Yavesía, Ixtlán, Oaxaca. México.	DQ793100.1
H. delicatulus (Hdeli22442)	Cañada de la Hermita, Dexcani Alto, Jilotepec, Estado de México, México.	DQ793099.1
H. delicatulus (Hdeli22440)	Cañada de la Hermita, Dexcani Alto, Jilotepec, Estado de México, México.	DQ793098.1
H. lophurus (Hlo98616S)	E Slope, Los Esesmiles, Departamento de Chalatenango, El Salvador.	DQ793097.1
H. lophurus (Hloph021CH)	Poblado Chiquihuites, Unión Juárez, Volcán Tacaná, Chiapas. México.	DQ793096.1
H. lophurus (Hloph022CH)	Poblado Chiquihuites, Unión Juárez, Volcán Tacaná, Chiapas. México.	DQ793095.1
H. lophurus (Hloph888)	Polígono 1, El Triunfo, El Triunfo, Angel Albino Corzo, Chiapas, México.	DQ793094.1
H. lophurus (Hloph11804)	4 km NW Santa Eulalia, Yayquich, Departamento Huehuetenango, Guatemala.	DQ793093.1
H. lophurus (Hloph13401)	La Cabaña, Sierra de la Minas, San Agustín Acasaguastlán, Guatemala.	DQ793092.1
H. lophurus (Hloph13354)	La Cabaña, Sierra de la Minas, San Agustín Acasaguastlán, Guatemala.	DQ793091.1
H. lophurus (Hloph13422)	6 km NNW San Lorenzo, Sierra de las Minas, Río Hondo, RB, Zacapa, Guatemala.	DQ793090.1
H. lophurus (Hloph)	Not available	U83863.1
Peromyscus boylii	Not available	U83864.1
P. slevini	Not available	U40248.1
P. melanotis	Not available	U40247.1
P. polionotus	Not available	U40254.1
P. maniculatus oreas	Not available	U40062.1
P. leucopus	Not available	U40252.1
P. eremicus	Not available	U83861.1
P. mexicanus	Not available	U83862.1
Osgoodomys banderanus	Not available	U83860.1
Podomys floridanus	Not available	U83865.1
Megadonthomys cryophilus	Not available	DQ793119.1
Neotoma floridana	Not available	U83827.1
Baiomys taylori	Not available	U83829.1
Onychomys leucogaster	Not available	U83858.1

Supplementary Material 2. Cladogenetic biogeographic events are inferred from biogeographical stochastic mapping (BSM) under the DEC+J model. The table shows the divergence time (Ma), process type, ancestral area, descendant areas (1 and 2), and the dispersal areas in founder events. A = X, B =, C =, D =, E =, F =, G =, H =, I = y J =. Geographic areas correspond to those shown in Figure 1: A = Isthmus of Tehuantepec + Chiapas + Guatemala; B = Honduras + El Salvador; C = Sierra Madre Oriental; D = Eastern Trans-Mexican Volcanic Belt; E = Northern Trans-Mexican Volcanic Belt; F = Southern Trans-Mexican Volcanic Belt; G = Cerro Zempoaltepetl; H = Sierra de Juárez; I = Baja California; J = Central Mexican Plateau.

Time (My)	Taxa	Process type	Ancestral area	Descendeant 1	Descendant 2	Dispersal area
4.69	((P. boylii + P. slevini) + ((H. simulatus 171 + H. simulatus 019TD) + ((H. schmidly 6071 + (H. schmidly 6060 + H. delicatulus 22440)) + ((H. ixtlani 00173O + (H. lepturus 35239 + (H. chinanteco 124129 + H. lepturus 33630)) + (H. lophurus 98616S + H. lophurus 022CH)))))	Founder	IJ	C	IJ	C
4.1	((H. simulatus 171 + H. simulatus 019TD) + ((H. schmidly 6071 + (H. schmidly 6060 + H. delicatulus 22440)) + ((H. ixtlani 00173O + (H. lepturus 35239 + (H. chinanteco 124129 + H. lepturus 33630)) + (H. lophurus 98616S + H. lophurus 022CH))))	Founder	C	F	C	F
3.47	((H. schmidly 6071 + (H. schmidly 6060 + H. delicatulus 22440)) + ((H. ixtlani 00173O + (H. lepturus 35239 + (H. chinanteco 124129 + H. lepturus 33630)) + (H. lophurus 98616S + H. lophurus 022CH)))	Founder	F	H	F	H
2.66	((H. ixtlani 00173O + (H. lepturus 35239 + (H. chinanteco 124129 + H. lepturus 33630)) + (H. lophurus 98616S + H. lophurus 022CH))	Founder	H	A	H	A
1.57	(P. boylii + P. slevini)	Subset sympatry	IJ	IJ	J	-
1.5	(H. ixtlani 00173O + (H. lepturus 35239 + (H. chinanteco 124129 + H. lepturus 33630)))	Founder	H	G	H	G
0.83	(H. lepturus 35239 + (H. chinanteco 124129 + H. lepturus 33630))	Narrow sympatry	G	G	G	-
0.82	(H. lophurus 98616S + H. lophurus 022CH)	Founder	A	B	A	B
0.77	(H. simulatus 171 + H. simulatus 019TD)	Founder	C	C	D	D
0.67	(H. schmidly 6071 + (H. schmidly 6060 + H. delicatulus 22440))	Narrow sympatry	F	F	F	-
0.37	(H. chinanteco 124129 + H. lepturus 33630)	Founder	G	H	G	H
0.27	(H. schmidly 6060 +H. delicatulus 22440)	Founder	F	E	F	E

THERYA, 2026, Vol. 17(2):295-306

DOI: 10.12933/therya.2026.6269 ISSN 2007-3364

Figure 1. Distribution areas of the species included in the analysis: Peromyscus boylii (Central Mexican Plateau), Habromys delicatulus (Northern Trans-Mexican Volcanic Belt), H. schmidlyi (Southern Trans-Mexican Volcanic Belt), H. simulatus (Sierra Madre Oriental and eastern Trans-Mexican Volcanic Belt), H. ixtlani and H. chinanteco (Sierra de Juárez), H. lepturus (Cerro Zempoaltepetl), H. lophurus (Isthmus of Tehuantepec, Chiapas, Guatemala, Honduras, and El Salvador). Peromyscus slevini (not illustrated) is found only on Santa Catalina Island in the Gulf of California.

Table 1. Parameters used in the calibration of each internal node. The ages used in the divergence time estimates for the nodes of the Habromys genus phylogeny were taken from León-Paniagua et al. (2007).

Nodes-Groups	Calibration
Neotominae	14.8 Ma (M = 7.1, S = 0.01, mean in real space = true, offset = 7.7)
Onychomys-Peromyscus-Habromys	10.3 Ma (M = 4.9, S = 0.15, mean in real space = true, offset = 5.36)
Peromyscus-Habromys	4.69 Ma (M = 2.25, S = 0.05, mean in real space = true, offset = 2.44)
Habromys simulatus	0.78 Ma (M = 0.4, S = 0.1, mean in real space = true, offset = 0.38)

Table 2. Summary of statistics for selecting the best model obtained in BioGeoBEARS. LnL = likelihood, d = dispersion parameter, e = extinction parameter, j = founder-event parameter, AIC = Akaike Information Criterion, and AIC_wt = weighted Akaike Information Criterion.

Model	LnL	Number of parameters per model	d	e	j	AIC	AIC_wt
DEC	-40.4	2	0.07	0.13	0	84.79	0.000023
*DEC+j	-28.92	3	1E-12	1E-12	0.31	63.85	0.8
DIVALIKE	-37.91	2	0.059	0.026	0	79.82	0.0003
DIVALIKE+j	-30.62	3	0.0064	1E-12	0.14	67.24	0.15
BAYAREALIKE	-46.09	2	0.15	0.54	0	96.19	0.000000076
BAYAREALIKE+j	-31.59	3	0.0007	0.031	0.17	69.18	0.055

Figure 2. Phylogenetic reconstruction dated for the genus Habromys and its external groups based on information from León-Paniagua et al. (2007)

Figure 3. Hypothesis of ancestral area estimation for the genus Habromys. The main process involved was the founder-event present in all nodes, with the exception of sympatry events in the species H. delicatulus and H. lepturus. The colors of the branches correspond to those of the areas in Figure 1.

Figure 4. Distribution of biogeographic processes counts across 50 biogeographical stochastic mapping (BSM) replicates under the DEC+J model. Panels show the frequency of inferred events for five processes: (A) Anagenetic dispersal, (B) Narrow sympatry, (C) Subset sympatry, (D) Vicariance, and (E) Founder events. The results indicate a strong predominance of founder-event speciation relative to other biogeographic processes.