Bothriechis bicolor (Bocourt, 1868)

Clause, Adam G., Luna-Reyes, Roberto, Lang, Noé Jiménez, Oca, Adrián Nieto-Montes de & Hernández, Luis Alberto Martínez, 2020, Problems with imperfect locality data: distribution and conservation status of an enigmatic pitviper, Amphibian & Reptile Conservation (e 246) 14 (2), pp. 185-197 : 188-192

Ambiguity in the Distribution of B. bicolor

This study highlights the ambiguity that can exist concerning species distributions. This ambiguity can potentially lead to erroneous biogeographical conclusions, and complicate conservation assessments. By exploring these issues as they relate to B. bicolor , several sources of ongoing scholarly confusion are resolved and the need for greater awareness of problems associated with imprecise biodiversity information are highlighted.

Perhaps the greatest ambiguity in the literature associated with B. bicolor is whether the species is known from Honduras. Historically, many authors placed the species in Honduras ( Bogert 1968; Meyer and Wilson 1971; Wilson and Meyer 1982; Wilson 1983; Wilson and Meyer 1985; Campbell and Lamar 1989; Crother et al. 1992; Wilson and McCranie 1994; McDiarmid et al. 1999). However, all Honduran material ascribed to B. bicolor by these authors was subsequently referred to the newly described species B. thalassinus ( Campbell and Smith 2000) . No new Honduran Bothriechis material has since been forthcoming other than Honduran populations announced as the newly described species B. guifarroi ,

2 All elevations are rounded to the nearest 10 m, and elevations preceded by “ca.” are rough estimates due to imprecise locality data.

3 Cuenca del Lago Atitlán (CDLA), Reserva de la Biósfera La Sepultura (REBISE), Reserva de la Biósfera El Triunfo (REBITRI), Reserva de la Biósfera Volcán Tacaná (REBIVTA), Área de Protección de Flora y Fauna La Frailescana (APFFLF), Zona Sujeta a Conservación Ecológica Pico El Loro-Paxtal (ZSCEPELP).

4 Colección Zoológica Regional de la Secretaría de Medio Ambiente e Historia Natural (CZR-HE); Field Museum of Natural History (FMNH); Natural History Museum of Los Angeles County photographic collection (LACM PC); Musuem of Comparative Zoology, Harvard University (MCZ); Museo de Zoologiá “Alfonso L. Herrera,” Facultad de Ciencias,Universidad Nacional Autónoma de México (MZFC-HE), Musuem of Vertebrate Zoology,University of California, Berkeley (MVZ); University of Michigan Museum of Zoology (UMMZ); Smithsonian Institution National Museum of Natural History (USNM); University of Texas at Arlington (UTA);University of Texas at El Paso Biodiversity Collections (UTEP).

5 Locality may actually lie outside the reserve boundary, but locality data are too imprecise for exact determination.

together with a possibly undescribed species ( Townsend et al. 2013). The taxonomic validity of either B. thalassinus or B. guifarroi has never been questioned in the literature. As such, although a number of authors ( Taggart et al. 2001; Wilson and McCranie 2002; Campbell and Muñoz-Alonso 2014; Pla et al. 2017; Mason et al. 2019) later attributed Honduran populations of B. thalassinus to B. bicolor , they either universally overlooked the description of B. thalassinus or mistakenly considered the two species roughly sympatric in Honduras. All other recent works ( Köhler 2008; Castoe et al. 2009; Townsend and Wilson 2010; Wilson and Johnson 2010; McCranie 2011; Townsend et al. 2013; Solís et al. 2014; Wallach et al. 2014; McCranie 2015) have not recognized B. bicolor as a member of the Honduran herpetofauna. Importantly, these works include all modern, authoritative treatments and checklists of the Honduran snake assemblage ( Townsend and Wilson 2010; McCranie 2011; Solís et al. 2014; McCranie 2015). Given the uncontroversial transfer of all Honduran B. bicolor material to the binomial B. thalassinus by Campbell and Smith (2000), and given that no Honduran B. bicolor vouchers have since been reported, we here affirm that B. bicolor is undocumented from Honduras. The nearest B. bicolor vouchers (Finca Rosario Vista Hermosa, Table 1) were obtained ca. 150 airline km west of the Honduras border.

Another major ambiguity relating to the geographic distribution of B. bicolor is the comparatively large number of problematic localities that have been treated inconsistently in the literature. Supplementing brief coverage by Bogert (1968), these seven problematic localities are discussed below because they encapsulate issues commonly posed by geographic data.

Bocourt (1868) gave the type locality for B. bicolor as “Des forêts de Saint-Augustin, département de Solola ( Guatémala), sur le versant occidental de la Cordillère. 610 mètres d’altitude.” Nonetheless, only a tiny corner of the department of Sololá lies at 610 m asl. The locality description conceivably refers to Finca San Agustín, department of Suchitepéquez, ca. 550–700 m asl on the slopes of Volcán Atitlán, ca. 8 km south of the border with the department of Sololá. This is consistent with the claim by several authors ( McDiarmid et al. 1999; Campbell and Lamar 2004) that the type locality probably lies on Volcán Atitlán. However, Wallach et al. (2014) erroneously georeferenced the type locality to the department of Sacatepéquez in the urban zone of the city of Antigua at ca. 1,530 m asl, adding additional confusion to the published literature. Assuming that Bocourt’s types did, indeed, originate from somewhere on Volcán Atitlán, they are also essentially topotypic with a specimen from the vague locality “cuesta de Atitlan im westlichen Guatémala ” that Müller (1877, 1878) used to describe “ Bothrops (Bothriechis) Bernoullii ” (see detailed discussion by McDiarmid et al. [1999]). Müller’s taxon was subsequently synonymized with B. bicolor , but the fact remains that the provenance of the types for both binomials is inexactly known.

In Chiapas, Juliá-Zertuche and Varela-Juliá (1978) reported a record from “Colonia Ejidal Morelos, Mpio. de Huixtla, Chis. […] y a unos 500 m. de altitud aproximadamente” as the type locality for another taxon, Bothriechis ornatus , that was also later synonymized with B. bicolor . The only community or site in the Municipio (Municipality) de Huixtla with the word “ Morelos ” in its name that we could identify is the hamlet of José María Morelos, but it sits at ca. 1,350 m asl, over 3 airline km from the 500 m contour. To our knowledge, no other Bothriechis vouchers have since reached a museum collection from anywhere within 15 airline km of the Municipio de Huixtla, leaving this locality vague and open to interpretation.

Three additional Chiapas localities cannot be confidently placed because they lack elevation data, no verbatim place names are identifiable, and they could plausibly correspond to two or more sites separated by over 10 airline km with imperfectly matching names. These three localities are as follows: “Catharinas (=Catarina la Grande?)” ( Greene 1971), “Chicharras” or variations thereof ( Smith 1941; Bogert 1968; Campbell and Lamar 2004), and “Finca La Lucha” ( Greene 1971).

Lastly, we are aware of an unvouchered 1994 sight record of a snake identified as B. bicolor from Rancho El Recuerdo in the Municipality of Jiquipilas, within what is now the Reserva de la Biósfera La Sepultura. If accurate, this would extend the range of the species ca. 40 km to the NW and would halve the distance between B. bicolor and known populations of its congener B. rowleyi near Cerro Baúl ( Bogert 1968). In 2018 and 2019, the authors unsuccessfully searched for Bothriechis near Rancho El Recuerdo on Cerro La Palmita. Several damaging wildfires had recently swept through this forested region ( Myers 2011), which might have influenced these survey results. However, until verifiable material reaches a museum, we consider the existence of Bothriechis in the vicinity of Cerro La Palmita uncertain.

Ambiguity in Elevation Range and Biogeography of B. bicolor

Intertwined with the problematic localities discussed above is ambiguity in the elevational range of B. bicolor . Based on material of sound provenance, the species is known from 900– 2,090 m asl ( Table 1). Yet, as indicated above, the problematic “Saint-Augustin” and “Colonia Ejidal Morelos” localities supposedly originate from 610 and 500 m, respectively. Additionally, Crother et al. (1992) list a minimum elevation of 457 m for specimens from Finca Rosario Vista Hermosa in Guatemala, but this was likely in error because museum catalogue data for those specimens list no elevation below 1,300 m asl. Despite prior authors consistently accepting 500 m as the lower elevation limit, for reasons articulated above, we consider the underlying data questionable. Confirmation of the geospatial validity of the seven problematic historical records, and thus of the true minimum elevation for B. bicolor , will necessitate targeted resurveys. Nonetheless, this could prove particularly challenging because lower-elevation habitats are more degraded relative to those at higher elevations ( Campbell and Lamar 2004; Campbell and Muñoz-Alonso 2014; Godínez-Gómez and Mendoza 2019), increasing the likelihood that low-elevation B. bicolor populations could now be extirpated. Climate change may have also pushed low-elevation B. bicolor populations upslope ( Elsen et al. 2020), which would further complicate resurveys to verify the lower elevation limit of the species.

The final source of confusion relating to the distribution of B. bicolor is biogeographical. All published sources indicate a strictly Pacific-versant range for B. bicolor , other than Clause et al. (2016) who were the first to explicitly state that B. bicolor occurs on interior-draining (Gulf of Mexico) slopes of the Sierra Madre de Chiapas. However, several prior authors beginning with Luna-Reyes (1997) had also reported localities from the Atlantic versant of that mountain range ( Meneses-Millán and García-Padilla 2015; Heimes 2016). Although Pacific drainages still harbor the majority of B. bicolor localities range-wide, our results emphasize that the species can no longer be accurately characterized as having a Pacific-versant distribution, at least in Mexico. We encourage field workers to be attentive to the possibility of encountering this species on both sides of the Continental Divide in the Sierra Madre de Chiapas. Future discovery of new B. bicolor localities will likely further improve understanding of how widely its range spans the Continental Divide, as would development of a rigorous ecological niche model for the species ( Wisz et al. 2008; Ríos-Muñoz and Espinosa-Martínez 2019).

More broadly, this work underscores the fact that the distribution of many organisms in southern Mexico remains poorly resolved, even at coarse spatial scales. The 68-km range extension for B. bicolor reported herein is one of several range extensions exceeding 50 km for highland squamates ( Morales et al. 2015; Hidalgo-García et al. 2018; Valdenegro-Brito et al. 2018) and salamanders ( Bouzid et al. 2015; Barrio-Amorós et al. 2016) reported in the last five years from Chiapas and Guatemala. Future survey efforts in remote, mountainous areas throughout Mesoamerica hold additional promise for wildlife discoveries of high biogeographical and conservation value.