nueva página del texto (beta)
Inglés (pdf)
Artículo en XML
Referencias del artículo
Traducción automática
Enviar artículo por email
Citado por SciELO 
Similares en
SciELO 
Permalink
Oaxaca, one of the culturally and biologically most distinctive states of Mexico, hosts an unusually large proportion of the plant diversity occurring in this country. With precise documented figures of 8,903 (García-Mendoza & Meave 2011) and 10,229 (Villaseñor 2016) species of vascular plants, this state’s flora only rivals with neighboring Chiapas, whose estimated richness ranges between 8,250-10,000 species, and exceeds estimated figures for other adjacent states like Guerrero (6,500-7,000 species) and Veracruz (8,000-8,500 species) (García-Mendoza 2004, González-Espinosa et al. 2005, Krömer et al. 2010, Villaseñor 2016). Many factors contribute to this high-order diversity. One of the most widely accepted is the highly complex topography of Oaxaca’s territory (Lorence & García-Mendoza 1989). The presence of deep valleys, isolated mountain peaks, regions of very high humidity almost adjacent to others where water is rather scarce, offer an extraordinary gamut of ecological scenarios for the evolution of lineages adapted to different climatic settings (see García-Mendoza [2004] for an extensive analysis of the state’s flora and its study). Indeed, the geo-climatic history of mountains of Mexico is deemed responsible for the intense diversification of many plant groups (e.g., Perry 1991, Rzedowski 1991, Nixon 1993, Villaseñor et al. 1998, Valencia-Ávalos & Nixon 2004).
A striking feature of the topography of Oaxaca’s territory is the massive and complex mountain formation located on its northern part. These mountains, collectively known as Northern Oaxaca Range (Sierra Norte de Oaxaca), actually form part of the Sierra Madre del Sur Physiographic Province, an extensive mountain complex that occupies much of southern Mexico (Ferrusquía-Villafranca 1993, Centeno-García 2004). Due to its NW-SE orientation, the Northern Oaxaca Range acts as a barrier that obstructs the movement of the moisture-laden Easterlies blowing from the Gulf of Mexico. Thus, on their leeward slopes most of the water transported inland is discharged, rendering this region the rainiest in the country and warranting its classification as hyper-humid (Shen & Chen 2010). The combination of this very high precipitation with an extremely rough topography, two efficient deterrents of human activities, likely explains why the Northern Oaxaca Range still hosts extensive tracts of very well-preserved forest. Such good conservation status has been more or less compatible with a relatively large population of native people, among which the Chinantec ethnic group is noteworthy, as they inhabit the most inaccessible and hard for human livelihood area of this region (Martin & Madrid 1992).
Expectedly, a region with such characteristics should host a very large biological diversity. However, efforts to inventory its plant component have been few and generally very localized. At the end of the first half of the 20th century the first accounts related to the flora of the Chinantec region were published (Reko 1949, Paray 1951, Matuda 1959). Later, the rise of the pharmaceutical industry centered on the production of contraceptives based on barbasco (Dioscorea composita Hemsl., Dioscoreaceae) attracted to the region research groups that pioneered the tropical ecological research in the country (Tamayo & Beltrán 1977 and papers therein). Such work concentrated on the study of lowland primary and secondary forests (Gómez-Pompa et al. 1964, Sousa-Sánchez 1964, Sarukhán-Kermez 1968). Yet, the flora typical of high-elevation forests remained largely unknown for years, until the construction of Highway 175 that connected Tuxtepec with Oaxaca City, the state’s capital. In the last three decades of the 20th century numerous botanists explored the region; regrettably, their findings were seldom published (e.g., Lorence & Torres-Colín 1988, Martin & Madrid 1992, Meave et al. 1996, Gallardo et al. 1998, Ibarra-Manríquez & Mendoza 2003). Some floristic studies were conducted alongside the study of vegetation (Boyle 1996, Arellanes-Cancino 2000, Romero-Romero et al. 2000). Two particularly important contributions to the knowledge of this region’s flora are the works of Rzedowski & Palacios-Chávez (1977) and Torres-Colín et al. (2009).
In this paper we report an initial floristic checklist of a relatively small but extraordinarily well-preserved portion of the Northern Oaxaca Range. This checklist results from an intense and systematic effort that concentrated mostly on undisturbed forests located in the 800-3,020 m a.s.l. range, although it also includes species collected at lower elevations and some areas of secondary vegetation derived from human activities.
Materials and methods
Study region. The study area is located in northern Oaxaca State, Mexico, and forms part of La Chinantla, a culturally defined region that roughly corresponds with the geographical range of the Chinantec ethnic group (Meave et al. 2006). Our botanical exploration took place mainly in the Perfume River basin, a low-order tributary of the Papaloapan River, near the village of Santa Cruz Tepetotutla, San Felipe Usila municipio (municipio is a second-order territorial unit sometimes translated as municipality but more closely equivalent to a county in other countries). Additional botanical expeditions were conducted around Nueva Santa Flora and Cerro Verde, in the same municipio, as well as in the surroundings of Arroyo Seco and Cerro Mirador (Valle Nacional municipio), and of La Esperanza and Cerro Pelón (Santiago Comaltepec municipio). Extreme coordinates encompassing all these localities are 17º 35’ 10” to 17º 59’ 30” N, and 96º 18’ 30” to 96º 34’ 43” W (INEGI 2017; Figure 1). The area explored during the botanical survey is approximately 265 km2 in size.
Figure 1 Map showing the approximate extent of the area covered by the floristic survey in La Chinantla region, Northern Oaxaca Range (Oaxaca State), southern Mexico. The red dots show the location of the 73 main localities where plant collections were done. Gray lines are 1,000 m contour lines; elevation of the highest regional peak (Cerro Pelón) is also shown. Main tributaries of the Papaloapan river are shown in blue; paved roads are shown in brown.
The uplifting of the Northern Oaxaca Range began around 14 Ma ago (Centeno-García 2004). This mountain range limits to the North with the Gulf of Mexico Coastal Plain, and from there it raises abruptly from almost sea level to ca. 3,200 m; more than 120 mountain tops are located above 2,500 m. The terrain is typically very steep, with around 45 % of the land having slopes between 18º and 45º or more (Ortiz-Pérez et al. 2004), which explains why massive landslides are common. Such steepness is readily perceived along the road connecting Valle Nacional (65 m) with the top of Humo Chico Mountain (3,200 m), two locations only ca. 30 km apart on the horizontal plane.
Lithosol is the prevailing soil type in the region (Alfaro 2004). The few studies on the regional soils emphasize their overall shallowness, but soils are highly heterogeneous regarding fertility, from those with relatively high organic matter contents to very infertile Oxisols (van der Wal 1996, 1998), which appears to be related to the presence of different parent materials. Main pedogenic processes occurring in the region are humus accumulation, in situ weathering, podzolization, and iron reduction due to water stagnation in mineral topsoil (Álvarez-Arteaga et al. 2008).
The climate of La Chinantla is highly variable, mostly owing to the large elevational gradient. Unfortunately, the scarcity of weather stations precludes detailed description of the regional climate. Between May and October the moisture-laden Trade Winds enter the Mexican territory from the Gulf of Mexico, resulting in a well-marked rainy season in the summer months. The adiabatic cooling of the air masses as they are forced to climb the Northern Oaxaca Range produces intense rainfall, particularly at middle elevations. Examples of this humped-shaped variation are the following localities: at Valle Nacional (65 m) annual precipitation is 3,590 mm; further up, at Vista Hermosa (1,450 m) annual rainfall is 5,956 mm, whereas at the high-elevation locality of Humo Chico (3,240 m), annual precipitation is reduced to levels similar to those recorded at the lowlands (3,616 mm) (Meave et al. 2006). Evaporation data are almost non-existent; only for Vista Hermosa there is a record of 1,131 mm, which implies that water availability for plants at this location is very high. During the relatively dry winter, La Chinantla still receives some rain during climatic events known as nortes, characterized by cool, moist winds. Moreover, fog condensation (a phenomenon also known as horizontal precipitation) contributes substantially to water entry into the system (Vogelmann 1973). For these reasons, La Chinantla is by far the rainiest region of Mexico and its climate may be classified as hyper-humid (Rowshan et al. 2007, Shen & Chen 2010). Excess water is drained through multiple creeks and rivers, eventually forming the Papaloapan River that discharges into the Gulf of Mexico (Trejo 2004). Air temperature also shows an extraordinary variation in the region, mostly seen as a sharp decrease with increasing elevation; at the lowest elevations climate is hot (mean annual temperature > 22 °C), but it changes to a semi-hot climate (18-22 °C), temperate (12-18 °C), and a cold one (5-12 °C) towards higher elevations (Meave et al. 2006).
The lush vegetation cover superimposed to this steep climatic gradient and intricate geomorphological mosaic is also highly complex (Rincón-Gutiérrez 2007). The native forests are more or less organized in parallel altitudinal belts, although this pattern is often broken by the presence of deep ravines, exposed ridges and different slope aspects. Broadly speaking, plant communities of La Chinantla may be classified as tropical rain, tropical evergreen and mesophyllous montane forests (Miranda & Hernández-X. 1963, Rzedowski 1978), or as lowland, pre-montane, lower montane and upper montane forests, according to the forest classification developed for the ecologically similar montane forest cline of Costa Rica (Kappelle 1996). Despite a generally very good conservation status of these forests, human activities are leaving an important footprint in the area, particularly below the 1,000 m contour, where much of the land has been transformed into agricultural fields, mostly for the cultivation of maize. At higher elevations coffee plantations are also common, and these are typically shaded by native trees (Bandeira et al. 2005). At these lower elevations, secondary vegetation stands with different times since abandonment intermingle with patches of primary vegetation (van der Wal 1998, Romero-Romero et al. 2000).
Checklist preparation. In assembling the checklist of the vascular flora for La Chinantla, we integrated the findings of various more or less independent projects in which the authors of this contribution participated. The majority of the information derives from a floristic survey conducted in the higher portions of La Chinantla between 1993 and 1997. In total, we made 16 trips that summed 89 days of plant-specimens collecting in this area. Previously, we conducted an initial phase of plant collecting and vegetation sampling that encompassed seven field trips (19 days). Moreover, during one year (1995-1996) 18 stands of secondary vegetation ranging from 5 to 50 years old were sampled (six field trips with 66 days in the field). In total, we defined 93 geo-referenced localities (Figure 1). The total elevational range covered by the entire floristic survey was 250-3,040 m a.s.l., although the strongest effort concentrated above 800 m.
Most specimens were collected in fairly pristine plant communities far away from roads and paths; while this made the botanical survey somewhat inefficient, it allowed us to include multiple species that are typical primary forest components. Being distant from laboratory facilities, we soaked the plant specimens in diluted alcohol to ensure its preservation (Calzada & Perales-Rivera 1990). Plant determination was initially done by the authors, but the reliability of the taxonomic information depended mostly on the judgment of expert taxonomists in different plant groups (Table 1). When the specimen processing and species determination phases were completed, specimens were deposited in several herbaria. The first and only complete set is deposited at MEXU (Instituto de Biología, UNAM, Mexico City), and further albeit incomplete sets were distributed in herbaria located in Mexico (XAL, OAX, CHAPA, IEB, SERO, ENCB, FCME) and abroad (MO, K). The taxonomic arrangement of the checklist follows APG III (Chase & Reveal 2009, The Angiosperm Phylogeny Group 2009, Christenhusz et al. 2011a, b, Reveal & Chase 2011, Stevens 2015). All names included in the checklist were verified in the Tropicos online database (http://www.tropicos.org) and The Plant List website (http://www.theplantlist.org). Information for some specimens deposited at MEXU could be updated through the UNIBIO database (unibio.unam.mx).
Table 1 Names of expert taxonomists that assisted in the identifications of plant specimens collected in La Chinantla region, Northern Oaxaca Range, Mexico. This list excludes some people who have provided/updated species names over two decades for specimens deposited at MO and MEXU.
| Alfonso Delgado Salinas | Helga Ochoterena Booth | Mario Sousa Sánchez ‡ |
| Adolfo Espejo Serna | Isidro Méndez Larios | Nelly Diego Pérez |
| Abisaí García Mendoza | Isela Rodríguez Arévalo | Oswaldo Téllez Valdez |
| Ana Rosa López Ferrari | Isolda Luna Vega | Patricia Dávila Aranda |
| Angélica Ramírez Roa | Jaime Alejandro Torres | Peter Fritsch |
| Alan Reid Smith | Jaime Jiménez Ramírez | Patricia Magaña Rueda |
| Aarón Rodríguez Contreras | José Luis Villaseñor Ríos | Rafael Fernández Nava |
| Charlotte M. Taylor | Jon Ricketson | Rafael Lira Saade |
| Daly | John Thomas Mickel | Ricardo de Santiago Gómez |
| Eduardo A. Pérez García | Lourdes Rico Arce | Rosa María Fonseca Juárez |
| María Esther León Velasco | Luz María González Villareal | Salvador Arias Montes |
| Enrique Ortíz Bermúdez | Lucio Lozada Pérez | Susana Valencia Ávalos |
| Ernesto Velázquez Montes | Marie-Stéphanie Samain | Sergio Zamudio Ruíz |
| Francisco G. Lorea Hernández | Miguel Ángel Soto Arenas ‡ | Thomas F. Daniel |
| Gerardo A. Salazar Chávez | Mónica Elías González | Verónica Juárez Jaimes |
| Gabriel Flores Franco | Mónica Palacios Ríos | Victor W. Steinmann |
| Hilda Belmont | Michael Nee |
‡ Deceased.
Results
We collected 2,653 specimens, each with at least three but generally with more than five duplicates. The checklist includes 1,021 species distributed in 471 genera and 162 families. Although the proportion of species that could be fully determined is high (84.8 %), 26 species (2.5 %) could be only determined to family, and a further 130 (12.7 %) were determined to genus level (all these cases were tallied as morphospecies). These figures correspond to a genus/family ratio of 2.91, a species/genus ratio of 2.03, and a species/family ratio of 6.30. The overall specimens/species ratio (2.6) reflects a satisfactory collecting effort, although for some groups this effort was slightly lower (Lycopodiophyta, 1.82 specimens/species). More than half of all species are Eudicots; behind them, with considerably lower species richness, Monocots, Pteridophyta and Magnoliidae also emerged as important components of the La Chinantla flora (Table 2).
Table 2 Distribution of vascular plant species from the La Chinantla region (Oaxaca, Mexico) among major plant groups.
| Plant group | Families | Genera | Species | Specimens | Specimens/Species |
|---|---|---|---|---|---|
| Lycopodiophyta | 2 | 3 | 11 | 20 | 1.82 |
| Pteridophyta | 20 | 53 | 120 | 246 | 2.05 |
| Gimnospermopsida | 4 | 5 | 9 | 27 | 3.00 |
| Magnoliidae | 8 | 21 | 105 | 304 | 2.90 |
| Eudicots | 108 | 317 | 608 | 1623 | 2.67 |
| Monocots | 20 | 72 | 168 | 433 | 2.58 |
| Total | 162 | 471 | 1021 | 2653 | 2.60 |
Overall, families with the highest genus richness also have the largest species richness (Figure 2). In this regard, the three most remarkable families were Asteraceae (36 genera and 68 species), Rubiaceae (18, 60), and Orchidaceae (28, 59). In strong contrast to these highly diverse families, the checklist includes a considerable proportion of families (51.6 %) with low richness of lower taxa (53 families with a single genus and a single species, 17 families with one genus only and two species, and 14 families with two genera and two species). There were three exceptions to the increasing trend of species with the increasing number of genera, namely Aspleniaceae, Begoniaceae and Piperaceae; in these families the number of species greatly exceeds the number of genera that they should have according to the general trend.
Figure 2 Relationship between genus-level richness and species-level richness in the families included in the checklist of vascular plants for the La Chinantla region. The names of the families with the largest numbers of genera and species are shown (bold typeface indicates those families encompassing more than 10 genera or more than 10 species). The numbers adjacent to the dots specify how many families possess such combination of numbers of genera and species.
The uneven distribution of taxon diversity observed at the family level also characterized the genus level, as a few genera concentrated large numbers of species: Peperomia (29), Miconia and Piper (25 each), Begonia (14), Psychotria and Solanum (13 each), Anthurium, Asplenium and Ocotea (12 each), Lepanthes (11), and Arachnotryx, Chamaedorea and Tillandsia (10 each). In contrast, a total of 305 genera (29.8 %) are represented by one species only.
During the floristic survey we recorded the growth form and growth habit for each specimen. When this information was condensed by species, it became clear that some species could be placed in more than one class of each of these two classifications. Table 3 shows the distribution of species among categories. Regarding growth form, herbs and trees were the most numerous categories, with 484 (47.3 %) and 360 (35.2 %) species, respectively; at the other end cycads and tree ferns, two rare or absent growth forms in many floras, had two and ten species, respectively. Distribution of species in growth habit categories was also very heterogeneous; although terrestrial plants strongly prevailed (874 species, 85.4 %), the number of epiphytic species was also considerable (172 species, 15.9 %).
Table 3 Distribution of species by growth form and growth habit categories. Percent values do not add to 100 % for all growth habit categories because some species were placed in more than one class according to these criteria.
| Category | Species | Percent | |
|---|---|---|---|
| Growth form | |||
| Cycad | 2 | 0.20 | |
| Tree fern | 10 | 0.98 | |
| Palm | 12 | 1.17 | |
| Liana | 47 | 4.59 | |
| Herb & Herb-shrub | 484 | 47.31 | |
| Shrub | 108 | 10.56 | |
| Shrub-tree & Tree | 360 | 35.19 | |
| Growth habit | |||
| Climber | 67 | 5.84 | |
| Epiphyte & Hemi-epiphyte | 172 | 15.00 | |
| Rupicolous | 34 | 2.96 | |
| Terrestrial | 874 | 76.20 | |
The distribution of species along the elevational gradient was also quite heterogeneous. The numbers of species recorded were smaller for the locations corresponding to the lowest (lowland, i.e., < 800 m) and the highest (upper montane, i.e., > 2,300 m) elevations, with 209 and 199 species, respectively. In contrast, the largest number of species (557) was recorded for the pre-montane forest belt (800-1,400 m), which was followed by the lower montane (1,400-2,300 m belt; 383 species). It is necessary to clarify that this distribution does not reflect truthfully different levels of plant diversity at different elevations; although the lower richness recorded at the highest elevations may be relatively real, the small number of species from the lowlands is rather due to a considerably smaller collecting effort there (Figure 3). Despite this caveat, it is clear that pre-montane and lower-montane forests belts concentrate very high diversity levels. In addition, these belts also concentrate the largest numbers of exclusive species (349 and 187 species, respectively). As expected, there was a considerable degree of floristic overlap between these belts, with the pre-montane and lower montane belts having the largest number of shared species (87), a figure closely followed by the intersection of the lowland intersection (73); the lower montane, upper montane intersection included much fewer species (23). Only three species occurred across the four elevational belts, and no single species was shared between the lowland and the upper montane forest only (i.e., absent from the two intermediate belts).
Figure 3 Numeric distribution of the species included in the checklist of the vascular plants for the La Chinantla region in four zones. The size of the colored ovals is representative of the differential species richness recorded at each elevational zone. Figures in brackets indicate the total size of the flora recorded at each zone. The remaining numbers indicate either how many species are exclusive of each elevational zone (figures for the non-overlapping sections of the alveoli), or are shared between them, either in pairs, triads or the single possible tetrad.
Discussion
The floristic survey conducted in La Chinantla and the continued study of the plant collections made there demonstrate beyond any doubt that, with over 1,000 species of vascular plants, this region in northern Oaxaca hosts a very rich and one of the most unique floras in the Mexican territory. Admittedly, however, the floristic survey conducted by us in this topographically complex territory is by no means exhaustive. For example, after many frustrated attempts, we were unable to reach some areas where plant collecting had been originally scheduled based on cartographic analysis (e.g., areas particularly distant from roads and human settlements). Moreover, the collecting effort devoted to the lower elevations (i.e., the lowland forests) was very limited compared to that made at higher elevations. Such bias in this floristic inventory may explain, at least partially, why our observed species/genus (2.03) and species/family (6.35) ratios were lower than the national means (3.5 and 8.2, respectively) reported by Villaseñor (2016).
Even though the total size of the flora thriving in La Chinantla is still unknown, an educated guess of the real richness based on a few back-of-the envelope calculations is possible. Most likely, the less accurate account reported by us is that for the lowland area (209 species). For the relatively close Los Tuxtlas region in Veracruz state, Ibarra-Manríquez & Sinaca-Colín (1995, 1996a, b) reported around 940 species; thus, a conservative estimate of the real richness in the La Chinantla lowlands would be 700 species (rainfall at Los Tuxtlas is higher than at La Chinantla lowlands). Then, by making the also conservative assumption that for each of the higher elevational belts we missed around 10 % of the plant species occurring there, we come up with the figure of 1,954 species (the sum of the so ‘corrected’ figures 700, 613, 422, and 219, to be compared with those in Figure 3). To this number we must subtract the potential floristic overlap (the numbers of observed shared species between belts increased each by 10 %). The result of these calculations is 1,647 species, a figure that is not unrealistic and that would be roughly equivalent to ca. 7 % of the Mexican flora (Villaseñor 2016).
If such speculative exercise is worth anything, this would be to underline the need to make sustained efforts to continue studying the flora of this region in the future, as we found evidence that it contains very important components both for the flora of Oaxaca state and the entire country. For example, our checklist contains 60 species that are absent from the most updated and comprehensive catalogue of vascular plant species produced for the entire Oaxaca state (García-Mendoza & Meave 2011), which resulted from the collaboration of dozens of experts in plant taxonomy and floristics; these species are indicated with an asterisk (*) in Appendix. In addition, our checklist includes 75 species that did not make their way into the recently published checklist of vascular plants of Mexico (Villaseñor 2016), an unprecedented, colossal work based on many years of thorough and critical examination of virtually all floristic information available for the country, including sources of difficult access in gray literature. Also relevant is the fact that our floristic survey demonstrated the worth in terms of endemism, as it hosts 39 taxa endemic to Oaxaca (García-Mendoza & Meave 2011), a figure that will likely increase in the future as our knowledge on this flora improves.
Despite these peculiarities of the La Chinantla floristic array, it is also noteworthy that it shares some attributes with the Mexican flora in general. This is exemplified by the prevalence in the La Chinantla flora of four of the six most speciose families in Mexico, namely Asteraceae, Leguminosae, Orchidaceae, and Rubiaceae (Villaseñor 2016). A detailed comparison with the floristic sets of other regions is a good way to uncover this flora’s singularities; unfortunately, selecting regions to perform such comparison with the results of our study is not easy, given the large differences in the size of the area covered by the surveys and the amplitude of the elevational gradient comprised in the different studies. Bearing these considerations in mind, here we make two comparisons that are particularly revealing. First, we compared the flora of La Chinantla (ca. 459,000 ha; de Teresa 1999) with the Omiltemi region, Guerrero (approximate area of 3,600 ha, 1,800-2,800 m a.s.l.); there, Jiménez-Ramírez et al. (1993) recorded a lower richness (698 species) than that of La Chinantla, but in both regions Orchidaceae was a prevalent family (103 species in Omiltemi; Salazar-Chávez 1993), in addition to Asteraceae and Leguminosae. A second comparison was made with Cerro Quetzal, located in the El Triunfo Biosphere Reserve, Chiapas (11,988 ha; 1,200-2,300 m a.s.l.); this region and La Chinantla host similar plant communities, typical of very humid areas. For Cerro Quetzal, Pérez-Farrera et al. (2012) also reported a smaller richness of vascular plant species (795) than at La Chinantla; interestingly, however, this region bears a higher similarity with La Chinantla than with Omiltemi regarding not only the most speciose families (Asteraceae, Leguminosae, Orchidaceae, Rubiaceae, Dryopteridaceae, Melastomaceae, Piperaceae and Polypodiaceae), but also the most speciose genera: Asplenium (Aspleniaceae), Begonia (Begoniaceae), Peperomia, Piper (both Piperaceae) and Psychotria (Rubiaceae). The prevalence of these groups may be an emergent pattern of Mesoamerican floras typical of hyper-humid regions.
In preparing the plant checklist for La Chinantla, we faced large difficulties for obtaining reliable determinations for our specimens. The ultimate implication was that this project required a much larger investment of time than anticipated. These obstacles were mostly due to the still incipient knowledge of the flora of this hyper-humid region, as it encompasses numerous taxa with imprecise systematic delimitation. Thorough monographic works for these taxa are lacking or are being currently prepared at best. Some species-rich families and genera of this flora, such as Asteraceae, Myrtaceae, Melastomataceae, and Begonia, Peperomia and Piper, provide examples of this situation. However, not only speciose families and genera posed difficulties; actually, the numerous families (53) represented by a single genus each (often by a single species) in the region represented another problematic group. As a rule, we were unable to locate expert taxonomists specialized in these families, perhaps because few botanists are attracted to them due to their low diversity. Therefore, determination of these specimens was very time consuming. Once the familiar adscription of such taxa gets solved, it is usually easier to find out the generic and specific identities. Given the considerable contribution of these species to the total flora, their inclusion in regional floras or field guides should be of high priority. In fact, this situation is not unique to the La Chinantla flora, as the Mexican flora comprises 44 families that are each represented by a single species (Villaseñor 2016). Examples of these taxa occur in all major groups like ferns (Plagiogyraceae), gymnosperms (Taxaceae), and angiosperms (Altingiaceae, Chloranthaceae or Winteraceae). Interestingly, many such families are typical representatives of forests occurring in very humid mountains (Rzedowski 1996).
For all these reasons, many years went by before we achieved an acceptable level of taxonomic resolution for our plant collection. Even so, we were unable to provide full determinations for many taxa and these remained catalogued as morphospecies. For some of them it was possible to assign their generic adscription but for others we went as far as family level only (for a few specimens not even the family could be recognized, and these were excluded from the checklist). Actually, we suspect that many —if not most— of these morphospecies will turn out to be new taxa for science. There are several reasons pointing to this possibility. Among Lauraceae, Ocotea sauroderma Lorea-Hern. and Licaria chinanteca Lorea-Hern. were described based on material collected during this project (Lorea-Hernández 1999, 2001); they represent two cases of species recognized as new taxa as soon as they were examined by an expert taxonomist in the family. Similarly, continuous visits to the internet-hosted databases of the Missouri Botanical Garden (http://www.tropicos.org) and the National Herbarium of Mexico (unibio.unam.mx) resulted in a ceaseless stream of species determinations thanks to the work of numerous taxonomists visiting these herbaria over the last two decades. In some cases, this work revealed that other specimens also represent undescribed species (e.g., Anthurium sp. nov., Araceae, A. Rincón 373). Curiously, the expert taxonomist of this family, Thomas B. Croat, determined the new species status for this collection based on a duplicate specimen deposited at XAL, while such recognition was not available for the specimens deposited in the larger —and more frequently visited— herbaria MEXU and MO. These cases confirm Villaseñor’s (2015) tenet that La Chinantla should be regarded as an area with high speciation rates (based on the analysis of the species/families and species/genera ratios); on this ground author highlighted this as a high priority region for the search of new taxa for science. In this regard, plant groups such as Anthurium, Arachnotryx, Asplenium, Begonia, Chamaedorea, Lepanthes, Miconia, Ocotea, Peperomia, Piper, Psychotria, Solanum, and Tillandsia are very promising and deserve particular attention.
An additional consideration about the group of morphospecies is that they do not represent just one or few plant types, as they included plants representing a range of growth forms and habits (Appendix); by growth form variously-sized trees are noteworthy (49 species), as well as many terrestrial and epiphytic herbs (32 and 22 species, respectively), and shrubs (23). The group of morphospecies also comprised a broad range of families (58), for some of which expert taxonomists are available (e.g., Bromeliaceae, Cyatheaceae, Fagaceae, Selaginellaceae and Solanaceae). This situation pinpoints once more the necessity to carefully examine these specimens by the specialists. However, it is also likely that at least in some cases they could not be properly determined due to the lack of diagnostic structures, as they did not always bear flowers or fruits. Alternatively, they may represent highly variable groups for which more collections are needed in order to critically assess their morphological differentiation from accepted known species. Whatever the case may be, these problems will only be solved with better and longer botanical surveys in La Chinantla; to this end, one plausible strategy is the training of professional parataxonomists living in the region, as this would very likely increase the efficacy of collecting plant specimens.
Another interesting attribute of our checklist is that many species included in it represent groups (of any taxonomic level) that are poorly represented at herbaria. In fact, our specimens still are the only representatives of several species either at MO or MEXU. This situation may be reflecting a high degree of endemism, but it result from the widespread strategy that characterizes most botanical surveys that consists in conducting plant surveys in readily accessible localities, or in zones with adequate infrastructure, such as some formally decreed natural protected areas. If we are ever going to get a more complete knowledge of the Mexican flora, this approach must be revised.
Regardless of these possibilities, it appears that rarity is a common phenomenon among the plant species of La Chinantla. Rabinowitz (1981) distinguished three components of rarity (population density, geographical range and ecological breadth). Plant species may be rare according to low values of any one of these three criteria, or to combinations of them (only high values for the three criteria do not result in rarity). Although we lack precise information to formally analyze the prevalence of rarity in the La Chinantla flora, there are indications (number of specimens in the herbaria, number of specimens collected by us, elevational spread of the species) to substantiate the hypothesis that a considerable proportion of these species possess some degree of rarity. The Red List of Mexican Cloud Forest Trees (González-Espinosa et al. 2011), allowed us to evaluate this attribute for some species included in the present study, because IUCN Red List categories are indicative of the degree to which their conservation is threatened. There are 123 species in the following categories: 10 Critically Endangered, 42 Endangered, 45 Vulnerable and 26 Near Threatened. In contrast, only 70 were assigned to the Least Concern category. Further evidence is provided by the study of the genus Peperomia in Veracruz state (Vergara-Rodríguez et al. 2017); there are 13 species shared between this state and La Chinantla, 11 of which were listed as Least Concern, whereas P. petrophila C.DC. was classified as Vulnerable and P. peltilimba C.DC. ex Trel. as Near Threatened. A final consideration on these two latter species and for five of those listed as LC is that they are epiphytes, which actually increases their conservation risk due to habitat loss, as they fully depend on their host plants for their establishment.
Information on species’ elevational ranges is useful in making an initial assessment of their ecological breadth. By dividing the area covered by our floristic survey in four elevational belts corresponding to distinctly different forest types (lowland, pre-montane, lower montane and upper montane), we could assess the proportions of species exclusive of each belt vs. species with broad elevational distribution (Figure 3). Only three species, namely one fern (Polypodium plebeium), and two woody vascular plants (Dendropanax arboreus and Hoffmania nicotianifolia) were collected in all four elevational zones. This is not surprising, as strong floristic gradients have been reported along the slopes of the mountains of southern and western Mexico (Vázquez-G. & Givnish 1998, Salas-Morales & Meave 2012). In fact, the reality may be even more complex; for example, Dendropanax arboreus exhibits strong morphological differentiation along the elevational gradient, growing as a tall, rather sturdy tree in the lowlands, and as a slim, short-statured shrub in the upper reaches, casting doubts on the recognition of these contrasting morphologies as a single species. More importantly, however, is the fact that for the total set of species collected at each elevational zone, the numbers of exclusive species are always larger (sometimes much more so) than the numbers of shared species between adjacent zones, whereas the numbers of species shared by three zones (2-23) are very low. Species common to non-adjacent elevational zones are even fewer. In addition, detailed analyses conducted for two plant groups (Quercus, Meave et al. 2006; Lauraceae, Rincón-Gutiérrez 2007) revealed narrow elevational ranges for all examined species. It would be worth to analyze with greater detail the specific shape of the floristic turnover along the entire gradient in the future, in order to identify those elevations at which the largest floristic changes take place.
An important shortcoming of this study is the uneven collecting effort along the elevation gradient. Despite this limitation, the reduction in floristic richness with increasing elevation is clear, particularly in those areas located above 2,300 m (upper montane), in strong contrast with the richness recorded for the pre-montane forest belt (800-1,400 m; 557 species), as well as in the lower montane forest (1,400-2,300 m belt; 383 species). The insufficient floristic knowledge for lower elevations prevents us from drawing more definite conclusions about the general floristic pattern at La Chinantla. Currently, two main patterns have been observed, namely a hump-shaped pattern with the largest richness at intermediate elevations, and a continuous decreasing pattern (Bhattarai & Vetaas 2003, Gould et al. 2006, Grytnes & Beaman 2006, Kluge & Kessler 2011, Salas-Morales & Meave 2012). At present, it is uncertain which of them will better match the floristic gradient at La Chinantla.
Our study confirms the large contribution of the Northern Oaxaca range to Mexico’s biological heritage. Except for those plant communities proper of lower elevations (< 1,000 m), many of which have been cleared to give way to productive systems, the various plant communities occurring in the region still have a good conservation status. This is noteworthy not only because of the area they occupy, but also because of their peculiar floristic composition and physiognomy. For example, the presence of very rare taxa such as Cyrilla, Juglans, Oreomunnea, Podocarpus, and Ticodendron (Martin & Madrid 1992, Meave et al. 1996, Gallardo et al. 1998), as well as the notoriously large richness of Lauraceae species, suggests that this is a very unique flora due to these antique and rare lineages (Rzedowski & Palacios-Chávez 1977), uncommon elsewhere in Mexico. Moreover, some plant communities of the region possess particular physiognomies, often characterized by very tall trees with large trunks, by the twisted appearance of trees in some upper montane forests, and frequently by the abundance of epiphytic plants, even at low heights on tree trunks, including numerous species of bromeliads, ferns, orchids and Peperomia. Last, but no least, the considerable area and continuity of these forests results in eye-catching landscapes rarely seen in other regions of the country.
Notwithstanding this comparatively favorable situation, and like many other forested regions of the planet, particularly in tropical mountains, the vulnerability of La Chinantla’s forest ecosystems needs recognition, as shown by relatively small actions that have strongly impacted the vegetation. One such action was the construction of the road connecting the isolated village of Santa Cruz Tepetotutla, located in the core of our study area. After several failed attempts to build this road through different routes, the final stroke obliterated a beautiful and well-preserved piece of elfin forest where many of our collections came from. Fortunately, the situation does not seem to be so serious and in fact it is promising, as in this region several community-based conservation initiatives and programs of payment for ecosystem services exist (Martin et al. 2011, Bray et al. 2012, Velasco-Murguía et al. 2014). Yet, the diversification of sources of income through activities that better match conservation goals, such as eco-tourism, fish production in ponds, or even biodiversity-friendly coffee plantations (Bandeira et al. 2005), would be desirable. These efforts require the largest possible support from societies in Mexico and abroad in order to guarantee the long-term protection of these unique forests and the biota they host. We hope that the floristic information produced in this study will contribute to the fulfillment of this important objective.
