Structure and woody species diversity of the Dasylirion cedrosanum (Nolinaceae) rosette scrub of central and southern Coahuila State, Mexico

Estructura y diversidad de especies leñosas del matorral rosetófilo de Dasylirion cedrosanum (Nolinaceae) del centro y sur del estado de Coahuila, México

Juan Antonio Encina-Domínguez^1,3, Jorge A. Meave² and Alejandro Zárate-Lupercio¹

¹ Departamento Forestal, Laboratorio de Sistemas de Información Geográfica. Universidad Autónoma Agraria Antonio Narro. Saltillo, Coahuila, México.

³ Author for correspondence: juanencina@gmail.com.

Received: August 24th, 2012
Accepted: January 2nd, 2013

Abstract

The most prominent vegetation type in the state of Coahuila, in northern Mexico, is Chihuahuan Desert Scrub. This plant formation encompasses the little known Dasylirion cedrosanum (sotol) rosette scrub, a community that extends over more than one fourth of Coahuila and whose most distinctive species is subjected to intense extraction from its native communities for a variety of purposes. Based on a highly replicated sampling procedure that included 131 plots located in the state's central and southern portions, we analyzed vegetation structure, floristics, and species diversity of this plant community. A cluster analysis allowed us to differentiate five variants (associations) of the Dasylirion cedrosanum rosette scrub, which together host a richness of 97 species of vascular plants distributed in 61 genera and 28 families. These communities occurred across an elevational range of 850-2,550 m a.s.l. Dasylirion cedrosanum density varied between 193 and 705 ind./ha, with the highest value occurring in the Quercus intricata-Dasylirion cedrosanum association. Other prominent species in these scrublands were Agave lecheguilla and Euphorbia antisyphilitica. The Agave lecheguilla-Dasylirion cedrosanum association was the most broadly distributed and it hosted the highest richness species, while the largest Shannon diversity index value was recorded for the Dasylirion cedrosanum-Viguiera greggii association. The effects of human utilization of this species on the structure of those plant communities where it occurs are largely unknown; further studies are needed to better understand its dynamics and to be able to suggest more adequate regulations for its commercial harvesting.

Key words: Chihuahuan Desert, cluster analysis, elevation gradient, natural resources utilization, sotol production.

La vegetación más prominente en el estado de Coahuila, en el norte de México, es el matorral desértico chihuahuense. Esta formación vegetal incluye el matorral rosetófilo de Dasylirion cedrosanum, comunidad poco conocida que cubre más de la cuarta parte del estado de Coahuila y cuya especie característica es sometida a una explotación intensa en sus comunidades nativas para distintos fines. Con base en un muestreo fuertemente replicado, que incluyó 131 unidades de muestreo ubicadas en el centro y sur del estado, se evaluó la estructura de la vegetación, su composición florística y la diversidad de especies de esta comunidad vegetal. Un análisis de clasificación numérica permitió diferenciar cinco variantes (asociaciones) del matorral rosetófilo de Dasylirion cedrosanum, que en conjunto albergan una riqueza de 97 especies agrupadas en 61 géneros y 28 familias. Estos matorrales se distribuyen en un intervalo altitudinal de 850-2,530 m s.n.m. La densidad de Dasylirion cedrosanum varió de 193 a 705 ind./ha; el valor más alto fue registrado en la asociación Quercus intricata-Dasylirion cedrosanum. Otras especies comunes en estos matorrales fueron Agave lecheguilla y Euphorbia antisyphilitica. La asociación Agave lecheguilla-Dasylirion cedrosanum tuvo la distribución espacial más amplia y la mayor riqueza de especies, mientras que el mayor valor del índice de diversidad de Shannon se obtuvo para la asociación Dasylirion cedrosanum-Viguiera greggii. Hay un gran desconocimiento en torno a los efectos del uso de esta especie por el ser humano sobre la estructura de las comunidades en las que ésta se presenta; por lo tanto, es necesario realizar nuevos estudios que permitan conocer su dinámica y establecer la normatividad más adecuada para su aprovechamiento comercial.

Palabras clave: análisis de clasificación, aprovechamiento de recursos vegetales; Desierto Chihuahuense; gradiente altitudinal; producción de sotol.

The most prominent vegetation type in the state of Coahuila, in northern Mexico, is Chihuahuan Desert Scrub. This plant formation is mostly represented by microphyllous scrub and rosette scrub (Henrickson and Johnston, 1986; Villarreal and Valdés, 1992-1993). Rosette scrub often referred to also as rosette-scrub forest, is a community dominated by evergreen, short and spiny shrubs, with rosette-forming leaves concentrated at the plant's base (Rzedowski, 1966; Box, 1981; Martorell and Ezcurra, 2002). in the Chihuahuan Desert, the rosette-dominated plant communities variously known as Lechuguilla scrub, Yucca woodland, Dasylirion scrub, or Dasylirion woodland, account together for approximately 12% (60,840 km²) of the total area of this biogeographical region (Henrickson and Johnston, 1986), while in Coahuila, rosette scrub extends over an area of 40,579 km², which represents more than one fourth (26.8%) of the state territory (ICE, 2001). While at first glance, its floristic composition and structure do not appear to vary greatly across space, in fact four variants have been described for this community (Henrickson and Johnston, 1986; González-Elizondo et al., 2007). One of them is the Dasylirion scrub, regarded as a physiognomically distinct plant community typical of the Chihuahuan Desert and of the entire North American Arid Region (Martorell and Ezcurra, 2002). This community is generally found in mountainous areas, at elevations between 1,000 and 2,000 m a.s.l., on lithosols and on areas with abundant limestone outcrops (Henrickson and Johnston, 1986).

The genus Dasylirion comprises 16 species, all of them distributed in mountainous arid and semi-arid regions of northern Mexico and southwestern USA (Bogler, 1995, 1998), with four of them occurring in Coahuila (Villarreal, 2001). Dasylirion cedrosanum Trel. is the most abundant species of this genus, and it is the physiognomically dominant component of the rosette scrub that occurs in south-central Coahui-la and adjacent areas of the states of Durango and zacatecas, Mexico (Henrickson and Johnston, 1997; Villarreal, 2001). This taxon represents one of the most important non-timber forest products in the state rural areas, as it is used for human and cattle nourishment, as well as a source of construction materials, handcraft making, and alcoholic distills (Marro-quín et al., 1981; Romahn de la Vega, 1992, Olhagaray et al., 2004). its relevance has increased recently as it is now being harvested for the commercial production of sotol, an alcoholic beverage obtained from the stems (Marroquín et al., 1981; Sierra et al., 2008). Regrettably, this activity lacks adequate management plans and there are no commercial plantations of this species; instead, the plants are extracted directly from their native ecosystems (Villavicencio et al., 2007). Therefore, its populations are likely to undergo severe reductions in the near future. Such population decline would probably also affect other important accompanying species, since Coahuila's rosette scrub hosts the largest number of endemic plant taxa and species included in the Mexican Official List of Endangered Species for this State (Villarreal-Quintanilla and Encina-Domínguez, 2005; SEMARNAT, 2010).

The goals of this study were to describe vegetation structure and floristics of the Dasylirion cedrosanum scrub, and to provide ecological information for this community in Coahuila, including the associated shrub species, in order to contribute to the existing knowledge about this species and to provide a stronger basis for its management.

Materials and methods

Study area. The study was conducted in the central and southern portions of Coahuila state, over an area of approximately 20,779 km² (ICE, 2001) (Figure 1), i.e. 4.1% of the 507,000 km² corresponding to the Chihuahuan Desert Region (CDR) (MacMahon and Wagner, 1985). The study area includes several smaller sierras (mountain ranges) that form part of the Eastern Sierra Madre (Sierra Madre Oriental) and the Mexican Plateau (Altiplanicie Mexicana) physiographic provinces: Arteaga, Los Alamitos, La Concordia, Las Delicias, La Fragua, La Gavia, Jimulco, La Madera, Menchaca, La Paila, Parras, San Marcos, zapalinamé. Elevation ranges from 900 m in central Coahuila to 2,500 m in the mountains of the state southeastern corner. The prevailing climates are of two types (García, 2004): very dry climates (BWhw) typical of the western portion of the study area, which is part of the Mexican Plateau, and semi-warm dry climates (BS₀hw), characteristic of the Eastern Sierra Madre, in the central and southeastern parts of the study area. Annual temperature oscillation is high; mean annual temperature ranges from 17 to 19 °C, with maxima of 25 to 32 °C from May through October, and minima of 1 to 7 °C from November to April; on average there are 12 days with frost each year. Total annual precipitation ranges between 350 and 450 mm, with rainfall concentrating in summer months. Lower Cretaceous sedimentary rocks, mostly limestone and conglomerates, characterize parental material in the area. Soils are mostly lithosols, calcareous regosols, and haplic xerosols, and they are shallow with low organic matter content, loamy texture, and good drainage. Villarreal and Valdés (1992-1993) classified regional vegetation as rosette desert scrub (matorral desértico rosetófilo), and highlighted Agave lecheguilla, Dasylirion cedrosanum, Euphorbia antisyphilitica, and Hechtia texensis as the most frequent plant species. At some localities in the SE part of Coahuila transitional communities exist among this vegetation type, sub-montane scrub and pinyon pine forest, where the most representative species are Quercus intricata and Pinus cembroides.

We constructed a matrix containing density values of 90 species occurring in all sampling sites, with which we performed a cluster analysis using Ward's method (Minimum Variance), an agglomerative, polythetic, and hierarchical classification procedure (Ward, 1963; Manly, 1986). The classification was performed with NTSYSpc ver. 2.0 (Rohlf, 1998). Between plot similarities were calculated with the Morisita Index. For each group of plots interpreted as a plant association, we calculated density, frequency, and crown cover for each species; relative values of these variables were summed in order to calculate Relative Importance Values (RIV; Mueller-Dombois and Ellenberg, 1974), according to the following formula:

We chose this conventional way of estimating the structural contributions of the species occurring in the community as it integrates in a single calculation the three basic attributes of vegetation structure. We also calculated log2 Shannon-Wiener's (H') diversity index (Magurran, 1988), and evenness (E) according to Pielou (1966); pairwise comparisons of diversity indices between associations was performed through Hutcheson (1970) modified t-test as described by Magurran (1988). Finally, we examined Dasylirion cedrosanum distribution along the altitudinal gradient by performing a simple linear regression (zar, 1999) of this species densities on sampling site elevation.

Results

Floristic composition. In total, 97 species belonging to 61 genera, and 28 families were recorded (Appendix). The richest families were Cactaceae (30 species), Fabaceae and Asteraceae (12 each), and Agavaceae (9). In turn, the richest genera were Opuntia (7 species), Acacia (5), Agave (4), and Viguiera, Leucophyllum, Quercus, and Yucca (3 species each). Eight species were endemic to Coahuila and adjacent regions in neighboring states, and seven have official protection status according to the Mexican environmental legislation (SEMARNAT, 2010), with three species being part of both groups (Table 1). The large majority of species recorded (91, 93%) were shrubs, while the few remaining taxa were trees. In addition to Dasylirion cedrosanum, the most abundant species were Agave lecheguilla, Euphorbia antisyphilitica, Hechtia texensis, and Viguiera greggii. Some dominant species in neighboring plant communities were also common here, such as Larrea tridentata, Lindleya mespiloides, Pinus cembroides, Quercus intricata, and Rhus virens.

Site classification. The classification of the sampling sites allowed us to differentiate five vegetation variants at the cutoff level of 55%, which were interpreted as plant associations of the Dasylirion cedrosanum rosette scrub (Figure 2). In general, these associations occurred along an elevational gradient that increased from top to bottom in the dendrogram, from 850 to 2,530 m a.s.l. The first three associations comprised 105 sites (80% of the total); their mean elevation was lower than 1,600 m a.s.l., and they were characterized by the dominance of low, rosette-like shrubs, in addition to some succulent species. Conversely, the remaining two associations occurred above the 1,600 m contour and were dominated mostly by microphyllous and sclerophyllous shrubs and trees.

Vegetation structure in the associations of the Dasylirion cedrosanum rosette scrub. (1) Hechtia texensis-Agave lecheguilla association. This association occurred in the most arid section (central-west) of the study area. Most sites concentrated on the lower slopes (mean 1,300 m a.s.l.; range 856-1,581 m a.s.l.) with southern aspect of the La Madera and El Fuste sierras. Vegetation consisted in desert scrub with an open canopy up to 0.5 m in height, and was dominated by H. texensis and A. lecheguilla; along with Euphorbia antisyphilitica, these species accounted for 55% of plant cover and 57% of total RIV. Dasylirion cedrosanum ranked fourth in RIV (11.52%), and its mean density and height (± 1 SD) in this association were 284 ± 6.24 ind./ha and 1.39 ± 0.23 m, respectively (Table 2). Individuals of this species were scattered in the vegetation and often they were the only prominent plants in the vegetation physiognomy. Because of the topographically low position of this association, some very common species of the microphyllous scrub typical of valley bottoms and flatter sites were abundant in it, such as Grusonia bradtiana and Larrea tridentata.

(2) Agave lecheguilla-Dasylirion cedrosanum association. This was the most widespread of all associations, as it occurred across much of the study area. However, it was particularly common in mid-slopes (mean 1,500 m a.s.l.; range 905 to 2,381 m a.s.l.) with northern aspect of the sierras of the central-eastern part of the study area (La Paila and La Gavia), and in those located in the southeast and southwest parts (zapalinamé, Parras, and Jimulco). Vegetation of this association was dominated by A. lecheguilla, a colony-forming species that accounted for as much as 30% of total plant cover and of total RIV (Table 2). The co-dominant species was D. cedrosanum, with a RIV of 13.4% and a mean density (± 1 SD) of 492 ± 16.9 ind./ha; mean vegetation height (± 1 SD) in this association was 1.37 ± 0.16 m. Additional common species were Fouquieria splendens and Euphorbia antisyphilitica (RIVs of 5.8 and 5.1%, respectively). Although less abundant, barrel cacti were also present, represented mostly by Ferocactus pilosus and Echinocactus platyacanthus which, along with D. cedrosanum and F. splendens, were the physiognomic dominants of this association.

(4) Dasylirion cedrosanum-Viguiera greggii association. This association occurs on mid- and high slopes with a northern aspect in the sierras of southern Coahuila (zapalinamé, Parras and Jimulco), with a mean elevation of 1,919 m a.s.l. (range: 1,375 to 2,527 m a.s.l.). Dasylirion cedrosanum was the species with the largest structural contribution, accounting for almost one third of total plant cover (Table 2); RIV of this species was 16.0%, and its mean density and height (± 1 SD) were 704 ± 17.92 ind./ha and 1.25 ± 0.28 m, respectively. Co-dominant species were V. greggii, Agave scabra, and Lindleya mespiloides, with RIVs between 6 and 8%. As a result of the elevational distribution of this association, often turns into an ecotone with the sub-montane scrub typical of higher elevations, which results in a considerable number of associated woody species; in fact, some low-statured individuals of Juniperus, Pinus, and Quercus were observed, whilst the number of succulent plants, in particular of cacti, was low relative to the former three associations.

(5) Quercus intricata-Dasylirion cedrosanum association. This association thrives on high slopes with northern aspect in the mountains located in southern Coahuila (Arteaga and Concordia sierras), with a mean elevation of 2,083 m a.s.l. (range: 2,009 to 2,130 m a.s.l.). The rosette scrub representing this association was mostly composed of colonies or patches of short individuals of Quercus intricata (1.5 m in height), a shrub species with RIV of 29.0% (Table 2). The single co-dominant in this association was D. cedrosanum (RIV of 10.77 %), which here had a mean density (± 1 SD) of 705 ± 30.36 ind./ha, and a mean height of 1.70 ± 0.26 m. Other species making significant contributions to vegetation structure were Leucophyllum langmaniae and Pinus cembroides, with RIV between 8 and 10%. This is the most mesic association among all five floristic variants and represents the upper altitudinal limit of the D. cedrosanum rosette scrub. At higher elevations oak and pine forests occur and, in fact, the floristic composition and the physiognomy of this association are more suggestive of a less humid variant of those forests than of a typical Dasylirion community.

Diversity indices. The largest values of H' and E were obtained for the Dasylirion cedrosanum-Viguiera greggii association (4.396 bits and 76.0%, respectively; Table 3); at the opposite end, values calculated for the Hechtia texensis-Agave lecheguilla association (2.472 bits and 45.0%) reflected a much less diverse community. In analyzing total species richness by association, we recorded the largest number of species (70) in the A. lecheguilla-D. cedrosanum association. However, this figure conceals important differences in the concentration of species on an area basis: the D. cedrosanum-V. greggii association clearly had the highest species density (3.05 species 500 m^-2 on average), followed distantly by the Quercus intricata-D. cedrosanum association, in which mean density was 2.12 species 500 m^-2 (Table 3). According to the t-tests performed to statistically compare diversity values between pairs of associations, most comparisons did not yield significant differences, except for the comparison between associations 1 and 4 (t - 3.62, d.f. - 11,466, P < 0.001), and between associations 2 and 4 (t -3.18, d.f. - 10,282, P < 0.01).

Density and distribution of Dasylirion cedrosanum. The elevational range displayed by D. cedrosanum in Coahuila is very broad (850-2,530 m a.s.l.). Along this gradient its density increases linearly at an approximate rate of 68 ind./ha for every 100 m increase in elevation (Figure 3). Mean density at elevations < 1,000 m was always < 200 ind./ha, whereas at higher elevations (2,000-2,500 m a.s.l.) densities ranged from 900 to 1,200 ind./ha. Dasylirion cedrosanum reached its highest abundance in the mountains located in southern Coahuila, on the Eastern Sierra Madre, whereas the lowest densities were recorded in the sierras of Central Coahuila, in the Mexican Plateau Physiographic Province.

Discussion

Floristic composition and diversity. The woody flora associated to the Dasylirion cedrosanum scrub comprises 97 species. This floristic richness is high considering that this is the least frequent variant of the rosette scrub (Henrickson and Johnston, 1986); furthermore, this figure is similar to the 90 shrub species reported for analogous xerophytic scrubs in the El Huizache area, state of San Luis Potosí, also in the Chihuahuan Desert (Huerta-Martínez and García-Moya, 2004).

Examples of species occurring in this area that are highly relevant for conservation are Astrophytum capricorne, The-locactus rinconensis var. nidulans, and Glandulicactus uncinatus. All of them are at risk of extinction (SEMARNAT, 2010) and the former two are endemic to Coahuila and adjacent areas (Villarreal-Quintanilla and Encina-Domínguez, 2005). In addition, the conservation of Ariocarpus re-tusus and Lophophora williamsii is a reason for concern, as these two species are subjected to an intense extraction from its natural milieu (Robbins, 2003). This information is important because Godínez-Álvarez and Ortega-Baes (2007) reported that the state of Coahuila has the highest species richness (> 125 species) and hosts more than 50 endangered species of cacti. Agave victoriae-reginae, a species that was also present in the studied community, perhaps holds the highest conservation priority, as it is classified as endangered, in addition to being endemic to the state and surrounding areas, and whose largest populations occur in our study area (Martínez-Palacios et al., 1999). Diversity indices are high according to Margalef (1991), with Shannon-Wiener index ranging from 1.5 to 3.5 bits, with exceptional values of 4.5 bits. These values are consistent with the notion that plant diversity in Mexico's arid regions is high in hilly areas, where the prevalence of shallow, stony soils is associated to increased water availability, ultimately resulting in higher species richness (Rzedowski, 1978). Moreover, in such areas micro-relief is more heterogeneous, which creates a higher diversity of habitats (Pérez et al., 2008). Shannon-Wiener index had a sigmoid pattern along the altitudinal gradient where Dasylirion cedrosanum thrives; it peaked at a mean elevation of 1,920 m, at the transition from rosette scrub to more temperate plant communities like pinyon-pine forest, which results in an increased diversity caused by the concurrence of these two vegetation types. Marroquín et al. (1981) also pointed to this elevational transition as the cause of a high floristic diversity in Coahuila rosette scrub. Interestingly, the largest values of Shannon-Wiener index were obtained for the Dasylirion cedrosanum-dominated associations, which may be due to the fact that the most important species in the remaining associations (those dominated by Hechtia texensis, Agave lecheguilla, and Quercus intricata) have large biomass accumulations and a more intense occupation of space derived from the vegetative reproduction that characterizes them (Marroquín et al., 1981; Reyes-Agüero et al., 2000). Unlike communities dominated by sotol, which reproduces mostly through seeds (Bogler, 1995), the dominance of clonal plants results in lower diversity and evenness in their associations (Margalef, 1991). Dasylirion cedrosanum-dominated stands are not only more diverse, but they also host more cacti species, some of which have a distinct conservation status.

Distribution and structure of the Dasylirion cedrosanum rosette scrub. The cluster analysis showed that D. cedrosanum does not form highly discrete units within its range, but rather that this species is variously mixed with those more typical of other rosette scrub associations, such as the sub-montane scrub and the pinyon pine forest, and with those from the microphyllous scrub, with which it alternates its dominance depending on elevation and water availability. Similar patterns were reported for the Catorce Range of San Luis Potosí (Granados-Sánchez and Sánchez-González, 2003). The distribution patterns observed in our area both for the associations and their most important species do not differ from those described elsewhere for several mountainous regions of Mexico deserts (Pavón et al., 2000; Martorell and Ezcurra, 2002). These patterns are closely related to elevation, a physical variable that serves as a surrogate of humidity and temperature (Daubenmire, 1982; Reynolds et al., 2000), and whose effect is most evident on the vegetation of arid and semi-arid zones (Montaña and Valiente-Banuet, 1998; Huerta-Martínez et al., 2004). The high densities of Agave lecheguilla, Euphorbia antisyphilitica, and Hechtia texensis, as well as of cacti and other xerophytic floristic elements of associations dominated by these three species, have been attributed to the scarce rainfall and higher temperatures that prevail at lower elevations (Marroquín et al., 1981; Pavón et al., 2000; Reyes-Agüero et al., 2000), whereas the groups containing more woody species, both shrubs and trees, at the highest reaches of the gradient concur with the prevalence of more humid and temperate conditions at those sites (Granados-Sánchez and Sánchez-González, 2003). However, Dasylirion cedrosanum-dominated associations occur at intermediate elevations, according to the findings of Martorell and Ezcurra (2007), who reported that species of Dasylirion, Nolina, and Yucca are abundant at mid-elevations of mountains within arid and semi-arid regions of Mexico where fog is a recurring phenomenon. Their abundance could be linked to the morphology of their leaves and their heights, both of which provide plants access to an increased water supply through fog interception. Though fog incidence has not been investigated in our study region, this phenomenon is frequent in the rainy season and occasional in the winter, with the incursion of cold fronts (pers. obs.), suggesting that this factor may have a significant influence on Dasylirion cedrosanum distribution and abundance.

Vegetation structure of the studied communities is similar to that described for the rosette scrub (sensu lato) regarding the most important species (Rzedowski, 1978; Marroquín et al., 1981; Henrickson and Johnston, 1986; Villarreal and Valdés, 1992-1993; González-Elizondo et al., 2007). A structural feature that we observed, which is common within the vegetation typical of many arid regions, particularly in mountainous areas, is that a few species with clonal growth make the largest contribution to plant cover (Rzedowski, 1978; Pérez-García and Meave, 2004). This was the case of Agave lecheguilla, Hechtia texensis, and Quercus intricata. Although Dasylirion cedrosanum is the most frequent species in all associations, its low density results in its low RIV ranking relative to clonal species. Therefore, localities where D. cedrosanum had the highest RIV were uncommon: only two associations, accounting for almost one third (29%) of the sampled surface, were both physiognomically and structurally dominated by this species. This result supports Henrickson and Johnston conclusions (1986), who reported that sotolares (Dasylirion-dominated scrub communities) are rare in the Chihuahuan Desert. This result also suggests that our study species is a poor competitor and that it could face regeneration problems should its populations continue to be impacted by human activity, considering its dependence on sexual reproduction, low growth rates, and long life cycle (Palma-Estrada, 2000; Sierra et al., 2008). This information, together with the structure of its populations across its geographical range, should be given thoughtful consideration in preparing management plans for this species.

Dasylirion cedrosanum density and ecological relations. In Coahuila D. cedrosanum occurs on mountains and hilly terrain, at elevations between 850 and 2,600 m a.s.l., on stony, shallow, calcareous soils and dry continental or dry, semi-hot climates. These environmental conditions are common in the Chihuahuan Desert (Henrickson and Johnston, 1997), where most species of this genus are present (Bogler, 1995). Although, D. cedrosanum is considered a typical element of the rosette scrub (Marroquín et al., 1981), in these communities it has a low density and in fact, its density increases considerably at higher elevations, in transitional areas with sub-montane scrub and pinyon pine forest, where it becomes one of the most common vegetation components (Henrickson and Johnston, 1986; Granados-Sánchez and Sánchez-González, 2003). The fact that the highest densities of this species were recorded in north-facing slopes and in xerophytic scrub communities of the Eastern Sierra Madre suggests that it responds positively to increasing humidity levels and decreasing temperatures, a possibility that future studies may confirm. Despite the considerable variation of D. cedrosanum density between the associations characterized by the presence of this species, in our study region its mean density (475 ind./ha) is almost 40% larger than the value reported (341 ind./ha) for the northeastern part of the state of zacatecas (Robles et al., 2008). Comparable differences exist for density between these two regions across elevations (at 1,900 m a.s.l. in Coahuila density was estimated at 855 ind./ha versus 574 ind./ha in Zacatecas). However, further research is needed to determine the underlying causes of the differences in density between these two regions, considering that in the subregion of Cedros and Mazapil, northern zacatecas, the harvesting of sotol individuals for preparing alcoholic beverages has been going on for over two decades already (Robles et al., 2008), while harvesting in Coahuila is still incipient.

To-date, the effects of the harvesting of this species for different purposes on the structure of its populations and the associated species are unknown; the same is true for the disturbance caused on the plant communities where this species thrives. undoubtedly, there is an urgent need to design management plans in order to minimize the impacts of human disturbance on this community, as well as to prevent a potential shrinkage of Dasylirion cedrosanum range; ultimately, these actions may be critical for the conservation of this species. In this regard, the use of micro-propagation techniques for this species is cause for some optimism, as it could increase its regeneration rates (Villavicencio et al., 2007). Also, we need to study the dynamics of its populations in all associations where this species occurs in order to ensure the most adequate regulation for its potential commercial extraction. Considering the lack of protected areas in this region, the conservation of the D. cedrosanum rosette scrub, as well as of all the endemic and threatened taxa associated to this community remains in peril under current land use practices. In order to be successful, the size and status of these areas should acknowledge the compromise between the needs of development and those of conservation (Riemann and Ezcurra, 2005).

Acknowledgments

We are grateful to the ANSM herbarium for the support provided for conducting this research. José Á. Villarreal kindly determined herb species collected during the project. We thank Felipe N. Hernández for drawing figure 1. Thanks to Helen M. Poulos for carefully reviewing the English of this manuscript. The authors also thank two anonymous reviewers for their comments and criticism on an earlier version that helped improve this paper.

Bogler D.J. 1995. Systematics of Dasylirion: taxonomy and molecular phylogeny. Boletín de la Sociedad Botánica de México 56:69-76.         [ Links ]

Bogler D.J. 1998. Three new species of Dasylirion (Nolinaceae) from Mexico and a clarification of the D. longissimum complex. Brittonia 50:71-86.         [ Links ]

Box E.O. 1981. Predicting physiognomic vegetation types with climate variables. Plant Ecology 45:127-139.         [ Links ]

Daubenmire R.F. 1982. Ecología Vegetal: Tratado de Autoecología de Plantas. 3rd ed., Limusa, Mexico City.         [ Links ]

García E. 2004. Modificaciones al Sistema de Clasificación Climática de Köppen. Instituto de Geografía. Universidad Nacional Autónoma de México, Mexico City.         [ Links ]

Godínez-Álvarez H. and Ortega-Baes P. 2007. Mexican cactus diversity: environmental correlates and conservation priorities. Boletín de la Sociedad Botánica de México 81:81-87.         [ Links ]

González-Elizondo M.S., González-Elizondo M. and Márquez-Linares M.A. 2007. Vegetación y Ecorregiones de Durango. CIIDIR-Instituto Politécnico Nacional and Plaza y Valdés, Mexico City.         [ Links ]

Granados-Sánchez D. and Sánchez-González A. 2003. Clasificación fisonómica de la vegetación de la Sierra de Catorce, San Luis Potosí, a lo largo de un gradiente altitudinal. TERRA Latinoamericana 21:321-332.         [ Links ]

Henrickson J. and Johnston M.C.. 1986. Vegetation and community types of the Chihuahuan Desert. In: Barlow J.C., Powell A.M. and Timmermann B.N. Eds. Second Symposium on Resources of the Chihuahuan Desert Region, United States and Mexico, pp. 20-39, Chihuahuan Desert Research Institute, Alpine.         [ Links ]

Henrickson J. and Johnston M. C. 1997. A flora of the Chihuahuan Desert Region. (Draft version). Published by J. Henrickson. Los Angeles.         [ Links ]

Huerta-Martínez F.M. and García-Moya E. 2004. Diversidad de especies perennes y su relación con el ambiente en un área semiárida del centro de México: implicaciones para la conservación. Interciencia 29:435-441.         [ Links ]

Huerta-Martínez F.M., Vázquez-García J.A., García-Moya E., López-Mata L. and Vaquera-Huerta H. 2004. Vegetation ordination at the southern Chihuahuan Desert (San Luis Potosí, Mexico). Plant Ecology 174:79-87.         [ Links ]

Hutcheson K. 1970. A test for comparing diversities based on the Shannon formula. Journal of Theoretical Biology 29:151-154.         [ Links ]

ICE. Instituto Coahuilense de Ecología. 2001. Ordenamiento Ecológico de Coahuila, México. Gobierno de Coahuila, Saltillo.         [ Links ]

MacMahon J.A. and Wagner F.H. 1985. The Mojave, Sonoran and Chihuahuan deserts of North America. In: Evenari M., Noy-Meir I. and Goodall D.W. Eds. Hot Deserts and Arid Shrublands, Vol. 12A, pp. 105-202, Elsevier, Amsterdam.         [ Links ]

Magurran A.E. 1988. Ecological Diversity and its Measurement. Princeton university Press, Princeton.         [ Links ]

Manly B.F.J. 1986. Multivariate Statistical Methods: A Primer. Chapman & Hall, London.         [ Links ]

Margalef R. 1991. Teoría de los Sistemas Ecológicos. universitat de Barcelona, Barcelona.         [ Links ]

Marroquín J.S., Borja L.G., Velázquez C.R. and de la Cruz J.A. 1981. Estudio ecológico dasonómico de las zonas áridas del norte de México. 2nd ed., Instituto Nacional de Investigaciones Forestales, Secretaría de Agricultura y Recursos Hidráulicos, Mexico City.         [ Links ]

Martínez-Palacios A., Eguiarte L.E. and Furnier G.R. 1999. Genetic diversity of the endangered endemic Agave victoriae-regi-nae (Agavaceae) in the Chihuahuan Desert. American Journal of Botany 86:1093-1098.         [ Links ]

Martorell C. and Ezcurra E. 2002. Rosette scrub occurrence and fog availability in arid mountains of Mexico. Journal of Vegetation Science 13:651-662.         [ Links ]

Martorell C. and Ezcurra E. 2007. The narrow-leaf syndrome: a functional and evolutionary approach to the form of fog-harvesting rosette plants. Oecologia 151:561-573.         [ Links ]

Montaña C. and Valiente-Banuet A. 1998. Floristic and life-form diversity along an altitudinal gradient in an intertropical semia-rid Mexican region. The Southwest Naturalist 43:25-39.         [ Links ]

Mueller-Dombois D. and Ellenberg H. 1974. Aims and Methods of Vegetation Ecology. John Wiley & Sons, New York.         [ Links ]

Olhagaray R.E.C., Esparza C.G. and Vega S.F. 2004. Producción y comercialización de licores de sotol (Dasylirion cedrosanum Trel.) en Durango, México. Revista Ciencia Forestal en México 29:83-90.         [ Links ]

Palma-Estrada J.I. 2000. Bases para la propagación del sotol (Dasylirion spp.) vía in vitro y por semilla. B.Sc. Thesis. Facultad de Ciencias Agrícolas y Forestales, Universidad Autónoma de Chihuahua, Ciudad Delicias. 89 pp.         [ Links ]

Pavón N.P., Hernández-Trejo H. and Rico-Gray V. 2000. Distribution of plant life forms along an altitudinal gradient in the semi-arid valley of zapotitlán, México. Journal of Vegetation Science 11:39-42.         [ Links ]

Pérez A., Mas J.F., Velázquez A. and Vázquez L. 2008. Modeling vegetation diversity types in Mexico based upon topographic features. Interciencia 33:88-95.         [ Links ]

Pérez-García E.A. and Meave J.A. 2004. Heterogeneity of xerophytic vegetation of limestone outcrops in a tropical deciduous forest region in southern Mexico. Plant Ecology 175:147-163.         [ Links ]

Pielou E.C. 1966. The measurement of diversity in different types of biological collections. Journal of Theoretical Biology 13:131-144.         [ Links ]

Pinkava D.J. 1984. Vegetation and flora of the Bolson of Cuatro Ciénegas region, Coahuila, Mexico: IV. Summary, endemism and corrected catalogue. Journal of the Arizona-Nevada Academy of Science 19:23-47.         [ Links ]

Reyes-Agüero J.A., Aguirre-Rivera J.R. and Peña-Valdivia C.B. 2000. Biología y aprovechamiento de Agave lechuguilla Torrey. Boletín de la Sociedad Botánica de México 67:75-88.         [ Links ]

Reynolds J.F., Kemp PR. and Tenhunen J.D. 2000. Effects of long-term rainfall variability on evapotranspiration and soil water distribution in the Chihuahuan Desert: a modeling analysis. Plant Ecology 150:145-159.         [ Links ]

Riemann H. and Ezcurra E. 2005. Plant endemism and natural protected areas in the Peninsula of Baja California, Mexico. Biological Conservation 122:141-150.         [ Links ]

Robbins C.S. 2003. Comercio espinoso: comercio y conservación de cactos en el Desierto Chihuahuense. TRAFFIC Norteamérica and Fondo Mundial para la Naturaleza, Washington, D.C.         [ Links ]

Robles E.A., España M.J.L. and Robles B.H. 2008. Biomasa y forraje, distribución espacial y abundancia de la planta del sotol (Dasylirion spp.) en el ejido El Jazmín, Mazapil, Zacatecas, México. Revista Investigación Científica 4:1-9.         [ Links ]

Rodríguez G.A., Villarreal Q.J.A. and Valdés R.J. 1994. Vegetación y flora de la Sierra de Parras, Coahuila. Agraria 10:79-109.         [ Links ]

Rohlf F. J. 1998. NTSYSpc. Numerical Taxonomy and Multivariate Analysis Version 2.0. Applied Biostatistics Inc., New York.         [ Links ]

Romahn de la Vega C.F. 1992. Principales Productos Forestales no Maderables de México. Universidad Autónoma Chapingo, Texcoco.         [ Links ]

Rzedowski J. 1966. Vegetación del Estado de San Luis Potosí. Acta Cientifica Potosina 5:5-291.         [ Links ]

Rzedowski J. 1978. Vegetación de México. Limusa, Mexico City.         [ Links ]

SEMARNAT-. Secretaría de Medio Ambiente y Recursos Naturales. 2010. Norma Oficial Mexicana NOM-059-SEMARNAT-2010. Protección ambiental-Especies nativas de México de flora y fauna silvestres-Categorías de riesgo y especificaciones para su inclusión, exclusión o cambio-Lista de especies en riesgo.         [ Links ]

Sierra T.S., Lara M.C.R., Carrillo R.R., Mendoza C.A., Morales N.C. and RoyoM.M.H. 2008. Los sotoles (Dasylirion spp.) de Chihuahua. Folleto técnico No. 20, Instituto Nacional de Investigaciones Forestales Agrícolas y Pecuarias. Centro de Investigación Regional Norte Centro. Chihuahua.         [ Links ]

Villarreal Q.J.A. 1994. Vegetación y flora de la Sierra de La Paila, Coahuila, México. Sida 16:109-138.         [ Links ]

Villarreal Q.J.A. and Valdés R.J. 1992-1993. Vegetación de Coahuila, México. Revista de Manejo de Pastizales 6:9-18.         [ Links ]

Villarreal Q.J.Á. 2001. Listados florísticos de México. XXIII. Flora de Coahuila. Universidad Nacional Autónoma de México, Instituto de Biología, Mexico City.         [ Links ]

Villarreal-Quintanilla J.Á. and Encina-Domínguez J.A. 2005. Plantas vasculares endémicas de Coahuila y algunas áreas adyacentes, México. Acta Botanica Mexicana 70:1-46.         [ Links ]

Villavicencio G.E., Cano P.A. and Juárez S.A. 2007. Guía para la micropropagación y producción in vitro de plantas de sotol (Dasylirion cedrosanum Trel.). Folleto técnico No. 37, Instituto Nacional de Investigaciones Forestales Agrícolas y Pecuarias. Centro de investigación Regional del Noreste. Saltillo.         [ Links ]

Ward J.H. 1963. Hierarchical grouping to optimize an objective function. Journal of American Statistical Association 58:236-244.         [ Links ]

Zar J.H. 1999. Biostatistical Analysis. 4th ed., Prentice Hall, Upper Saddle River.         [ Links ]