Biogeografía

 

Is geographical rarity frequent among the cacti of the Chihuahuan Desert?

 

¿Es la rareza geográfica frecuente entre las cactáceas del Desierto Chihuahuense?

 

Héctor M. Hernández^*, Carlos Gómez–Hinostrosa and Gibrán Hoffmann

 

Departamento de Botánica, Instituto de Biología, Universidad Nacional Autónoma de México, Apartado postal 70–233, 04510 Mexico D. F., Mexico.

 

^*Correspondent:
hmhm@ibiologia.unam.mx

 

Recibido: 16 febrero 2009
Aceptado: 08 julio 2009

 

Abstract

With the aim of assessing the extent of geographical rarity of Mexican Cactaceae, we calculated the distribution size (area of occupancy) of 142 species from the Chihuahuan Desert. In addition, using 2 variables (number of localities and range size), we preliminarily assessed their conservation status using the current IUCN Red List criteria. The results showed enormous variation in the areas of occupancy, although from the biogeographic and conservation perspective the most exceptional group comprises the extremely narrow endemics (42 species), whose range is restricted to areas smaller than 10 km2. Our results reinforce the reputation of this plant family as exceptionally rare geographically. We suggest that geographical rarity of Cactaceae in the Chihuahuan Desert is a natural phenomenon; however, we propose that the range of several species has been influenced by human activities. Regarding the conservation status of the species, 75 of them are categorized as Least concern. The remaining 67 species (47.2%) fall in 1 of the 3 categories of threat (27 Vulnerable, 11 Endangered, and 29 Critically endangered). These figures confirm the critical conservation status of Mexican Cactaceae.

Key words: areas of occupancy, cacti, Chihuahuan Desert, distribution size, Mexico, IUCN Red List.

 

Resumen

Se calculó el tamaño de la distribución (área de ocupación) de 142 especies de cactáceas del Desierto Chihuahuense, con el objeto de evaluar su grado de rareza geográfica. Además, mediante el uso de 2 variables (número de localidades y tamaño de distribución), se estimó de manera preliminar su estado de conservación usando los criterios actuales de la Lista Roja de la UICN. Los resultados mostraron gran variación en las áreas de ocupación. Sin embargo, desde una perspectiva biogeográfica y de la conservación, el grupo de especies más excepcional corresponde a las endémicas restringidas (42 spp.), cuyas áreas de distribución son menores de 10 km². Los resultados fortalecen la reputación de las cactáceas de ser una familia de plantas excepcionalmente rara geográficamente. Se sugiere que la rareza geográfica en cactáceas es un fenómeno natural; sin embargo, se propone que la distribución de algunas de las especies ha sido influenciada por actividades humanas. En lo que respecta al estado de conservación de las especies, 75 pueden ser consideradas bajo la categoría de Preocupación menor, mientras que las restantes 67 (47.2%) caen en alguna de las categorías de amenaza (Vulnerable = 27 spp., Amenazada = 11 spp., Críticamente amenazada = 29 spp.). Los resultados confirman el grave estado de conservación de las cactáceas mexicanas.

Palabras clave: áreas de ocupación, cactáceas, Desierto Chihuahuense, México, tamaño de distribución, Lista Roja de la UICN.

 

Introduction

Rare species are usually defined as those having low local abundance, small range size, and/or high habitat specificity (Rabinowitz, 1981; Gaston, 1994). Rarity and conservation are closely linked, because rare species have a greater likelihood of extinction than common ones (Gaston, 1994). In fact, the degree of rarity of the species has often been a useful parameter for their assignment to different categories of threat (IUCN, 2001). Investigations on the Cactaceae from the Chihuahuan Desert Region (CDR) have provided a considerable amount of quantitative and qualitative data suggesting that a significant number of cactus species is particularly rare geographically and ecologically (Hernández and Godínez, 1994; Hernández and Bárcenas, 1995, 1996; Bárcenas, 1999; Gómez–Hinostrosa and Hernández, 2000; Hernández et al., 2001, 2004, 2008; Hernández and Gómez–Hinostrosa, 2004; Goettsch and Hernández, 2006). Specifically, geographical rarity, recognized as the occurrence of particular species in relatively small areas, has been conceived as a recurring fact in Mexican Cactaceae. However, we are just starting to understand the real magnitude of this phenomenon and the way it is expressed in the CDR.

Essentially confined to the New World, the cactus family comprises an estimated number of 1 500 species (Anderson, 2001; Hunt, 2006). Mexico has unequivocally been identified as the main centre of diversity of Cactaceae and a focal point for the conservation of members of this plant family. This country, primarily its dry areas, is home of almost 40% of the global diversity. In particular, the CDR, the largest desert region in North America, is home of the richest assemblage of Cactaceae worldwide. Hernández et al. (2004) reported 329 cactus species and 39 genera for the region, with 70% of the species and 43% of the genera being strictly endemic.

In a recent investigation (Hernández and Navarro, 2007), we calculated the distribution size of some Chihuahuan Desert cacti, confirming that some species have extremely limited geographical ranges; however, we still lack comprehensive information about the extent of geographical rarity of Cactaceae in the region. In this paper we estimate the area of occupancy (AOO) of a representative selection of Chihuahuan Desert cacti, ranging from widely distributed species to extremely narrowly endemic ones. The results allowed us to obtain a broad perspective of the nature and magnitude of geographical rarity of the Cactaceae in the Chihuahuan Desert. In addition, using 2 biogeographical variables (number of localities and AOO), we preliminarily assessed the conservation status of all of the studied species using the IUCN Red List current criteria.

 

Material and methods

Study area. The Chihuahuan Desert is a vast area extending from central Mexico, in the states of Guanajuato, Querétaro, and Hidalgo north to southern Texas, New Mexico, and a small part of Arizona. To the west and east, the region is bordered by the Sierra Madre Occidental and the Sierra Madre Oriental, and it has an approximate extension of 533 600 km² (the area was recalculated by us in this paper, but see Hernández and Gómez–Hinostrosa, 2005; Hernández, 2006). From the conservation perspective, the CDR stands among the 3 most remarkable deserts in the world, thanks to its relatively high biodiversity, numerous endemic species, and the existence of extensive well preserved areas (Dinerstein et al., 1999; Mittermeier et al., 2002). Thorough descriptions of the climatological, ecological, biogeographical, and vegetational characteristics of the CDR have been provided by numerous authors (e. g., Shreve, 1942; Johnston, 1977; Schmidt, 1979; Medellín–Leal, 1982; Henrickson and Johnston, 1986; Hernández and Gómez–Hinostrosa, 2005; Hernández, 2006).

Species selection. A total of 142 cactus species from the CDR were selected for this study, of which 129 (85.2%) are endemic to the region (Appendix 1). The species were chosen because they were reasonably well defined taxonomically and adequately represented by herbarium collections. We excluded several species from the analysis, either because they were taxonomically problematic or because they were underrepresented in the database. This was the case of several extremely wide–ranging species, such as Opuntia lasiacantha, O. rastrera, O. streptacantha, Cylindropuntia imbricata, and C. tunicata, as well as several narrowly distributed ones (e.g., Coryphantha wohlschlageri, Mammillaria longiflora, M. stella–de–tacubaya, M. roseoalba, etc.). Data on the general geographic distribution of all of the studied species may be found in Hernández et al. (2004) and are also summarized in the Appendix 1. Taxonomic nomenclature follows Hunt (2006).

Definitions. We followed the IUCN (2001) definition of area of occupancy, as the area occupied by a taxon within its more general extent of occurrence. The concept recognizes the fact that organisms usually are absent in some areas within its usually broader extent of occurrence, where uncolonized or unsuitable habitats exist. We consider AOO as a synonym of distribution or range size. Also, we define locality as a graphic point on a map, taken from a geo–referenced herbarium record, and separated by its nearest neighbour by at least 797.9 m, the radius of a 2–km² circle. Points located within this radius were considered part of the same locality. In concordance with the UICN (2001) convention, we considered a 2–km² area as a reasonable measure to quantify individual localities.

Data sources. Geo–referenced specimen data were taken from a database of cactus collections from North and Central America. The database contained at the time of this study over 26 300 herbarium records of Cactaceae from 35 herbaria, including our own collections resulting from extensive fieldwork conducted in the CDR over the last 15 years. A total of 6 496 geo–referenced records were taken from the database for this study (Fig. 1). The specimens included in the database had been collected between 1823 and 2007, with 35.4% collected between 1950 and 1989, and 41.9% since 1990 to the present. We estimate that in the CDR the collecting effort is one herbarium specimen of Cactaceae per every 61.1 km² (total area = 533 600 km2).

Cartographic method by conglomerates. For the calculation of the species geographic ranges, we used the cartographic method by conglomerates (CMC) proposed by Hernández and Navarro (2007). This method incorporates elements of the areographic and cartographic methods, previously used by different authors to calculate distribution size. The CMC allows the recognition of discrete aggregations of 2 or more points (conglomerates) as well as isolated, single localities (satellites), both of which are clearly and objectively identifiable. The CMC procedure consisted of the following steps:

1. The geo–referenced data points of each species were plotted over a digital map. All available records of each species were used to plot the distribution maps.

2. Points were interconnected to form an open, minimum spanning tree (Rapoport, 1975), where all points were united by their shortest distance to their nearest neighbor. In minimum spanning trees all the points are interconnected and no loops should be formed (Fig. 2a).

3. Minimum distances between pairs of points were measured and the average distance (mean propinquity index) was calculated. The resulting average figure was used as a radius to trace a circle around each point (Fig. 2b). Thus, the size of the circles surrounding each point depended on the spatial configuration of the localities of each particular species.

4. At this stage of the procedure the disjunct conglomerates of 2 or more partially overlapping circles and the single, isolated satellites were clearly identifiable. In the example of Figure 2b, Astrophytum capricorne had 3 distinct conglomerates and 2 satellites.

5. The area of each conglomerate was calculated individually by superimposing a grid over the map containing the recorded localities. However, an appropriate scale (grid cell size) must be previously defined, due to the fact that area calculation is scale–dependent (see Hernández and Navarro, 2007). To determine the scale for each conglomerate, the distance between the 2 more distant points in that conglomerate was measured (Fig. 2c), and 10% of that maximum distance defined the grid cell size in each conglomerate. Thus, in the example of Astrophytum capricorne (Fig. 2d) the largest distance between points within each conglomerate (54.8, 98.3, and 19.96 km) determined 3 different scales: 5.48 for the western (C1) conglomerate (grid cell size = 5.48 × 5.48 = 30.03 km²), 9.83 for the central (C2) conglomerate (grid cell size = 9.83 × 9.83 = 96.629 km²), and 1.996 for the eastern (C3) one (grid cell size = 1.996 × 1.996 = 3.984 km²).

6. Finally, the area occupied by the species in each conglomerate was calculated by adding the grid cell areas where the species was present. As for the satellites, each was assigned a constant area of 2 km². The AOO of the species was then calculated as the sum of the areas of all of its conglomerates and satellites.

The CMC was applied to all of the species whose distribution area was represented by at least 2 localities on a map. However, we found that our method was inapplicable to several species known to have extremely discontinuous distributions that very often inhabit extremely specialized habitats, such as gypsum and other unusual soil formations. These exceptional cases are exemplified by Ariocarpus kotschoubeyanus, a species known from some 35 discrete locations, in 6 different states, across a large portion of the CDR. This plant is an edaphic specialist growing almost exclusively in silty, dry lake beds, which are found widely scattered throughout the desert areas in northern Mexico. The indiscriminate application of our method to species such as A. kotschoubeyanus would produce unrealistically large AOO estimations, in this particular case AOO= 3 104.9 km². In cases such as this one we decided as an alternative to consider each individual locality as a 2–km² satellite. Consequently, the resulting area for A. kotschoubeyanus upon applying this criterion, and considering that some of the satellites overlap partially, is 67.95 km², a more convincing result for this rare and ecologically specialized species (see Hernández and Navarro, 2007). This same principle was also applied to the species known from a single locality. The species in Appendix 1 marked with a cross (†) are those to which all their localities were considered as satellites.

A Geographic Information System (ArcView 3.2) was used at all stages of the AOO calculation. This tool allowed us to plot the point distribution maps, measure all distribution parameters, and generate the grids with maximum accuracy. Minimum spanning trees were constructed with the aid of the ArcView extension "Trazos 2004" by Camilo Rojas (http://panbiog.infobio.net/?Software). For more information on the CMC application, readers are requested to see Hernández and Navarro (2007).

 

Results

Number of localities. The number of herbarium records available for a particular species depends on several factors, such as the size and accessibility of its geographical range, its local abundance, its visibility (small, cryptic species vs. large, conspicuous ones), and its maneuvrability (spineless or weakly–spined species vs. rigidly–spined species). In other words, the probability of a cactus plant reaching a herbarium cabinet is higher when it is geographically and ecologically common, conspicuous, and easy to process into a herbarium specimen. Considering these factors, the number of known localities per species showed substantial variation, ranging from a single locality, as was the case of 21 species, to 267 in the widespread, treelike Myrtillocactus geometrizans (Appendix 1). The average number of localities per species was 41.9 (± S.D. = 63.1, n = 142).

A good preliminary indicator of the degree of geographical rarity of the Cactaceae from the CDR is the number of localities known for each particular species. In fact, a simple linear regression model showed that there is a highly significant correlation between the number of localities (log₁₀ transformed) of each particular species and their estimated log₁₀ AOO (r² = 0.789, F_(1,140) = 522.59, P < 0.001), suggesting that the number of herbarium records may be used to predict range size. If the high correlation between the number of localities and AOOs is considered, our results suggest a clear tendency among cacti in the CDR to have restricted geographical ranges. Thus, almost half of the studied species (67 spp. or 47.8% of the sample) are known from 10 or less discrete localities (Fig. 3); of these, 33 species (23.2%) occur only in 3 or fewer localities, and 21 species (14.8%) in a single locality. In contrast, only 17 species (12%) are represented by a number of localities ranging from 100 to 200, and just seven wide–ranging species (4.9%) have been registered in 201 or more localities: Echinocactus platyacanthus, Echinocereus pentalophus, Mammillaria crinita, M. formosa, Myrtillocactus geometrizans, Opuntia stenopetala, and Sclerocactus uncinatus.

Areas of occupancy. As shown in the Appendix 1, the estimated AOOs vary continuously from tiny areas measuring less than 1 km² (e. g., Ariocarpus bravoanus and Mammillaria carmenae) to larger ones covering extensive portions of the CDR [e.g., Cylindropuntia kleineae (AOO= 132 702 km²), Peniocereus greggii (AOO= 126 337 km²), and Sclerocactus uncinatus (AOO= 102 881 km²)]. The average AOO was 8 015 km² (± S.D. = 20 421; n = 142).

The great majority of the species (117 spp. or 82.4%) are restricted to areas smaller than 10 000 km², roughly the surface of a one–degree grid square at 35–36° latitude (Fig. 4). Among these, 59 spp. have ranges smaller than 100 km², representing 41.6% of the studied species. Remarkably, 42 species, nearly a third of the total sample (29.6%), are extremely narrow endemics whose geographic range is restricted to areas smaller than 10 km2 (Appendix 1 and Fig. 4).

IUCN Red List. The data presented in the Appendix 1 are particularly useful in evaluating the conservation status of the species, using the Red List criteria (IUCN, 2001). In connection to this, we preliminarily categorized the conservation status of the 142 cactus species studied in this investigation. We accomplished this by comparing the AOO estimations and the number of known localities given in the Appendix 1 with the thresholds set by the IUCN (criterion B) for Vulnerable species (< 2 000 km², and < 10 localities), Endangered (< 500 km², and < 5 localities), and Critically Endangered (< 10 km², and 1 locality). Thus, according to these 2 variables, 75 species, those having the widest geographic ranges and the highest number of localities are categorized as Least concern. The remaining 67 species (47.2%) fall in any one of the 3 categories of threat (Vulnerable = 27 spp., Endangered = 11 spp., and Critically endangered = 29 spp.).

 

Discussion

Performance of the Cartographic method by conglomerates. Just as the typical cartographic (point–to–grid) method is greatly influenced by sampling effort, the method used here to calculate AOO also produced underestimates when sampling was poor or spatially biased (Graham and Hijmans, 2006). However, we believe that the relatively high number of records available for this study assure the relative accuracy of our AOO estimations (Fig. 1). As indicated in the methods, 6 496 geo–referenced records were used in our study, of which only 519 (8%) were redundant (same species, same locality); thus, we used in the mapping process 5 977 non–redundant records. On this basis, we consider that the database is complete enough to provide a reasonably accurate picture of the geographic ranges of the species, irrespective of whether they are widespread or narrow endemics. It has to be recognized, however, that there are extensive areas in the CDR that still are insufficiently explored (Fig. 1), as compared to intensively sampled regions such as Huizache and Mier y Noriega, for example (Hernández et al., 2001; Gómez–Hinostrosa and Hernández, 2000).

After using the CMC with a high number of species, we maintain Hernández and Navarro's (2007) argument that the results derived from the CMC are more accurate than the areographic (Rapoport, 1975) and cartographic methods. One of the limitations of the latter 2 methods is that they assume that the geographic range of the species is continuous (uninterrupted by disjunctions), which is rarely seen in nature (Brown and Lomolino, 1998; Gaston, 2003). As a result, these methods overestimate AOO calculations when compared to the CMC, because they incorporate extensive unoccupied areas. In their study of 10 cactus species, Hernández and Navarro (2007) reported that the average AOO calculated by the aerographic and cartographic methods was 3.5 and 5.5 times larger respectively than that estimated by the CMC.

Consequently, we believe that the CMC is more precise because it is sensitive to the presence of spatial disjunctions, excluding from the AOO calculation areas where the species are absent. In this respect, we have recently shown that geographical discontinuities are the rule among the Cactaceae in the CDR; these are expressed at different scales, from local patchiness to large geographical disjunctions (see Figs. 3 and 4 in Goettsch and Hernández, 2006, and Hernández et al., 2008, respectively).

Number of localities and areas of occupancy. We found a highly significant correlation between the number of localities and the estimated AOO. Thus, knowledge of the number of localities known for a particular species could be used as a predictor of their range size. Using this criterion, consideration of the locality data alone in this investigation (Appendix 1) suggested the high frequency of cactus species in the CDR with small AOO's.

The use of this model could prove to be practical for estimating range size of other groups of plants and other organisms inhabiting desert and semi–desert regions. However, convincing results would be produced only if the most complete set of documented data of the global distribution of the species is used. Obviously, incomplete sets of information would result in erroneous conclusions.

Conservation status. The proportions given in the results for the different Red List categories are an indication of the conservation status of the Cactaceae of the CDR, and probably reflect the extent of endangerment of the entire family in Mexico, at least for the species inhabiting desert and semi–desert areas. In connection to this, it is worth noticing that the percent of threatened Cactaceae calculated here (47.2%) is similar to that reported for the cycads of the world (51.9%) in the 2007 version of the IUCN Red List (http://www.redlist.org). The cycads are an assemblage of 3 plant families (Cycadaceae, Stangeriaceae, and Zamiaceae), entirely evaluated for the Red List, having the reputation of being highly endangered. Unfortunately we do not know how the conservation status of Mexican Cactaceae compares to that of other critical groups of the Mexican succulent flora (e. g., Agavaceae, Crassulaceae, Nolinaceae, etc.) as no comprehensive information on their status exists for these groups.

A total of 43 species in the Appendix 1 has been previously evaluated for the Red List (http://www.redlist.org) by various advisors, using different sets of data. Of these, 39 are categorized as threatened (VU, EN or CR), and it is interesting that 34 of them (89.5%) are also classified as threatened according to our data. However, 1:1 coincidences (either VU, EN or CR) decrease to 38.5%. We judge that this level of correspondence with the current Red List is a confirmation that our approach harmonizes reasonably well with the Red List criteria established by the IUCN (Hernández and Navarro, 2007).

The influence of man. There are numerous documented cases of natural populations being virtually eliminated as a consequence of habitat destruction, unsustainable use, or poaching, resulting in the reduction of the species geographical range (del Castillo and Trujillo, 1991; Anderson et al. 1994). However, some cactus species are able to tolerate moderate amounts of disturbance (Martorell and Peters, 2005, 2009), and what is more, it is likely that the geographic range of several species has been enlarged by human activities.

For example, evidence exists that the AOO of some species of prickly pears (e. g., Opuntia megacantha, O. hyptiacantha, O. robusta, O. streptacantha, etc.) has been expanded in the CDR by the widespread dissemination of their seeds by man and his domestic animals. Edible fruits of Opuntia have been consumed by the inhabitants of the CDR since prehistoric times (Pimienta, 1990; Casas and Barbera, 2002). In addition, since their introduction to this region more than 3 centuries ago, cows, horses, and burros very likely have contributed to the dispersal of these species. A similar process of range expansion probably has occurred in the widespread Myrtillocactus geometrizans, a large columnar species that produces edible fruits. Furthermore, cholla species (Cylindropuntia leptocaulis, C. kleineae, C. tunicata, and C. imbricata) evidently have enlarged their range as a consequence of human activities. They even behave like weeds, in cases becoming extremely abundant after disturbance. The first 3 cholla species have detachable stem segments facilitating vegetative reproduction and dispersal.

How rare are cacti? The results presented in the Appendix 1 confirm previous statements suggesting that cacti very often inhabit small, frequently extremely restricted, areas of the CDR (Gómez–Hinostrosa and Hernández, 2000; Hernández et al., 2001, 2004, 2008; Hernández and Gómez–Hinostrosa, 2004; Goettsch and Hernández, 2006), and reinforce the reputation of this plant family as exceptionally rare geographically. Unfortunately, however, we lack comparative data allowing us to test whether geographical rarity, as seen in the Cactaceae, is shared with other plant groups inhabiting dry regions in Mexico and elsewhere.

Our perception of a high number of narrowly endemic cactus species in the CDR is not an artefact created by poor or spatially biased collecting. The presence of geographically rare cactus species has been corroborated by decades of exploration by amateur and professional botanists, including our systematic collecting programs carried out during more than 15 years. Geographical rarity in the CDR obviously is a natural phenomenon, likely produced by drastic climatic changes during the last 15 000 years (Van Devender, 1990), coupled with the fact that many of the species appear to have extremely limited dispersal abilities (Goettsch and Hernández, 2006; Hernández et al., 2008). Unfortunately, during the last decades, habitat deterioration and illicit collecting of individuals of particular species to become collectors' items have exacerbated the critical conservation status of rare cacti.

From the biogeographic and conservation perspective, the most important group of species (59 spp.) are the rarest ones occurring in areas smaller than 100 km² (41.6% of the sample), in particular the narrow endemics, restricted to areas smaller than 10 km² (Appendix 1 and Fig. 4). With a few exceptions, these microendemic species are well known and highly attractive to cactus collectors, most of them belonging to genera endemic to the CDR, such as the charismatic Ariocarpus, Astrophytum, Aztekium, Geohintonia, and Turbinicarpus, or to popular genera such as Mammillaria and Echinocereus. The extremely small area occupied by these species, in conjunction with the pressures exerted by illegal collectors, place these species in a critical conservation status. In fact, this group of microendemics is the rarest and most endangered species assemblage in Mexico.

 

Acknowledgments

We thank Fernando Chiang and 2 anonymous reviewers for constructive criticism of the original manuscript, Julieta Rosell for statistical advice, Alfredo Wong for technical support in database management, Julio César Montero for advice in the preparation of the figures, and the curators of the following herbaria for making their respective collections available to us: ARIZ, ASU, B, CAS, CICY, CIIDIR, CHAPA, DES, ENCB, F, GH, HNT, HUAA, IBUG, IEB, K, LL, MICH, MO, NMC, NY, POM, RSA, SLPM, SRSC, TEX, UAT, UC, UNM, US, UTEP, and XAL. Fieldwork associated to this project was carried out under collecting permits FLOR–0004 and FLOR–0094, provided by Secretaría del Medio Ambiente y Recursos Naturales, Mexico.

 

Literature cited

Anderson, E. F., S. Arias and N. P. Taylor. 1994. Threatened cacti of Mexico. Royal Botanic Gardens, Kew. 136 p.        [ Links ]

Anderson, E. F. 2001. The cactus family. Timber Press, Portland, Oregon. 776 p.        [ Links ]

Bárcenas, R. T. 1999. Patrones de distribución de cactáceas en el estado de Guanajuato. Thesis, Facultad de Ciencias, Universidad Nacional Autónoma de México, México, D. F. 25 p.        [ Links ]

Brown, J. H. and M. V. Lomolino. 1998. Biogeography. 2^nd edition, Sinauer Associates, Inc. Sunderland, Massachusetts. 691 p.        [ Links ]

Casas, A. and G. Barbera. 2002. Mesoamerican domestication and diffusion. In Cacti, biology and uses, P. Nobel (ed.). University of California Press, Berkley. p. 143–162.        [ Links ]

del Castillo, R. and S. Trujillo. 1991. Ethnobotany of Ferocactus histrix and Echinocactus platyacanthus (Cactaceae) in the semiarid central Mexico: past, present and future. Economic Botany 45:495–502.        [ Links ]

Dinerstein E, D. Olson, J. Atchley, C. Loucks, S. Contreras–Balderas, R. Abell, E. Iñigo, E. Enkerlin, C. Williams and C. Castilleja (eds.). 1999. Ecoregion–based conservation in the Chihuahuan Desert: a biological assessment and biodiversity vision. WWF, CONABIO, PRONATURA and ITESM, Washington, DC. 122 p.        [ Links ]

Gaston, K. J. 1994. Rarity. Chapman and Hall, London. 205 p.        [ Links ]

Gaston, K. J. 2003. The structure and dynamics of geographic ranges. Oxford University Press, Oxford. 266 p.        [ Links ]

Goettsch, B. and H. M. Hernández. 2006. Beta diversity and similarity among cactus assemblages in the Chihuahuan Desert. Journal of Arid Environments 65:513–528.        [ Links ]

Gómez–Hinostrosa, C. and H. M. Hernández. 2000. Diversity, geographical distribution, and conservation of Cactaceae in the Mier y Noriega region, Mexico. Biodiversity and Conservation 9:403–418.        [ Links ]

Graham, C. and R. Hijmans. 2006. A comparison of methods for mapping species ranges and species richness. Global Ecology and Biogeography 15:578–587.        [ Links ]

Henrickson, J. and M. C. Johnston. 1986. Vegetation and community types of the Chihuahuan Desert. In Second Symposium on the Resources of the Chihuahuan Desert Region, United States and Mexico, J. C. Barlow, A. M. Powell and B. Timmermann (eds.). Chihuahuan Desert Research Institute and Sul Ross State University, Alpine, Texas. p. 20–39.        [ Links ]

Hernández, H. M. and H. Godínez. 1994. Contribución al conocimiento de las cactáceas mexicanas amenazadas. Acta Botanica Mexicana 26:33–52.        [ Links ]

Hernández, H. M. and R. T. Bárcenas. 1995. Endangered cacti in the Chihuahuan Desert. I. Distribution patterns. Conservation Biology 9:1176–1190.        [ Links ]

Hernández, H. M. and R. T. Bárcenas. 1996. Endangered cacti in the Chihuahuan Desert. II. Biogeography and Conservation. Conservation Biology 10:1200–1209.        [ Links ]

Hernández, H. M., C. Gómez–Hinostrosa, and R. T. Bárcenas. 2001. Diversity, spatial arrangement, and endemism of Cactaceae in the Huizache area, a hot–spot in the Chihuahuan Desert. Biodiversity and Conservation 10:1097–1112.        [ Links ]

Hernández, H. M. and C. Gómez–Hinostrosa. 2004. Studies on Mexican Cactaceae. IV. A new subspecies of Echinocereus palmeri Britton and Rose, first record of the species in the Chihuahuan Desert. Bradleya 22:1–8.        [ Links ]

Hernández, H. M., C. Gómez–Hinostrosa, and B. Goettsch. 2004. Checklist of Chihuahuan Desert Cactaceae. Harvard Papers in Botany 9:51–68.        [ Links ]

Hernández, H. M. and C. Gómez–Hinostrosa. 2005. Cactus diversity and endemism in the Chihuahuan Desert region. In Biodiversity, ecosystems and conservation in Northern Mexico, J. L. Cartron, G. Ceballos and R. Felger (eds.). Oxford University Press, New York. p. 264–275.        [ Links ]

Hernández, H. M. 2006. La vida en los desiertos mexicanos. Fondo de Cultura Económica, México, D. F. 188 p.        [ Links ]

Hernández, H. M. and M. Navarro. 2007. A new method to estimate areas of occupancy using herbarium data. Biodiversity and Conservation 16:2457–2470.        [ Links ]

Hernández, H. M., B. Goettsch, C. Gómez–Hinostrosa and H. Arita. 2008. Cactus species turnover and diversity along a latitudinal transect on the Chihuahuan Desert Region. Biodiversity and Conservation 17:703–720.        [ Links ]

Hunt, D. (compiler). 2006. The new cactus lexicon. DH Books, Milborne Port, England. 373 p.        [ Links ]

IUCN. 2001. IUCN Red List categories and criteria: version 3.1. IUCN Species Survival Commission. IUCN, Gland, Switzerland and Cambridge. 30 p.        [ Links ]

Johnston, M. C. 1977. Brief resume of botanical, including vegetational, features of the Chihuahuan Desert Region with special emphasis on their uniqueness. In Transactions of the Symposium on the biological resources of the Chihuahuan Desert Region, United States and Mexico, R. H. Wauer and D. H. Riskind (eds.). National Park Service, Washington, DC. p. 335–359.        [ Links ]

Martorell, C. and E. M. Peters. 2005. The measurement of chronic disturbance and its effects on the threatened cactus Mammillaria pectinifera. Biological Conservation 124:199–207.        [ Links ]

Martorell, C. and E. M. Peters. 2009. Disturbance–response analysis: a method for rapid assessment of the threat to species in disturbed areas. Conservation Biology 23:377–387.        [ Links ]

Medellín–Leal, F. 1982. The Chihuahuan Desert. In Reference handbook on the Deserts of North America, G. L. Bender (ed.). Greenwood Press, Westport, Connecticut. p. 321–372.        [ Links ]

Mittermeier, R. A., C. G. Mittermeier, P. Robles–Gil, J. Pilgrim, G. A. B. da Fonseca, T. Brooks and W. R. Konstant (eds.). 2002. Wilderness: Earth's Last Wild Places. CEMEX, Mexico City. 576 p.        [ Links ]

Pimienta, E. 1990. El nopal tunero. Universidad de Guadalajara, Guadalajara, Jalisco. 246 p.        [ Links ]

Rabinowitz, D. 1981. Seven forms of rarity. In The biological aspects of rare plant conservation, H. Synge (ed.). Wiley, New York. p. 205–217.        [ Links ]

Rapoport, E. H. 1975. Areografía: estrategias geográficas de las especies. Fondo de Cultura Económica, Mexico, D. F. 214 p.        [ Links ]

Schmidt, R. H Jr. 1979. A climatic delineation of the "real" Chihuahuan Desert. Journal of Arid Environments 2:243–250.        [ Links ]

Shreve, F. 1942. The desert vegetation of North America. Botanical Review 8:195–246.        [ Links ]

Van Devender, T. 1990. Late Quaternary vegetation and climate of the Chihuahuan Desert, United States and Mexico. In Packrat middens. The last 40,000 years of biotic change, J. Betancourt, T. Van Devender and P. Martin (eds.). The University of Arizona Press, Tucson, Arizona. p. 104–133.        [ Links ]