Living Rocks of Mexico
Ariocarpus Revisited
  home [ Cultivars ] [ Species ]  


This paper is reproduced form Haseltonia 5: 1-20.(1997) with the  kind permission of the Editor and Authors, Edward F. Anderson and W. A. Fitz Maurice.

Edward F. Anderson, Desert Botanical Garden, 1201 N. Galvin Parkway, Phoenix, AZ 85008, U.S.A.


W. A. Fitz Maurice, CANTE A C , Mesones 71, 37700 San Miguel de Allende, Gto.  Mexico

Summary. The authors propose that A. fissuratus var. hintonii be placed as a subspecies of A. bravoanus. They suggest that the subgenera of Ariocarpus- subgen. Ariocarpus and subgen. Roseocactus (Aegopadothele) be eliminated, as well as the infraspecific taxa of A. fissuratus var. fissuratus and var. lloydii. They consider Ariocarpus (Anhalonium) elongatus to be a synonym of A. retusus. On the basis of widespread sharing of characteristics and their recent identification of a large hybrid swarm between A. retusus and A. trigonus, the authors conclude that A. trigonus is a subspecies of A. retusus. A new key of the species and subspecies reflecting the proposals outlined above is included. Conservation status of the species within Ariocarpus is variably classified from Endangered to Low Risk.

Resumen. Los autores proponen que A. fissuratus var. hintonii sea puesta como una subespecie de A. bravoanus Se sugiere que los suhgéneros de Ariocarpus el subgénero Ariocarpus y el subgenero Roseocactus (Aegopadothele), sean eliminados, asi como los taxa infraespecificos de A. fissuratus var. fissuratus var. lloydii. Se considera que Ariocarpus (Anhalonium) elongatus es un sinónimo de A. retusus, Se concluye que A. trigonus es una subespecie de A. retusus, basándose en un comun y amplio espectro de caracterIsticas y en la reciente identificación de una población con un amplio espectro de hibridación, "a hybrid swarm", entre A. retusus y A. trigonus Se incluye una clave para la identificacion de las especies y las subespecies, reflejando las propuestas arriba sitadas El estado de conservación de las especies incluidas dentro de Ariocarpus tiene una clasificación variable, desde peligro de extlnclón hasta bajo riesgo.

Introduction (top)

More than 35 years ago one of us (Anderson) published a series of papers dealing with the taxonomy and ecology of Ariocarpus (Anderson, 1958, 1960, 1962, 1963, 1964). Since the time of that study, much has been written about the genus and several new taxa have been proposed. We have also been able to study many additional populations of Ariocarpus in the field, including all taxa that have been proposed in the last 35 years. This paper presents new data and interpretations of the genus based on our recent laboratory work, studies of plants in the field, and analysis of the literature.

The Taxonomic Status of A. bravoanus and A. fissuratus var. hintonii (top)

Two new taxa of Ariocarpus have been described within the past seven years. Stuppy and Taylor (1989) described a new taxon of Ariocarpus. The original 'discovery' of this plant was at the experimental garden near Matehuala, S.L.P., operated under the direction of Universidad Autonoma Agraria Antonio Narro, Saltillo, Coahuila. The caretaker was selling collected plants which included the new taxon. A few years later, George Hinton discovered a locality for the plant about 25 kilometers south of the experimental garden. One of us (Fitz Maurice), with Tom Davis, looked for additional locations and found the place near the experimental garden (Fitz Maurice and Davis, 1987). At that time, the local people told of substantial collecting at that site by foreigners. This same site is almost certainly the one located earlier by the caretaker at the experimental garden. This new taxon seemed most closely to resemble A. fissuratus, a widespread, highly variable species found throughout much of northern Mexico and into Texas. Plants were given to Stuppy and Taylor (1989) by one of us (Fitz Maurice), who described it as A. fissuratus var. hintonii.

Only a few years after publication of A. fissuratus var. hintonii, Hector Hernández discovered a new and different population of Ariocarpus growing farther south in the state of San Luis Potosi (Map 1). It was described by Hernández and Anderson (1992) as A. bravoanus.

In the summer of 1993 we visited the habitats of these two new taxa in the company of George S Hinton. Ariocarpus fissuratus var. hintonii is presently known from only two localities at elevations at about 1400 meters, separated by a distance of only about 25 kilometers, and growing in typical Chihuahuan Desert, in the state of San Luis PotosI (Fig 1, Map 1). Numerous plants were found at both localities, though they were well camouflaged in the ground (Fig 2). Ariocarpus bravoanus is known from only a single locality at an elevation of 1500 meters in somewhat modified Chihuahuan Desert, also within the state of San Lois PotosI (Fig 3). This population consists of no more than a few thousand individuals, restricted to a known area of less than one square kilometer (Fig.4). The only other species of Ariocarpus that occur in the area are A. retusus (in similar habitats) and A. kotschoubeyanus (growing in very different localities, typically silt flats). Ariocarpus agavoides, which Hernández and Anderson (1992) suggested is closely related to A. bravoanus, occurs more than 130 kilometers away in the state of Tamaulipas (Map 2). Ariocarpus fissuratus var. hintonii occurs in the state of San Luis Potosi at a distance of 85 kilometers. Unfortunately, living plant material of A. fissuratus var. hintonii was not available to either Hernández or Anderson at the time they described A. bravoanus, nor had they seen it in the field.

The recent opportunity to observe the natural variations in the populations of A. bravoanus and A. fissuratus var. hintonii, as well as to dissect several individuals to study the structure of the tubercles and areoles and to examine the seeds with the scanning electron microscope, has enabled us to draw some conclusions regarding the probable relationship of these plants to other taxa of Ariocarpus. Hernández and Anderson (1992) commented that A. bravoanus "possesses a combination of morphological characters not corresponding to any of the known species of this genus" and that its "features, in fact, place it intermediate to the subgenera Ariocarpus and Roseocactus". Our studies of 1993-96 reinforce this statement and lead us to reinterpret the ideas of Anderson (1960, 1962) regarding the development of areoles in Ariocarpus.

As shown in Figure 5, Anderson (1960, 1962) described four schemes of areole development in Ariocarpus. A) A. fissuratus and A. kotschoubeyanus, in which elongation occurs beneath the areole, thus creating an elongated structure within a furrow, B) A. retusus, in which the spiniferous and floral portions of the areole are separated in the process of elongation, creating dimorphic areoles, C) A. trigonus and A. scaphirostris, in which elongation of the tubercle occurs distally to the entire areolar region, and D) A. agavoides, in which elongation of the tubercle occurs only basally to an undivided areolar area, but, like all species, after flowering has occurred in the young tubercle at the center of the plant Anderson (1960) commented that sometimes these schemes do not develop uniformly, thus producing individuals of A. fissuratus, for example, which lack areolar grooves (Fig 6). Nonetheless, he felt that there were still sufficient differences between those species having areolar grooves and those without to have separate subgenera. We now question that decision and discuss it in the next section of this paper.

New evidence suggests that Schemes B and C are not always distinct and that sometimes individual plants may have both types, this is discussed further in a later section. Also, to the four schemes described above, there must he added a fifth, representing the areole development of the two recently-described taxa, A. bravoanus and A. fissuratus var. hintonii. However, this scheme is considerably more complicated than that observed in the other groups. As shown in Scheme E of Figure 5, the tubercle may develop (and elongate) either behind or under the areolar area, producing a condition similar to either A or D. Again, elongation occurs primarily after flowering. Intermediate conditions also occur, creating areolar areas in the middle of the tubercle, with varying degrees of elongation.

Further examination of the two new taxa suggest that they are more closely related to one another than to any other taxa of Ariocarpus. Not only are their distributions closer to each other than to other possible relatives, such as A. fissuratus or A. agavoides, but they share a syndrome of similar characters. They are diminutive in size, probably an indication of neoteny, such as is also seen in A. agavoides and A. scaphirostris. They both possess epidermal papillae to varying degrees (Figs. 7-10), these are absent in all other taxa except A. fissuratus. Scanning electron micrography shows that their seeds are similar in both size and the cuticular nature of the verrucose testa (Hernández and Anderson, 1992), but they definitely differ from seeds of A. fissuratus. With the exception of A. retusus and A. kotschoubeyanus, which have extensive north-south ranges, A. bravoanus and A. fissuratus var. hintonii tend to be the southernmost taxa of the genus. These taxa occur over a range of about 85 kilometers (50 miles), but they are more than 200 kilometers (120 miles) away from the closest known locality of A. fissuratus. They also have similar tubercles and flower structure when compared with other Ariocarpus taxa. The shared tendency for vertical orientation, tubercle size, and fine papillae are shared characters common to the two taxa and different from A. fissuratus. Finally, an examination of individuals from all the known populations shows intermediate forms, particularly with regard to the development of the areole. Some individuals of A. bravoanus, for example, have a partially developed areolar groove (Fig 11), whereas some specimens of A. fissuratus var. hintonii nearly lack the groove (Fig 12).

We conclude that the two new taxa represent a single species, with characters intermediate between A. agavoides and A. fissuratus. We propose that the infraspecific taxon hintonii be retained, for the known populations of this taxon are geographically isolated from any other populations of Ariocarpus. If further field work should locate new populations of this species intermediate in location to those presently known, and with individuals demonstrating even greater heterogeneity of tubercles and areolar characters, then it is possible that an infraspecific taxon should not be formally recognized. This would make A. bravoanus similar to A. fissuratus and A. retusus, which are widespread species with great variation of tubercle structure (see the following sections). However, we propose the following new combination, believing that it best indicates our present understanding of relationships within the genus: Ariocarpus bravoanus H. Hernandez & E.F. Anderson subsp. hintonii (W. Stuppy & N.P. Taylor) E.F. Anderson & W.A. Fitz Maurice, comb. et stat nov.. (Basionym = Ariocarpus fissuratus var. hintonii W. Stuppy & N.P. Taylor, Bradleya 7:841989).

The Status of the Subgenera of Ariocarpus (top)

Berger (1925) proposed the genus Roseocactus as a segregate of Ariocarpus, arguing that the areoles were of completely different types and that the place of the origin of the flower was different. Anderson (1960) accepted Berger's first argument but demonstrated that, in fact, the place of flower origin was essentially the same in all species of Ariocarpus (including Roseocactus). He therefore accepted W.T. Marshall's (1946) classification and relegated Roseocactus to the rank of subgenus. Hunt and Taylor correctly pointed out in an editors' footnote in Hernández and Anderson (1992) that the correct name for the subgenus Roseocactus (A. Berger) W.T. Marshall should be subgenus Aegopodothele K Schumann, as it was published 48 years earlier (Schumann, 1898).

In re-examining the validity of distinguishing between the two subgenera of Ariocarpus, by whatever names they would be called, we conclude that there is no longer a justification for such a dichotomy, considering the areolar structure of A. bravoanus. There simply is no basis for considering that there are two distinct types of areole in the genus, because A. bravoanus and its subspecies hintonii have numerous intermediate forms.

Infraspecific Taxa of A. fissuratus (top)

Anderson (1963) and others (Barthlott, 1979; Hunt, 1989; Bravo H and Sanchez-Mejorada, 1991) accepted two varieties of A. fissuratus var. fissuratus (Fig 13) and var. lloydii (Fig 14). Stuppy and Taylor (1989) proposed var. hintonii as a third variety of A. fissuratus, but in the previous section of this paper we recombined it as a subspecies of A. bravoanus.

One can readily distinguish between plants of the most northern populations in the Big Bend area of Texas- var. fissuratus- and those of the southern population- var. lloydii- located near Parras, Coahuila, and occurring westward to Nazas, Durango. Stuppy and Taylor (1989), in describing the new variety hintonii, discussed the characteristics and distributions of the two varieties of A. fissuratus and the problem of intermediate forms They concluded that "these two varieties can be distinguished over most of their ranges and seem worth maintaining".

We, too, have been troubled by the presence of intermediate forms of A. fissuratus from Cuatro Ciénegas to Estación Marte in Coahuila. Moreover, much of this area in northern Mexico has been poorly explored. The intermediate form from Coahuila was probably the plant referred to by Backeberg and Kilian (1960), though all they said was that it came from "Nordmexiko" and its exact locality was not known. Unfortunately, one of their photographs (on page 149) labeled as Roseocactus intermedius is almost certainly A. kotschoubeyanus. Anderson (1963) considered R. intermedius a nomen nudum and relegated it to the synonymy of A. fissuratus var. lloydii, However, he did not see living material of the intermediate form until 1979 (EFA #5018, US, DES). Both of us have now observed this intermediate form at several sites (Fig 15). The presence of an intermediate form over a north-south range of nearly 150 kilometers and between the localities of the two varieties, suggests to us that A. fissuratus consists of intergrading geographical races We conclude that the taxon A. fissuratus consists of a continuous variation of tubercle characters in shape and configuration of tubercles in a north-south cline and that it is inappropriate to designate portions of the continuum as infraspecific taxa. No other characters differ sufficiently within the species to support a division at the infraspecific level.

Orthography of A. scapharostrus (top)

Little needs to be added to the discussion by Hunt (1991), who argues correctly that a better spelling for this taxon would be scaphirostris. We accept his correction and refer to the taxon with his spelling throughout this paper.

The Status of Ariocarpus (Anhalonium) elongatus Salm-Dyck (top)

Anderson (1964) listed the specific epithet elongatus as a synonym of A. retusus, the type species of the genus, which was described by Scheidweiler in 1838. Scheidweiler did not preserve or designate a type, so Anderson (1964) designated a neotype for A. retusus based on a specimen he collected east of the city of San Luis Potosi (EFA #1200, POM; a specimen is also at DES) Bravo H and Sànchez-Mejorada (1991) concurred with Anderson that the epithet elongatus is a synonym of A. retusus.

Recently Hunt (1991) discussed the disputed status of the specific epithet elongatus, accurately summarizing the various treatments of the epithet by numerous authors. The epithet was first used by Salm-Dyck (1850) in describing a new species of Anhalonium, a generic name published by Lemaire in 1839. As explained by Anderson (1964), Anhalonium is a later synonym of Ariocarpus, having been published for the same plant (A. retusus) described by Scheidweiler a year earlier. Unfortunately, Salm-Dyck did not include an illustration, a locality from which the plant came, nor a description of the flower, any one of these almost certainly would have made clear to what plant he was referring. Neither did he designate or preserve a type specimen. The problem, as correctly stated by Hunt (1991), is that Salm-Dyck referred to the tubercles of A. elongatus as "reminiscent of the leaves of Crassula perfoliata (an apt description of the tubercles of A. trigonus), whereas those of Anhalonium prismaticum (now Ariocarpus retusus) resembled leaves of Aloe retusa." Lee (1937) believed that the taxon elongatus deserved recognition as a separate species. Likewise, Hunt (1992) provisionally accepted it as a separate species. Wettstein (1933), on the other hand, believed that it probably referred to A. trigonus, as did Hunt (1989). Hunt (1991) further pointed out that if this were true, then the taxon presently known as A. trigonus must become A. elongatus because the name was published 40 years earlier!. Haage (1977), in his expansion of Backeberg's Kakteenlexikon, was equivocal, stating that "this sp,, or variety of A. trigonus, is disputed". Interestingly, he went on to say that the taxon was from Jaumave, Tamaulipas, though this locality certainly was not stated by Salm-Dyck in 1850. How could Haage know the original source of the plant to which Salm-Dyck referred? If he were, indeed, correct in stating that the taxon elongatus came from Jaumave, then quite likely it is the same as the taxon currently known as A. trigonus.

We have re-examined this problem and, in the absence of a stated locality by Salm-Dyck, believe the identity of the taxon can be determined from another part of his description of A. elongatum. In the second paragraph of the description of Anhalonium elongatum he states "apice pulvillo tomentoso instructa," which, translated from the Latin, means "apex [of tubercle] provided with a tomentose cushion". We believe he is referring to the small areolar pad near the tip of the tubercle, which is a distinguishing characteristic of A. retusus. Ariocarpus trigonus never has such terminal areolar pads on its tubercles in the adult plant, they are present only on the tubercles of seedlings.

The dilemma regarding the shape and length of the tubercles described for A. elongatum by Salm-Dyck can be resolved by examining individuals of A. retusus growing at the Junction of highways 57 and 80 near El Huizache in the state of San Luis PotosI. Some of these plants, one of which Anderson (1964) designated as the neotype of A. elongatum (EFA #1077, POM), have notably long, tapering tubercles which are up to 5 centimeters (two inches) long and with a similar shape to those of A. trigonus (Fig 16). Some even have a slight curve, like those of A. trigonus. However, they are distinctly blue-green in color, rather than the yellow-green of A. trigonus. If the El Huizache population is extensively examined, plants are found with widely varying tubercle shapes. ranging from 1:1 to 4:1 in length to width ratio. In fact, the A. trigonus-like tubercles tend to he the exception rather than the rule. In other words, this population shows a range of tubercle variation from that of "typical" A. trigonus to that of the northern form of A. retusus. The El Huizache population does not consist of two species growing sympatrically but is, rather, a population of A. retusus showing high variability of tubercle shape. Moreover, the tips of the tubercles, especially when young, usually bear areolar pads. Thus, although Hunt (1991) believed that Anderson's (1964) designation of his collection #1077 as the neotype of A. elongatus was "decidedly unsafe," we feel that it is as accurate a representative of Salm-Dyck's limited description that can he found and is representative of one of the many variants of A. retusus.

The A. retusus - A. trigonus Dilemma (top)

Anderson (1964), Bravo H and Sánchez-Mejorada (1991) and most others have accepted A. retusus and A. trigonus as separate species. In 1994 we visited a site near Aramberri, Nuevo Leon, with a population of Ariocarpus its discoverer had identified as A. trigonus. Flowers of some of the plants from this population were reported as red, which is clearly different from those of either A. retusus or A. trigonus. The discovery of this population of Ariocarpus with "red" flowers has raised questions regarding the distinctness of the two taxa - or whether there might be a new taxon. Plants from the Aramberri area have now been investigated by both of us. It was the first time that one of us (Anderson) hesitated to say that a plant was either A. retusus or A. trigonus. The population is not clearly A. trigonus, nor is it clearly A. retusus, the only two species they resemble. The following characters have been the basis for distinguishing A. retusus from A. trigonus, but the Aramberri and several other populations show that there are exceptions or intergradations. The characters are:

Epidermis color: A. retusus is usually gray-green but is sometimes a darker green without gray, as seen in populations at San Rafael and south of Villa Juárez, S.L.P. Plants of A. trigonus are typically olive-green to yellow-green These two usually distinct color forms merge in the population of Ariocarpus near Aramberri.

Tubercle shape: A. retusus tubercles range from broad and short (1:1), with an angled keel to, less frequently, more slender and long (1:4) with no angled keel. The latter are found at Entronque Huizache, S.L.P., San Roberto, N.L. , and other locations. Tubercles of A. trigonus range from slender and long (1:4) to, less frequently, broad and short (1:1), never with an angled keel. While the A. retusus tubercles usually have a triangular cross-section with sharp edges, those of A. trigonus are more or less rounded. However, these two types of edges tend to merge in some localities, such as in Aramberri and San Rafael, S.L.P.

  Areoles: Both taxa have terminal spiniferous areoles present in the seedling stage. This areole is usually present on tubercles of mature plants of A. retusus, but is lacking in A. trigonus. The location of this areolar pad, when present in A. retusus, varies from well below up to the distal tip. The Aramberri population of Ariocarpus has individual plants bearing both terminal areolar pads and lacking pads altogether. Even immature tubercles of some mature plants lack areolar pads.

Flowers: There are no significant differences in flower characters between the two taxa except for color. Flowers of A. retusus have been described as white to pinkish-white and those of A. trigonus as light yellow or cream-colored. Plants from the Aramberri area have flowers that range from white to creamy-white to pink-tipped to deep pink or magenta (the same color as in A. fissuratus or A. agavoides). In October 1995 one of us (Fitz Maurice) visited a site north of Aramberri where more than 100 plants were observed, 30 of them flowering. One plant's flowers were deep pink in color, two plants' flowers had pink perianth-tips, while all the others were white with an occasional hint of creamy white. However, these plants, though possessing the more or less typical white flowers of A. retusus, had morphological features more suggestive of A. trigonus than of A. retusus. That is, many of the plants had tubercles curving up toward the center, length-to width ratios of 3:1 or greater, dark-colored tubercles with rounded edges, and no terminal areolar pad (Fig. 17). Other plants appear to be clearly intermediate, with tubercles shaped like those of A. retusus but with the yellow-green color of A. trigonus and lacking a terminal spiniferous areolar pad. In a second area to the north of Aramberri there were more than 100 plants in flower. More than 95% of these plants' flowers were white or near-white, the remainder ranging from pink tipped to full deep pink. Again, many of the plants appeared morphologically more similar to A. trigonus than to A. retusus. In a third area visited, west of Aramberri, there were several hundred plants, about 30% of which were flowering. About 99% of these flowers were pink to deep pink in color (Fig 18), a reversal of the condition in the two populations to the north. It is important to emphasize that all populations in the vicinity of Aramberri have individuals with tubercle configurations varying from typical A. retusus to typical A. trigonus, but these variations do not correlate in any way with flower color. Occasional intermediate forms between A. retusus and A. trigonus have been observed in horticulture. Such intermediate forms have now been observed at several localities, principally in the Aramberri area, but also in San Luis Potosi at El Huizache and San Rafael. We believe there are other populations with intermediate forms.

Seeds: The seeds of A. retusus and A. trigonus are alike in essentially all aspects and can be microscopically differentiated from seeds of other Ariocarpus species.

Geographic distribution: A. retusus is widespread throughout the Chihuahuan Desert of northern Mexico at elevations over 1000 meters, with the highest known locality at over 2000 meters near San Luis Potosi. Ariocarpus trigonus occurs primarily in Thorn Forest vegetation at elevations below 1000 meters. One known locality for A. trigonus, northwest of Palmillas, Tamaulipas, is at 1400 meters (Glass & Foster #665), and it occurs in modified Chihuahuan Desert vegetation in the Valley of Jaumave at elevations of less than 1000 meters. Near Liera, Tam , it occurs at 100 meters elevation. Thus, in general the two taxa are geographically separate. Map 3 shows the geographical distribution of collections, known to the authors, of A. retusus and A. trigonus. However, the populations near Aramberri are only 15-20 kilometers (9-12.5 miles) from typical A. trigonus habitat and a small but uncertain distance from what is generally considered to be the area for A. retusus. Other areas, such as at Villa Juarez, San Rafael, and Entronque Huizache, also show a merging of characters.

There are three possible options for dealing with the A. retusus - A. trigonus dilemma:

The first option is to accept A. retusus and A. trigonus as representing the extremes of a continuous cline, such as seen in A. fissuratus and in the north-south cline of A. retusus as described in an earlier paper (Anderson, 1964). However, throughout much of their ranges A. retusus and A. trigonus remain distinct If, indeed, a continuous clime exists, then all populations should be recognized as A. retusus, and A. trigonus relegated to synonymy. Considering the Aramberri populations, we do not believe this best represents the situation.

The second option is to continue to recognize A. retusus and A. trigonus as separate species, despite the evidence of their hybridizing at Aramberri (Map 3). Occasional intermediate forms appear in horticulture (Hirao,1979), but, with the exception of the Aramberri populations, it is virtually impossible to confuse the two taxa in the field. The two taxa, as noted above, tend to have definite syndromes of characters, which include different areole development. Hybridization studies by Lux and Stanik (1983) and observations of numerous tubercles from field-collected plants support the hypothesis that this may be a population sharing genes of both species. Moreover, the Aramberri populations are separated from the range of A. trigonus by only about 20 kilometers (12.5 miles) and there is a canyon through the Sierra at that point which could facilitate genetic exchange. No other species of Ariocarpus are known to hybridize in situ; in fact, some species even occur sympatrically. Therefore, it may not be appropriate to continue to recognize the two taxa as distinct species.

The third option is to recognize A. trigonus as an infraspecific taxon of A. retusus, reflecting the fact that the two taxa can interbreed. Thus, A. trigonus would be a subspecific taxon of the highly variable species, A. retusus, differentiated by geographical location, overlapping but lower altitudes, the absence of terminal spiniferous areolar pads, and flowers with a consistent yellow or creamy-white color. The Aramberri plants are thus a hybrid swarm in the west to east cline between the typical Chihuahuan Desert form of A. retusus and its subspecies or variety trigonus, usually found at lower elevations in Thorn Shrub vegetation.

Thus, the A. retusus - A. trigonus dilemma is not easily resolved. What is seen at Aramberri clearly is an intergrading of the two sets of characters. If this is not more than intergrading geographical races, then the dilemma is whether to formally recognize the extreme form (the typical A. trigonus as an infraspecific taxon. We have already argued that Ariocarpus consists of highly variable taxa. This was shown in an early paper by Anderson (1964) with the variability of tubercles of A. retusus from north to south. In this paper we have suggested that, likewise, A. fissuratus is so variable that it should not be subdivided into infraspecific taxa.

We conclude that A. retusus and A. trigonus are another instance of great variability and that they should not be retained as separate species. Rather, we propose that A. trigonus be relegated to the infraspecific taxon of subspecies Ariocarpus retusus Scheidweiler subsp. trigonus (F A C Weber) Anderson and Fitz Maurice, comb. et stat nov. (Basionym = Anhalonium trlgonum F.A.C. Weber, Dict. Hort. Bois 9O, 1893).

Conservation Status of Ariocarpus (top)

All species of Ariocarpus are included in Appendix 1 of CITES as a result of action taken at the meeting of the CITES parties in Kyoto, Japan, in 1992. However this does not mean that all are threatened or endangered, for one of the justifications for the inclusion of all species in Appendix 1 was because the more common species could be easily confused with the very rare ones. The following summarizes our observations concerning the conservation status of Ariocarpus species. Our categories of threat follow the new Species Survival Commission system of the IUCN (1994).

A. agavoides: (Fig 19, Map 2) This species is one of the most restricted in its known distribution and it continues to be affected by habitat a!teration and illegal collecting. Hernández-Barrera (1992) conducted a demographic survey at the known localities of A. agavoides and found several hundred plants. Recently a new site for this species has been found at a considerable distance from the type locality. Further field work is necessary to determine the distribution of this species, but we feel that it is appropriately classified as endangered (EN).

A. bravoanus: (Fig 3, Fig.4, Map 1) this new species, with its subsp. hintonii, is at present known to occur in only three restricted localities. One location is very near a village, and there is a serious threat of habitat alteration. The site is also known to collectors and there is definite evidence of their activities. In fact, plants of this subspecies have been confiscated in Europe. The other two sites are far from villages or other human activities, so there is less likelihood that those localities will be affected by habitat alteration. However, the number of individual plants in each locality is only in the thousands, so collectors could quickly decimate any of these small populations. In fact, we recently visited the only known site of A. bravoanus subsp. bravoanus and found evidence of illegal collecting. Local people also use several Ariocarpus species, including A. bravoanus, for medicinal purposes. Further field work is essential to determine if this species occurs in other areas. At present we feel this species should he considered endangered (EN) because of its very restricted distribution and vulnerability.

A. fissuratus:(Fig 13, Fig.14, Fig.15, Map 1) one of the most widespread species of the genus, there is little evidence of impact by humans. Some habitat destruction may affect local populations such as in the area near Parras, but, in general, this species is appropriately considered as low risk (LR).

A. kotschoubeyanus:(Fig 20, Map 2) another relatively common species with a wide distribution, this taxon is probably safe. However, agricultural development is occurring on many of the silt flats where A. kotschoubeyanus is commonly found, so local populations will certainly he impacted. We consider this species to he vulnerable (VU).

A. retusus: (Fig 16, Fig.17Fig.18Fig.21, Map 3) this species is both common and widespread throughout the Chihuahuan Desert. There is virtually no danger that it will he eliminated or adversely affected by human activities, though some local populations could he destroyed by habitat destruction. Much of the habitat of subspecies trigonus has been altered for agriculture. However, it still has a fairly wide distribution, both east of the Sierra Madre Oriental and within the Valley of Jaumave. Local populations have been destroyed by human activities, but Martinez et al (1993), have shown that the plant is abundant in many areas in the Valley of Jaumave. This species should be considered low risk (LR).

A. scaphirostris: (Fig 22. Map 2) although this species is known to occur only in one valley in the state of Nuevo Leon, the only possible threat to its existence is collecting. The distinctive habitat is not appropriate for any human activities, including agriculture However, because of its restricted distribution we consider this species to be vulnerable (VU).

Key to the Species and Subspecies of Ariocarpus (top)


Mature plants not usually exceeding 9 cm in diameter



Tubercles divergent, not tightly compacted at base




Tubercles at least twice longer than broad, surface not verrucose, not in San Luis Potosi





Tubercles bearing a spiniferous areolar pad, Tamaulipas - 1..A..agavoides





Tubercles not bearing a spiniferous areolar pad, Nuevo Leon - 2..A..scaphirostris




Tubercles less than twice longer than broad, surface verrucose. San Luis Potosi - 3..A..bravoanus





Tubercles strongly erect, lacking an areolar groove - subsp.. bravoanus





Tubercles not strongly erect, bearing an areolar groove - subsp..hintonii



Tubercles not divergent, tightly compacted at base - 4..A.. kotschoubeyanus


Mature plants usually exceeding 9 in cm diameter,  



Tubercles bearing an areolar groove, verrucose, rounded at tip - 5..A..fissuratus



Tubercles not bearing an areolar groove, not verrucose, pointed at tip - 6. A.. retusus  




Tubercles gray- or blue-green, usually bearing a spiniferous areole near the tip, not curved towards tip, flowers white to pinkish-white - subsp.. retusus




Tubercles yellow-green, never bearing a spiniferous areole near the top, curved towards tip, flowers yellow to creamy white - subsp.. trigonus

Acknowledgments (top)

We wish to thank the following for providing facilities and financial support for this project. Cante A C, Desert Botanical Garden, World Wildlife Fund, and the Cactus and Succulent Society of America Research Fund. We also thank Dr Hector Hernández for making scanning electron micrographs of the seeds of several species of Ariocarpus. George Hinton developed the computer software for generating the maps, his expertise and advice is much appreciated. A portion of this study was conducted under the direction of Cante A C, San Miguel de Allende, Gto, with the authorization of La Dirección General de Aprovechamiento Ecologico de los Recursos Naturales del Instituto Nacional de Ecologica, SEDESOL, oficios numero 01516 y 05860 We also wish to thank the Mexican government for issuing collecting permit #D00 (2) 2532 to allow us to collect a small number of Ariocarpus plants for study, especially during flowering season. The plants are presently growing at Cante A C and the Desert Botanical Garden.

References (top)

Anderson, E.F. 1958. A recent field trip in search of Ariocarpus Cact. SuccJ. (U.S.): 30:171-174.

Anderson, E.F. 1960. A revision of Ariocarpus (Cactaceae) I The status of the proposed genus Roseocactus Amer. J. Bot. 47(7):582-589.

Anderson, E.F. 1962. A revision of Ariocarpus (Cactaceae) II The status of the proposed genus Neogomesia Amer. J. Bot. 49(6):615-622.

Anderson, E.F. 1963. A revision of Ariocarpus (Cactaceae) III Formal taxonomy of the subgenus Roseocactus Amer. J. Bot. 50(7):724-732.

Anderson, E.F. 1964. A revision of Ariocarpus (Cactaceae) IV Formal taxonomy of the subgenus Ariocarpus Amer. J. Bot. 51(2):144-151.

Backeberg, C. 1966. Kakteenlexikon Veb Gustav Fischer, Jena.

Barthlott, W. 1979. Cacti Stanley Thornes, Cheltenham.

Berger, A. 1925. Roseocactus, a new genus of Cactaceae J. Wash.(D.C.) Acad. Sci. 15:43-48.

Bravo H.,H., and H. Sánchez-Mejorada 1991. Las Cactáceas de Mexico Vol 2, p. 260. Universidad Nacional Autónoma de Mexico, Mexico City.

Fitz Maurice, W.A., and T. E. Davis. 1987. Fieldnotes Cact. Succ. J. (U.S.): 59:144-145.

Haage, W. 1977. Curt Backeberg's Cactus Lexicon Trans. of 3rd ed. Blandford Press, Poole.

Hernández, H., and E.F. Anderson. 1992. A new species of Ariocarpus (Cactaceae) Bradleya 10:1-4.

Hernandez-Barrera, J.T. 1992. Aspectos biologicos de Ariocarpus agavoides (Castañeda) Anderson. Cact. Suc. Méx. 37:40-45.

Hirao, H. Color encyclopedia of cacti .Tokyo: Seibundo Shinkosha PubI. Co., Ltd.

Hunt, D. 1989. Cactaceae. In S.M. Walters et al., eds. The European Garden Flora. Vol. 3, p 277. Cambridge University Press.

Hunt, D.1991. Notes on miscellaneous genera of Cactaceae (with N.P. Taylor). Bradleya 9:83.

Hunt, D.1992. CITES Cactaceae checklist. Royal Botanic Gardens, Kew.

IUCN. 1994 IUCN. Red List categories. IUCN. Gland.

Lee. M.H. 1937. Ariocarpus Cact. Succ. J. (U.S.): 9:71-77.

Lemaire, C. 1839. Cactearum genera nova et species nova en horto Monville. Paris.

Lux, A., and R. Stanik. 1983. Hybridisation in der Gattung Ariocarpus Scheidweiler Kakt. Sukk. 34:267- 271.

Marshall, W.T. 1946. Revision of the genus Ariocarpus Cact. Succ. J. (U.S.): 18:55-56.

Martinez, J.G., H. Suzán A., and C.A. Salazar O. 1993. Aspectos ecologicos y demograflcos de Ariocarpus trigonus (Weber) Schumann. Cact. Suc. Mex 38:30-38.

Salm-Dyck, J. de. 1850. Cacteae in horto Dyckensi cultae. Anno 1849. Apud Henry & Cohen, Bonn.

Scheidweiler, M.J. 1838. Descriptio diagnostica nonnullarum Cactearum quae a domino Galeotti in provinciis Potosi et Guanaxato regni Mexicani inveniuntur. Genus Ariocarpus. Bull. Acad. Sci. Brux. 5:491- 492.

Schumann, K. 1898. Gesamtbeschreibung der Kakteen. J. Neumann, Neudamm.

Stuppy, W., and N.P. Taylor. 1989. A new variety of Ariocarpus fissuratus (Cactaceae). Bradleya 7:84-88.

Wettstein, R. von. 1933. Die Bildung von Assimilationsorganen bei Kakteen und die Erscheinung der Irreversibilitat. Kakteenkunde 1933:11-13.

---------- end of page ----------