Ariocarpus agavoides (Castañeda) E. F. Anderson, Amer. Jour. Bot. 49: 615. 1962. Neogomesia agavoides Castañeda, Cact. Succ. Jour. 13: 98. 1941.

In 1941, the late Marcello Castañeda, an engineer for the state of Tamaulipas, Mexico, discovered near the town of Tula what he believed was a new genus and species of cactus (fig. 1). Aided by the late W. T. Marshall of the Desert Botanical Garden at Tempe, Arizona, he published the name Neogomesia agavoides (originally spelled agavioides) in the Cactus and Succulent Journal (Castañeda, 1941 ). Castañeda believed this plant constituted a distinct genus closely related to Ariocarpus. He said: 'This genus is nearest to Ariocarpus from which it differs in its long-tubed flower and its clavate, red fruit which is prominently in sight from the first' (Castañeda, 1941).

Until recently Neogomesia has been rare in cactus collections and quite expensive. Cactus enthusiasts are reported to have paid up to $35 apiece for these plants. I had the pleasure of meeting and traveling with Marcello Castañeda a short time before his death. Through his assistance and the aid of other persons interested in Mexican cacti I had the opportunity to visit the type locality of Neogomesia and to obtain specimens for study. The purpose of my research project was to determine the relationship of Neogomesia to the earlier proposed genus Ariocarpus which I have been studying for several years (Anderson, 1958, 1960, 1961 ). I presently consider Ariocarpus to be composed of six species: A. trigonus, A. fissuratus, A. kotschoubeyanus, A. retusus.. A. scapharostrus, and A, agavoides. The first four species listed above were discussed in earlier articles (Anderson, 1960, 1961 ). I collected the fifth species, A. scapharostrus. only recently, and further study of it is necessary. However, there is little doubt that the plant is an Ariocarpus.

Research on this paper was conducted at Pomona College and the Rancho Santa Ana Botanic Garden, both in Claremont, California. The type specimen of Neogomesia agavoides was borrowed from the Dudley Herbarium, Stanford University, and studied along with materials which I collected in Mexico. My collections of A. agavoides, numbered 1186, 1616, and 1736, are deposited in the Herbarium of Pomona College. Specimens of the other species of Ariocarpus were compared with the plants of Neogomesia. Both similarities and differences were noted, and conclusions were then drawn at the completion of the study.

Seeds

Plants collected in Mexico had a few seeds hidden in the trichomes, and other seeds were obtained later from plants in cultivation. Buxbaum (1955) believed certain seed characters to be of 'taxonomic value', such as the hilum and strophiole, the shape, the color of the integument, and the structure of the outer testa. Previously the internal structure of the testa has not been analyzed.

The seeds of Ariocarpus agavoides are 1-1.5 mm long, black, pyriform except for the presence of the large flattened hilum, and verrucose (fig. 2). They appear to be similar to those of the other species of Ariocarpus (fig. 3). The testa of the seed is 3-layered (fig. 15, 16). The 2 innermost layers are thin-walled and flattened. The enlarged outer cells of the testa (seed coat) form the verrucose surface of the seed. The layer consists of hemispheroidal cells with thickened walls which have small ridges on their outer surfaces. Numerous pits radiate outward from the lumen (center) of each outer cell. The proximally located lumen and pits appear blackish. The testa structure of A. agavoides differs little from that of the other species of Ariocarpus, but the pits radiating from the lumina extend slightly farther in the former species. The seed sizes of Ariocarpus may be compared in Table 1.

Table 1. Seed Sizes of Ariocarpus

Seedlings

Because seeds were not available, seedlings of A. agavoides have not been studied by other authors. However, Buxbaum (1950), Meyran (1956), and Anderson (1960, 1961) have studied the seedlings of other species of Ariocarpus

From the emergence of the embryo from the seed coat through the development of several tubercles the seedlings of A. agavoides (fig. 4) resemble those of other species of Ariocarpus (fig. 5). The cotyledons are united and the tubercles arise between them. Three to six minute spines occur at the tip of each tubercle and persist there for several months.

Habitat

 Ariocarpus agavoides is known from only the type locality at an altitude of 4,000 feet on the western edge of the Sierra Madre Oriental near the village of Tula, Tamaulipas. Like other species of Ariocarpus. A. agavoides grows on rocky limestone hills sometimes shaded by associated plants (fig. 6). Because of its small size it is difficult to see except when it flowers (fig. 7) (fig. 8). The soil is nearly neutral, pH 7.1. Associated plants are scattered on the hillside, and they, include common Chihuahuan Desert species Such as Prosopis sp., Koeberlinia spinosa, Jatropha spathulata, and Condalia sp. Associated cacti are Thelocactus tulensis and Neolloydia conoidea.

Significant internal Characters

Extensive mucilage-containing structures, believed to be unique to Ariocarpus (Anderson, 1960, 1961) occur also in A, agavoides (fig. 19). Each plant has several canals and a large central reservoir at the base of the stem. The mucilage is probably an aid to water storage, for during the dry months it practically disappears. This mucilage was analysed for its chemical constitution. A test for pectic substances using 0.5% ammonium oxalate was negative, but the Benedict's solution test for reducing sugars was positive. Paper chromatographic studies by David L. Walkington of Orange State College (oral communication) showed that these sugars are mainly glucose and arabinose. They are probably the main component of the mucilage. Tests were also made to ascertain the presence or absence of free nuclear material, starch grains, crystals, and lipids. Relative amounts of these substances in the mucilage of the species of Ariocarpus are shown in Table 2.

Table 2. Chemical Contents of the mucilage of Ariocarpus

Some genera of cacti have distinctive features of the outer cell layers such as unequal size of cells, heavy thickening of walls, and the presence or absence of druses (Boke, 1959, 1960). The outer layers of cells vary in Ariocarpus. The epidermis of A. fissuratus (fig. 9) is composed of thin-walled irregularly-shaped cells. The hypodermis may consist of more than one layer of cells with thickened walls. Ariocarpus retusus. and A. trigonus are similar to A. fissuratus. In A.agavoides the epidermal cells have thickened walls; some hypodermal cells also show thickening, but generally the hypodermal layer is indistinct from the cortex (fig. 10). None of the species of Ariocarpus have crystalline material except in the mucilage.

Areoles

Earlier studies (Anderson, 1960, 1961 ) showed that although there is some variation of areolar structure, all flowers are borne from areolar regions at the bases of the tubercles (fig. 11). Buxbaum (1950) discussed the development of the tubercle and the areole of A. agavoides and believed that the area of the tubercle distal to the areole should be termed the leaf. He thought the development of tubercles in this species to be different from that in the other species of Ariocarpus. Backberg (1959) noted the peculiar changing of position of the areolcs on young and mature tubercles.

Observations of collected specimens showed that the areolar region, indicated by a mass of trichomes, varied in distance from the base of the tubercle according to its age. Occasionally 2-3 small, flexible spines develop from the areolar region on mature tubercles (fig. 12). These spines differ from the firm spines (feste Stacheln) reported earlier by Backeberg (1958). However, he fails to mention this character in a later volume (Backeberg, 1961). On young tubercles the areolar region is at the base, and flowers and fruits arise from it during this period (fig. 13) (fig. 14). These observations indicate that the spines and the flowers are produced from the same areolar region (fig. 14). This is in contrast to the highly dimorphic condition of the areoles of A. retusus. reported earlier (Norman H. Boke personal correspondence; Anderson,1960, 1961). Measurements of the distance of the areolar region from the tubercle tip in relation to the total tubercle length show little change in the distance between areole and tip in young and fully mature tubercles (fig. 17). This indicates that growth and elongation of the tubercles occur between the areolar region and the base of the tubercle. Figure 18 is a diagrammatic cornparison of the development of areoles and tubercles of Ariocarpus. The basic difference within the group is the place of growth and elongation of the tubercles. However, the original areolar areas are the same, and the method of areole development and tubercle elongation in A. agavoides is probably one of several variations of tubercle development in Ariocarpus.

Chemistry

Several workers reported the presence of alkaloids in some species of Ariocarpus (Ewell, 1896; Schultes, 1937a 1937b; Reti, 1950). No studies of the chemistry of A. agavoides have been published. Gordon A. Alles (oral communication) made paper chromatograms of the alkaloids of several species of Ariocarpus including A. agavoides. He found that hordenine (anhaline) is the primary alkaloidal constituent of these plants. At present, hordeine is known to occur in few other cacti, so its presence in A. agavoides and other species of Ariocarpus indicates chemical similarities. Recently I have made comparative paper chromatograrns for Ariocarpus agavoides, A. retusus., and A. fissuralus. These chromatograms were of amino acids and 'secondary substances'. No major differences could be noted, although the chromatographic patterns of Ariocarpus could be distinguished readily from those of other genera such as 0bregonia, Thelocactus. and Lophophora. This is further evidence of chemical similarities among the several species of Ariocarpus.

Flowers

Castañeda (1941) did not indicate the exact place of origin of flowers in A. agavoides except "flower and fruit arising from the lower part of the areoles of nascent tubercles;. Opinions concerning relationships within the Cactaceae based on the position of the flower have varied. Backeberg (1958, 1961) placed Neogomesia agavoides in his "Boreoechinocacti" along with Echinocactus, Astrophylum. Leuchtenbergia, and Thelocactus because the flowers arise from the spine-bearing areoles. Castañeda (1941), Marshall and Bock (1941), and Buxbaum, (1957, 1958), on the other hand, believed Neogomesia was related to Ariocarpus because the flowers arose from the lower parts of the tubercles.

Mature specimens of A. agavoides with flowers and fruits were dissected to determine the point of flower origin. Flowers arise from within the masses of trichomes of the areolar regions at the bases of young tubercles which at the time are not over 15mm in length (fig. 13). In the other species of Ariocarpus the point of origin of the flower is from a flower bearing areolar region at the base of the tubercle (fig. 11). In A. agavoides the undivided areole is also at the base of the tubercle at the time of flowering and fruiting (fig. 13) (fig. 14). Later growth displaces the areolar area to a more distal position. All species of Ariocarpus (including A. agavoides) flower from September through early December following the rainy season. Fruits mature early the following summer after the rains have begun again. Flowers of A. agavoides were described by Castañeda (1941), who believed their shape distinguished them from those of Ariocarpus. He reported the flower to be long-tubed, The author measured, photographed, and compared flowers of A. agavoides (fig. 13) with those of other species of Ariocarpus (fig. 11), and he could note little difference between the 2 groups, and especially so if the flowers were fully opened. Flowers tightly surrounded by tubercles were prohibited from opening completely. Flower characters which all species of have in common are diurnal anthesis, persistence of anthesis for 1 or a few days, naked ovary except for the trichomes arising from the arcole, eritire perianth margins, attenuate perianth tips, and similar color of stamens and pistils. Ariocarpus agavoides, A. fissuratus. most plants of A. kotschoubeyanus. and A. scapharostrus have magenta flowers. Ariocarpus retusus. has white flowers and A. Irigonus has yellow flowers.

Pollen

The pollen of A. agavoides is tricolpate (having 3 elongate apertures) (Fig. 20A) and rounded or slightly ovate. The hump-like pila form a pattern on the surface of the exine (Fig. 20 A, B), and the colpae are 4-5 tinies longer than wide (Fig. 20C). Pollen of the other species of Ariocarpus is similar. However, there are variations in the patterns of the pila among the several species. Ariocarpus retusus  has a well formed reticulate pattern, whereas the patterns of the other species including A. agavoides is more linear than reticulate ( Fig, 20 D, E, F, G). Comparative sizes of pollen grains of Ariocarpus are shown in Table 3. There is sufficient variation of pollen grains among genera of the Cactaccae so that similar pollen may indicate a natural affinity.

Table 3. Diameters of pollen grains of Ariocarpus

Chromosomes

Previous chromosome counts in the cacti indicate a consistent basic number of x=11 (Johansen, 1933; Beard, 1937; Katagiri, 1953; Darlington and Wylie, 1955). The author (1960, 1961) reported the root-tip chrornosome numbers of A. fissuratus and A. retusus. to be 2n=22. Root-tip chromosomes of A.agavoides were also found to be 2n=22 (Fig. 21). No variation in this number was noted.

Fruits

Castañeda (1941) stated that Neogomesia differed from Ariocarpus because it had a 'clavate, red fruit which is prominently in sight from the first', whereas the fruit of Ariocarpus is usually white or greenish and appears only at maturity.

Fruits of A. agavoldes are visible from the second or third week after fertilization, and they remain fleshy and brightly colored for about 8 months (Fig. 22). As they mature, the fruits become brownish and dry, and soon they disintegrate, permitting the seeds to escape (fig. 24) The differences between the fruits of A. agavoides and the other species of Ariocarpus are in color and sometimes shape. All of the species have fleshy fruits with persistent perianth parts (Fig. 23). Occasionally pink or reddish fruits are found on A. retusus.. A. fissuratus. and A. trigonus. but the usual color is white or light green. Ariocarpus kotschoubeyanus and A. agavoides do not have the extensive development of trichomes found in the other species, and their fruits are exposed to the environment soon after fertilization. The development of anthocyanins in the fruits of A. kotschoubeyanus and especially A. agavoides is probably an adaptation protecting the fruits from the sun. The white, greenish, or pink fruits, in contrast, remain buried and protected within the trichomes for a greater part of their development. The color of the immature fruit can often be useful as a diagnostic character on the specific level. Some Ariocarpus fruits protrude from the trichome mass at maturity (Fig. 25) and are dry and brown. The clavate shape of the fruits of A. agavoides is not different from the shape of the other fruits of Ariocarpus. The fruits of A. agavoides have been reported to be rounded, but all fruits measured by the author were at least twice as long as wide (Fig. 14).

Compatibility. Most species of Ariocarpus were intercrossed in controlled experiments, and all were found to be compatible with one another. Ariocarpus agavoides was crossed with A. kotschoubeyanus, A. fissuratus, and A. retusus. In all cases fruits and seeds were formed. Selected samples of seeds were planted and were found to be viable. This indicates that the species of Ariocarpus are interfertile and that they have close genetic relationship.

Conclusions

Ariocarpus agavoides is similar to the other species of Ariocarpus in the characters of seeds, seedlings, habitat, some internal characters, chemistry, flower structure, flowering period, pollen, fruit structure, and compatibility. It differs in areole structure, but I believe that its areole development is similar to the areole development patterns in the other species of Ariocarpus. The presence of a mucilage system known only in Ariocarpus is indicative of a close affinity within the group. Castañeda's (1941) 2 main criteria for establishing the genus Neogomesia are not sufficient to warrant a genus separate from Ariocarpus, and I conclude that Neogomesia should be included with the genus Ariocarpus, creating a new specific combination A. agavoides.

Literature Cited

ANDERSON, E. F. 1958. A recent field trip in search of Ariocarpus. Cact. Succ. Jour. 30: 171-174.

ANDERSON, E. F. 1960. A revision of Ariocarpus (Cactaceae). 1. The status of the proposed genus Roseocaclus. Amer. Jour. Bot. 47: 582-589.

ANDERSON, E. F. 1961. A study of the proposed genus Roseocactus. Cact. Succ. Jour. 33: 122-127.

BACKEBERG, C. 1958. Die Cactaceae. Gustav Fischer,Jena. 1: 1-638.

BACKEBERG, C. 1959. Notes on Neogomesia, Pediocactus. Utahia, Navajoa, and Pilocanthus (Cactaceae). Nat. Cact. Succ. jour. 14: 63-67.

BACKEBERG, C. 1961. Die Cactaceae. Gustav Fischer, Jena. 5: 2631-3543.

BEARD, E. C. 1937. Some chromosome compliments in the Cactaceae and a study of meiosis in Echinocereus papillosus. Bet. Gaz. 99: 1-21

BOKE, N. H. 1959. Endomorphic and ectomorphic characters in Pelecyphora, and Encephalocarpus. Amer. Jour. Bot. 46: 197-209.

BOKE, N. H. 1960. Anatomy and development in Solisia. Amer. Jour. Bot. 47: 59-65.

BUXBAUM, F. 1950. Morphology of cacti. Section 1. Roots and stems. Abbey Garden Press, Pasadena, California.

BUXBAUM, F. 1955. Morphology of cacti. Section III. Fruits and seeds. Abbey Garden Press, Pasadena, California.

BUXBAUM, F. 1957. Die systematische Einteilung. In H. Krairiz, Die Kakteen. System ( 1-4).

BUXBAUM, F. 1958. The phylogenetic division of the subfamily Cereoideae, Cactaceae. Madroño 14: 177-206.

CASTANENDA, M. 1941. A new cactus. Cact. Succ. Jour. 13: 98-99.

DARLINGTON, C. D. and A. P. WYLIE. 1955. Chromosome atlas of flowering plants. George Allen and Unwin Ltd. London..

EWELL, E. E. 1896. The chemistry of the Cactaceae. Jour. Amer. Chem. Soc. 18: 624-643.

JOHANSEN, D. A. 1933. Recent work on the cytology of the cacti. Cact. Succ. Jour. 4: 356.

KATAGIRI, S. 1953. Chromosome numbers and polyploidy in certain Cactaceae. Cact. Succ. Jour. 25: 141-143.

MARSHALL, W. T. and T. M. BOCK, 1941. Cactaceae. Abbey Garden Press, Pasadena, California.

MEYRÁN, J. 1956. Notas sobre plántulas de Cactáceas. Cact. Succ. Mex. 1: 107-112.

RETI, L. 1950. Cactus alkaloids and some related compounds. Progress in the chemistry of organic natural products. Springer. Wien.

SCHULTES, R. E. 1937a. Peyote and plants used in the peyote ceremony. Bot. Mus. Leafl. Harvard Univ. 4: 129-152.

SCHULTES, R. E. 1937b. Peyote (Lophophora williamsii) and plants confused with it.  Bot. Mus. Leafl. Harvard Univ. 5: 61-68.

1. A more technical article dealing with Neogomesia was published earlier in Amer. Jour. Bot. 49, 615-622, 1962, under the title A revision of Ariocarpus (Cactaceae) II The status of the proposed genus Neogomesia.

2. This study was made possible through a grant to Pomona College, Claremont, California by the late Dr Gordon A Alles of Pasadena, California. The author also wishes to thank Dr Lyman Benson for his advice and criticism.