Tagua


Related species
Description and phenology
Distribution, abundance and ecology
Uses and economic potential
Souvenir manufacturing
Tagua as a substitute for ivory
Collection methods and yields
Propagation and cultivation methods
Research contacts
Commercial contacts



Rodrigo G. Bernal & Gloria Galeano

Family: Palmae (Arecaceae)

Species: Phytelephas aequatorialis Spruce

Synonyms: Palandra aequatorialis (Spruce) O.F. Cook

Common names: Tagua, corozo, coroso (past trade name); marfil vegetal, vegetable ivory (English); avorio vegetate (Italian); Elfenbeinnuss, vegetablische Elfenbein (German); ivoire végetal (French); tagua, corozo (Colombia, Ecuador); jarina (Peru, Brazil).

Related species


The genus Phytelephas belongs to the palm subfamily Phytelephantoideae, containing the three genera Ammandra, Aphandra, and Phytelephas (Barfod 1991b), all of which produce vegetable ivory. All the species that were commercially exploited in the past, and also the best known in the group, belong to the genus Phytelephas, which includes 4 species: Phytelephas aequatorialis Spruce, P. macrocarpa R. & P., with three subspecies - ssp. macrocarpa, ssp. tenuicaulis Barfod, and ssp. schottii H. Wendl.; P. tumacana O.F. Cook; and P. seemannii O.F. Cook. Most vegetable ivory came from P. aequatorialis (Ecuador), P. tumacana and P. macrocarpa ssp. schottii (Colombia), and P. seemannii (Colombia and Panama).

The genus Ammandra contains two species: Ammandra decasperma O. F. Cook, from the Pacific lowlands of Colombia, and Ammandra dasyneura (Burret) Barfod from northwestern Amazônia Colombia. The genus Aphandra includes one species, Aphandra natalia (Balslev L Henderson) Barfod, from Amazonian Ecuador. None of these species played a significant role in the vegetable ivory trade.

Description and phenology


Phytelephas aequatorialis is a solitary, unarmed, erect palm, up to 12 m tall. The stem is 25-30 cm in diameter, and is conspicuously marked by prominent leaf scars spirally arranged. The stem is crowned by up to 30 large, pinnate (feather-like) leaves, to 8 m long, the dead leaves hanging for some time under the crown; each leaf has 100-140 narrow leaflets on each side, the largest ones up to 90 × 6.5 cm, regularly arranged on one plane, or middle leaflets inserted in groups of 2-7, and then arranged on several planes.

All species in the subfamily Phytelephantoideae are dioecious, with male and female inflorescences produced by different trees. Inflorescences are produced among the leaves, and are enclosed in bud in two sheaths (bracts): a basal, outer, coriaceous, flattened profil, and an inner, coriaceous peduncular bract that encloses the flowers until flowering time. Male and female inflorescences are very different in shape and structure. In Phytelephas male inflorescences are long, cylindrical, fleshy, and spike-like, up to 150 cm long, with 300-500 cream-colored flower clusters spirally arranged and densely crowded; each cluster has two pairs of male flowers inserted on a short branch; individual flowers measure 1 cm long, and have reduced petals and 500-700 stamens. Female inflorescences have a 40-50 cm long, club-shaped stalk, with 20-30 stalkless female flowers crowded at the apex; each female flower has 6-8 sepals up to 15-20 cm long, a pistil with a long style and 5 stigmas up to 9 cm long.

The infructescence is almost spherical and head-like, 30 cm in diameter, usually with 15-20 densely crowded fruits. Individual fruits are conical, often five to six-edged by mutual pressure, 10-15 cm in diameter. The epicarp is thick and woody with numerous spiny projections; the mesocarp is thin, fleshy, oleaginous and yellow to orange. Each fruit has 5-6 seeds, 5 × 3 cm on average, often wedge-shaped, but very variable in size and shape; the endosperm is homogenous, fluid when young, later gelatinous and finally very hard and white, ivory-like at maturity, with a small central cavity (which is often lacking). Seedling leaves are feather-like.

Other species in the genus are quite similar in general appearance. Phytelephas seemannii is very distinctive in having a prostrate, creeping stem; P. macrocarpa ssp. macrocarpa has a prostrate or erect stem; ssp. schottii has a subterranean or creeping stem; and ssp. tenuicaulis is very distinctive in having slender, clustered stems. Besides P. aequatorialis, all other species have leaflets regularly arranged in one plane. The size of the central cavity is variable in the different species, but seems to be smaller in P. aequatorialis.

Flowering and fruiting of P. aequatorialis are reported as continuous year round (Acosta Solís 1948), although Barfod (1991b) has suggested that flowering of Phytelephas species is synchronized with the dry season in areas with a seasonal climate. Actually, in areas where tagua nuts are gathered, production is continuous, although a peak in fruit production does exist in the dry season.

Distribution, abundance and ecology


Species of Phytelephas are distributed along the Pacific lowlands of Panama, Colombia, and Ecuador, the valley of the Magdalena River, in Colombia, and northwestern Amazônia in Colombia, Ecuador, Peru, and Brazil. P. aequatorialis is distributed in western Ecuador, from the border with Colombia south to the Province of Azuay. P. tumacana is restricted to the Tumaco area, in southwestern Colombia, where it is now endangered (Bernal 1989). P. seemannii grows along the lowlands of western Colombia (Chocó), and eastern and Central Panama. P. macrocarpa is distributed in northwestern Amazônia in Colombia, Ecuador, Peru and Brazil, and one subspecies, ssp. schottii, is endemic to the Magdalena River basin in Colombia and adjacent areas of the Catatumbo River basin (Barfod 1991b).

Phytelephas species grow mostly on alluvial soils under 500 meters, with soil temperatures usually over 18C but P. aequatorialis and P. macrocarpa ssp. schottii often reach 1,000-1,200 m. All species are better adapted to humid and shady areas, mostly with over 2,500 mm of year rainfall, but P. macrocarpa ssp. schottii is often found on steep slopes in rather dry areas in northeastern Colombia.

On alluvial, often seasonally flooded soils in the Pacific lowlands and northwestern Amazônia species of Phytelephas often form large, rather homogeneous stands, called taguales in Colombia and Ecuador. These taguales range in area from less than one hectare to 25 hectares or more, with up to 240-500 palms per hectare. Estimates of the areas covered with taguales in Colombia and Ecuador have not been made. Although the extension of taguales was probably favored by man during the early boom of tagua, flooding rivers are perhaps the major dispersers of the heavy seeds along the floodplains. During an overnight flooding of; a small river in western Colombia, about thirty seeds were deposited in a 0.1 ha plot of taguale (R. Bernal, pers. obs.).

Rodents, like paces (Agouti paca) and agoutis (Dasyprocta), carry the seeds away from the taguale, and then eat the fleshy mesocarp, or bury the seeds for later retrieval. This behavior of rodents probably accounts for dispersal of tagua beyond the floodplains.

In some areas palms are left in pastures after the forest is cleared, and they become the woody component of simple silvopastoral systems (Borgtoft-Pedersen & Balslev 1990). Under these conditions they continue to set fruit, but they do not regenerate. With the exception of some areas in the Santiago River basin, in NW Ecuador (Acosta Solís 1944a), and on the Mira River, in SW Colombia (More Mora 1990), Phytelephas species have never been extensively cultivated. In western Colombia, large areas of wild taguales were burnt in the last forty years to establish rice fields.

Uses and economic potential


Principal use



Principal use


The principal use of the tagua palm lies in the endosperm of its seeds. This material is hard and heavy, and has a cream color; when polished, it: is quite similar to true ivory (hence the name vegetable ivory), although the materials differ in their properties. Tagua softens when hydrated, recovering its hardness with drying, and dissolves when soaked in water for long periods (Acosta Solís 1944a); ivory remains hard in water. Tagua nuts are easy to polish and dye, and are suitable for carving figurines, chess pieces, scrimshaws, handles, and other items. The most important use in the past, however, was for button production (Acosta Solís 1944ab, 1971; Barfod 1989, 1991a).

The first statistics of tagua production come from Colombia for the period 1840-1841, when tagua "made up a minor percentage of Colombian exports" (Ocampo 1984). After the 1860's tagua became one of the 5 major export products of Colombia (Tovar Zambrano 1989) and one of the 5 major forest products of Ecuador (Acosta Solís 1944a). In the period 1875-1878, tagua made up 3.1 percent of Colombian exports (Tovar Zambrano 1989). For Ecuador, tagua became even more important. The first statistics for this country appear in 1865 (Acosta Solís 1944a). In 1929, during one of its peaks, tagua exports from Ecuador amounted to 25,791 MT, worth US$ 1.2 million (Acosta Solís 1944a). In today's figures, this would mean about US $15 million. Exports of tagua from Colombia declined in the 1920's and disappeared about 1935. Ecuador continued to produce tagua, but exports declined after 1941, and tagua disappeared from trade almost completely by about 1945 (Barfod 1989).

Tagua button manufacturing was a major industry during the late 19th century and early 20th century. In the 1920's, 20% of all buttons produced in the United States were made of tagua (Acosta Solís 1944a). By the 1930's, plastics began to replace tagua for button manufacture, until tagua industry almost completely disappeared (Barfod 1989). Button production never died off completely, however, and a small button industry survived in Ecuador. This continued to produce the tagua disks from which tagua buttons were manufactured in Japan, West Germany, and Italy (Barfod et al. 1990).

In 1990 Conservation International, based in Washington, D.C., launched the Tagua Initiative, a program aimed toward linking the tagua producers in priority rainforest areas, with the international markets. Two clothing companies in the United States initially joined the program by buying a first lot of one million buttons. Other companies have joined the program since. Revenues from the Tagua Initiative are reverted through the local partner NGOs into conservation and sustainable development programs in the areas where tagua is produced. This program is being developed now in two areas: the Santiago River basin, in NW Ecuador, by CIDESA; and the Pacific coast of NW Colombia, by Fundación Inguedé. Community management of tagua exploitation, as promoted by the Tagua Initiative, offers an attractive economic alternative requiring forest conservation by local inhabitants.

Souvenir manufacturing


Even after tagua disappeared almost completely from the button industry, it continued to be used by small-scale souvenir industries in Colombia and Ecuador. The Ecuadoran souvenir industry is based in the Andean town of Riobamba, where many different figurines are produced. These are sold in town, and in souvenir shops in other cities. In Colombia, the tagua souvenir industry is based in the Andean city of Chiquinquirá, where many different designs are produced. These souvenirs, however, are not sold outside Chiquinquirá (Bernal, in press). It is noteworthy that both in Colombia and Ecuador the tagua handicraft industry is best developed in areas where tagua palms do not grow, raw material being brought from other areas. The Ecuadoran industry has a tradition of hand carving not developed in Colombia, where most products are made in lathes. Several talented artisans are currently exploring tagua again, both in Colombia and Ecuador, and high-quality handicrafts are beginning to be produced.

Tagua as a substitute for ivory


Secondary uses



The "rediscovery. of a sustainable source of vegetable ivory in the times of the international ban on elephant ivory has aroused a growing worldwide interest in this exotic material. Fashioned into jewelry, tagua is gaining reputation, and watches, earrings, bracelets, and necklaces featuring tagua are now obtainable in some luxury shops. Also for small sculptures, tagua seems ideal as a substitute, at least in part, for elephant ivory. Vegetable ivory is not only quite similar to elephant ivory, but also, like true ivory, it comes from an exotic source in remote, wild areas. Indeed, the use of a substitute for ivory, which not only protects elephants, but also helps to save critical areas of South American rainforest, is quite attractive.

The endosperm of tagua, the vegetable ivory, is composed of large, thick-walled cells, whose main components are two long-chain polysaccharides - mannan A (45-48%) and mannan B (24-25%), cellulose (6-7.5%), and other cell contents (Aspinall et al. 1953, 1958; Timell 1957).

Secondary uses


Tagua shavings resulting from button production are ground into a flour, which is exported from Ecuador to the United States and Japan. Prices for flour in 1988 were US$ 1.50 per 100 lb (Barfod et al. 1990).

The leaves of Phytelephas aequatorialis and other species of Phytelephas are a very durable and widely used material for thatching.

The orangish, thin mesocarp that surrounds the nuts of all species in the genus is edible, and is considered a delicacy where tagua palms are abundant. It's flavor is vaguely reminiscent of coconut. This mesocarp is also actively sought by game rodents, which play an important role in the diet of local communities. Thus, commercial exploitation of the mesocarp might not be advisable, as it would further reduce an important source of protein for rural inhabitants.

No information on the nutritive properties of the mesocarp is available. Although both the outer and inner mesocarp are edible, it is the inner, leathery layer that is really appreciated by local inhabitants, since the outer, powdery mesocarp is rather insipid. Although some oil is present in the mesocarp (Patiño 1977), the thinness of this layer makes it economically unattractive compared to other potential oil palms. Barfod (1991b) states that unless this oil proves to have valuable properties, its exploitation is probably not profitable. It is probably better for tagua gatherers to leave the mesocarp to the rodents, receiving meat and peeled nuts in exchange.

The endosperm of tagua is a clear liquid when the seed is very immature, and is a refreshing drink in the forest. When the fruit is maturing it becomes thicker and gelatinous, and is also edible, tasting like coconut in the same stage of development; it is also considered a delicacy in the areas where palms are abundant.

The palm heart of Phytelephas aequatorialis has been reported as a delicacy consumed in some areas of western Ecuador (Acosta Solís 1944a).

Collection methods and yields


Tagua nuts are produced only by the female palms. When ripe, the fruit breaks apart and the woody-spiny epicarp disintegrates, dropping the nuts onto the ground, still covered by the orange fleshy mesocarp. This mesocarp is eaten by rodents in situ, or some nuts are taken away and hidden as reserves. Thus, tagua is gathered from the ground, after rodents have taken their part. The heavy nuts are collected in sacks, or in baskets woven with forest fibers, and are taken home, sometimes in dugout canoes when the taguale is far away.

Processing of the nuts involves drying, which can be done outdoors, under the house, where nuts are piled, or, in wetter areas, indoors. Drying may take several weeks (at least 4 weeks in tropical wet climates, but artisans in the highlands of Colombia claim that less than one year of shade drying would crack the material). After drying, the outer shell of the nut is brittle, and can be removed by beating the nut at its point of attachment (a prominence familiarly referred to as the navel). Small pieces of shell sometimes remain adhered to the peeled seed, and they are removed with a knife or a machete. After peeling, the seed is still covered by the seed coat, a thin, chocolate-brown layer firmly adhered to the endosperm (the vegetable ivory).

Tagua in Ecuador is sold in the shell to factories, where peeling is done by hand, or by means of peeling machines developed by local industries.

For the button industry, each tagua nut is cut into three slices (sometimes only two are usable depending on the size of the central cavity) with a disk saw, and then dried overnight in an oven heated with tagua wastes. Each slice is then cut with a lathe into a disk, which is the basis for the button itself.

No detailed study of yields were made during the "golden age" of tagua. Nevertheless, figures given by Acosta Solís (1944a, 1948) for P. aequatorialis indicate a production of about 30 kg of dry, peeled tagua per palm per year. This figure, however, is probably the upper limit of production of wild palms, and the average may be one-half to one-third that amount.

With a density of 240-500 palms per hectare, one-half of which are females (assuming a sex ratio of 1: 1), yields of 1.27.5 MT/ha/yr can be expected from a wild, unmanaged taguale in the Ecuadoran lowlands. An intermediate figure of about 4 MT/ha seems a reasonable estimate. Claès (1925) estimated a minimum production of 2.25 MT/ha/yr for tagua in the Magdalena River basin, in Colombia. On the Pacific coast of Colombia tagua produces a low number of seeds per fruiting head, and the total production for this species is only 1-2 MT/ha.

After tagua lost its value in the 1930's and 1940's, large areas of taguales were replaced by bananas in Ecuador, and by African oil palm and rice in Colombia. To date, no figures are available on the extent of the surviving native stands of tagua in either country, so total production is hard to estimate. However, figures of tagua exports at its highest point can give an idea of the upper limit expected: in 1929, Ecuador exported 25,791 MT of tagua, roughly one-third of it to the United States. In the same year, Colombia exported 2,970 MT of tagua to the United States. Since many of the areas deforested in the last decades were extremely rich in tagua (El Oro province in Ecuador, the Tumaco and Urabá regions in Colombia), current potential production is well below these figures.

Assessments of the current extent of the taguales and the perspectives for market expansion are necessary before considering plantations as an option. Present demand is met by production from wild stands, but demand is increasing.

Propagation and cultivation methods


Except for small areas in the Santiago River in Ecuador (Acosta Solís 1944ab), tagua has never been cultivated and no information on cultivation techniques or procedures is available. Before the fall of tagua in the 1940's, campesinos in NW Ecuador scattered seeds in old agricultural plots. When the seed sprouted, they removed the weeds and regarded the taguale as established (Acosta Solís 1948). Under these conditions palms usually began to produce fruit after 15 years. Once the palm begins to set fruit, it will remain productive for up to one century.

Germination takes 4 to 9 months or even more, and the first seedling leaves are rather large for palms. No seedlings have been observed under palms growing in open areas, which suggests that shade is necessary for early development. Regeneration in taguales is abundant, and it offers a good basis for management.

It is probable that some cultural practices may easily increase tagua yields. Male palm trees are often eliminated in Ecuador, in order to leave room for more female palms in the taguale, a practice that apparently does not reduce yields. In dense stands, there is often a high percentage of leaf overlapping, a factor that may reduce seed production. Tagua seems to produce well under partial illumination (as under the thin canopy found in riverside forest). However, palms growing in full light usually have smaller leaves than palms growing in shade. Much basic research on the development and production of tagua under different conditions remains to be done. Much of this research is currently being conducted or planned in Colombia and Ecuador.

In Ecuador, tagua is usually associated with tree crops such as cacao (Theobroma cacao), breadfruit (Artocarpus altilis), and with timber trees like Cedrela odorata and Cordia alliodora. These associations are the result of agroforestry in what were probably large taguales along the alluvial lowlands. In western Colombia, tagua grows under the shade of mixed riverine forest, including a wide array of species used for timber (Apeiba aspera, Chlorophora tinctoria, Carapa guianensis, Tabebuia sp., Virola sp. and several Lauraceae), for firewood (Inga, Gustavia, and others), for canoes (Anacardium excelsum), and for wild fruits or nuts (Spondias mombin, Dipteryx panamensis, Inga sp., and others). One of the common species associated with tagua is black rubber (Castilla elastica), formerly an important forest product. No crops are usually associated with tagua in Colombia, although tagua is occasionally left in small agroforestry systems with chontaduro palm (Bactris gasipaes), almirajó (Patinoa almirajo) and bananas.

The management of tagua at high densities in agroforestry systems in Ecuador suggests that this could be a good alternative for increasing cultivation, if expanding demand requires it. Agroforesty not only has the advantage of offering local inhabitants other crops and forest resources (including game), but it is also the system that has been used for decades for tagua in Ecuador, and for other crops throughout the Neotropics.

Monocultures, on the other hand, would be more susceptible to pests and diseases, some of which already attack other cultivated palms in tagua producing areas. As a matter of fact, a beetle very similar (and perhaps identical) to the palm beetle (Rhynchophorus palmarum) attacks the stems of tagua, killing the palm (Acosta Solís 1948). R. palmarum is also the vector of the red-ring disease, that has destroyed thousands of coconut palms along the Pacific coast of Colombia and Ecuador. This disease is reported by local inhabitants of the Chocó, Colombia, to occur also on tagua, and it would be one of the potential threats to monocultures.

Research contacts


Anders Barfod, Botanisk Institut, Aarhus Universitet, 68 Nordlandsvej, DK 8240 Risskov, Denmark.

Rodrigo G. Bernal, Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Apartado 7495, Bogotá, Colombia. Phone 2442387, Fax 2682485.

Henrik Borgtoft Pedersen, Botanisk Institut, Aarhus Universitet, 68 Nordlandsvej, DK 8240 Risskov, Denmark.

Rodrigo Calero, CIDESA, Apartado 608, Suc. 12 de Octubre, Quito, Ecuador. Phone 433788, Fax 593-2502399.

Gloria Galeano, Instituto de Ciencias Naturales, Universidad Nacional de Colombia, Apartado 7495, Bogotá, Colombia. Phone 2442387, Fax 2682485.

Karen Ziffer, Conservation International, 1015 18th St. NW, suite 1000, Washington, DC, 20036, USA. Phone (202) 4295660, Fax (202) 887-5188.

Commercial contacts


Rodrigo Calero, CIDESA, Apartado 608, Suc. 12 de Octubre, Quito, Ecuador. Phone 433788, Fax 593-2502399.

Fundación Inguedé, Apartado 41595, Bogotá, Colombia. Fax 2353353.

Karen Ziffer, Conservation International, 1015 18th St. NW, suite 1000, Washington, DC, 20036, USA. Phone (202) 429-5660, Fax (202) 887-5188.