Pejibaye

Related species
Description and phenology
Distribution, abundance and ecology
Uses and economic potential
Collection methods and yields
Propagation and cultivation methods
Research contacts
Commercial contacts

Charles R. Clement

Family: Palmae (Arecaceae)

Species: Bactris gasipaes Kunth

Synonyms: Bactris gasipaes H.B.K., Guilielma speciosa Martius, G. utilis Oersted, G. chontaduro Triana, G. gasipaes (HBK) L.H. Bailey.

Common names: pejibaye (Costa Rica), pupunha (Brazil), chontaduro, cachipay (Colombia), chontaruro (Ecuador), pijuayo (Peru), gachipaes (Venezuela), peach palm, pewa nut (Trinidad).

Related species

The genus Bactris contains roughly 200 species: Numerous Bactris sp are occasionally used as edible fruit in northern South America, although none of them attain the importance of pejibaye.

Description and phenology

The pejibaye is a caespitose (multistemmed) palm that may attain 20+ m in height. Stem diameter varies from 15 to 30 cm and internode length from 2 to 30 cm. The internodes are armed with numerous black, brittle spines, although spineless mutants occur and have been selected for in several areas. The stem is topped by a crown of 15 to 25 pinnate fronds, with the leaflets inserted at different angles; the tender unexpanded leaves in the center of the crown form the palm heart, an important economic product. Among the axils of the senescent fronds the monoecious inflorescences develop. After pollinization the bunch may contain between 50 and 1000 fruits and weigh between 1-25 kg. Numerous factors can cause fruit abortion: poor plant nutrition, poor pollinization, drought, crowding, insects and diseases - which also contribute to low average bunch weights. Table 1 presents bunch composition data. The fruit that ripen have a starchy to oily, humid mesocarp; a fibrous red, orange or yellow epicarp; and a single endocarp with a fibrous, oily white endosperm. Individual fruit may weigh between 10 and 250 g and are potentially and traditionally the most important economic product. Table 2 presents fruit composition data.

Mora Urpí & Clement (1988) classified 8 landraces (primitive varieties) of pejibaye in Amazônia. and Clement (1986) classified one more and classified the Occidental pejibayes as a race. Fruit mesocarp composition vary considerably among the races, as occurs with other plant, bunch and fruit characters.

In central Amazônia. the pejibaye flowers between August and October, with the main harvest between December and March. There is frequently a minor harvest, with flowering in May-June and fruiting in August-September. In the Costa Rican Atlantic coast the main fruiting season is between August and November (More Urpí 1984).

Table 1. Average bunch composition data from several sources.

Table 2. Average fruit composition data from several sources.

Distribution, abundance and ecology

The pejibaye was distributed by the Amerindians in the pre-Columbian period, from its probable center of origin in southwestern Amazônia. and its undoubted center of domestication in western Amazônia (Clement 1988). At contact (1500 AD) it was found from southeastern Honduras in the north, down through Central America and around the Caribbean shores (with a small disjunction east of the mouth of the Magdalena River, where conditions are too dry) to the Atlantic port of São Luiz de Maranhão (Brazil), from where its distribution turns inland along the middle reaches of the eastern Amazon River tributaries, then up the Madeira River into Bolivia and reaches its southern most point near Santa Cruz de la Sierra, from where it extends northward along the Andes. On the Pacific coast it first appears in southwestern Costa Rica and extends down to extreme northern Peru. It thus has a disjunct distribution: Central America and the Pacific coast of South America, with a small area between Darien (Panama) and the Magdalena River (Colombia); northern South America and Amazônia, with some small areas of nonoccurrence due to local edapho-climatic conditions, principally dryness (Clement 1988).

The pejibaye is a truly domesticated species (Clement 1988). As such, it only occurs where man has planted it, although its phylogenetic relationships with other species are starting to be worked out and this research may identify one of the wild species as the original pejibaye. Until this is resolved, it can be considered a cultigen, i.e. with no positively identified wild ancestor in the local flora. Abundance therefore depends upon its local importance to man. Where it has a long history of use, for example in western Amazônia, and southern Central America, it can be extremely abundant, i.e. several thousand palms in the fields around Amerindian villages. Patiño (1963) reports on its importance in southern Costa Rica shortly after contact (1500 AD), when some Spanish mercenaries cut 50,000 stems of pejibaye at one site, for which they were later punished by the recently installed royal judicial system.

Because it is a domesticated species its ecological tolerance is much wider than that of any probable ancestor. It fruits in the following conditions: from 2 m above sea level to more than 1,200 m; with rainfall from 1800 to 5000 mm (although drought of more than 3 months reduces fruiting, therefore, rainfall distribution patterns are important); in moderately sandy to very clayey soils if these are well drained (bad drainage is unacceptable}, even where nutrient levels are poor, although it obviously yields better on richer soils; in full sun, since it does not tolerate shading. Because of rapid stem growth, however, it can be combined with other species, especially shade-tolerant:, smaller statured ones.

Uses and economic potential

Principal uses
Secondary uses

Clement & Mora Urpí (1987) and Clement & Arkcoll (1989) have discussed 5 products obtained from pejibaye with potential in the modern tropical market economy. Several of these uses are directly' related to fruit mesocarp composition (Tables 3, 4). What follows is condensed from the above mentioned references.

Principal uses

Fruit for direct human consumption

This is the traditional use for the pejibaye fruit and is the only one known in most places. The whole fruit is separated from the bunch and boiled in salted water for 30 to 60 minutes, to improve flavor and eliminate the irritating oxalate crystals (Arkcoll & Aguiar 1984) and trypsin inhibitor (Murillo et al. 1983). Piedrahita & Velez (1982) report that cooking for 15 to 20 minutes deactivates the peroxidase enzyme activity. The fruit are then peeled, halved and pitted, and are ready for consumption. Pejibaye fruit are frequently consumed at breakfast, or as an appetizer before later meals. When the fruit are dry, starchy and low in oil, they are tastier when accompanied by mayonnaise or a sauce. The cooked fruit can also be used whole in stews, or ground to a flour for use in a variety of preparations and pastries. About 40 pejibaye recipes have been collected by Calvo (1981).

Table 3. Average mesocarp composition data from several sources. (% dry weight)

While the flavor of fruit is quite variable, the typical pejibaye has a distinctive bland to strong flavor (depending upon carotenoid and oil content) of its own. A liking for this is acquired fairly easily but it is not exciting enough to generate new markets without good quality control and clever marketing (FAO 1986). Several authors have mentioned that pejibaye may taste like European chestnut (Castanea sativa) (Popenoe L Jimenez 1921; NAS 1975). Arkcoll & Aguiar (1984) found fruit with flavors similar to potato, maize and cassava and some which were sweet.

Table 4. Vitamins present in pejibaye mesocarp, from several sources. (in 100 g fresh mesocarp)

Mesocarp texture is also variable, ranging from that of a soggy potato to that of a good raw cashew (Anacardium occidentale). Texture is determined by water, starch, fiber and oil content, with watery, low starch types being soggy and starchy, low water types being floury to crunchy, even after cooking. Preferences vary from region to region. In Costa Rica the "best" fruit are dry, high-starch ones with a nutty texture; in Manaus the "best" fruit are moderately starchy, somewhat oily and humid, with a less nutty, though still firm, texture.

In Costa Rica whole or halved, pitted or unpitted, peeled or unpeeled fruit are marketed in brine in 500 and 1000 g jars or cans. Quality is extremely variable but they sell well when fresh fruit are unavailable. In Colombia dehydrated fruit have been prepared for market.

Fruit for animal ration

This is the major alternative use being studied at the moment. Because its dried fruit can partially or completely substitute maize for many use-, the pejibaye may be considered a tree cereal. This is especially true for animal rations, as pejibaye flour can substitute for the maize base that is generally used. Costa Rican researchers are currently leading this effort and Murillo & Zumbado (1986) have reviewed recent work with chicken feeds. Their team (Soto 1983, Zumbado & Murillo 1984, Cooz 1984, Loynaz 1985, Facuseh 1986, Espinoza 1986) has studied pejibaye preparation (autoclaving, extrusion, sun drying, etc.) and maize substitution levels for starter and primary rations for layers and broilers.

Using second-quality fruit (first-quality having high market value) their results showed that pejibaye can be-used as the principal energy source in primary rations, but should be used leas intensively in starter rations. Beat treatment is essential, especially for starter rations, to deactivate the trypsin inhibitor (Murillo et al. 1983). Heat treatment by extrusion is cheaper than other methods tested and is recommended. Moderate to high levels (30-100% substitution of maize) also allows the production of a cheaper meal (both for starter and for primary rations), at Costa Rican costs for maize and second-quality pejibaye.

One attractive option, that avoids drying and heat treatments, is to ensile the fruit on the farm and feed it directly to stock, possibly with a protein supplement. There is some suggestion that acid ferment may break down the antinutritional factors (Sangil 1985) and is appreciated by pigs (I. Araújo, pers. com.). Ensiling would also be an excellent way to cheaply store pejibaye fruit for animal feeds.

Fruit for flour

To avoid saturating the high-value fresh fruit market, it is interesting to consider the potential for developing other products for human consumption. Among Calvo's (1981) 40 recipes are several for making breads, cakes and other pastries from pejibaye flour. Tracy (1986, 1987) tested pejibaye flour, mixed with wheat flour, for bread in Costa Rica. Ninety percent wheat and 10 percent pejibaye gave bread dough with excellent baking characteristics, slightly less protein, more energy (from the oil) and more vitamin A (b-carotene). Eighty-five percent wheat and 15 percent pejibaye gave a slightly heavier dough, similar to whole wheat bread. Both were acceptable to a small group of consumers. This product might develop well in the "natural foods" market.

Tracy (1986) also tested several cake recipes with good results, both in the kitchen and on the market. Pejibaye flour must be mixed with wheat flour to make a good cake in Costa Rica. In the Manaus region, however, some pejibayes can be used pure, with excellent results. This may be due to the higher oil content.

Clement & Mora Urpí (1987) point out that pejibaye flour is quite similar to yellow cassava or maize flour and could substitute for these in many areas, with nutritional advantages over the cassava flour. This was one of the alternative products developed by the Amerindians, who appear to have domesticated the Putumayo and Vaupés "macrocarpa" landraces specifically for this purpose (Clement 1988). These landraces have extremely high starch and low oil levels in large fruit, excellent for making flour that stores well for long periods. This flour can also be fermented to make caissuma (Brazil) or masato (Peru), a soupy-gruel with a peachy flavor and mild alcoholic level. Flours can also be extruded to make a variety of attractive snacks that might find a large market.

Fruit for oil

Arkcoll & Aguiar (1984) were the first to point out pejibaye's oil potential. By searching for high oil fruit these authors eventually found fruit with 62% oil in the dry mesocarp and 34% oil on bunch weight. Clement & Arkcoll (1985) later pointed out that the oily fruit are more frequent in the more primitive (i.e. less selected) populations, especially in the "microcarpa" racial group. This seems to be due to Amerindian selection for starch, which, being negatively correlated with oil, means that the more selected landraces have low oil levels in the mesocarp. While per hectare oil yields [extrapolated from the best plants and assuming in vitro cloning (Clement & Arkcoll 1991)] are lower than established crops text African oil palm (Eleais guineensis) yields 5 T/ha/yr of vegetable oil], they are higher than these were at: a similar stage of development, suggesting that modern breeding and biotechnology methods could produce a new oil crop quickly (Clement & Arkcoll 1991).

Although there already are excellent oil crops for the wet tropics (i.e. E. guineensis and Cocos nucifera), the need for greater crop diversity is well demonstrated by the diseases and pests that limit or prevent their use in several countries (Meunier 1976). Pejibaye has an advantage over these two in that it will also provide a significant quantity of a high quality meal after oil extraction, suitable for humans and animals.

Unsaturated fatty acids are more common in pejibaye mesocarp oil than in African oil palm (Table 5), although the caiaué (Elaeis oleifera) has similar oils and is now being introgressed with oil palm to improve oil quality in that species (Hartley 1977). The fact remains, however, that unsaturated fats have a good market value at present, and are interesting from both a nutritional and industrial point of view.

As pointed out by Arkcoll & Aguiar (1984) and emphasized by Clement & Arkcoll (1985, 1991), most pejibayes have oil-separation problems when pressure is extracted. The oil, starch and water form an emulsion that must be solvent extracted. There are plants that have good separation characteristics, but using only these would severely reduce the genetic base of any improvement program. The oiliest pejibayes are relatively dry, however, so that separation may not be a severe problem with improved materials.

Table 5. Fatty acid composition data of pejibaye mesocarp oil. from several sources (% oil)

Heart of palm

Palmito, or palm hearts, are the only product for which pejibaye is currently grown on a commercial scale. There are more than 2000 ha planted in Costa Rica (Clement & Mora Urpí 1987) and already 300 ha in Amazônia, (Acre state). The palmito market is very volatile because of quality control in the main palmito species, the asset palm (Euterpe oleracea). This is extracted nearly free of cost from enormous natural populations in the Amazonas River estuary, which are being severely devastated by most companies. Quality control varies considerably from company to company; those practicing management rather than devastation having better control. Both of these factors may open a larger market for pejibaye, as plantation quality control has been good and plantations are managed rather than devastated.

Processing technology for the pejibaye palmito has been developed in Costa Rica and in Brazil. The most detailed studies are by Ferreira et al. (1982ab) at the Instituto de Tecnologia de Alimentos (ITAL), in Campinas, São Paulo, Brazil. Recently the University of Costa Rica Food Technology Research Center (CITA), in conjunction with a French research institute, has started working on technology to package, ship and commercialize fresh palmito (J. Mora Urpí, pers. com.), which should significantly expand the palmito market, as there are many more uses for fresh palmito than for the currently processed product.

Most pejibayes are spiny, both on the trunk and on the leaf petiole and rachis, which complicates extraction of the palmito. Several spineless populations have been found in western Amazônia, especially around Yurimaguas, Peru (Clement et al. 1988). Many other western Amazonian populations show variable frequencies of spinelessness or reduced spininess. There is also a spineless population in Costa Rica. This germplasm forms the genetic base for the improvement programs currently planned in Brazil and Costa Rica.

Both the apical and basal residue from palmito extraction have some potential use and require more intensive research to find new uses. The basal residue, just below the plant's apical meristem, is very tender and has a crispy texture. This could be made into a cream soup or be thin-sliced (transversally) as a substitute for bamboo shoots or deep fried to make chips. The apical residue is slightly fibrous leaf and petiole material, which can serve as a vegetable.

Secondary uses

Mora Urpí (1984) reports that the wood from the outer stem (the inner part is pith) is an attractive black with yellow fibers, strong, durable, easy to work and takes a good polish and suggests its use as parquet. Patiño (1963) reports numerous Amerindian uses for this wood, some of which are adapted for the local handicrafts industry. Because pejibaye grows so tall so quickly, plantations must periodically be renewed from stem offshoots. The wood obtained from these periodic renewals could supply these use options without endangering fruit production.