Among the options is pejibaye (Bactris gasipaes Kunth), also known as the peach palm (Hamilton 1987; Roecklein and Leung 1987; Clement 1990). The pejibaye is the Neotropic's only domesticated palm and is well suited to modern agriculture. Its palm heart has potential as a gourmet fresh vegetable. The palm heart is composed of the tender leaves that originate in and grow from the palm meristem and are consumed before they expand and green. Palm heart can supply both the tourist's desire for exotic foods and west coast markets for exotic vegetables. Its multiple potentials in modern agriculture have been examined by Mora Urpí (1984) and Clement and Mora Urpí (1987).
The pejibaye fruit has four uses: cooked fruit for human consumption, flour for bread and confectionaries, vegetable oil, and animal ration. Its unique flavor and texture, however, is unknown to the public outside of tropical America and would require a concerted marketing effort to promote acceptance.
The palm heart, in contrast, is well known and its consumption is expanding rapidly worldwide (Mora Urpí et al. 1991). Palm heart production and canning is already a rapidly growing industry in Latin America. The pejibaye palm heart has been on international markets since 1978 and is expanding in market share in both Europe and the United States (Mora Urpí et al. 1991). The United States currently (1989) imports 2000 t of canned palm heart (up from 900 t in 1980), valued at $4.5 million, of which 700 t is from pejibaye (U.S. Dept. Commerce 1990).
The palm heart market is based upon a processed product, which, although popular, has limited culinary potential because it is pre-cooked in a slightly salty, strongly acidic solution (Quast and Bernhardt 1978). Nonetheless, the value of current world trade in canned palm hearts is estimated to be close to $50 million. World consumption is much greater, because humid tropic populations consume the bulk of locally available fresh palm heart. Brazil, for example, is both the largest producer and largest exporter (70% of world trade): in 1988, 100,000 t of fresh palm heart were produced, but only 9,500 t were exported (Coradin and Clement 1989). Brazil's production is based upon the palm Euterpe oleracea, which occurs in large natural populations in the Amazon River estuary. E. oleracea has several characteristics that limit its potential for the fresh market. The most important of these is the presence of enzymes that discolor the palm heart upon contact with air.
The popularity of fresh palm heart in Latin America suggests that a strong demand can be created for this product in countries that currently know only the canned product. The pejibaye is ideally suited for the fresh market, because its palm heart does not discolor upon cutting and has good shelf life. Production of pejibaye for the fresh market could permit higher returns to growers. This includes the Hawaiian farmer, who has excellent growing conditions, an efficient agro-industrial infrastructure, and a large tourist market in search of novel foods.
In this paper we will discuss the commercial potential of pejibaye, the current American palm heart market, and the role that Hawaii can play in the crop's further expansion and commercialization.
In spite of these advantages, international trade in palm hearts is based upon the genus Euterpe, of South and Central America. Brazil started exporting E. edulis in the 1950s (Renesto and Vieira 1977). This agro-industry was based upon extraction of wild palms from the Atlantic forests of southern Brazil and soon decimated the wild populations, since it is a single-stemmed species with very slow growth rates. The industry then shifted to the estuary of the Amazon River to exploit E. oleracea, a smaller statured, caespitose species (Calzavara 1972). The majority of Brazil's exports (>90%) are now derived from this species, although Brazilians still consider E. edulis to be the premier palm heart (Coradin and Clement 1989).
The pejibaye palm heart is yellower than that of E. edulis (Ferreira et al. 1982a), a slight disadvantage in a market accustomed to a bone-white product. The texture of pejibaye is somewhat firmer than that of E. edulis, due to slightly lower water and higher fiber contents.
A Brazilian test panel accustomed to E. edulis found the pejibaye acceptable (7.0 vs 8.5 for E. edulis, on a 1 to 9 scale) (Ferreira et al. 1982a). Several restaurant chefs in Ubatuba, Sao Paulo, Brazil (in the heart of E. edulis' distribution), were extremely enthusiastic about pejibaye's culinary potential after evaluating this new product in direct comparison with E. edulis.
E. edulis must be planted with light to moderate shade during the first 3 to 5 years, and suffers high plant mortality during establishment (Bovi et al. 1988). At very high plant densities (6,666 and 10,000 plants/ha), yields are excellent (2.9 and 2.5 t/ha, respectively), but the palm hearts are small (200 to 250 g). At lower densities, yields are lower (1 to 2 t/ha), but palm hearts are larger (to 600 g) (Bovi et al. 1988). Harvest size is only attained at 6 to 8 years and, since the plants have only one stem, the plantation must be replanted (Bovi et al. 1988). Annual yields, in uneven aged natural stands, are therefore estimated at 0.5 t/ha (2.9 t/ha after 6 years).
E. oleracea grows faster than E. edulis, attaining harvest size in 4 to 6 years, both in its natural, nutrient rich ecosystem and in plantation (Bovi et al. 1988). This species should be planted in light shade, which can be removed within a year. Compared to E. edulis, plant mortality during establishment is low in humid soils (Calzavara 1972), but high on plateau soils in Amazonia (Gomes 1983). Yields are similar to those of E. edulis and show similar size to density trends. Because it is a caespitose palm, management for continuous cropping is possible (Calzavara 1972). Each clump yields another palm heart after 18 to 24 months (Bovi et al. 1988). Annual yields of E. oleracea are estimated at 1.4 t/ha after the first harvest.
Pejibaye can be planted in full sun, after light shade in the nursery. If correctly handled, it does not suffer plant mortality during establishment. In agroecosystems, it grows rapidly and responds readily to applications of fertilizers and other inputs, and attains harvest size in 18 to 30 months (Mora Urpí 1984). Yields and size to density trends are similar to the Euterpe species. Like E. oleracea, it is caespitose and is managed for continuous cropping and each clump yields another palm heart after 9 to 15 months (Clement et al. 1988). Annual yields are 2 t/ha after the first harvest in Costa Rica (Mora Urpí et al. 1991).
Although Brazil has a strong Euterpe tradition, there is currently strong interest in planting pejibaye in high density monocultures, because of the above mentioned advantages. In Acre state, 300+ ha are in production (A. Vieira, BONAL SA pers. commun.). Numerous agribusinessmen from Sao Paulo, Espirito Santo, and Bahia are starting to plant and have created an over-heated market for seeds of spineless pejibaye. The only national producer of these seeds, located in Manaus, received requests for 3+ million seed in 1989, but could only supply 0.5 million. Recent (1991) Brazilian government environmental regulations require management plans for extraction of Euterpe species from natural stands. This new factor will probably encourage palm heart agribusinesses to plant more pejibaye.
In the United States market, imports of E. oleracea have been extremely erratic over the last two decades. In 1977, the Organization of American States reported poor quality control as the major factor limiting imports of Euterpe palm hearts. As Mora Urpí et al. (1991) explain, exploitation of wild populations of E. oleracea by poorly trained harvesters results in an extremely variable product, frequently too fibrous or discolored for canning (Quast and Bernhardt 1978). These defects are not always handled adequately in the canning plants, and fibrous or discolored palm hearts are occasionally marketed. This lack of rigorous quality control in the canning plant has not been addressed by most Brazilian exporters to date, so that considerable volatility can continue to be expected.
Pejibaye palm heart entered the American market in 1978. Since then it has continued to expand its market share, except in 1988, when it contracted as a result of Brazil's introduction of a minimum price for its palm heart, which increased prices for all tropical American palm hearts and lowered demand. The current Brazilian government eliminated this price support in 1990. In 1989, Costa Rican pejibaye accounted for 22.5% of the domestic US market, versus only 3% in 1982. During this period, the American palm heart import market more than doubled in size
Fig. 2 presents the palm heart imports observed during the 1980s in the American market. Euterpe imports are extremely volatile, but did not previously have competition from another high quality palm heart. This new factor may be fatal for the Brazilian Euterpes, unless the Brazilian exporters start practicing good quality control. The pejibaye trend, based only upon Costa Rican exports, is strongly upward. If the Brazilian Euterpe exporters do not institute quality control and Costa Rican exports continue to expand, Fig. 2 suggests that pejibaye could become the major palm heart in the American market by the late 1990s.
An ideal way to penetrate this market is through the chef's clubs in Honolulu and on the Island of Hawaii. The imagination of these master chefs in developing new culinary uses for fresh palm heart could be essential. Through their restaurants, an initial test of the tourist and local markets can be made.
The state of Hawaii recently initiated a crop diversification program, coordinated by the Council for Agricultural Product Expansion (CAPE), an advisory group to the Governor's Agricultural Coordinating Committee. The University of Hawaii and an active agribusiness sector are also involved. Continued diversification will depend upon the imagination, energy and cooperation of these three groups.
In the early 1980s, the United States Agency for International Development sponsored a series of pan-Amazonian germplasm explorations that collected and mapped a considerable portion of the genetic diversity inherited from the Native Americans in the Neotropics (Clement and Coradin 1988). In the final report of that project, Mora Urpí and Clement (1988) classified and mapped 8 landraces of pejibaye and numerous hybrid populations. Clement and Mora Urpí (1987) and Clement (1988) proposed improvement programs for pejibaye's five major potential uses based upon different landraces and hybrid populations.
The spineless hybrid population found at Yurimaguas and the associated Pampa Hermosa landrace were suggested for use in the palm heart improvement program, combined with spineless germplasm available from the Benjamin Constant population of the Putumayo landrace and the San Carlos population of the Central American complex (Clement et al. 1988). With this germplasm identified and available from several Latin American germplasm collections, a new introduction of pejibaye into Hawaii was organized.
Open-pollinated progenies from each of the two major Amazonian spineless populations identified above have been introduced. Seed from Benjamin Constant and Yurimaguas included both selected and non-selected spineless genotypes from the collections of the National Research Institute for Amazonia--INPA, of Brazil's National Research Council. Additional germplasm from San Carlos, Costa Rica, is expected in 1992. This germplasm will provide a wide genetic base from which to identify promising accessions for the diverse Hawaiian agroecosystems.
This germplasm will be maintained at two sites, the National Clonal Germplasm Repository's Waiakea station, at Hilo, on the island of Hawaii, and the University of Hawaii's Waimanalo Experiment Station, on the island of Oahu. The environmental conditions at Hilo are characterized by thin soils of recent volcanic origin. The climate provides abundant rainfall and lacks a pronounced dry season, although any dry spell on these shallow soils can become serious, if prolonged for more than a week or two. Environmental conditions at Waimanalo are unique in Hawaii, with a rich loamy soil, abundant rainfall (October-March) with a pronounced dry season and strong trade winds. Nonetheless, pejibaye does well there, if sheltered from the wind.
Progeny x density trials are being planted with germplasm from each base population. The densities used are 3,333, 5,000, and 6,666 plants/ha. Current commercial density in Costa Rica is 5,000 plants/ha, but larger or smaller palm hearts may be required for the fresh market. After the first harvest, the plants will be managed to have two stems each, thus doubling densities.
The progeny trials are located in environments representative of major Hawaiian agricultural zones. The choice of sites provides information that can be extrapolated to almost any likely plantation site for pejibaye in Hawaii, Latin America, or elsewhere in the Pacific basin.
Growth and yield data will be collected at 6 month intervals. This will permit timely evaluation of pejibaye populational adaptation to the conditions chosen and will permit early modification of cultural practices to enhance growth, should this prove necessary. Several physiological parameters of interest in plant improvement and agronomy will be estimated from the growth and yield data (Corley 1983). Genotype-environment interactions will be examined to improve the extrapolation of results from this experiment to other areas.
Yet, Hawaiian agriculture depends upon herbicides for weed control, as labor is both scarce and expensive (DeFrank 1990). The Environmental Protection Agency regulations are constantly narrowing the number of herbicides that can be used in the United States and registration for new crops is becoming increasingly difficult and expensive. Consumers are also demanding less pesticide residues in their foods and are willing to pay more for products produced in herbicide-free environments (DeFrank 1990). To evaluate the effect of herbicides on pejibaye in Hawaii and look at "organic" alternatives, a second project was designed and approved by the CAPE. The major points to be evaluated include:
Three herbicides (Gramoxone, Goal, and Surflan) will be tested with pejibaye at commercial density (5,000 plants/ha). A non-bearing EPA approval will be requested for the most efficient.
Ground covers, including Desmodium ovalifolium, D. heterophyllum, Arachis pintoi, and Cassia rotundifolia, will be tested with pejibaye at the same commercial density. These species are prostrate non-climbers and relatively shade-tolerant. D. ovalifolium has given good results at pejibaye fruiting densities (400 plants/ha) in Brazil and Peru, but has not been tested at palm heart densities. If one or more of these species forms a good cover and does not compete with the crop, it will help control weeds, as well as offer the other agroecological advantages expected from cover crops (DeFrank 1990).
Black plastic mulch sheets have a long tradition in Hawaii as a crop establishment aid for controlling early weed growth. Recent advances in plastic technology allow mulch to remain in the field for up to five years for efficient weed control (DeFrank and Easton-Smith 1990). This mulch will be combined with both the herbicides and the ground covers and evaluated as an aid in further reducing costs.