Specialty Crops for Pacific Island Agroforestry (http://agroforestry.net/scps) Farm and Forestry Production and Marketing Profile for Giant Swamp Taro (Cyrtosperma chamissonis) By Harley I. Manner USES AND PRODUCTS Caladium cordifolium Hartzer Caladium sagittifolium Gaud., nom. nud. Giant swamp taro is the dominant aroid on the atoll islands Crytosperma merkusii var. giganteum Nadeaud of the Pacific. The primary product of this crop is the un- Cyrtosperma edule Schott ex Engle derground corm, which varies in characteristics with culti- Cyrtosperma edule Schott var and age. Plucknett (1977) reports that the young leaves Cyrtosperma edulis Schott ex Seem. and inflorescences are eaten as vegetables and the petioles Cyrtosperma merkusii sensu Drake yield a fiber for weaving. Merlin and Juvik (1996) wrote that Cyrtosperma nadeaudianum J.W. Moore during WWII, starving Chuukese would eat the peeled and Xanthosoma sagittifolium sensu Luke non (L.) Schott chopped stalks in soups. The leaf is used as a food wrapper and cover for the earth oven (um, uhmw) and the plant has Common names been used in traditional medicine in many of the high and Caroline Islands: muen (Mokil/MoakilloaAtoll), fulah (Ant low islands of Micronesia. In Kiribati, Catala (1957) was Atoll) told that specialists used a yellow mold from sliced and sun- Chuuk: fanan, pashok, pashon, pula, pwula, bula dried corms to treat skin infections. Tuvalu (Ellice Islands): brokka brokka There is no commercial production of this species and very English: giant swamp taro, swamp taro little international transshipment of the corm. However, Fiji: via, viakana cooked and frozen shipments of the corm are often sent by French Polynesia: ‘apeveo, taa faa individuals in the Federated States of Micronesia (FSM) to French: taro des atolls their relatives and friends living in Guam, Hawai‘i, and the Kiribati (Gilberts): te-babai, babai, tamu mainland U.S. Hawai‘i: maota Ifaluk: pulax As a food, the corm of the giant swamp taro can be roasted, Kapingamarangi: puraka, bulaga boiled, or baked whole, or mashed or grated and combined Kosrae: pashok, pashon, pasruk with other starches for eating. According to Merlin et al. Lamotrek: bulokh (1994) the Marshallese combine the corm with staple foods Marianas Islands: baba in preparing: Marquesas Islands: kape taataa, ta‘o- kape- taa-taa Wūden—with cooked and pounded breadfruit, Colocasia Marshall Islands: buroro, kaliklik, iaraj, iaratz, iarej, iarij, taro, bananas, or nuts mixed with grated coconut wan Jebwater—with grated Colocasia taro mixed with coconut Mortlock Islands: tepuraka milk, wrapped in taro leaves and baked in the oven Nukuoro: bulaga Palau: brak Totaimon—with Colocasia taro grated and mixed with coco- Philippines: galiang (Bicol); palau (Cebu Bisaya); and pa- nut oil and coconut sap lauan (Samar-Leyte Bisaya, Panay Bisaya) Kōmākij—with mashed taro or potato Pingelap: muang, mwang, muhang, muiang, mweiang Jukjuk—with pounded Colocasia taro mixed with coconut. Pohnpei: muang, mwang, muhang, muiang, mwang, mwahng, mwong Puluwat pwula, bula BOTANICAL DESCRIPTION Raiatea, Society Islands: opevea Satawal: pula Preferred scientific name and author Solomon Islands: kakake Tahiti: moata, maota Cyrtosperma chamissonis (Schott) Merr. after Smith (1979) Tonga: pula‘a and Fosberg et al. (1987). This species is often considered Ulithi: bwolok, bwolokh, pwolok, puns, pura synonymous with C. merkusii (Hassk.) Schott. Western Polynesia: pulaka, pula‘a, puraka Family Woleai: bwolog Araceae (aroid family), subfamily Aroideae Yap: lok, lak Brief botanical description Non-preferred scientific names Very large, stemless (acaulescent) herbaceous plant, which Arisacontis chamissonis Schott by one account can reach a height of 5 m and is the larg- Arum cordifolium Wilkes et al., nom. nud., pro parte est in the Aroideae family. Vickers (1982) suggests that this Arum costatum of Christian, pro parte, non Wall species is the largest plant in the world that yields an ed- Arum sagittaefolium Chamisso ex Schott ible corm. In some varieties, the corm can weigh as much as Arum sagittifolium sensu Chamisso non L. 100–120 kg if left to grow for a number of years (Untaman Farm and Forestry Production and Marketing Profile for Giant Swamp Taro by Harley I. Manner 2 Left: Close up of plant in Yap. June 2007. Middle: Cataphylls, or reduced leaves are found on some varieties of giant swamp taro. This is a two-year-old Palauan variety from the Agana Swamp, Guam. March 2003. Right: A 2- or 3-year-old giant swamp taro plant with inflorescences in various stages of maturity at the Agana Swamp. March 2003. 1982). The leaves are large (reaching a length of 1 m), erect, Current distribution worldwide and saggitate to hastate in shape with two long, acute basal This species is most widely distributed and cultivated in lobes. Petioles are large, sometimes with prickles, spiny to- Micronesia and the western Pacific, in particular, on atoll ward the base and reaching 3 m in length in some variet- islands where it is either the first or second most important ies (Pursglove 1975). Cataphylls (reduced leaves) are found cultivated aroid. It is less common in the eastern Pacific, al- on the underside of the leaf in some varieties. The spathe is though recent archaeological evidence shows that the spe- thick, yellowish with green veins while the spadix is yellow cies was introduced about 1451 CE to Henderson Island to orange (Smith 1979). (24° 22’ S, 128° 19’ W) of the Pitcairn group in southeastern Polynesia (Hather and Weisler 2000). Much of the species’ DISTRIBUTION current distribution can be inferred from the list of com- mon names presented earlier. The species is grown in In- Native range donesia, the Philippines, Papua New Guinea, Solomon Is- According to Pursglove (1975), giant swamp taro grows wild lands, Fiji, Tahiti, Cook Islands, Tokelau, Samoa, Palau, Yap, in the Indo-Malesian region and was introduced into many Chuuk, Pohnpei, Kosrae, Marshall Islands, Guam, Kiribati, Pacific islands in pre-European times. In contrast to other Tuvalu, the Marquesas Islands and most, if not all, of the aroids, e.g., Colocasia esculenta and Xanthosoma spp., Purs- inhabited atolls of Micronesia and Melanesia. It is a major glove (1975:58) considers Cyrtosperma an aroid that is culti- crop in most atolls and low islands of the Pacific, but other vated today to a “more limited extent.” than in Yap, Pohnpei, Chuuk, and Palau, it is a minor crop Plucknett (1977) says that this species is probably native to today in the high islands of the Pacific. The species is an Indonesia. Smith (1979:451) suggests that the origin of the aboriginal introduction to most Pacific islands except for giant swamp taro cultivated in the Pacific islands could be French Polynesia where Plucknett (1977) says that it is prob- “from wild stock in or around northern New Guinea.” He ably a post-European discovery-era introduction. In Tonga, suggests that New Guinea is a center of speciation. giant swamp taro is now very scarce. The species was not cultivated, but was used for food only during times of food shortage (Prescott and Folaumoetu‘i 2004). Specialty Crops for Pacific Island Agroforestry (http://agroforestry.net/scps) 3 ENVIRONMENTAL PREFERENCES AND Elevation, rainfall, and temperature TOLERANCES lower: sea level Climate upper: Untaman (1982) indicates that the species can grow up to an elevation This species is well adapted to moist tropical climates. It of 200 m in Yap. French (2004) says that also does well in warm, seasonally moist climates that have this species grows up to an elevation of 150 m in Papua New Guinea. However, a short dry season and variable precipitation. In Koppen’s based on the temperature/elevation classification, the species does well in the A climates where relationships, this species can grow at precipitation exceeds evapotranspiration on an annual basis. Elevation range even higher elevations, up to about 600 A main factor for its growth is a continuous supply of water, m, although the required conditions for although it can tolerate short periods of dryness. growth may not be available. This spe- cies is a component of some Papua New Based on temperature information from the Marshall Is- Guinea agricultural systems that extend lands, this species easily tolerates maximum temperatures from sea level to 600 m elevation (al- though the species may not be planted at of 35–38°C. The discovery of subfossil leaf fragments from the higher elevations). Henderson Island indirectly suggests that swamp taro is lower: It probably cannot survive where able to withstand a minimum monthly mean temperature annual rainfall is unable to support a of approximately 15.5°C. more or less constant fresh water supply. For atoll islands, the rainfall and size Soils of the islet must be able to maintain a freshwater lens despite tidal fluctuations Giant swamp taro is a water loving plant (hydrophyte) and evapotranspirational losses. adapted to fresh to brackish water conditions in coastal upper: Upper and lower amounts of marshes, natural and man-made swamps, and pit depres- Mean annual rainfall rainfall are not relevant as the species sions. Deep soils are preferable, as local taro experts stress must grow in marshy or swampy land that giant swamp taro grows both upward and downward in where the water table is near the surface (where there is a continuous supply of contrast to Colocasia taro which grows upward only (Engl- water). It can grow in swiftly flowing berger 2009). Few, if any studies, have focused on the soil streams or even on stream banks that tolerances of giant swamp taro. An idea of the pH tolerance are subject to swiftly flowing water, but it range of this species can be inferred by referencing the loca- is more commonly found where stream flow and erosion potential are less. tion of this species with known soils. In Yap, for example, giant swamp taro is cultivated in a bottomland soil known Rainfall pattern A continuous water supply is required either from rain or other sources. as Mesei (which is also the Palauan word for a taro patch). lower: 23°C Mesei soil is a very deep, poorly drained mucky peat derived Mean annual temperature upper: 31°C from organic materials overlying a silt loam or silty clay Mean maximum tempera- loam of alluvial origins (Smith 1983). The pH of the topsoil 38°C ture of hottest month and subsoil are 4.5–5.5 and 5.6–6.5 respectively. Both ho- Mean minimum tempera- rizons are very permeable, with low shrink-swell potential. ture of coldest month 15°C The organic matter percentage in the topsoil is almost 100%. 10°C, on the basis of minimum tem- Minimum temperature Another bottomland soil in which giant swamp taro is cul- perature at Henderson Island (Pitcairn tolerated tivated is the Dechel soil, which is a deep, poorly drained group) mucky silt loam. The pH range of the Dechel soil is between 5.1 and 7.3. Dechel is also a Palauan word for a marsh as well evidence for its greater tolerance is not readily available. as a less intensive system of cultivation in which the vegeta- Mourits (1996) found giant swamp taro crop failure and pit tion is cleared, but not turned under as in a mesei system. In abandonment in at Kiebu and Butaritari, Kiribati at conduc- the Agana Swamp on Guam, Palauan taro growers use the tivities of about 3300–5000 µS/cm (electrical conductivity dechel system to cultivate taro. of water is directly related to its concentration of dissolved On Ulithi Atoll, giant swamp taro is cultivated in depres- salt ions). Webb (2007) provides a photograph of a possibly sions in marshy lands dominated by a soil called the Ngede- salt tolerant variety of Colocasia taro growing in the Fonga- bus Variant. The Ngedebus Variant is a very permeable, fale pulaka pits on Funafuti, where giant swamp taro cul- gravelly loamy sand with a pH range of 6.6–8.4 for the top- tivation was abandoned because of chronic problems with soil and subsoil, respectively. saline incursion” during “natural high water events.” Webb’s In comparison to Colocasia taro, giant swamp taro is re- (2007) survey of salinity and observations of pulaka in Tu- puted to be more salt tolerant (Plucknett 1977), although valu found similar results as shown in the following table. Farm and Forestry Production and Marketing Profile for Giant Swamp Taro by Harley I. Manner 4 More problems with salinization of taro pits are anticipated Te anga (arm’s length)—7 years of age. Babai of this size is because of global warming and sea level rise. required for certain rituals. Te bonaua (breastbone) 10+ years of age. Hard, very large Table 1. A guide for salinity tolerance ranges for giant tuber grown mainly for presentations (e.g., by a young man’s swamp taro in Tuvalu pulaka pits (Webb 2007) family to that of the girl he is to marry). Conductance (µs/cm) Condition of pulaka Flowering and fruiting ≤ 1,000 Ideal Growing Conditions 1,000–2,000 Tolerable growing conditions The giant swamp taro produces inflorescences in the leaf ax- ≥ 3,000 Crop decline and failure ils of plants beginning around the second year of growth and continuing on for a number of years thereafter. The inflores- Webb (2007) concluded that continuous monitoring of the cence is large, with the open spathe 25–65 cm or more in taro pits should provide greater technical accuracy and understanding of the relationship between sa- linity and the growth conditions of giant swamp taro. GROWTH AND DEVELOPMENT There are few details about the growth rate of this species. In all likelihood, the rate of growth is cul- tivar dependent. Most varieties mature between one and 2 years of age. The Chuukese variety Onou maram is said to be harvestable after its name- sake, in 6 months. The Kiribati cultivar Te ikaraoi can remain in the ground for up to 12–15 years at which time the corm can weigh up to 90 kg. Koch (1986) says that the maximum time in the ground is 10 years, at which time the corm has become fi- brous and acrid. Citing others, Englberger (2009) Palauan mixed tree gardens on the hillside and taro patches on the stream reported that in Pohnpei, giant swamp taro can be valley bottom from Arakabesan, Meyungs, Palau, 50 m above sea level. Co- kept in the ground for 10–15 years or longer and locasia taro (light gray green in color) is flanked by giant swamp taro. Ba- that on Mwoakilloa Atoll some corms are reported nanas (Musa spp.), coconuts, and betel nut trees can be seen on the opposite to be more than 20 years old. In Yap, the plant is slope. June 2006. said to be mature and harvestable when flowers appear or when new leaves are reduced in size and the main corm rises above the surrounding corm- lets (Untaman 1982). Time to flowering depends on the variety. For many, flowering occurs between one and two years. Some information concerning the growth and de- velopment of giant swamp taro comes from Catala (1957), who recorded five growth stages, measured by the I-Kiribati with their arms: Te kunei—a 9-month-old plant. The tuber has a length equal to ½ of a forearm. The corm is very tender at this stage. The variety Te katutu is eaten at this stage. Side shoots are cut from the main plant at this time for replanting. Te namatanibura (forearm length)—at 3 years of age, a fully mature babai (plant). Some varieties Left: Close-up of Palauan taro field showing intercropping of giant swamp are relished at this stage while others are too bitter. taro and Colocasia taro, as well as a new planting mulched with banana Etan tenamatanibura (¾ arm’s length)—5 years leaves in the foreground. Right: Giant swamp taro thriving on edge of taro field in Palau. June 2006. Specialty Crops for Pacific Island Agroforestry (http://agroforestry.net/scps) 5 length. The spadix is tubular to cylindric, about 20 cm long, and has both male and female flowers (Plucknett 1977). The seeds are generally infertile. AGROFORESTRY AND ENVIRONMENTAL SERVICES The species is often interplanted with Colocasia taro. In Yap, after an area is cleared for planting, Colocasia taro is planted first at a spacing of 1–1.5 m apart. A day or two later, giant swamp taro is planted between the Colocasia plants. Within a year, the Colocasia taro is harvested and replanted in the emptied space. After harvesting of the second Colocasia crop, it is replanted again. No replanting of Colocasia occurs after the third Colocasia harvest (Untaman 1982). Giant swamp taro is often planted as a fringe species be- tween an open field or patch of Colocasia taro and the for- est as giant swamp taro tolerates shade quite well. On atolls, the smaller taro depressions are heavily shaded by adjacent trees. PROPAGATION AND PLANTING This species is propagated using setts, which are suckers, the top of the corm with about 30 cm (12 in) of petiole, or corm- lets, which are young, immature corms produced by a more mature plant. Planting methods and techniques vary greatly depending on the habitat. The simplest systems can be found in the freshwater marshes of high islands. At the Agana Swamp in Guam, Palauan migrants/residents use the dechel cultiva- tion system. In this method, the marsh is cleared of its veg- etation (mainly a reed, Phragmites karka) and then planted with setts from previously harvested giant swamp taro. The plantings may be single plants, short rows of 6–10 plants, or interplanted among Colocasia taro. In Palau, a more labor intensive mesei system was used to cultivate Colocasia taro. This system in which the marshy soil was overturned and mulched is rarely practiced today. If marshes were not available, streams were diverted and taro patches similar to the Hawaiian lo‘i were constructed. Taro swamps were also created adjacent to streams and giant Palauan taro cultivation at the Agana Swamp, central Guam. swamp taro planted in the slower flowing water. Top: Palauan residents of Guam have cultivated brak (Giant swamp taro) and kukau (Colocasia esculenta) in the peaty soil In the atolls, more elaborate systems of cultivation were of the swamp for more than 25 years. April 2001. Middle: Palau- used. Giant swamp taro and Colocasia taro were planted ans refer to the taro patch as the mesei, even though the dechel either in mulched depressions, raised beds of organic mat- method of cultivation is practiced there. In a traditional Palau- ter, or as in Kiribati, the “bottomless basket,” which may be an mesei, the soil is overturned to bury organic matter. In the one of the most complex systems of cultivation devised. De- dechel system, the vegetation is cut and cleared; the soil is not scriptions of this system have been presented by many (e.g., overturned prior to planting. March 2006. Bottom: Only women Catala (1957), Koch (1986), Vickers (1982), Loumala (1974) cultivate the swamp. Men are responsible for digging the ditches, and Lambert (1982)). Briefly, according to Vickers (1982), lining the pathways with old roofing tin or wood to ease walking after a pit was excavated down to the water table, the soil in in the mucky soil and for building the small shacks. Each culti- the pit was prepared by digging a hole about 60 cm3 in vol- vator has a shack for resting, talking, and eating lunch. March 2006. Farm and Forestry Production and Marketing Profile for Giant Swamp Taro by Harley I. Manner 6 ume. The hole was filled with chopped Guettarda speciosa and Tournefortia argentea leaves, then covered with whole Guettarda leaves and a layer of black humic sand. This was all trodden upon. A babai plant or sett was placed in the middle of this so that its upper roots were at the water level. Each plant was surrounded by a circlet of woven pandanus or coconut fronds in the form of a bottomless basket, then covered with several layers of chopped leaves and soil. The circlet basket was held in place by Guettarda stakes (Koch 1986). Compost was added as the mixture rotted and more organic matter was added as each new leaf emerged. Each plant was supplied with compost at least four times a year until the corm was harvested. On Ulithi Atoll, FSM, in addition to being grown in taro pits, this species is also grown either alone or in combina- tion with Colocasia taro in rectangular cement block tanks of variable dimensions. On the upraised limestone island of Fais in Yap, giant swamp taro is grown only in these tanks because the freshwater lens is 20 m or so below the surface. CULTIVATION Variability of species On atolls, the number of cultivars is high and new cultivars are often introduced. Luomala (1974) states that there are 40 native named varieties of giant swamp taro in Kiribati. At Eita on Tabiteuea, 10 of the 14 varieties were brought there after the arrival of the British in 1892. For Puluwat, Elbert’s (1972) Puluwat dictionary listed 33 varieties; Manner and Mallon (1989) found 24 varieties on the atoll, of which 11 had been introduced after 1972. Elsewhere Raynor (1991) listed 24 for Pohnpei and Englberger et al. (2004) reported 22 for Mwoakilloa Atoll. At a workshop, Mortlock Island- ers helped identify 32 varieties of giant swamp taro and donated 21 varieties for planting in the taro genebank on Pohnpei (Wagner 2008). No commercial varieties have been identified, although a cultivar said to be originally from Yap is currently being exported from Palau to Guam. A cautionary note Many traditional Pacific island agroforestry systems are Top: Roadside giant swamp taro cultivation in a small stream characterized by high species and cultivar diversity, which in the periurban area of Kolonia, Pohnpei, FSM. The edge of some experts believe fosters agricultural sustainability and the bridge can be seen in the lower left. October 2001. Middle: stability. Counting the number of giant swamp taro plantings on a ma‘a or taro islet, Puluwat, FSM. These islets are created in both natu- For example, Altieri (1999: 29) wrote that correct “biodiver- ral and excavated depressions in an islet’s interior. A wide range sification results in pest regulation through restoration of of culturally useful plants are grown on these islets. June 1988. natural control of insect pests, diseases and nematodes and Bottom: Repairing a ma‘a, Puluwat Atoll. The anthropic organic also produces optimal nutrient recycling and soil conserva- origin of the soil is evident in this photo. The woman’s pride in tion by activating soil biota, all factors leading to sustainable her taro is evidenced by the care and attention to details. Note, yields, energy conservation, and less dependence on exter- for example, the woven coconut frond which helps keep the islet nal inputs.” While the relationship between biodiversity and intact. June 1988. stability is appealing, Lebot’s (1992) study is instructive; Ta- Specialty Crops for Pacific Island Agroforestry (http://agroforestry.net/scps) 7 Left: Alex Laungowa and his friend Meldin measuring the dimensions of a community taro tank on Fais Island, FSM. The water table on Fais, an upraised limestone island, lies 20 m below the surface. As giant swamp taro requires a constantly moist soil, it is solely planted in concrete block tanks filled with water and organic matter. June 2008. Right: An example of the “bottomless” Pandanus basket method of giant swamp taro cultivation at Bonriki, Tarawa, Kiribati. September 1982. ble 2 from his study is a listing of the number of cultivars for esculenta var. colocasia, and Cyrtosperma chamissonis), sug- the major vegetatively propagated food plants of the Pacific arcane, yam, seedless breadfruit, and bananas are the clones Islands of SE Asian and/or Papua New Guinean origins. of vegetatively propagated plants which do not produce vi- able seeds. The many different cultivars of these traditional According to Lebot (1992: 310), many of the traditional food plants in Polynesia and Micronesia are the clonal descen- plants of the Pacific are losing their positions in the tradi- dants of the very few zymotypes. As an example, only three tional cropping systems because of historical and environ- zymotypes were identified in a group of 149 Polynesian cul- mental factors and their genetic vulnerability to pests and tivars of Colocasia taro, an indication that their morphologi- pathogens which results in the “rapid deterioration in yield cal variation is controlled by very few genes” (Lebot 1992: potential and agronomic performance.” These traditional 313). food plants, namely taro (Alocasia macrorrhiza, Colocasia Table 2. Geographic distribution and approximate number of cultivars per species New New Solomon Vanuatu Fiji Caledo- Tonga Samoas Tuvalu Cooks Tahiti Hawai‘i Pohnpei Guinea Islands nia Colocasia esculenta 452 262 154 72 82 14 28 13 91 35 82 15 Cyrtosperma chamissonis NA NA 1 1 0 0 12 23 0 0 0 24 Alocasia macrorhiza 2 4 3 2 3 9 19 2 2 1 1 10 Dioscorea alata 159 238 136 89 111 16 12 1 8 4 2 157 Dioscorea esculenta 94 117 17 16 23 1 2 2 1 2 1 13 Dioscorea nummularia 8 31 9 1 12 0 6 0 0 4 1 7 Piper methysticum 4 0 82 12 0 7 6 0 1 4 12 2 Artocarpus altilis NA 140 132 20 16 10 25 12 8 30 1 50 Musa spp. 420 NA 52 26 NA 25 28 4 28 18 23 55 Saccharum officinarum 244 5 4 13 20 2 2 1 NA 4 31 16 Source: Lebot 1992. NA = Not Available. Farm and Forestry Production and Marketing Profile for Giant Swamp Taro by Harley I. Manner 8 Polyculture and intercropping of different crop species pro- Successful commercial production will require a reduction vides some protection because pest and pathogens are not in the intensity of work effort in order to be economically able to increase in populations to destructive levels on iso- feasible. This may mean simplifying certain current meth- lated individuals of a species. Abandonment of the subsis- ods such as planting in bottomless baskets. Ecological fac- tence garden to fallow further keeps pest populations low. tors such as a constant fresh water supply which is free from However, these pests and pathogens, some of which are po- saltwater intrusion are necessary. Infrastructure such as lyphagous and have different hosts, remain in relict plants cheap and reliable transportation facilities, particularly for in surrounding the area for future population increase when the atolls, are likewise necessary. the garden site is replanted (Lebot 1992). He also argues that Advantages and disadvantages of growing in the selection of cultivars for disease resistance by traditional polycultures farmers is ineffective and inefficient because the planting materials and the agroecosystems themselves are infected by The ecological advantages and benefits of polycultures (the pathogens from previous cultivation cycles. This results in synergistic interactions between species) apply also to the poor agronomic performance of the traditional food plants taro patch. In many atolls, the giant swamp taro pits are of the Pacific islands and its replacement by higher yield- fringed with a tree cover of wild and cultivated trees and ing crops such as sweetpotato, cassava, and Xanthosoma other species. Often the giant swamp taro planted near the taro. Other factors, including continuous cropping, the loss pit edges and in the shade of the trees are taller and more of natural and socio-cultural barriers, the improvements in vigorous while those planted in the middle of the taro field inter-island transportation systems, to name a few, also lead are smaller and yellowish brown. to the spread of pathogens and diseases with infected plants. As Cyrtosperma and Colocasia taros have different rates of In brief, the majority of Polynesian and Micronesian culti- maturation, by growing these species in a polyculture, a vars of taro which were derived from a narrow genetic base farmer can have a continuous harvest for up to 3 years from are very susceptible to the pests and pathogens, for example, the same plot beginning with an initial harvest after 5–6 the viral diseases of alomae and bobone, which have led to months of planting. However, as Cyrtosperma and Coloca- severe disruptions of Colocasia taro agriculture in parts of sia taros are both aroids and dominate the patch, by their Papua New Guinea and the Solomon Islands (Lebot 1992). abundance they represent a potential food source for a pest population. Growing a wide range of cultivars may be effec- Basic crop management tive in reducing the effects of pest predation on taro. According to Catala (1957), leaves used for composting and fertilizer were in order of importance: Sida fallax, Guettarda PESTS AND DISEASES speciosa, Messerschmidia [Tournefortia] argentea, Artocar- As relatively little research has been conducted on giant pus spp., Boerhavia diffusa, Wollastonia biflora, and Cordia swamp taro, our knowledge of the effects of pests and patho- subcordata. Triumfetta procumbens and Hibiscus tiliaceus gens is incomplete. Various reports indicate: were used less frequently. • “Dry rot” and boring damage of the corm by a nema- The addition of Sida fallax commanded special attention. In tode Radopholus similis on Yap (Murukesan et al. 2005). most instances, the leaves were used dried as the compost- ing green leaves release heat that can kill the babai. Direct • Boring damage by the taro beetle Papauana huebneri contact between the green Sida fallax leaves and the babai Faimaire reduces the edible corm and allows for inva- was avoided. A layer of sand was used to top-off the com- sion by secondary organisms and eventual death of the post, said to reduce the effects of heat released by the de- plant in Kiribati (Vickers 1982). composing organic matter. • The leaf-eating pests include Aphis gossypii Glover, mealybugs Pseudococcus sp. Nr. adonium L. and Fer- Commercial production risiana virgata Ckll., and an unidentified bagworm (Ps- Commercial production of giant swamp taro is unlikely on chidae) in Kiribati (Vickers 1982). atolls because of the lack of large, nearby markets, or intra- • Pythium rot, as has been identified in giant swamp taro island transshipment infrastructure, and the work required in the Trust Territory (Jackson and Firman 1984). for production. • DMV (Dasheen mosaic virus) infects giant swamp taro For high islands, traditional cultivation methods of proven on Kiribati (Jackson and Firman 1984). scientific worth need to be followed, rather than introduc- • Crabs have been reported to cause major damage in the ing methods and techniques from more commercial/mod- Mortlock Islands of Chuuk (Levendusky et al. 2006). ern economies (e.g., artificial fertilizers), which may cause more instability and problems for the system. Specialty Crops for Pacific Island Agroforestry (http://agroforestry.net/scps) 9 Left: A giant swamp taro planting next to a washing shack at back of the Mechitiw Elementary School and Community Hall, Mechitiw Village, Weno, Chuuk State, FSM. This patch is located at the base of Atarafar Ridge where runoff and seepage water collect in a coastal swampy lowland. Since 1988 more than 80% of the giant swamp taro plantings in this village have been replaced by housing and other urban functions. October 2009. Right: Homegarden planting in Yap. June 2007. Left: Reynolds Albert in a Mokilese backyard garden at Sohkes, Pohnpei Island. In this part of the island settlers from the atolls of Pingelap, the Mortlocks, and Mokil have adapted to the high island environment as shown by the compostion of their gardens. The giant swamp taro reaches about 4 m in height. August 1989. Right: Typical Yapese planting of giant swamp taro in agroforest of betel nut palm, breadfruit, and banana. June 2007. Farm and Forestry Production and Marketing Profile for Giant Swamp Taro by Harley I. Manner 10