O. Mertz

27

J. Trop. Agric. and Fd. Sc. 27(1) (1999): 27–37

Preliminary study on the cultivation potential of wild vegetablesEtlingera elatior, E. punicea and Commelina paludosa of Sarawak(Kajian awal tentang potensi penanaman sayur liar Sarawak Etlingera elatior,E. punicea dan Commelina paludosa)

O. Mertz*

Key words: wild vegetables, Etlingera elatior, E. punicea, Commelina paludosa, Sarawak

AbstrakBiasanya sayur liar penting kepada masyarakat luar bandar dari segi pemakanandan keselamatan makanan. Akan tetapi potensi penanaman sayuran ini masihbelum diteliti. Pucuk muda, bunga dan buah Etlingera elatior dan Etlingerapunicea serta pucuk Commelina paludosa biasanya dijadikan sayur olehpenduduk di Asia Tenggara. Kajian terhadap tindak balas baja dan nilai pasaranuntuk menentukan potensi penanaman telah dijalankan di Sarawak, Malaysia.Bagi C. paludosa, tindak balas terhadap naungan juga dikaji. Etlingera elatiormengeluarkan hasil pucuk muda yang tinggi dengan aras baja yang rendah. Olehitu, penanamannya akan menguntungkan jika permintaan daripada pasaran bandarboleh ditentukan. Bagi kudup bunga yang kerap dipasarkan, hasilnya tidak dinilaisepenuhnya. Cara-cara untuk meningkatkan pengeluaran bunga patut dikaji.Etlingera punicea didapati tidak sesuai ditanam dalam keadaan kajian yangdijalankan. Akan tetapi ia harus diuji pada jenis tanah lain kerana pucuk mudanyakerap dipasarkan. Di bawah naungan C. paludosa mengeluarkan hasil yangtinggi sejajar dengan peningkatan baja. Spesies ini amat sesuai ditanam dihalaman rumah yang bernaungan dan berpotensi sebagai tanaman kontan yangbernilai jika diperkenalkan dalam pasaran bandar. Petani sepenuh masa mungkinmendapat manfaat daripada ketiga-tiga spesies jenis saka ini yang mungkin bolehdituai berterusan dengan tenaga buruh yang sedikit.

AbstractWild vegetables often play an important role in rural communities in terms ofnutrition and food security but their potential for cultivation remains largelyunexplored. The young shoots, flowers and fruits of Etlingera elatior andEtlingera punicea, and the shoots of Commelina paludosa are commonlyconsumed in Southeast Asia. Studies on fertiliser response and market value todetermine the cultivation potential were carried out in Sarawak, Malaysia.Response to shade was also tested for C. paludosa. Etlingera elatior produceshigh yields of young shoots with low levels of fertiliser, and cultivation isprofitable if the urban demand can be ensured. The yield of the more frequentlymarketed flower buds was not fully evaluated and ways to increase flowerproduction should be investigated. Etlingera punicea is not suitable forcultivation under the present trial conditions, but should be tested in other soiltypes as the young shoots are frequently marketed. Commelina paludosa

*Institute of Geography, University of Copenhagen, Øster Voldgade 10, 1350 København K, DenmarkAuthor's full name: Ole Mertz©Malaysian Agricultural Research and Development Institute 1999e-mail: om@geogr.ku.dk

Cultivation potential of Sarawak wild vegetables

28

produced high yields that increase linearly with increasing fertiliser applicationsunder shaded conditions. This species would be excellent for cultivation inshaded home gardens and could be a potentially valuable cash crop if introducedinto urban markets. Subsistence farmers may benefit from all three species asthey are perennial and may be harvested continuously with low labour input.

Plate 1. Field trial with Etlingera elatior. Unfertilised plot in foreground

IntroductionConsumption of wild vegetables iswidespread in most tropical countries andnumerous studies have shown theimportance of wild plants when used byindigenous communities (Salafsky et al.1993; Etkin 1994; Leaman et al. 1996).Moreover, urban demand driven bytraditional food preferences and fear ofpesticide residue in cultivated vegetables arevery often high, and many wild species haveconsiderable marketing potential (Peters etal. 1989; Burgers 1993). Several studiespoint to the advantage of developing localspecies for vegetable production (de Forestaand Michon 1993; Colfer and Soedjito 1996;Christensen 1997) rather than using highyielding, but often disease prone exoticcrops in resource poor communities. Withincreasing pressure on forest resources,certain species are also likely to disappear ifnot taken into cultivation.

Very little attention has been given tothe potential of domesticating wildvegetables in the tropics, and the objectiveof this paper is to determine the possibilityof growing three wild vegetable speciescommonly used for subsistence andsometimes marketed in Southeast Asia.Fertiliser and shade trials at a researchstation and a market survey were employedin the study and related to farmerassessment of the species as reported inMertz (1999 a, b).

Etlingera elatior (Jack.) R.M. Smithand E. punicea (Roxb.) R. M. Smith, both inZingiberaceae (Plate 1 and Plate 2) werechosen for study as they are very popularvegetables in Sarawak for subsistence andmarketing. They are found all over Sarawakand gathered mainly in secondary forest(DOA 1992; Christensen 1997). They arealso common in other parts of SoutheastAsia, particularly E. elatior whose flowers

O. Mertz

29

Plate 2. Flower buds of Etlingera elatior (right) and young shoots of E. punicea(left) sold at the market in Kuching, Sarawak

Plate 3. Harvesting Commelina paludosa at the field trial (left). The leaves of the young shoots areremoved to obtain the edible stem tip (right)

are used in Indonesia and West Malaysiawhere they are regularly cultivated in homegardens (Ochse 1977; Siemonsma andKasem Piluek 1993). Both species are tallherbs up to 5 m high (Smith 1986). All partsof the plants are strongly scented. The heartsof young shoots are boiled as vegetable, andthe young flowering shoots (flower buds) ofE. elatior are commonly used as acondiment (Burkill 1935; Wong et al. 1993).

The fruits of both species are eaten rawwhen ripe.

Commelina paludosa vel aff. Bl. inCommelinaceae (Plate 3) is a creeping herbcommon in wet areas in the highlands ofSarawak, often associated with irrigationcanals for wet rice. It was chosen because ofits popularity with the highland people andits high production potential. It is found inother parts of Southeast Asia in association

Cultivation potential of Sarawak wild vegetables

30

with C. benghalensis L. and C. nudiflora L.both of which are commonly eaten(Caldwell 1972; Lugod and Padua 1979;Siemonsma and Kasem Piluek 1993). Theseand other Commelina spp. are also collectedin many African communities (Malaisse andParent 1985; Mwajumwa et al. 1991). Thestem tips of C. paludosa are peeled andboiled.

There are, to my knowledge, no studieson cultivation of E. elatior and E. puniceadespite their relative importance in the dietof Southeast Asian populations. TheDepartment of Agriculture, Sarawak, hasestablished successful observation trials withboth species in 1993, but no yields havebeen recorded (Chai, C.C., DOA pers.comm. 1996). No studies on C. paludosahave been reported.

Materials and methodsFertiliser and shade trials were established atRampangi (1° 40’N, 110° 20’E), a researchstation of the Department of Agriculture,Sarawak situated on the flat coastal plain 16km north of Kuching and carried out duringthe period from June 1995 to July 1996. Thesoils are drained acid sulphate soils (haplicsulfaquents, USDA Soil Taxonomy)characterised by 22–28% clay content, a pHof 3–3.7, low levels of exchangeable plantnutrients and CEC, and high content of Al.The climate is humid tropical with annualmean rainfall of 4 600 mm (DOA 1994).July 1995 was drier than average with only127 mm, and dry spells occurred inDecember 1995 and February-March 1996.

Vegetative plant material was collectedfrom wild populations in various parts ofSarawak and was kept moist before plantingdirectly in the field. The plants were mixedrandomly with respect to origin andmorphological variations.

The land was prepared with a powertiller 3 weeks before planting. Drainagecanals 0.7 m wide were dug between plotsand a base liming of Dolomite at 300 kg/hawas applied. One trial was established for E.elatior and one for E. punicea, each

consisting of 16 plots. The plot size was 16m2 (4 m x 4 m) with 16 plants per plot, aplant spacing of 1 m x 1 m, and 0.5 mbetween the edge of the plot and the nearestplants. Two trials were established forC. paludosa, one consisting of 16 shadedplots and one of 16 unshaded plots. Plot sizewas 7.3 m2 (2.6 m x 2.8 m) with 20 plantsper plot, a spacing of 0.4 m between plantsand 0.6 m between rows, and 0.5 m betweenthe edge of the plot and the nearest plants.The shaded plots were completely enclosedby black polyethylene netting providing50% shade. The net was placed at a heightof 2.5 m.

Four levels of compound fertiliserNPK-Mg+TE (12:12:17:2 + TE) at 0, 200,400, and 600 kg/ha were applied monthly toE. elatior (totalling 0, 2 600, 5 200 and7 800 kg/ha) and E. punicea (totalling 0,1 800, 3 600 and 5 400 kg/ha), and onceevery 3 weeks to C. paludosa (totalling 0,3 200, 6 400 and 9 600 kg/ha, basefertilisation of 400 kg/ha excluded). Eachtreatment was replicated four times and theplot layout was completely randomised. Thehigh fertiliser levels were chosen to measurethe maximum potential productivity of thespecies.

Irrigation was carried out manuallywith a hose and sprinkler whenevernecessary. Each plot in a trial was given thesame amount of irrigation water. Manualirrigation was chosen as the automatedsprinkler systems available would give asystematic irregularity in water distribution.Weeding was carried out manually. Sprayingagainst aphids with a synthetic pyrethrum(2.27% alphacypermethrin) was necessaryfor C. paludosa (3 October 1995 and 15December 1995).

The number of plants was recorded on3 August 1995 for all species. Etlingeraelatior and E. punicea plants were alsocounted on 15 December 1995 and 5 June1996. Etlingera elatior was replanted on 15August 1995 and the E. punicea trial had tobe restarted on 20 September 1995 due to

O. Mertz

31

very poor survival. All C. paludosa plotswere also replanted on 21 August 1995.

Plant heights were measured on 23August 1995, 10 January 1996 and 7 June1996 for E. elatior, and on 5 June 1996 forE. punicea. The longest shoot of each plantwas measured and the average height ofliving plants per plot was determined.

Harvesting of E. elatior was carried outon 5 June 1996 and 16 July 1996. Youngshoots longer than 55 cm and shorter than105 cm were cut 5 cm above the ground andweighed. Samples of the shoots were peeledand the heart weighed to determine theweight of the edible portion. The floweringshoots were cut at the ground and the budswere cut to a standard length of 15 cmbefore weighing. Etlingera punicea was notharvested.

Commelina paludosa was harvestedevery 3 weeks between 26 September 1995and 18 May 1996 with a total of 12harvests. All young shoots longer than 20cm were harvested by cutting 20 cm belowthe tip of the longest leaf. The leaves wereremoved from samples of the shoots, thestem peeled and the edible portion weighed.

The number of surviving plants, plantheight and harvest results were subjected toan analysis of variance (ANOVA). Linearregressions were calculated using ANOVAwith treatment sum of squares partitionedfor regression analysis (Petersen 1994).

The vegetable markets in the five urbancentres of Bintulu, Miri, Kuching, Sibu andSri Aman were surveyed every Sunday(Miri, Saturday) for one year, starting April

1995. The markets were chosen for size andgeographical location. All vendors werevisited and the number, average weight andprice of products originating from thevegetables were recorded. Each vendor wasvisited at approximately the same time ofthe day every week and only bundlesavailable at that time were included in thesurvey. On one occasion, all vendors wereasked whether the produce had been bought,cultivated, or collected from the wild.

ResultsTwo months after planting, only about 60%of the E. elatior and 20% of the E. puniceahad germinated (Table 1). Etlingera elatiorwas replanted and subsequent plant countsafter 4 and 10 months yielded survival ratesof about 90% or more. The E. punicea trialwas restarted, but the average survival ratenever exceeded 70% and growth wasgenerally very irregular. No significantcorrelation was found between the survivalrate and fertiliser levels.

Survival rate of C. paludosa after 2months was relatively good in the shadedplots (Table 1), and after replanting and basefertilisation it was no longer possible todistinguish individual plants. In theunshaded plots, survival was considerablylower but it was decided to maintain the trialas reference even though mortality remainedhigh throughout the trial and most plants inthe unfertilised plots eventually died.

Seven and 12 months after planting ofE. elatior, increasing fertiliser applicationsin the 0 to 400 kg/ha range led to significant

Table 1. Survival rates of Commelina paludosa, Etlingera elatior andE. punicea at three dates (planting dates 6–7 June 1995)

Crop Survival rate (%)

3 Aug. 95 15 Dec. 95 5 June 96

Commelina paludosa, unshaded 43.1* – –Commelina paludosa, shaded 72.8* – –Etlingera elatior 62.5* 93.0 91.0Etlingera punicea 20.3 60.2** 68.4

*Followed by replanting 15–21 August 1995. Further plant counts of C. paludosa not possible**Trial restarted 20.9.1995, replanting on 17.1.1996

Cultivation potential of Sarawak wild vegetables

32

Table 3. Average cumulated yields of Commelina paludosa and Etlingera elatior at four levels offertiliser applications

Crop No. of Mean yield (kg/ha) at F-value C.V. SEharvests four fertiliser levels (kg/ha)

0 200 400 600(1)

C. paludosa, unshaded 12 836 18 603 14 329 39 918 7.0** 66.4 8 668C. paludosa, shaded 12 10 555 40 616 60 845 69 551 31.2** 20.8 6 648E. elatior, shoot yield 2 1 286 3 142 3 481 3 679 19.2** 17.3 356E. elatior, flower yield 2 9 114 194 202 1.4 ns 118.8 106

(1) Fertiliser application once every 3 weeks for C. paludosa and once a month for E. elatior*significant at 5% level (F0.05 = 3.49)**significant at 1% level (F0.01= 5.95)ns = non-significantC.V. = coefficient of varianceSE = standard error between two treatments

Table 2. Plant heights of Etlingera elatior at fourlevels of fertiliser applications

Fertiliser Mean plant height (cm)level(kg/ha) 2.5 mth 7 mth 12 mth

0 43 96 152200 55 129 212400 48 142 252600 48 142 232

F-value 2.6ns 11.5** 40.5**C.V. (%) 12.4 10.1 6.4SE 4.3 9.1 9.6

increases in plant heights (Table 2).Additional fertiliser had no effect after 7months and even negative effects on plantheights after 12 months.

The shoot yields of E. elatior confirmthe same trend showing a marked increase(2.5 times) between 0 and 200 kg/hafertiliser levels, whereas the increases inyields in the 200–600 kg/ha fertiliser rangeare relatively small (Table 3). The yields arebased on the weight of unpeeled shoots ofwhich the edible heart constitutes 20–25%.

Flowering of E. elatior commenced bythe end of May 1996 but occurred veryirregularly. Consequently, the flower budyields show very high variability betweenreplications although a tendency of very lowyields in unfertilised compared with

fertilised plots is visible. The edible part ofthe flowering shoots is the undevelopedpetals and receptacle which constitute 45%of the total weight.

Conclusive production measurementsof E. punicea were not obtained as the plantdeveloped poorly throughout the trial. Noserious pests or diseases were observed inE. elatior and E. punicea. Some leaves wereattacked by undetected leaf-cutting insects inMay 1996, but without significant damageto the plants.

Cumulated shoot yields of unshadedC. paludosa show high variability betweenreplications. There is clearly some responseto fertiliser, with considerably higher yieldsin fertilised plots compared with unfertilisedplots, and in plots with the highest fertiliserlevel compared with intermediate levels.Yields in unfertilised plots fell to zero at theend of the trial as most plants were dying.

The cumulated yields in shaded plotsare significantly higher than in unshadedplots at the same fertiliser levels except at600 kg/ha, where exceptionally high yieldsin one of the unshaded plots render thedifference non-significant. Yields of theshaded C. paludosa increase linearly withincreasing fertiliser levels. The yields arebased on the weight of shoots beforeremoving the leaves. The edible core of thestem constitutes approximately 46% of theweight of measured yields.

O. Mertz

33

Other than the relatively light attack ofaphids, no serious pests or diseases wereobserved in the shaded and unshadedC. paludosa. Weed growth was quite severein the shaded trials but became manageableas the plants developed a fuller groundcover.

The following market data are based on1996 prices when RM1 was equivalent toUS$0.4. The most frequently marketedproducts of E. elatior are the flower budswith an average price of RM3.6/kg and ofE. punicea the unpeeled shoots (RM1.4/kg).The unpeeled shoots of E. elatior (RM1.5/kg)are less common and the fruits of bothspecies are only found occasionally.Commelina paludosa was not recorded inthe markets.

The survey of 21 vendors in Bintulu,Miri and Kuching in April 1996 sellingvarious products of E. elatior revealed thatnine (43%) of the vendors were growing ormaintaining small wild gardens of E. elatior,generally with no inputs other than land andlimited labour for weeding. Only two (14%)of 14 vendors selling E. punicea hadcultivated the plant themselves.

Discussion and conclusionsThe relatively poor survival rate of allspecies at the beginning of the trial wasprobably caused by an intolerance toexcessive heat and, despite regularirrigation, recurrent drying of the soilexperienced during the dry months of June-July 1995. The liming rate of 300 kg/ha mayalso have been too low.

The reasons for the continued poorgrowth and frequent rotting of unsproutedrhizomes of E. punicea were not determined,but as it occurs naturally mainly in hilly andriverine areas of interior Sarawak, the soilconditions in Rampangi may be unsuitableeither in terms of low pH and Al toxicity orpoor drainage. Moreover, shading may benecessary for sprouting of rhizomes andthese may also be more sensitive totransplantation than E. elatior.

The plant height measurements ofE. elatior suggest good response at thelowest level of fertiliser (200 kg/ha/month)compared with unfertilised plots and thathigh doses are unnecessary. This isconfirmed by the harvest data which couldnot be obtained over a longer period becauseof the slow establishment of plants.However, with shoot yields of the secondharvest 2.7 to 4.5 times higher than the firstharvest, there is considerable potential foryield improvements with time.

The natural occurrence of C. paludosain partly shaded, wet locations explains thepoor survival of unshaded plants. However,soil fertility may be an equally importantfactor as the survival in the unfertilised plotsof the unshaded trial was significantly lowerthan in fertilised plots (significant at 5%level, F = 5.15*, F0.05 = 3.49). Because ofthe high variability in the unshaded C. paludosatrial, it is difficult to establish firmconclusions. Yields may be rather high,notably with high dosages of fertiliser butgenerally the cultivation of this specieswithout shade under the present trialconditions must be considered unsure.

Shaded C. paludosa, on the other hand,produces high yields. Fluctuations betweenindividual harvests and a tendency for yieldsat high fertiliser levels to decline during theharvesting period (Figure 1) indicate someharvesting stress. The fluctuations observedamong the first four harvests follow thesame trend at all fertiliser levels with morethan 50% reductions in yields between thefirst and second harvests. The only plausibleexplanation seems to be that all harvestswith low yields were immediately precededby weeding which may have caused lightphysical damage to young shoots. In termsof yield stability, the fertiliser level at 200kg/ha seems most appropriate.

A much longer harvest period wouldhave been desirable as all plants areperennial, but this was not possible in thisstudy. However, a careful extrapolation ofthe harvest data of E. elatior to reach aconservative estimate of annual yields is

Cultivation potential of Sarawak wild vegetables

34

0

2 000

4 000

6 000

8 000

10 000

12 000

●

●

●

●

● ● ● ● ● ● ● ● ●

◆

◆

◆

◆

◆

◆ ◆ ◆◆

◆◆

◆ ◆

▲

▲

▲

▲

▲

▲

▲

▲▲

▲

▲

▲

▲

0 kg/ha200 kg/ha600 kg/ha

18.530.46.419.323.22.212.119.1228.118.1117.1026.9

Shoo

t yie

ld (k

g/ha

)

Harvest date (Sept. 1995–May 1996)

Fertiliser applications every 3 weeks

Figure 1. Shoot yields of Commelina paludosa at different harvesting dates and fertiliser levels. Yieldsat the fertiliser level of 400 kg/ha are omitted for clarity as they follow the tendency of the 600 kg/halevel and are only slightly lower.

0

5 000

10 000

15 000

20 000

25 000

30 000

35 000

●

●

●

●

●

●

◆

◆

◆

◆

◆

◆�

�

�

�

�

�

▲

▲

▲

▲

▲

▲0 kg/ha200 kg/ha400 kg/ha600 kg/ha

2.52.01.51.00.5

Monthly fertiliser application

Inco

me

less

fer

tilis

er (

RM

/ha/

year

)

Price of shoots (RM/kg)

Figure 2. Annual income after subtraction offertiliser costs from shoot production of Etlingeraelatior at different shoot price and fertiliserlevels. Only fertiliser costs occurring during theharvesting period are included (price of fertiliserin Kuching 1995 = RM0.94/kg or US$0.38/kg).

0

20 000

40 000

60 000

80 000

100 000

120 000

●

●●

● ●●

◆

◆

◆

◆

◆

◆�

�

�

�

�

�

▲

▲

▲

▲

▲

▲

0 kg/ha200 kg/ha400 kg/ha600 kg/ha

1.251.00.750.50.25

Fertiliser applications every 3 weeks

Price of shoots (RM/kg)

Inco

me

less

fer

tili

ser

(RM

/ha/

year

)

Figure 3. Annual income after subtraction offertiliser costs from shaded Commelina paludosashoots at different fertiliser and hypotheticalshoot price levels. Only fertiliser costs occurringduring the harvesting period are included (priceof fertiliser in Kuching 1995 = RM0.94/kg orUS$0.38/kg)

obtained by multiplying the cumulatedyields by four, i.e. assuming that themeasured yields obtained from two harvestsduring 1.5 months can be repeated at leastevery 3 months. Multiplying with variousmarket price levels and subtracting the costsof fertiliser at various fertiliser levelsproduces the potential annual income from

shoot production (Figure 2). Depreciatedcosts of establishment (value of land, landpreparation, planting material, base limingand fertilisation), running costs (mainlylabour for weeding, fertiliser application andharvesting), and marketing costs (transportand rental of market stalls) were not

O. Mertz

35

included in the calculations and must besubtracted from the income levels.

At the current average market price ofRM1.4, the sale of unpeeled shootproduction may provide an annual income ofmore than RM15 000/ha with monthlyfertiliser applications of 200 kg/ha. Higherfertiliser dosages are unnecessary, even athigh price levels. On top of this comes thesale of flower buds and the added value ofselling peeled shoots. Cultivation of E.elatior could be a very profitable enterpriseprovided there is a sufficiently large marketfor the products.

The market survey indicate, however,that shoots of E. elatior are not sold asregularly as the flower bud, which the plantproduces in far lower quantity than theshoot. Possible ecologic or agronomicsolutions for enhancing flower budproduction should therefore be investigated.The economic importance of E. elatior isconfirmed by the fact that farmers alreadycultivate the plant as a cash crop and that itwas well received in on-farm trials, notablyin a more inaccessible area where it is notso easily accessed in the wild (Mertz 1999a,b). There is substantial scope for developingmore intensive cultivation of this speciesboth for subsistence and commercialpurposes.

No economic analysis was possible forE. punicea, but as its shoots are sold muchmore widely than E. elatior, production islikely to be profitable if cultivation takesplace under suitable conditions. The speciesis, however, fairly abundant in the wild andclear benefits of cultivation would have tobe demonstrated for farmers to domesticatethis plant.

Commelina paludosa is occasionallyfound in rural markets in Sarawak butotherwise this species is not marketed to anysignificant extent. However, given itspopularity in the highland communities andthe general popularity of leaf and shootproducts in markets, I will attempt aconservative estimate of the value of C. paludosaif it were to be marketed.

Shaded C. paludosa produces arelatively bulky harvest and up to 60% ofthe shoots is not used. This implies arelatively low market price and in Figure 3the potential annual income levels afterreduction of fertiliser costs and at marketprices of RM0.25–1.25/kg are shown. TheFigure is based on extrapolation ofcumulated yields obtained during theharvesting period, and it is assumed that theyield levels can be maintained.

Even at very low prices, the annualincome potential of C. paludosa issubstantial with as much as RM40 000 at anaverage market price of RM0.5/kg and witha fertiliser level of 400 kg/ha. The problemremains whether a market can be createdand whether production can be maintained ata sufficiently high level. An alternativemarket for C. paludosa may be for livestockfeed as other Commelina spp. are known fortheir excellent palatability mainly forruminants (Ibrahim et al. 1988; Gaullier1990; Siemonsma and Kasem Piluek 1993).

Commelina paludosa was introduced totwo Iban communities in Sarawak whichreceived the crop well as it grew vigorouslyin swampy areas. While commercialproduction may not be appropriate atpresent, the crop has good potential forcultivation in home gardens. Small perennialplots near streams or in swampy areaswould be ideal for C. paludosa where, withsmall amounts of fertiliser, it could providea continuous supply of vegetables tohouseholds. Although the plant is mainlyknown in the Highlands, introduction in thelowlands seems feasible if monitored withrespect to acceptability in the diet andpotential occurrence as weed, notably in wetrice systems (Mertz 1999a, b).

The species studied as well as otherrelated wild vegetables are all used in otherparts of Southeast Asia, where theyconstitute an important contribution to thediet of rural communities and urbanpopulations. The domestication, cultivation,development and marketing of these cropsmay therefore be envisaged in other

Cultivation potential of Sarawak wild vegetables

36

countries where, as in Malaysia, there is anincreasing awareness of food quality andenvironmentally sound production methods.Moreover, subsistence farmers with scarcelabour supply may benefit from theperennial growth of most wild vegetablesand maintain permanent groves that requirelittle maintenance in terms of landpreparation and weeding.

AcknowledgementsThis project was funded by the DanishCouncil for Development Research. I wouldlike to thank the State Government ofSarawak for granting permission to carry outthe field work; the staff at the Department ofAgriculture, Sarawak, for invaluable supportof this project, notably Mr Chai Chen Chongand Madam Chin Siew Phin, Mr Siong YewHong and Mr Ronnie Chu, and Mr Dona akSinos. I am also grateful for the commentsand advice of my supervisor, ProfessorSofus Christiansen, Institute of Geography,University of Copenhagen.

ReferencesBurgers, P. P. M. (1993). Rainforest and rural

economy in Sarawak. The Sarawak MuseumJournal 46: 19–44

Burkill, I. H. (1935). A dictionary of the economicproducts of the Malay Peninsula London,UK: Crown Agents for the Colonies

Caldwell, M. J. (1972). Ascorbic acid content ofMalaysian leaf vegetables. Ecology of Foodand Nutrition 1: 313–7

Christensen, H. (1997). Uses of plants in twoindigenous communities in Sarawak,Malaysia. Ph.D. Thesis. Institute of Biology,University of Aarhus, Denmark

Colfer, C. J. P. and Soedjito, H. (1996). Food,forests, and fields in a Bornean rain forest:Toward appropriate agroforestry development.In Borneo in transition: people, forests,conservation, and development (Padoch, C.and Peluso, N. L., ed.) p. 162–86. KualaLumpur, Malaysia: Oxford University Press

de Foresta, H. and Michon, G. (1993). Creation andmanagement of rural agroforests in Indonesia:potential applications in Africa. In Tropicalforests, people and food (Hladik, C. M.,Hladik, A., Linares, O. F., Pagezy, H.,Semple, A. and Hadley, M., ed.) p. 709–24.Paris, France: UNESCO

DOA (1992). Wild fruits and vegetables in SarawakKuching, Sarawak: Department of Agriculture

–––– (1994). Annual Report of Research BranchKuching, Sarawak: Department of Agriculture

Etkin, N. L. (1994). The cull of the wild. In Eatingon the wild side (Etkin, N. L., ed.) p. 1–21.Tucson, USA: The University of ArizonaPress

Gaullier, P. (1990). Cattle rearing on industrial oilpalm plantations in Cameroon. I. Study of theflora and biomass determination. Revued'Elevage et de Medecine Veterinaire desPays Tropicaux 43: 401–8

Ibrahim, M. N. M., Tamminga, S. and Zemmelink,G. (1988). Nutritive value of some commonlyavailable ruminant feeds in Sri Lanka.Ruminant feeding systems utilizing fibrousagricultural residues 1987 p. 137–49

Leaman, D. J.,Yusuf, R.,Sangat-Roemantyo, H. andArnason, J. T. (1996). The contribution ofethnobotanical research to socio-economicand conservation objectives: an example fromthe Apo Kayan Kenyah, p. 245–55. SeeColfer, C. J. P. and Soedjito, H. (1996)

Lugod, G. C. and Padua, L. S. D. (1979). Wild foodplants in the Philippines. Los Baños:University of the Philippines

Malaisse, F. and Parent, G. (1985). Edible wildvegetable products in the Zambezianwoodland area: a nutritional and ecologicalapproach. Ecology of Food and Nutrition 18:43–82

Mertz, O. (1999a). The potential of wild vegetablesfor permanent cultivation or as fallowimprovement crops in shifting cultivation,Sarawak, Malaysia. In Proc. RegionalWorkshop on Indigenous Strategies forIntensification of Shifting Cultivation inSoutheast Asia, 23–27 June 1997, Bogor,Indonesia. Bogor: ICRAF (in press)

––––. (1999b). Wild vegetables as potential newcrops in farming systems of Sarawak,Malaysia. In Proc. International Workshop onSustainable Management of Non-Wood ForestProducts 14–17 October 1997, Serdang,Malaysia. Kuala Lumpur: Universiti PutraMalaysia Press (in press)

Mwajumwa, L. B. S., Kahangi, E. M. and Imungi,J. K. (1991). The prevalence and nutritionalvalue of some Kenyan indigenous leafyvegetables from three locations of MachakosDistrict. Ecology of Food and Nutrition 26:275–80

Ochse, J. J. (1977). Vegetables of the Dutch EastIndies (English edition of Indische groenten1931). Amsterdam: A. Asher & CO. B.V

O. Mertz

37

Peters, C. M., Gentry, A. H. and Mendelsohn, R. O.(1989). Valuation of an Amazonian rainforest.Nature 339: 655–6

Petersen, R. G. (1994). Agricultural fieldexperiments. Design and analysis New York:Marcel Dekker, Inc.

Salafsky, N., Dugelby, B. L. and Terbough, J. W.(1993). Can extractive reserves save the rainforest? An ecological and socioeconomiccomparison of nontimber forest productextraction systems in Peten, Guatemala, andWest Kalimantan, Indonesia. ConservationBiology 7: 39–52.

Siemonsma, J. S. and Kasem Piluek (1993). PlantResources of South-East Asia. No 8Vegetables Wageningen, The Netherlands:Pudoc Scientific Publishers

Smith, R. M. (1986). A review of BorneanZingiberaceae: II (Alpineae, concluded).Notes from the Royal Botanic GardensEdinburgh 43: 439–466

Wong, K. C., Yap, Y. F. and Ham, L. K. (1993).The essential oil of young flower shoots ofPhaeomeria speciosa Koord. Journal ofEssential Oil Research 5: 135–8

Accepted for publication on 3 March 1999