Top Banner
Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh: Land Use and Articial Selection 1 RUBY KHAN 2 ,NYREE ZEREGA * ,3,4 ,SALMA HOSSAIN 5 , AND M. I. ZUBERI 2 2 Department of Botany, University of Rajshahi, Rajshahi, Bangladesh 3 Chicago Botanic Garden, Glencoe, IL, USA 4 Northwestern University, Plant Biology and Conservation, Evanston, IL, USA 5 Gono University, Savar, Dhaka, Bangladesh *Corresponding author; e-mail: [email protected] Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh: Land Use and Articial Selection. Bangladesh is often recognized as a secondary center of diversity for jackfruit (Artocarpus heterophyllus Lam.) and is expected to harbor important genetic diversity of this underutilized tree crop. Unfortunately, genetic erosion is occurring before this rich local resource has been documented and utilized for potential crop improvement. The aim of this study was to carry out a village-based survey and make morphological measurements to document and assess jackfruit diversity across trees associated with a gradient of three locations/habitats (homesteads, public lands, and forest or fallow lands). We also tested the hypotheses that cultivated jackfruit found in homesteads exhibited positive selection pressure for characteristics desirable in the market, and that the tree location reected its history of origin and human selection. This was accomplished using 28 standardized morphological descriptors and represents the rst large-scale assessment (900 trees) of jackfruit diversity in multiple locations (nine villages). Among the descriptors studied, those most closely associated with jackfruit marketability showed a trend for selection among the trees located in homesteads. Both fruits and leaves were larger in plants on homesteads compared to plants in forest/fallow lands, and fruit quality was signicantly higher in homestead trees compared to the other two location categories. However, 18.7% and 23.7% of the fruits found on public lands and forest/fallow lands, respectively, were still considered to be of excellent fruit quality and tree vigor was signicantly higher in jackfruit in forest/fallow lands compared to the other location categories. The combination of high tree vigor and presence of excellent fruit quality among jackfruit in forest/ fallow lands suggests that the wildBangladeshi jackfruit germplasm, which is considered inferior for market and is being negatively impacted, has valuable genetic diversity to contribute to jackfruit cultivation. Furthermore, sampling across a gradient of habitats may enable the detection of possible trends resulting from domestication pressures. 1 Received 14 July 2009; accepted 1 March 2010; published online 2 April 2010. Economic Botany, 64(2), 2010, pp. 124136. © 2010, by The New York Botanical Garden Press, Bronx, NY 10458-5126 U.S.A.
13

Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

Dec 09, 2016

Download

Documents

phungnguyet
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

Jackfruit (Artocarpus heterophyllus Lam.) Diversityin Bangladesh: Land Use and Artificial Selection1

RUBY KHAN2, NYREE ZEREGA*,3,4, SALMA HOSSAIN

5, AND M. I. ZUBERI2

2Department of Botany, University of Rajshahi, Rajshahi, Bangladesh3Chicago Botanic Garden, Glencoe, IL, USA4Northwestern University, Plant Biology and Conservation, Evanston, IL, USA5Gono University, Savar, Dhaka, Bangladesh*Corresponding author; e-mail: [email protected]

Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh: Land Use and ArtificialSelection. Bangladesh is often recognized as a secondary center of diversity for jackfruit(Artocarpus heterophyllus Lam.) and is expected to harbor important genetic diversity of thisunderutilized tree crop. Unfortunately, genetic erosion is occurring before this rich local resourcehas been documented and utilized for potential crop improvement. The aim of this study was tocarry out a village-based survey andmakemorphological measurements to document and assessjackfruit diversity across trees associated with a gradient of three locations/habitats (homesteads,public lands, and forest or fallow lands). We also tested the hypotheses that cultivated jackfruitfound in homesteads exhibited positive selection pressure for characteristics desirable inthe market, and that the tree location reflected its history of origin and human selection.This was accomplished using 28 standardized morphological descriptors and represents the firstlarge-scale assessment (900 trees) of jackfruit diversity in multiple locations (nine villages).Among the descriptors studied, those most closely associated with jackfruit marketabilityshowed a trend for selection among the trees located in homesteads. Both fruits and leaveswere larger in plants on homesteads compared to plants in forest/fallow lands, and fruit qualitywas significantly higher in homestead trees compared to the other two location categories.However, 18.7% and 23.7% of the fruits found on public lands and forest/fallow lands,respectively, were still considered to be of excellent fruit quality and tree vigor was significantlyhigher in jackfruit in forest/fallow lands compared to the other location categories. Thecombination of high tree vigor and presence of excellent fruit quality among jackfruit in forest/fallow lands suggests that the “wild” Bangladeshi jackfruit germplasm, which is consideredinferior for market and is being negatively impacted, has valuable genetic diversity to contributeto jackfruit cultivation. Furthermore, sampling across a gradient of habitats may enable thedetection of possible trends resulting from domestication pressures.

1 Received 14 July 2009; accepted 1 March 2010;published online 2 April 2010.

Economic Botany, 64(2), 2010, pp. 124–136.© 2010, by The New York Botanical Garden Press, Bronx, NY 10458-5126 U.S.A.

Page 2: Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

Key Words: Jackfruit, Artocarpus heterophyllus, Bangladesh, genetic erosion, on-farmconservation, underutilized crops.

IntroductionPlant Genetic Resources (PGR) represent the

raw materials that farmers and plant breeders useto improve the quality and productivity of theircrops. There is worldwide concern over the ever-increasing loss of PGR diversity, especially of

underutilized crops (Williams and Haq 2002). Amajor reason for this is the replacement ofdiverse, genetically variable landraces with a fewgenetically uniform modern varieties (Brush1991; Harlan 1992; Hawkes 1983; NRC 1993).On-farm or in situ conservation has been advo-cated as an important way to conserve diversity

125KHAN ET AL.: JACKFRUIT DIVERSITY IN BANGLADESH2010]

Page 3: Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

(Oldfield and Alcorn 1987; Brush 1991; IPGRI2000). On-farm conservation of PGR is thecontinued cultivation and management of a diverseset of crop populations by farmers in the agro-ecosystems where the crop has evolved. It isdynamic and is aimed at maintaining the evolu-tionary processes that continue to shape diversity.Conservation, sustainable use, and the openexchange of underutilized crops will be crucial infeeding the world’s growing population. Bangladeshand surrounding regions are among the mostpopulous in the world and are rich in importantPGR. Cultigens and wild genotypes of underutil-ized crops in this region must be identified andassessed for genetic diversity and ensured for in situconservation for long-term sustainable use andconservation (Haque 1991).Jackfruit (Artocarpus heterophyllus Lam.), the

national fruit of Bangladesh, is a good example ofa highly diverse local resource whose genetic baseis being threatened. It is a large monoecious treeproducing often enormous (up to 50 cm×100 cmand weighing up to 50 kg, Jagadeesh et al. 2006;Jarrett 1959) multiple fruits that are cauliflorous(Fig. 1). The tree produces high quality timber,

leaves are used for fodder, and the fruit isconsumed by humans. It may be eaten fresh,cooked, or processed into juices, ice creams, orchips. In Bangladesh, uniform and high yieldingexotic genotypes are replacing genetically diversestands, and genetic variation is gradually beinglost for the sake of increased production (Haque1991). In 1992, the Commonwealth ScienceCouncil (CSC) and the International Centre forUnderutilized Crops (ICUC) held a workshop inDhaka, Bangladesh that was designed to identifyand develop plans for promoting the cultivation,processing, and conservation of underutilizedcrops that are already cultivated locally. Jackfruitwas identified as deserving priority attention inBangladesh because it is already an important anddiverse crop there (Azad et al. 2007; Hossain1996; Saha et al. 1996), and Bangladesh isconsidered a secondary center of jackfruit diver-sity (Arora 1998; Dhar 1998; Hossain 1996).This study focuses on assessing morphologicaldescriptors of jackfruit in Bangladesh across arange of locations to better understand jackfruitdiversity as well as inform in situ conservationstrategies.

JACKFRUIT ORIGINS

The place of origin and wild ancestor ofjackfruit is unknown. Jackfruit has been culti-vated for millennia and was referred to as early as300 B.C.E. by Theophrastus (Hort 1916). It isnow so widely cultivated that the region in whichit is indigenous and its wild progenitor is unclear.Today it can be found in cultivation at lowelevations from the Indian subcontinent throughBangladesh, Myanmar, into southern China,Laos, Vietnam, Cambodia, Thailand, Malaysia,Indonesia, and Oceania. It is also commonlycultivated in the Philippines, and has beenintroduced throughout Africa and the Neotropics.It has been under cultivation for so long thatJarrett (1959) speculated it would not be possibleto identify the wild progenitor. It is, nevertheless,believed to be native to the Indo-Malaysianregion (Beddome 1873; Brandis 1906; Gamble1902; Kanjilal et al. 1940; Talbot 1911; Wight1843). More specifically, jackfruit is often con-sidered to originate from the Western Ghats ofIndia as Wight (1843) reported finding treesgrowing in primary forest away from humanhabitation. Barrau (1976) suggested that itoriginated in Malaysia due to the great diversityof jackfruit cultivars found there. While jackfruit

Fig. 1. Cauliflorous male and female inflorescencesof a jackfruit tree. Scale bar is 5 cm.

126 ECONOMIC BOTANY [VOL 64

Page 4: Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

is important and widely cultivated in Malaysia, itwas likely introduced there and never found inthe wild (Jarrett 1959).

Bashar and Hossain (1993) suggest that wildjackfruit relatives inhabit the Andaman Islandsand report that it is found only under cultivationin Bangladesh. However, it is often found inforests in Bangladesh away from human habita-tion (Zuberi, pers. observation). In many districtslike Tangail, Mymensingh, Dhaka, and adjoiningareas, many jackfruit trees grow in the wild nearand in forest edges and fallow lands. Many ofthese trees grow tall and bear relatively small butnumerous fruits that are, in most cases, inedibleor of very poor quality. The local people call these“jangli kathal” (wild jackfruit).

A wide and undocumented diversity of jack-fruit has been observed throughout Bangladesh,but little attention has been paid to the doc-umentation and conservation of its geneticresources. This documentation is crucial to betterunderstand the history of jackfruit domestication.If Bangladesh is a “secondary center of diversity”for jackfruit, this implies that it was domesticatedelsewhere and subsequently introduced to Ban-gladesh, where the crop was then diversified. Ifthere truly are wild jackfruit trees in Bangladesh,then it is possible that Bangladesh is part of abroader center of origin of the crop (a “non-center,” sensu Harlan 1971). On the other hand,perhaps the trees growing in forests and fallow landsthat exhibit “wild” characteristics are not trulyancestrally wild, but rather represent naturalizedferal populations of trees that in the absence ofhuman selection evolved back toward wild charac-teristics after the crop was introduced to the areamillennia ago. Before these questions can beaddressed, a better understanding of the morpho-logical diversity of Bangladeshi jackfruit is necessary.

JACKFRUIT DIVERSITY

Being a multipurpose tree that yields food, fodder,timber, and fuel, jackfruit has played an importantrole in the rural economy of Bangladesh. Jackfruittrees are common in almost every household inBangladesh. However, reports indicate that a mod-erate level of genetic erosion of jackfruit diversity hasalready occurred in Bangladesh (Khan 2008; Zuberi,pers. observation). In addition to the loss ofjackfruit trees due to logging and clearing land foragriculture, market demand for jackfruit may leadto the replacement of local diversity with uniform

exotic genotypes and to the replacement of localconsumption with sales to large urban markets. As across-pollinated and seed-propagated species, jack-fruit population diversity results from the breedingsystem and natural selection associated with localenvironmental differences (evolution) or fromhuman selection and the preferences of the localcommunity cultivating them (domestication). As anunderutilized crop, jackfruit has escaped intensiveselection and cultivation. For these reasons, a widerange of genetic and morphological variation hasbeen reported for jackfruit (Azad 1999; Azadet al. 2007; Hossain 1996; IPGRI 2000; Jagadeeshet al. 2006; Saha et al. 1996; Schnell et al. 2001;Shyamalamma et al. 2008; Ullah and Haque2008). All of these studies indicate that bothwithin and between population variation exists.These studies have, however, been limited ingeographical scope and in the number of individ-uals examined, as very few germplasm collectionsof jackfruit are known (IPGRI 2000).

OBJECTIVES

The genetic diversity of jackfruit is a valuableresource for the present and future. Bangladesh isexpected to be home to rich morphological andgenetic variability, and possibly harbor wild jack-fruit. The documentation of this genetic resourceis a necessary first step in understanding andconserving the diversity for long-term sustainableuse. The aim of this study was to carry out avillage-based survey within a region where a richvariability of jackfruit is expected and to docu-ment jackfruit morphological diversity across treesin a gradient of locations (homesteads, publiclands, forest/fallow lands). This was accomplishedusing morphological standardized descriptors(IPGRI 2000), and it represents the first large-scale in situ assessment (900 trees) of jackfruitdiversity in multiple locations (nine villages). It ishypothesized that jackfruit populations will showgenetic diversity reflected in morphological varia-tion as adaptations to different local environ-mental (i.e., location) and human selectionpressure. We hypothesize that jackfruit intention-ally planted on homesteads will exhibit positiveselection pressure for characteristics desirable inthe market, and that the location of a tree willreflect its history of origin and human selection(e.g., trees near forested areas and fallow lands aremore likely to be plants with little or no humanselection, whereas in villages with new and more

127KHAN ET AL.: JACKFRUIT DIVERSITY IN BANGLADESH2010]

Page 5: Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

concentrated homesteads “wild” trees have beenremoved and new more desirable types have beenplanted). Examination of morphological varia-tion will help determine the phenotypic expres-sion of the genetic variation present. Theresults will provide baseline data on diversityof jackfruit genetic resources in Bangladesh,shed light on morphological variation due tothe pressure of cultivation and selection in thevillages, and will be used to develop theframework for a molecular level study ofBangladeshi jackfruit diversity in an effort tocombat the erosion of genetic diversity.

Material and MethodsSTUDY SITE

The study was conducted in the district ofTangail of Bangladesh (Fig. 2) in nine villages(Table 1) where the authors have good rapportwith villagers and where jackfruit trees growwidely in villages and forest edges. The villagesform a transition from forested areas with veryscattered human settlements to densely populatedvillages away from forest and fallow lands. Landuse types include farming (about 50%), forest(20%), and homesteads (15%). The study areafalls broadly within the region of semi-deciduousforest dominated by sal (Shorea robusta Roxb. exGaertn. f.). This area has a long history of humanhabitation, so it is not possible to know exactlythe past history of land use in areas that are nowforest or fallow lands. It is possible that theselands were managed in the past. However, theindigenous people still inhabit the study area andthey did little alteration to the forest until the20th century. The homesteads included in thisstudy represent “settlers” who cleared parts of theforests for croplands and settlements starting inthe 20th century. It is likely that the jackfruit andother tree species in and around forests areremnants from before this time.

Village Interviews

In the selected areas, 120 villagers were askedin unstructured interviews about the historyand origin of individual jackfruit trees in theirhomestead, neighborhood, fallow lands, and inthe forest (Table 1). Both males and females ofvarying ages were surveyed. However, given thatmany people do not know their age and womenare reluctant to answer questions about their age,

the exact ages of individuals were not gathered.Villagers were asked questions from a checklistin informal group discussions, regarding aparticular tree. In the survey questions, stresswas placed on the age and qualitative character-istics of the trees following recommendationsfrom the International Plant Genetic ResourcesInstitute’s “Descriptors for Jackfruit” (IPGRI2000). They were also asked about fruit quality,taste, and size as well as about their use of jackfruit.This was accomplished by having the group tastethe available fruit of the tree they were examiningand rank its quality following IPGRI descriptors(2000). Tree vigor was measured following therecommendations of IPGRI (2000) on a threepoint scale (low, medium, high). Tree vigor wasassessed by the first author after she had done afield survey and standardization trial.Assessments of jackfruit trees were made based

on three broad location categories: Trees foundon small homesteads (known to be intentionallyplanted by humans), trees found on public land(not known whether or not they were intention-ally planted), and trees found away (at least 2/3 km) from the homesteads in fallow lands,forests and forest edges (were not planted by thevillagers). The three categories will be referred toas “Homestead,” “Village,” and “Forest/Fallow,”respectively. These three location-based catego-ries were used in the present study to assessmorphological diversity of jackfruit across agradient of trees and habitats in a narrowgeographic range. Within each category, 300adult jackfruit trees were chosen randomly fromacross the nine villages, for a total of 900 treessampled (Table 1).

Assessment of Variation

Morphological variation among trees wasassessed using jackfruit descriptors from theInternational Plant Genetic Resources Institute(IPGRI 2000). Descriptors included 20 qualita-tive and eight quantitative measures of vegetativeand reproductive characters (Tables 2 and 3). Thequalitative tree characters were measured byobservation and drawing upon the knowledge ofthe local villagers. Quantitative characters weremeasured in centimeters (Table 3). Statisticalanalyses were performed using JMP 5.1.2 (SASInstitute, Cary, North Carolina, USA). Thequantitative data were analyzed using one-wayANOVA and Principal Components Analysis

128 ECONOMIC BOTANY [VOL 64

Page 6: Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

Fig. 2. Map of Bangladesh indicating Tangail district in dark gray.

129KHAN ET AL.: JACKFRUIT DIVERSITY IN BANGLADESH2010]

Page 7: Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

(PCA). For qualitative data, frequency distribu-tion analyses and chi-square tests were performed.

ResultsVILLAGE INTERVIEWS

Interviews in nine villages in the TangailDistrict of Bangladesh indicated that the diversityof tree species in the homesteads of this districthad declined drastically during the last fivedecades. Many homesteads reported that theyhad sold most of their large trees for timber and thatthe naturally growing trees in the forest and fallow

lands (ones that were not planted by humans) hadalmost disappeared. The tall, mature jackfruit treeswere cut as part of land clearance efforts andmarketed as timber. Increased demand and highprices convinced villagers to sell the large maturetrees, and they were not replanted. Instead, exotictrees like mahogany, Eucalyptus, and Acacia wereplanted due to their promotion by the authorities.Interviews also demonstrated the importance of

jackfruit and the recognition of wild diversity ofthe jackfruit (kathal) tree. Villagers ranked it asthe second most important tree crop on theirhomesteads after mango, and villagers were

TABLE 1. VILLAGES AND VILLAGERS SURVEYED FOR JACKFRUIT DIVERSITY.

Unions Villages Area (hectares) Population

Villagers Interviewed

Young Males Old Males Females

Arankhola Chunia 1,392 4,041 7 5 9Aushrara Idilpur 176 738 3 3 6

Haldia 2,167 1,472 6 4 5Mahishmara 125 5,844 9 6 13

Alokdia Trngri 153 1,285 3 3 4Akadi 406 1,617 2 1 2Raniad 113 3,152 1 1 2Sathibari 217 921 4 3 5Digarbaid 4,972 1,699 5 4 4

Total 9 9,722 20,769 40 30 50

TABLE 2. QUALITATIVE CHARACTERS INCLUDED IN STUDY.

Character Name Character States

Tree vigor Low, medium, highAge of tree Juvenile (5–10 y), young (11–19 y), medium (20–29 y), old (30–39 y), mature (40+ y)Canopy structure Pyramidal, broadly pyramidal, spherical, oblong, semicircular, elliptical, irregularBranching density Sparse, medium, denseBranching type Slender main trunk with few branches on top, main trunk medium with several

thick branches, main trunk short with many branches from baseTrunk surface Smooth, rough, very roughBranching pattern Erect, opposite, verticillate, horizontal, irregularLeaf blade shape Obovate, elliptic, broadly elliptic, narrowly elliptic, oblong, lyrateLeaf apex shape Acute, acuminate, retuse, obtuseLeaf base shape Oblique, rounded, cuneate, shortly attenuateLeaf texture Thick smooth, thick rough, medium, thin smooth, thin roughLeaf color Light green, green, dark green, pinkish greenFruit bearing position Main trunk, primary branch, secondary branchFruit shape Obloid, spheroid, ellipsoid, clavate, oblong, irregularStalk attachment to fruit Depressed, flattened, inflatedFruit rind color Green, greenish yellow, yellow, reddish yellowFruit surface Smooth, spinyShape of fruit spine Sharp pointed, intermediate, flatSpine density Sparse, denseFruit attraction Poor, intermediate, good, excellent

130 ECONOMIC BOTANY [VOL 64

Page 8: Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

especially cognizant of the wide variation in fruitquality. Several names were used to describe thisvariation, including “chaila kathal” (hard, tastelessjackfruit), “roachhara kathal” (jackfruit without“roa” or pulp), “hazari kathal” (bearing a thou-sand small fruits), “khokra kathal” (wavy, irregularshaped, tough fruit), and “neisha kathal” (verysoft, sweet, small-sized pulp). These namesindicate plants that are somewhat “wild” or notentirely domesticated. These are considered infe-rior in quality and generally are not preferred orhave a low price in the market. However, suchtrees are still kept for timber and wood, and poorpeople use these inferior fruits as “green vegeta-bles” before they reach maturity.

The villagers typically plant jackfruit trees thatproduce high quality fruits for human consump-tion in their homesteads. They keep the naturallygenerated jackfruit trees of poor fruit quality in thepineapple orchards and edges of farmlands fortimber because they grow rapidly and achieve tallstature. The villagers also noted that jackfruit treesin “Chala” (Uplands) of the region near forestsoccurred naturally when they came to settle there,but they have since been removed for timber.

ASSESSMENT OF VARIATION

All of the qualitative characters measuredexhibited wide variation. For each category(homestead, village, and forest/fallow), everypossible character state was present for each ofthe 20 qualitative characters. For most of thecharacters, the patterns were similar across plantcategories (data not shown) and there were noapparent trends. However, fruit attraction andtree vigor were among the few characters withsignificant differences between categories. Forforest/fallow and village trees, “good” qualityfruits were the most numerous, while among

homestead trees “excellent” quality fruits were themost numerous (Fig. 3a). The number of “excel-lent” fruits was significantly higher (p=0.0043) inhomestead trees compared to both village andforest/fallow trees, between which there was nosignificant difference (Fig. 3b). For tree vigor,forest/fallow trees were significantly (p=0.0194)more vigorous than either village or homesteadtrees (Fig. 4).

Among the four quantitative tree and leafcharacters, all except one (leaf blade length)showed significant differences among tree catego-ries (Table 3). For the diameter at breast height(dbh), there was a significant difference in size;forest/fallow trees were the largest and homesteadtrees were the smallest. Although heavily fruitinghomestead trees may have somewhat less vegeta-tive growth than trees with less fruit load, this stillsuggests that the forest/fallow trees are older andthe homestead trees are more recently planted.This agrees with the age assessments of the treesbased on interviews with villagers. The other twoleaf characters showing significant differences areleaf blade breadth and leaf tip length with aprogression from largest to smallest from home-stead to village to forest/fallow.

Among the four quantitative fruit characters,all showed significant differences among treecategories. In all cases the homestead trees hadlarger fruits and fruit stalks, followed by village,and the forest/fallow trees had the smallestdimensions (Table 3).

A principal components analysis was per-formed on the quantitative data in order tovisualize how the individual trees in the threecategories are associated with one another and todetermine what characters best explain thisassociation. Since dbh is indicative of the age ofa tree, this character was excluded. Based on the

TABLE 3. SUMMARY OF VARIATION OF QUANTITATIVE CHARACTERS AMONG THREE CATEGORIES OF JACKFRUIT TREESBASED ON ANOVA. THE MEANS (STANDARD DEVIATION) ARE INDICATED IN CENTIMETERS FOR EACH CHARACTER.

Character Homestead Village Forest/Fallow Significance

Tree (dbh) 79.89 (39.36) 114.54 (42.24) 132.64 (42.86) <0.0001*Leaf blade length 13.13 (2.73) 13.08 (2.19) 13.16 (2.52) 0.9281 n.s.Leaf blade breadth 11.32 (2.53) 9.70 (2.29) 8.97 (2.03) <0.0001*Leaf tip length 3.23 (2.79) 2.13 (1.64) 1.96 (1.08) <0.0001*Fruit stalk length 7.60 (1.65) 6.80 (1.09) 6.67 (0.93) <0.0001*Fruit stalk diameter 13.53 (1.77) 12.11 (1.88) 9.12 (2.35) <0.0001*Fruit length 56.82 (16.91) 47.38 (18.31) 35.11 (9.85) <0.0001*Fruit diameter 165.18 (43.02) 117.23 (45.58) 66.61 (17.40) <0.0001*

131KHAN ET AL.: JACKFRUIT DIVERSITY IN BANGLADESH2010]

Page 9: Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

remaining seven characters, trees from each categoryoverlap, but there is a clear progression from forest/fallow to village to homestead trees, with fruitlength and fruit diameter explaining 93.6% and5.8% of this association, respectively (Fig. 5). Asthere are positive correlations between fruit lengthand fruit diameter (r2=0.364) as well as betweenfruit length and fruit stalk length (r2=0.280), thedata are not completely independent. However,when fruit length is removed from the PCA, theresults are the same, that is, there remains a clear

progression from forest/fallow to village tohomestead trees. But rather than fruit lengthexplaining the majority of this association, fruitstalk diameter and leaf breadth explain 31.2% and20.7%, respectively.

Discussion and ConclusionBecause jackfruit is almost entirely cross-

pollinated, this long-lived, woody tree species isexpected to exhibit large within-population varia-tion (Hamrick and Godt 1990). Not surprisingly,

Fig. 3. Levels of jackfruit fruit attraction. A) Distribution of fruit attraction levels among 300 each (total 900)of forest/fallow, village, and homestead jackfruit trees. The y-axis indicates the number of trees. B) Histogram ofthe number of trees with excellent fruit attraction from among 300 each (total 900) of homestead, village, and forest/fallow jackfruit. Bars indicate standard error. Different letters above bars indicate a significant difference (p<0.05).

Fig. 4. Tree vigor in jackfruit. A) Distribution of tree vigor levels among 300 each (total 900) of forest/fallow,village, and homestead jackfruit trees. The y-axis indicates the number of trees. B) Histogram of the number oftrees with high tree vigor from among 300 each (total 900) of homestead, village, and forest/fallow jackfruit. Barsindicate standard error. Different letters above bars indicate a significant difference (p<0.05).

132 ECONOMIC BOTANY [VOL 64

Page 10: Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

great variability was observed both within andbetween categories for most of the characters.However, the data presented here demonstratesignificant differences across the three categoriesof trees, in six of seven quantitative measurements(excluding dbh, which is indicative of tree age)(Table 3). This supports the categorization ofjackfruit trees in this region into three morpho-logically distinct classes associated with theirlocation (homesteads, village – close to habita-tions but on public land, and forest/fallow). Italso suggests that there is unique genetic diversityharbored in each of these populations. Thisdiversity should be targeted for conservation andmay be indicative of the history of origin of thetrees. The differences between the populationscould possibly be due to the “wild” and “semi-wild” trees (of the forest/fallows and villages,respectively) having escaped from the constraintsof human selection such that they revert to “wild”phenotypes. However, since the area of origin fordomesticated jackfruit remains unknown, the pres-ence of this “wild” germplasm is significant and

could truly represent semi-wild (village) and wild(forest/fallow) jackfruit germplasm, respectively.

Considering both quantitative and qualitativedata, a trend for selection in homestead jackfruitpopulations was observed. The plants from thethree categories showed a pattern of variation thatmay be explained by domestication pressures.Specifically, both leaves and fruits were larger incultivated plants on homesteads compared to “wild”plants in forests and fallow lands (Table 3), andfruit quality was significantly higher in homesteadtrees compared to the other two categories (Fig. 3).As the leaves of some tree species may have largerleaves in younger individuals, correlation betweenage (as indicated by dbh) and the leaf sizecharacters were tested. There were no significantdifferences between these characters, suggestingthat the trend toward larger leaves in the home-steads is not age related. As the villagers aregrowing trees for market on their homesteads,fruit quality and size are important and thereappears to be a strong selection for larger, higherquality fruits among the homestead trees.

Fig. 5. Principle component analysis of seven quantitative characters. Symbols are as follows: homestead (opencircles), village (gray stars), and forest/fallow (black dots).

133KHAN ET AL.: JACKFRUIT DIVERSITY IN BANGLADESH2010]

Page 11: Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

Nonetheless, 18.7% and 23.7% of the villageand forest/fallow fruits, respectively, were stillconsidered to be of excellent fruit quality(Fig. 3b), indicating that these categories of treesstill have potential for the market. Interestingly,tree vigor, which is important for the long-termsurvival of a tree, does not seem to be underselection by humans among the homestead trees.Forest/fallow trees are significantly more vigorousthan either homestead or village trees (Fig. 4).With the combination of high tree vigor andpresence of some excellent quality fruit amongforest/fallow jackfruit, this suggests that the“wild” germplasm, which is considered inferiorfor market, has valuable genetic diversity tocontribute to jackfruit cultivation and should beconserved.While there are a few key characters, described

above, that seem to be under positive humanselection, most of the characters show no clearpattern or trend across forest/fallow, village, andhomestead trees. This indicates that jackfruit is ahighly variable tree crop and its domestication hasnot progressed much in Bangladesh. However,the results suggest that the process of domestica-tion is ongoing and supports the hypothesis thatthe location of trees can reflect their history ofhuman selection with trees away from humanhabitation (i.e., in forested areas and fallow lands)being under little or no human selection, butharboring valuable “wild” genetic diversity, suchas high tree vigor. These valuable genetic resourcesmay be incorporated in cultivation through plantbreeding efforts. Unfortunately, in villages withnew and more concentrated homesteads, “wild”trees are often replaced by a few, more genet-ically uniform varieties desirable for market.Given that Bangladesh has identified jackfruitas a priority underutilized tree crop for improve-ment, identifying and utilizing the full range ofits genetic diversity will be important. Samplingalong transects with a direction of transition fromconcentrated human settlements to areas away fromhuman habitation may result in the identification ofvaluable “wild” jackfruit genetic diversity. In thecurrent study, the sampling from three differenttypes of habitats enabled the detection of possibletrends resulting from domestication.Indigenous knowledge was of crucial impor-

tance in this study. Interviews indicated thatvillagers clearly use characters of fruit size andquality to identify jackfruit trees that have notbeen intentionally planted. The villagers defi-

nitely prefer the fruits of the homestead trees tothose of the naturally regenerating “wild” trees.They remove the wild trees from their home-steads and replace them with trees producinghigher quality, larger fruits. The findings high-light areas of concern and promise for jackfruit inBangladesh. The small, inferior fruit types facehigh rates of genetic erosion as people opt fortaste and size. Old, naturally generated trees withlarge tree structure and high tree vigor are rapidlybeing lost because of their market value fortimber, yet they harbor valuable genetic diversity.In rural villages throughout Bangladesh, smallfarmers depend largely on the local naturalresources, comprising an integrated agroecosys-tem of diverse field vegetables, fruit crops, andtrees. Sustainable agriculture and developmentunder complex, diverse, poverty stricken, riskprone, village-based conditions, like monsoonAsia, requires a more resilient traditional systemthat depends upon diversity, both in terms ofgenetic variability as well as diversity over timeand space in the farming system. Thus, the use oflocal crops and trees, like jackfruit, with highgenetic variability in the traditional systemsshould be preserved and revitalized.The current results and villagers’ knowledge

base indicate that Bangladesh could harbor wildjackfruit diversity and represent a noncenter orsecondary center of diversity. This highlights theneed in Bangladesh for A) further documentationand evaluation of jackfruit fruit types, B) develop-ment of a subset of representative descriptors fordifferent jackfruit plant and fruit types that willfacilitate ease of use by local researchers andfarmers, C) identification of preferred jackfruitcultivars, D) molecular assessment of jackfruitdiversity in a larger context compared to otherregions, and E) establishment of jackfruit germ-plasm collections as well as participatory on-farmconservation programs. Most of the villagersexpressed their willingness to participate in anyplant conservation program, but they emphasizedthat it would have to be part of a well-defined,long-term program with the involvement of thegovernment or non-governmental organizationsto insure long-term sustainability.

AcknowledgementsThe authors are grateful to the villagers in the

Tangail District for information and help duringthe field survey and would also like to thank the

134 ECONOMIC BOTANY [VOL 64

Page 12: Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

three reviewers for their insightful comments thatimproved this manuscript.

Literature CitedArora, R. K. 1998. Genetic Resources of Native

Tropical Fruits in Asia: Diversity, Distributionand IPGRI’s Emphasis on the Conservationand Use. Pages 42–53 in R. K. Arora and V.R. Rao, eds., Tropical Fruits in Asia—Diver-sity, Maintenance, Conservation and Use.International Plant Genetic Resource Institute,New Delhi.

Azad, A. K. 1999. Genetic Diversity of Jackfruitand Development of Propagation Methods. Ph.D. Thesis, University of Southampton, U.K.

———, J. G. Jones, and N. Haq. 2007.Assessing Morphological and Isozyme Varia-tion of Jackfruit (Artocarpus heterophyllusLam.) in Bangladesh. Agroforestry Systems71:109–125.

Barrau, J. 1976. Breadfruit and Relatives. Pages201–202 in N. W. Simmonds, ed., Evolutionof Crop Plants. Longman, London.

Bashar, M. A. and A. K. M. A. Hossain. 1993.Present Status of Jackfruit in Bangladesh. APaper Submitted to the International Centrefor Underutilized Crops (ICUC), Southamp-ton University, U.K.

Beddome, R. H. 1873. Forester’s Manual ofBotany for Southern India. Gantz Brothers atthe Adelphi Press, Madras.

Brandis, D. 1906. Indian Trees. Bishen SinghMahendra Pal Singh, Dehra Dun.

Brush, S. B. 1991. A Farmer Based Approach toConserving Crop Germplasm. Economic Bot-any 45:153–165.

Dhar, M. 1998. Techniques of Vegetative and invitro Propagation of Jackfruit. Institute ofPostgraduate Studies in Agriculture, Salna,Gazipur, Bangladesh.

Gamble, J. S. 1902. A Manual of IndianTimbers. An Account of the Growth, Distri-bution, and Uses of the Trees and Shrubs ofIndia and Ceylon with Descriptions of TheirWood Structures. Sampson Low, Marston,and Company, London.

Hamrick, J. L. and M. J. W. Godt. 1990.Allozyme Diversity in Plant Species. Pages43–63 in A. D. H. Brown, M. T. Clegg, A. L.Kahler, and B. S. Weir, eds., Plant PopulationGenetics, Breeding and Germplasm Resources.Sinauer, Massachusetts.

Haque, M. A. 1991. Village and Forestry inBangladesh. Joint Publication of BangladeshAgricultural University and SAARC Docu-mentation Center, New Delhi.

Harlan, J. R. 1971. Agricultural Origins: Centersand Noncenters. Science 174:468–474.

———. 1992. Crops and Man. AmericanSociety of Agronomy and Crop ScienceSociety of America, Madison.

Hawkes, J. G. 1983. The Diversity of CropPlants. Harvard University Press, Cambridge.

Hort, A. F. 1916. Theophrastus Enquiry intoPlants. Harvard University Press, Cambridge.

Hossain, A. K. M. A. 1996. Status Report onGenetic Resources of Jackfruit in Bangladesh.International Plant Genetic Resources Insti-tute Regional Office, Singapore.

IPGRI. 2000. Descriptors for Jackfruit (Artocar-pus heterophyllus). International Plant GeneticResources Institute (IPGRI), Rome.

Jagadeesh, S. L., B. S. Reddy, G. S. K. Swamy, K.Gorbal, L. Hegde, and G. S. V. Raghaven. 2006.Chemical Composition of Jackfruit (Artocarpusheterophyllus Lam.) Selections of Western Ghatsof India. Food Chemistry 102:361–365.

Jarrett, F. M. 1959. Studies in Artocarpus andAllied Genera. III. A Revision of ArtocarpusSubgenus Artocarpus. Journal of the ArnoldArboretum 40(113–155):298–368.

Kanjilal, U. N., P. C. Kanjilal, and A. Das. 1940.Flora of Assam. Published under the Authorityof the Government of Assam, India.

Khan, R. 2008. Character Differences Associatedwith Habitats of Jackfruit (Artocarpus hetero-phyllus) in Three Unions of Tangail District.M.S. thesis, Department of Botany, Universityof Rajshahi, Motihar, Bangladesh.

NRC (National Research Council). 1993.Agricultural Crop Issues and Policies /Committee on Managing Global GeneticResources: Agricultural Imperatives, Boardon Agriculture. National Academy Press,Washington.

Oldfield, M. L. and J. B. Alcorn. 1987. Conserva-tion of Traditional Agroecosystems. Bioscience37:199–208.

Saha, M. C., M. G. Sha, M. A. Rahman, M. I.Nazrul, A. Quasem, N. K. Halder, and A. F. M.E. Haque. 1996. Variability in Jackfruit, Inter-nal Research Review Workshop, 1–4. Horticul-ture Research Center, BARI, Gazipur.

Schnell, R. J., C. T. Olano, R. J. Campbell, andJ. S. Brown. 2001. AFLP Analysis of Genetic

135KHAN ET AL.: JACKFRUIT DIVERSITY IN BANGLADESH2010]

Page 13: Jackfruit (Artocarpus heterophyllus Lam.) Diversity in Bangladesh ...

Diversity within a Jackfruit Germplasm Collec-tion. Scientia Horticulturae 91:261–274.

Shyamalamma, S., S. B. C. Chandra, M. Hegde,and P. Naryanswamy. 2008. Evaluation ofGenetic Diversity in Jackfruit (Artocarpus hetero-phyllus Lam.) Based on Amplified FragmentLength Polymorphism Markers. Genetics andMolecular Research 7:645–656.

Talbot, W. A. 1911. Forest Flora of BombayPresidency and Sind, 2. Government ofBombay, Poona.

Ullah, M. A. and M. A. Haque. 2008. Studies onFruiting, Bearing Habit and Fruit Growth ofJackfruit Germplasm. Bangladesh Journal ofAgricultural Research 33:391–397.

Wight, R. 1843. Icones Plantarum Indiae Orientalis.J. B. Pharoah, Madras.

Williams, J. T. and N. Haq. 2002. Global Researchon Underutilized Crops. An Assessment ofCurrent Activities and Proposals for EnhancedCooperation. ICUC (International Centre forUnderutilized Crops), Southampton.

136 ECONOMIC BOTANY [VOL 64