Indigenous fruit trees of Madagascar: Potential components of agroforestry systems to improve human nutrition and restore biological diversity

Abstract

Biodiversity in Eastern Madagascar is threatened by slash and burn agriculture, which is resulting in species extinction,
land and soil degradation and rural impoverishment. An ethnobotanical study was undertaken to determine the domestication
potential of indigenous fruit tree species as components of agroforestry systems. Four major selection criteria were used:
nutritional and income needs of the population, diversification of the agroecosystem, and protection of plant and animal diversity.
At three sites, Andasibe, Masoala and Ranomafana, in the humid primary forest region of Eastern Madagascar, a total of 150
wild fruit species from 82 genera and 42 families, of which 85% were indigenous and 92% of woody habit, were identified. In
contrast to most of the deforested areas in Madagascar, the rural population in these areas possess an intimate knowledge
of indigenous plant resources. Most of the indigenous fruits are collected from the forest but for a few species, domestication
is initiated by managing naturally established species or by planting individual trees in agricultural fields. Wild fruits
supplement the daily diet, substitute for exotic fruits, gain importance during periods of food shortage and are most appreciated
by children. Commercialization of wild fruits is mainly undertaken by the poorer section of the population. Gender related
differences in knowledge and preferences on species were identified and related to the respective household responsibilities.
A list of the 26 priority species was established based on the preferences of children, women and men at the three sites.
Local, fruit-eating lemur species are also highly dependent on indigenous fruit trees and are crucial for successful regeneration
of forest vegetation.

Full text

Indigenous fruit trees of Madagascar: potential components
of agroforestry systems to improve human nutrition and
restore biological diversity
E. STYGER1,*, J. E. M. RAKOTOARIMANANA2, R. RABEVOHITRA3
and E. C. M. FERNANDES4
1
PASN/SNGF (Projet d’Appui au SNGF/Silo National des Graines Forestières); B.P. 5091,
Ambatobe, 101 Antananarivo, Madagascar; 2SNGF (Silo National des Graines Forestières);
B.P. 5091, Ambatobe, 101 Antananarivo, Madagascar; 3DRFP (Département des Recherches
Forestières et Piscicoles), FOFIFA, B.P. 904, 101 Antananarivo, Madagascar; 4 Department
of Soil, Crop and Atmospheric Sciences, Cornell University; 620 Bradfield Hall, Ithaca 14853,
New York, USA (*Author for correspondence: Cornell University, 624 Bradfield Hall, Ithaca
14853, New York, USA. E-mail-address: eds8@cornell.edu)
Key words: biodiversity, domestication, ethnobotany, gender, indigenous knowledge, non-
timber forest product
Abstract. Biodiversity in Eastern Madagascar is threatened by slash and burn agriculture, which
is resulting in species extinction, land and soil degradation and rural impoverishment. An
ethnobotanical study was undertaken to determine the domestication potential of indigenous fruit
tree species as components of agroforestry systems. Four major selection criteria were used:
nutritional and income needs of the population, diversification of the agroecosystem, and pro-
tection of plant and animal diversity. At three sites, Andasibe, Masoala and Ranomafana, in the
humid primary forest region of Eastern Madagascar, a total of 150 wild fruit species from 82
genera and 42 families, of which 85% were indigenous and 92% of woody habit, were identi-
fied. In contrast to most of the deforested areas in Madagascar, the rural population in these
areas possess an intimate knowledge of indigenous plant resources. Most of the indigenous
fruits are collected from the forest but for a few species, domestication is initiated by managing
naturally established species or by planting individual trees in agricultural fields. Wild fruits
supplement the daily diet, substitute for exotic fruits, gain importance during periods of food
shortage and are most appreciated by children. Commercialization of wild fruits is mainly
undertaken by the poorer section of the population. Gender related differences in knowledge and
preferences on species were identified and related to the respective household responsibilities.
A list of the 26 priority species was established based on the preferences of children, women
and men at the three sites. Local, fruit-eating lemur species are also highly dependent on indige-
nous fruit trees and are crucial for successful regeneration of forest vegetation.
Introduction
Madagascar has one of the richest diversity of flora and fauna in the world,
with more than 12,000 plant species, of which 80% are endemic (Perrier
de la Bathie, 1936; Humbert, 1959; Dejardin et al., 1973; Schatz et al., 1996).
The flora and fauna are threatened by deforestation for expansion of new
agricultural land. In recent years, the rate of deforestation has increased
dramatically resulting in high biodiversity loss and species extinction (Green
Agroforestry Systems 46: 289–310, 1999.
1999 Kluwer Academic Publishers. Printed in the Netherlands.

and Sussman, 1990), soil degradation and rural impoverishment (Ministère de
l’Economie et du Plan, 1990). There is an urgent need to stop this environ-
mental, economic and social degradation. Strategies and solutions need to be
developed to stop widespread slash and burn practices and to replace them
with sustainable resource management.
The flora of Madagascar has provided local people with firewood, timber,
medicinal plants, fiber, oil, resins, fodder, vegetable, nuts and fruits for cen-
turies. Wild fruits and nuts were always of special interest as food sources
for local people (Ferraro, 1994; Isaia, 1995). Information on indigenous fruit
trees of Madagascar is scarce and infrequently mentioned in botanical works
and inventories (Abadie, 1953; Turk, 1995). Essential knowledge about the
use of forest resources in Madagascar lies with the rural population. There is
major potential to harness the food and cash crop potential of wild species
(Leakey and Newton, 1994).
This ethnobotanical study was developed (1) to obtain a better under-
standing of the knowledge of rural populations on indigenous fruit tree species,
(2) to identify the potential of these species to improve nutrition, generate
new income, to conserve and safeguard the genetic resources and (3) to con-
tribute to a sustainable use of the natural resource base.
The objectives of this study were: (a) to identify and characterize the indige-
nous fruit tree species in the Eastern Region of Madagascar, (b) to under-
stand the role and importance of wild fruits for human consumption and local
markets, (c) to determine the level of species exploitation and (d) to priori-
tize the species by local peoples’ preferences. The aim was to rank the 10 to
20 most preferred species at different sites of the eastern region as an essen-
tial step to prioritizing their domestication.
Study area and methodology
Madagascar, the world’s fourth largest island, lies in the Indian Ocean
(11
°57′–25°35′S and 43°14′–50°27′E) (Davis et al., 1994). The study was
realized within the eastern zone of Madagascar which is characterized by a
humid to perhumid, temperate to warm climate. Altitudinal range lies between
0 to 1000 m above sea level. The average annual rainfall exceeds 1500 mm
with two dry months. The mean annual temperature lies between 21°to
24 °C. The forest vegetation is classified in Humbert’s phytogeographic
division as dense humid forest at low and medium altitude (Humbert, 1965;
Koechlin et al., 1974; Razanaka, 1989).
Subsistence farming is the dominant agricultural activity. The staple food
crop is rice, either irrigated in valley bottoms or cultivated on uplands.
Traditionally primary or secondary forests are slashed and burned (tavy) to
produce upland rice. After the first year of rice production, maize, beans, sweet
potatoes and manioc are planted for the two following years before the fields
are abandoned. Permanent fields contain banana intercropped with coffee,
290

exotic fruit trees and a number of various annual crops. Cattle are raised exten-
sively. Main cash crops are rice, coffee, exotic fruits, ginger, vanilla, clove
and some non-timber forest products.
Three different sites were selected for the field work in collaboration with
the on site development and conservation projects. The sites were Masoala,
Andasibe and Ranomafana (Figure 1). Characteristics for the three sites are
given in Table 1. The criteria for village selection were: (1) proximity to the
primary forest, (2) forest extraction essentially for domestic uses, and (3)
villages at a given site were selected in different altitudes in order to obtain
a wider range of rainforest vegetation composition. Two to three villages were
visited in each site.
The different research steps and methods used were:
•Species inventory: collection and identification of botanical specimens of
indigenous woody species (including trees, shrubs, palms, and epiphytes).
•Characterization of the species with the aid of local botanical experts
(project staff and people in the villages): botanical characteristics, ecolog-
ical requirements and multiple uses of the species etc.
•12 individual interviews (50% women and 50% men) in each village
(resulting in 24 interviews per site and 72 interviews in total) were held.
The dialogue focused on the importance of wild fruits in nutrition, inten-
sity of exploitation, the place of wild fruit trees in the agricultural system,
commercialization activities with wild fruits, opinion and preferences on
the uses of the identified species. The questionnaire was semi-structured
to allow a dynamic, lively discussion.
•Children, women and men were grouped separately to classify species and
to identify the most preferred species according to criteria selected by each
group (Gueye and Schoonmaker Freudenberger, 1991)
•Informal market surveys in the regional centers and market places (Martin,
1995).
Results
Species diversity
In total, 150 species from 82 genera and 42 families were identified (Table 2).
The exhaustive list of species, their occurrence in the three sites and their ver-
nacular names are reported in Table 3. This is the first inventory of species
with edible fruits as mentioned by the local population. In each of the three
sites a similar number of species, genera and families were found. Figure 2
indicates that out of the total 150 species, 102 species have been identified
solely either in Andasibe, Masoala or Ranomafana. Only 20 species occurred
in all three sites and 28 species were represented in two of the three sites.
This is an indication of high biodiversity and site specificity of species
distribution.
291

292
Figure 1.Map of Madagascar.

Thirty-seven species were identified only to the genus level, due to the
absence of botanical reference material in Madagascar. The most important
families include Apocynaceae, Clusiaceae, Moraceae, Myrtaceae, Rubiaceae
and Sapotaceae. Eighty percent of the species are endemic, 5% are indige-
nous and 15% are of exotic origin but occur spontaneously in the natural
vegetation. Seventy-one percent of the species were trees or shrubs, 9% vines,
5% palms, 4% belong to the family of Pandanaceae, 3% were epiphytes and
8% had a herbaceous habit. Exotic and herbaceous species appear in the inven-
tory because villagers described these species in the discussion on ‘wild’
fruit species.
Many genera of the identified endemic species can be found in other
parts of the world. Examples are Landolphia sp., Vangueria sp., Gambeya sp.,
Uapaca sp. (Africa), Carissa sp. (India, Africa), Garcinia sp. (Borneo,
Malaysia, South America and East and West Africa), Syzygium sp., Pandanus
sp. (South East Asia), Canarium sp. (Borneo), Eugenia sp., Rheedia sp.,
Omphalea sp., Salacia sp. and Mimusops sp. (South America) (FAO, 1983;
Verheij and Coronel, 1992; Maghembe, 1994; Villachica, 1996).
293
Table 1.Characteristics of the field research sites (Andasibe, Masoala and Ranomafana), Eastern
region of Madagascar.
Characteristic ANDASIBE MASOALA RANOMAFANA
Altitude (m) 850–1100 0–1200 400–1400
Rainfall (mm) 1700 2200–4000 2300–3000
T mean (°C) 19 °C24 °C21 °C
Vegetation Humid midland Humid lowland Humid low- and
primary forest primary forest mid-land primary
forest
Ethnic groups Betsimisaraka Betsimisaraka Betsileo and Tanala
Villages, altitude Volove, 680 m, Tanandavaely, 12 m , Bevoahazo, 850 m,
(m.a.s.l.), 16–20 km 30 km 10 km
distance to Akondromora, Ampokafo, 290 m, 40 km Miaranony, 650 m,
market (km) 643 m, 11–16 km Manakambahiny, 250 m, 10 km
35 km
Table 2.Number of indigenous fruit tree species, genera and families inventoried in the three
sites Andasibe, Masoala, Ranomafana, Eastern region, Madagascar.
Sites Number of species Number of genera Number of families
TOTAL 150 82 42
Andasibe 066 49 32
Masoala 079 57 31
Ranomafana 073 48 33

294
Table 3.Inventory of woody and herbaceous, indigenous and naturalized exotic fruit species in the three sites of Andasibe (AND or A), Masoala (MAS
or M) and Ranomafana (RAN or R) with indication of occurrence (x) in each of the sites, Eastern region, Madagascar.
Scientific name Family AND MAS RAN Vernacular name in Malagasy
Poupartia orientalis Anacardiaceae x Sakoanala (M)
Sorindea madagascariensis Anacardiaceae x x Voasirindrina (M), Voasindrirna (R)
Anisophyllea fallax Anisophylleaceae x Hazoharaka (R)
Polyalthia ghesquiereana Annonaceae x Ambavy (R)
Cabucala cryptophlebia Apocynaceae x Toalanosy (M)
Carissa edulis Apocynaceae x Hazolahy (A) syn* Monty
Carissa edulis var microphylla Apocynaceae x Hazolahy (A) syn Monty
Carissa sessiliflora Apocynaceae x Voatsikopika
Landolphia gummifera Apocynaceae x Voaheny mavo GF (M)
Landolphia myrtifolia Apocynaceae x x Vahimpingotra (A), Voaheny (R)
Landolphia myrtifolia crassipes Apocynaceae x Voaheny madinidravina (M)
Landolphia sp Apocynaceae x Voaheny medium leaves (R)
Mascarenhasia arborescens Apocynaceae x Herotra hazo (R)
Tabernaemontana retusa Apocynaceae x Montafara (A), syn. Montakala
Typhonodorum lindleyanum Araceae x x x Via (A,M,R)
Beccariophoenix madagascariensis Arecaceae x Angolafa (M)
Dypsis ceracea Arecaceae x Lafaza (M)
Dypsis sp1 Arecaceae x x Bedoda (A), Sira (R)
Dypsis sp2 Arecaceae x Tsingovatra (M)
Masoala madagascariensis Arecaceae x Lontana (M)
Raphia ruffia Arecaceae x x Rofia (A), Rafia (M)

295
Colea fusca Bignoniaceae x x x Antsasakampanarato (A) Akondrohazo, Goamargoza
Ophicolea floribunda Bignoniaceae x Sifontsoa (A)
Phyllarthron madagascariensis Bignoniaceae x x Zahana (A) – Zahamboangy (R) syn Antohiravina
Bromelia ananas Bromeliaceae x x Mananasy gasy (M,R)
Canarium madagascariense Burseraceae x Aramy be ravina (M)
Canarium sp1 Burseraceae x Ramy be (A)
Canarium sp2 Burseraceae x Ramy madinidravina (A)
Canarium sp3 Burseraceae x Ramy (R)
Cordyla haraka Cesalpiniaceae x Haraka syn Voahonda (M)
Dialium unifoliolatum Cesalpiniaceae x Zana PF (M) syn Zana mena
Senna occidentalis Cesalpiniaceae x Tsaramasondrano (R), syn Tsotsornangatra
Garcinia verrucosa Clusiaceae x Voaditsaka, Vongo (A)
Ochrocarpos madagascariensis Clusiaceae x Jabo (A)
Ochrocarpos orthocladus Clusiaceae x Vongo (A)
Rheedia aphanophlebia Clusiaceae x Voamalambotaho (R)
Rheedia madagascariensis Clusiaceae x Voavohitra (M)
Rheedia mangorensis Clusiaceae x Voatskimbakimba (R)
Rheedia sp1 Clusiaceae x –
Rheedia sp2 Clusiaceae x Voamalambotahy lahy (R)
Rheedia sp3 Clusiaceae x Vakamasina (R)
Symphonia louvelii Clusiaceae x Voandrapotika (M)
Symphonia urophylla Clusiaceae x x Azinina (M), Voatsititinjaza (R)
Calopyxis sp Combretaceae x Voamatavy, syn Tamenaka (M)
Combretum coccineum Combretaceae x Manikibonga (M)
Terminalia catappa Combretaceae x Antafana (M)
Terminalia ombrophila Combretaceae x Mantady (M)
Dichapetalum chlorinum Dichapetalaceae x Vahimazana (A) syn Vahintenako
Dichapetalum leucosia Dichapetalaceae x Voandavenona
Dichapetalum pachypus Dichapetalaceae x Voanjofo, Voandavenona (A), Vahindavenona

296
Table 3. Continued.
Scientific name Family AND MAS RAN Vernacular name in Malagasy
Dichapetalum sp1 Dichapetalaceae x Voamatavy (M)
Dichapetalum sp2 Dichapetaladeae x Voajofo (M)
Diospyros haplostylis Ebenaceae x Hazomainty (A)
Omphalea biglandulosa Euphorbiaceae x x Salehy (A), Voandramoha (R)
Uapaca ferruginea Euphorbiaceae x Voapaka
Uapaca louvelii Euphorbiaceae x Voapaka mena PF (R)
Uapaca sp Euphorbiaceae x Vopaka mena (M)
Uapaca sp2 Euphorbiaceae x Voapaka fotsy PF (M)
Uapaca thouarsii Euphorbiaceae x x Voapaka (A), Voapaka mena (M)
Aphloia theaeformis Flacourtiaceae x x x Rambafotsy (A,M), Fandramanana (A,R) Voafotsy
Ludia ludiaefolia Flacourtiaceae x Fantsikakoho
Ludia madagascariensis Flacourtiaceae x Lalangiala (A)
Ludia scolopioides Flacourtiaceae x Fandirana (A), syn Hasambihy
Scolopia madagascariensis Flacourtiaceae x x Ravinavetro (A)
Flagellaria indica Flagellariaeceae x x Viko (A), Vahipiki
Psorospermum lanceolatum Hypericaceae x Fanerana (R)
Salacia madagascariensis Hippocrateaceae x x Voantsimatra (A), Voamasoandro (R)
Beilschmiedia velutina Lauraceae x x Voakoromanga (A), Sarivanana (R)
Bakerella sp Loranthaceae x Voatakasina (R)
Acridocarpus sp Malpighiaceae x Tsararanonandroana (M)

297
Clidemia hirta Melastomataceae x x x Kelymazana (A), Mazambody, Sombatra
Medinilla longifolia Melastomataceae x x Vatakasina (M), Kalamasambarika GF (R)
Medinilla macrophyma Melastomataceae x Kalamasambarika (R)
Tristema virusanum Melastomataceae x x x Voatrotroka
Tambourissa purpurea Monimiaceae x Oditrovy (A), syn Ambora
Tambourissa trichophylla thouvenotii Monimiaceae x Disohasaka (R), Amborasavoka
Ficus lutea Moraceae x x Amontana (M;R)
Ficus pyrifolia Moraceae x x Nonosay (M), Nonoka PF (R)
Ficus sp1 Moraceae x Famakilela GF (R)
Ficus sp2 Moraceae x Famakilela PF (R)
Ficus tiliifolia Moraceae x x x Voara (M,A, R), Voara dambo (R), Voara tenany (R)
Ficus torrentium Moraceae x Voara rano (R) syn Fohomponina
Morus alba Moraceae x Voaroihazo (R)
Streblus dimepate Moraceae x Hodipaso (M)
Treculia madagascarica Moraceae x x Ampalibeala (M) syn Avoha
Ravenala madagascariensis Musaceae x x Ravenala (A,M,R)
Maesa lanceolata Myrsinaceae x x x Voarafy (A,M), Radoko (A),
Mauloutchia chapelieri chapelieri Myristicaceae x Rara (M)
Eugenia jambolana Myrtaceae x x Rotra vazaha syn Robazaha (R)
Eugenia jambos Myrtaceae x x x Zamborozana (A, R), Varotra (M)
Eugenia lokohensis Myrtaceae x x x Gavoala (A,M), Hompa gavo (M)
Eugenia sp1 Myrtaceae x Gavoala be ravina (A)
Eugenia sp2 Myrtaceae x Voabe (R)
Psidium cattleianum Myrtaceae x x Goavitsinahy (A,R)
Psidium guajava Myrtaceae x x x Goavibe (A), Gavo (M), Goavy gasy (R)
Syzygium bernieri Myrtaceae x x Varodalitra (M), Tsindrodrotra, Voaroipoitra (R)
Syzygium emirnensis var cuneifolia Myrtaceae x Robary (R)

298
Table 3. Continued.
Scientific name Family AND MAS RAN Vernacular name in Malagasy
Syzygium emirnensis Myrtaceae x Rotra mena, Rotra ravimboanjo (R)
Syzygium sp1 Myrtaceae x x x Rotra fotsy (R), Rotra (M), Robary (A)
Syzygium sp2 Myrtaceae x x Rotra mena GF (A,R), Robary (R)
Syzygium sp3 Myrtaceae x Varodalitra (M)
Syzygium sp4 Myrtaceae x Rotra laro (R)
Syzygium sp5 Myrtaceae x Rotra amboa (R)
Syzygium sp6 Myrtaceae x Rotra be ravina (A)
Syzygium sp7 Myrtaceae x Rotra be (M)
Syzygium sp8 Myrtaceae x Rotra mena vaventy ravine (R), syn Roipoitra
Noronhia sp Oleaceae x Tsilaitra (A)
Pandanus concretus Pandanaceae x x Rambo (M), Vakoana (A)
Pandanus freycinetioides Pandanaceae x Tsirika (R)
Pandanus pervilleanus Pandanaceae x Akomorika (M)
Pandanus pulcher Pandanaceae x –
Pandanus sp Pandanaceae x Karaboboka (R)
Pandanus vandamii Pandanaceae x x Kararaka (M)
Paropsia madagascariensis Passifloraceae x Fanonahona (M), syn Paiso ala
Passiflora foetida Passifloraceae x Bongambosy (M)
Passiflora incarnata Passifloraceae x x x Garana (A), Kilelaka (R) syn Voaloboka
Passiflora quadrangularis Passifloraceae x Barbajina (M) syn Ngaringadinina (M)
Piper sp1 Piperaceae x Tsimperfery (A)
Piper sp2 Piperaceae x Sakarivo vahy
Dilobeia thouarsii Proteaceae x x x Vivaona (A), syn Ramandriona (A,R), Vintanona (M)
Volombodimbaona (A)
Rubus apetalus Rosaceae x Voaroy fotsy, syn. Voaroy mainty (R)
Rubus myrianthus Rosaceae x Voaroy saka (R) syn Voaroitsaka

299
Rubus rosaefolius Rosaceae x x x Voandroy (A,M), Voaroy mena
Rubus sp Rosaceae x Takohaka (A)
Breonia madagascariensis Rubiaceae x x Valompangady (A), Voakiringy (R)
Canthium sp Rubiaceae x Tsifo (M)
Fernelia macrocarpa Rubiaceae x Toalanosy (M) syn Kafeala
Mapouria apoda Rubiaceae x Nofotrakoho (M)
Mussaenda arcuata Rubiaceae x x x Anendengona (A), syn Anandaingo, Lenga (M)
Mussaenda lanciolata Rubiaceae x Anambahy (R)
Rothmania poivrei Rubiaceae x Mantalanimbe (M)
Rothmania sp Rubiaceae x Taolanana (A)
Rothmania talagnignia Rubiaceae x Mantanaly fotsy (M) syn Mantalanimpotsy
Sabicea diversifolia Rubiaceae x x Voaseva (A), Sevalahy, Sevatrandraka (A, R)
Vangueria edulis Rubiaceae x Voavandrika (M), Vavandrika (M)
Citrus aurantifolium Rutaceae x Tsioha (M), Voasary ala
Citrus aurantium Rutaceae x x x Voangdy (M), Voangy ala (A,R)
Citrus medica Rutaceae x x Voasary gasy (R), Voatolongo (A,R)
Haplocoelum perrieri Sapindaceae x Lanary madinidravina (R)
Plagioscyphus jumellei Sapindaceae x x Soretry (M), Lanary (R) syn. Valanary
Tinopsis apiculata Sapindaceae x Volanary (A)
Faucherea manongarivensis Sapotaceae x Vasihy (M)
Faucherea sp1 Sapotaceae x Nanto (M)
Faucherea sp2 Sapotaceae x Nanto – Makaka (M)
Chrysophyllum boivinianum Sapotaceae x x x Famelona (A,M,R), syn Rahiaka (M)
Labramia costata Sapotaceae x x x Todinga (A, M),
Mimusops lohindri Sapotaceae x Voarantoala (M)
Vaccinium emirnense Vacciniaceae x Voaramontsina (A)
Lantana camara Verbenaceae x x x Radriaka
Aframomum angustifolium Zingiberaceae x x x Longoza (A,R) Lingoza (M)

Species management and utilization
The high fruit species diversity raises the question of their importance to
the rural population. Are these species used and if so to what extent and
intensity?
Three levels were considered in our analysis: (i) the collection of fruits
from wild stands in the forest, (ii) the management and protection of natu-
rally established species on agricultural land and, (iii) the cultivation of species
in farmers fields. In the three sites, all farmers collected fruits from wild trees,
while 60% were managing and maintaining wild fruit trees in the agricul-
tural environment. Only few farmers planted wild fruit trees. Exotic fruit trees
with established markets were present on 100% of the farms. The number of
species exploited in the three sites is indicated in Figure 3.
Exotic fruit tree cultivation is diverse with a range from13 to 27 species
in the three sites. Main exotic species were oranges and other citrus species,
lychee, bananas, mangos and jack-fruit. For the wild fruit species, the inten-
sity of species exploitation and management differs considerably. Major
exploitation is on an extractive level with 36 to 77 species collected from the
300
27
44
31
7
12
9
20
Andasibe (And) Masoala (Mas) Ranomafana (Ran) And and Mas And and Ran Mas and Ran And, Mas and Ran
Figure 2.Distribution of the total 150 species among the three sites Andasibe (And), Masoala
(Mas) and Ranomafana (Ran), Eastern region of Madagascar. The sectors show the number of
species in one, two or all three of the sites.

wild. The number of managed species was between nine and 27. Nevertheless,
we recovered 10 wild species that have been planted on farms.
The number of species utilized also differs among the three sites. Villagers
in Masoala exploit the largest number of species, followed by people in
Andasibe and Ranomafana. Explanations and reasons remain at the stage of
speculation, and research should be initiated to understand this phenomenon
better. Some possible explanation could relate to the time of immigration and
the duration of settlement in proximity to primary forest which would result
in differences in the level of people’s knowledge on indigenous vegetation
and its utilization.
Planted species are either naturalized, rare exotic species (Citrus auran-
tifolium, C. aurantium, C. medica, Syzygium jambos), or rare forest species
particularly appreciated for their multiple uses (Sorindea madagascariensis,
Carissa sessiliflora, Ficus sp., Eugenia sp.). Managed species are essentially
found in the banana fields or along field boundaries and paths. In Andasibe
and Ranomafana the main practice of land conversion is based on slash and
burn, and the managed species are part of secondary vegetation regeneration.
301
0
10
20
30
40
50
60
70
80
90
Ex pla Ind pla Ind man
Ind col
number of species
Andasibe Masoala Ranomafana
Figure 3.Number of exotic and indigenous fruit species used in the three sites Andasibe,
Masoala, Ranomafana: Number of exotic species planted (Ex pla), indigenous species planted
(Ind pla), managed (Ind man), collected (Ind col), Eastern region of Madagascar.

In Masoala, on the other hand, selective clearing is chosen for some of the
fields (vanilla fields), and thus primary forest tree remnants are encountered
as well as species regenerating after clearance. Management practices include
cutting back of branches and weeding. The decision to manage a species or
not is based on quality of fruits, multiple products of the species, manage-
able size and rarity of the species in natural vegetation.
Different collection intensities from wild tree stands exist ranging from
(i) systematic collection in baskets every fruiting season, to (ii) occasional
collection when encountered in the forest and brought back to the village in
pockets, and to (iii) eating them in the forest without bringing them back to
the village.
Taking into account the appreciation for fruits, their location and abun-
dance, fruits with highest collection intensity were found to have a very good
taste and came from frequently encountered species growing on fallow land
or secondary forest near the village. Rare collection was noted for less tasty
fruits regardless of their location and abundance. Favored fruits coming from
rare primary forest species were also seldom collected. Reasons for this are
the long distances to the trees, their low numbers and the difficulty of har-
vesting fruits from tall trees. Often, the lemurs picked the fruits before the
people. We hypothesize that appreciated fruits from rare forest species (e.g.
Breonia madagascariensis, Cabucala cryptophlebia, Dichapetalum sp.,
Landolphia sp., Ophalea biglandulosa, Rheedia sp., Salacia madagas-
cariensis) are not optimally available to population.
Wild fruits represent an important supplement to the daily diet, which is
mainly carbohydrate-rich and lacking important vitamins and micronutrients
(Hardenbergh, 1997). Wild fruits also substitute exotic fruits during periods
when they are not available. All family members eat wild fruits, but children
are by far the main consumers. People regularly consume wild fruits, while
traveling through the forest or working in the fields. Wild fruits are collected
between October and May, the period of food shortage. During this time dif-
ferent types of foods collected from the forest, such as roots, tubers, honey,
help in balancing the diet.
Wild fruit commercialization
In the villages near the forests, collected fruits are shared and not sold among
the people. Farmers indicated that the main reasons for the lack of commer-
cialization of these species were large distance to markets (10 to 40 km),
problems with fruit conservation, and the low prices. Despite these statements,
wild fruits can be found in the markets of regional centers. They are obtained
principally from areas under fallow vegetation and not from primary forests.
More then 20 species were inventoried in the different markets during our
survey. The trade is not organized and prices are low. Unlike the exotic fruit
sellers who buy the products from farmers, wild fruits are sold by the col-
lector. Poor farmers who lack the cash to purchase exotic fruit for sale can
302

collect wild fruit and raise cash for family needs. The demand for wild fruits
can be considerable because of their relative rarity on the market and the low
prices. Fruits from primary forests may have a commercial potential if they
can be processed in the village thereby adding value to the product and
improving preservation. ‘There have been relatively few systematic studies
to characterize fruit products with commercial potential from the farmer-iden-
tified priority species’ (Leakey, 1999).
Local knowledge, perceptions and preference for indigenous fruits
We asked men and women at the different sites to identify the factors that
determined their preferences for native fruit species.
The factors determining species preferences were:
• the type of food produced (fresh fruit, vegetable, beverage, spice, oil)
• the means of preparation (direct consumption, oil extraction)
• fruits preferably consumed by adults or children
• preferences on consumption (very good fruits, non liked fruits)
• parts of the fruit consumed (seed, flesh, skin)
• medicinal value of fruit
• other edible parts (e.g. stem, leaves, flowers)
We observed differences in knowledge and perception between men and
women. Men provided more specific information on ecological requirements
of a very wide range of species including the rare species from the primary
forests. Information included growth habit, phenology, association with plants
and wild animals (lemurs, birds), species distribution and abundance in the
area and exact location of individual trees. They had precise knowledge on
wood properties for construction and tools. Men preferred sweet tasting fruits
as well as oily seeds prepared by roasting, grinding, and cooking, which are
served as a side dish or sauce with rice. Wild fruits that mitigate hunger were
favored especially during food shortage periods where they can become essen-
tial parts of a meal.
Womens’ knowledge focused more specifically on the species found in
the surroundings of the village. Rare forest species, mentioned by men, were
not always recognized by women. Preference for a species was based on its
multiple uses, among which medicinal properties and nutritional quality of
fruits including other edible plant parts (stem, leaves and flowers) were
favored. Unlike the men who preferred sweet fruits, women preferred more
acid-tasting fruits.
Discussion and decision-making process differed between the groups of
men and women. Men tended to have a lot of arguments and long discus-
sions and had difficulty achieving consensus. Women discussed in a more
intensive and concentrated way, and came up faster with more precise and
coherent decisions.
Children had a high knowledge of wild fruits. Groups of girls and boys
303

at ages between six and 14 years recognized 75% to 80% of all the species
presented to them in the form of a branch with leaves. All the fruit species
from the fallows were known. The unknown species were primary forest
species. Although boys recognized more of the forest species than the girls
did, they were often unable to describe the taste of the fruits. Children
consume a wider range of wild fruits than adults, including the small, less
tasty fruits. Adults recognize the special role of wild fruits in the nutrition of
children. This was observed during one of the women’s group discussion
where they identified children’s fruits as a separate category. Wild fruit trees
are often protected in the landscape on the basis that children like to eat the
fruits.
Gender differences in knowledge on specific natural resources is attrib-
uted to the gender related division of responsibilities, labor and expertise,
control and interest at the intra-household, inter-household and community
level (Rocheleau, 1988). These differences in knowledge related to responsi-
bilities were well recognized in our study, for example, women don’t travel
as often and as far into the primary forest as men. Their activities are more
concentrated around the village. Being responsible for childcare, women paid
more attention to the medicinal values of a species.
Priority species
Each group of men, women and children were separately asked to select the
ten most preferred species out of the locally present wild fruit species and to
explain their choice. For the six priority species, the exercise of ‘two by two
comparison’ of all possible combinations was done, identifying the preferred
among the two species and by stating the arguments for and against each
species (Gueye and Schoonmaker Freudenberger, 1991). Compared to the
botanical inventory, where the exotic naturalized species represented 15% of
all species, their part in the nominated preferred species were 35% in Masoala,
50% in Andasibe and even 60% in Ranomafana. Rare forest species didn’t
find as much attention in this preference ranking as the nearby abundant, nat-
uralized exotic species growing in secondary vegetation. Despite the fact that
these frequent species are preferably used and accepted by rural population,
rare species with good domestication potential may not have been equally
considered during this procedure.
Out of 150 species, 26 priority species have been selected based on pref-
erences expressed by all the society groups at all three sites. Of the 26 species
17 species are indigenous and nine species are exotic and naturalized species
(Table 4). A ranking of species is not applicable at this stage, as other criteria
need to be addressed and assessed, such as propagation potential, species pro-
ductivity, fruit quality and processing potential.
304

Peoples view on domestication of indigenous fruit species
Despite an already high diversity of planted exotic fruit trees, farmers
expressed a desire to increase fruit tree cultivation for self-sufficiency needs
especially for their children, and also for commercialization purposes.
Preferences were given to exotic species whereas the following observations
were made about the constraints to planting indigenous trees: (1) Indigenous
fruit trees are present in abundance in the forest and even in the Savokas
(fallowland). Fruits can therefore be obtained to a satisfactory level by col-
lection; (2) Planting and management techniques for forest species are not
305
Table 4.Priority indigenous fruit tree species and priority exotic, naturalized fruit tree species
identified by the rural population of Andasibe, Masoala, Ranomafana in the Eastern region of
Madagascar.
Species Family Vernacular name
a) Indigenous species
First choice, no ranking
01. Labramia costata Sapotaceae Todinga
02. Landolphia myrtifolia Apocynaceae Voaheny
03. Sorindea madagascariensis Anacaridiaceae Voasirindrina
04. Carissa sessiliflora Apocynaceae Voatsikopika
05. Salacia madagascariensis Hypocrateaceae Voamasoandro
06. Symphonia urophylla Clusiaceae Azinina
07. Rheedia aphanophlebia Clusiaceae Vaomalambotaho
08. Raphia ruffia Arecaceae Raffia
09. Eugenia sp. Myrtaceae Rotra
10. Syzygium sp. Myrtaceae Rotra
Second choice, no ranking
11. Uapaca sp. Uapacaceae Voapaka
12. Faucherea sp. Sapotaceae Vasihy
13. Treculia madagascarica Moraceae Ampalibeala
14. Vangueria edulis Rubiaceae Vavandrika
15. Calopyxis sp. Combretaceae Voamatavy
16. Dilobeia thouarsii Protaceae Vivaona
17. Ficus tiliifolia Moraceae Voara
b) Exotic and naturalised species (no ranking)
01. Psidium guajava Myrtaceae Goavy be
02. Psidium cattleianum Myrtaceae Goavitsinahy
03. Passiflora incarnata Passifloraceae Garana
04. Passiflora quadrangularis Passifloraceae Barbajina
05. Citrus aurantium Rutaceae Voangidy
06. Citrus aurantifolia Rutaceae Tsioha
07. Citrus medica Rutaceae Voatolongo
08. Eugenia jambolana Myrtaceae Rotra vazaha
09. Eugenia jambos Myrtaceae Zamborozana

well known; (3) Some farmers weren’t aware of the commercialization
potential and, therefore, didn’t see an incentive for planting the trees; (4)
Ancestors did not plant forest trees and farmers were hesitant to change these
traditions.
Advantages of indigenous fruit tree planting were seen (1) in a higher
benefit from multiple uses of rare species; (2) in the high nutritional value of
the fruits; (3) in the provision of appreciated food to children; and (4) in the
adaptation of species to the local ecological conditions which facilitates a
successful cultivation of the trees in agroforestry or reforestation activities.
Ecological importance of indigenous fruit species
In addition to their contribution to family nutrient resources, native fruit
species have a major role to play in maintaining the process of forest regen-
eration.
Madagascar’s primary vegetation species have a weak natural regenera-
tive ability after slash and burn agriculture (Humbert, 1927; Lowry et al.,
1997). The disturbed vegetation is quickly invaded by a few aggressive exotic
species such as Pteridium sp., Clidemia hirta, Lantana camara, Psidium cat-
tleianum, that suppress the regrowth of native species. Under such condi-
tions floristic diversity of secondary vegetation may reach a pseudoclimax
stage and never attain its original floristic diversity (Humbert, 1927; Lowry
et al., 1997).
Madagascar has a low bird diversity and has only four frugivore bird
species (Landgrand, 1990). Nectar-feeding bat species, which are common
throughout most of the tropics, are also very rare in Madagascar with only
three species. In this absence of vertebrate competitors, lemurs have devel-
oped since the middle Cretaceous and Paleocene, a coevolutionary relation-
ship with flowering plants and play a crucial role as pollinator and seed
disperser of many primary forest species (Sussman and Raven, 1978; Kress,
1993). Close relationships have also been reported between seed ingestion
by lemurs and germination success in Malagasy monsoon rainforests (Scharfe
and Schlund, 1989). Smith and Ganzhorn (1996) observed that 70% of the
lemurs in 10 sites of evergreen forest of Madagascar were dominantly frugi-
vore. The distribution of the primate Aye-Aye Daubentonia madagascariensis
is closely linked with that of Canarium sp. (Iwano et al., 1991).
The complex interrelationships between mammals as pollinators and seed
dispersers have a direct impact on the regeneration potential of many primary
forest species. We hypothesize that increasing the indigenous fruit tree fre-
quency in the landscape would not only provide food for rapidly dwindling
lemur populations but would also facilitate and enhance fallow and forest
regeneration of indigenous species due to increased seed dispersal by the
lemurs. There is also the potential for increasing ecotourism activities and
thus farmer incomes as a result of enhanced lemur activity in the study area.
306

Summary and recommendations
During this ethnobotanical study, 150 wild fruit species belonging to 82 genera
and 42 families have been identified in the sites of Masoala, Andasibe and
Ranomafana of the Eastern Region of Madagascar. Eighty percent of species
were endemic and 92% were woody species. Two thirds of the species
occurred solely in one of the three sites, indicating a locally restricted species
distribution and a very high biodiversity.
We noticed in our study that there is not only a lack of botanical reference
specimens in Madagascar (25% of our botanical samples could only be iden-
tified to the genus level), but also a major knowledge gap on potential uses
of the primary vegetation species. The knowledge still resides with the rural
population living in close relationship with the forest. We also observed that
loss of local resource knowledge is underway where people live far away from
primary forest areas. The loss of knowledge is expressed in a considerable
knowledge difference between younger and older generations. Reduced
knowledge about forests and their uses leads to alienation and indifferent
attitudes of local people towards these resources and this results in haphazard
exploitation of the forest resources. The vicious circle of primary forest
destruction continues while traditional consciousness and natural resource
management skills are being progressively lost.
This study has shown that indigenous fruit trees are used and appreciated
by local people living close to primary forest. The major results of our study
are:
(i) A large variety of fruits are extracted from wild stands. Domestication
efforts were recognized for some selected species, by either protecting and
managing naturally grown species or by planting individual trees on field
boundaries.
(ii) Highest collection intensities for wild fruits were achieved close to the
habitations. These locations refer to homegardens, perennial crop fields (coffee
– banana fields, vanilla fields, clove fields), fallows and secondary vegeta-
tion. Fruits from the primary forest, even if most appreciated, were less often
collected and are therefore not optimally available to population.
(iii) Preferences towards the indigenous fruit tree species varied within
the community. Indigenous fruits can provide cash income for the poorer
farmers. Wild fruits have a supplementary nutritional role in the daily diet,
and contribute to food security in case of food shortages. Wild fruits are cer-
tainly most appreciated by children who collect and eat them regularly pro-
viding them with nutrients and vitamins. Adults preferred the fruit tree species
having various multiple uses. Differences in gender preference were related
to responsibilities in the household. While evaluating the potential of indige-
nous fruit trees as agroforestry species, the needs and preferences of the dif-
ferent users have to be taken into consideration, leading to different designs
of tree integration in the agroforestry system such as integration into home-
gardens, perennial fields, forest gardens etc.
307

Future research on evaluation of the domestication potential for the 17
indigenous priority species should include assessments on:
(i) Tree growth and fruit productivity
This includes identification of ecological requirements (soil, climate,
position in landscape and vegetation, and symbiotic relationships with
beneficial microorganisms such as mycorrhizas) for optimal growth and
fruit production, and the analysis of the potential of these species to accu-
mulate and concentrate essential nutrients for humans in their fruits.
Based on this information specific management interventions can be
proposed and tested for the cultivation of the species in order to increase
their productivity and the density of bioavailable nutrients in the agro-
forestry system.
(ii) Establishment and propagation
Needed baseline information involves reproduction biology, phenology
and nature of seeds. Next to establishment experiments on seed propa-
gation, vegetative propagation techniques (especially cuttings and air-
layerings) will be of importance, making the propagation of plants
independent on fruiting seasons and allowing the multiplication of
selected superior genotypes.
(iii) Economic considerations
Economic considerations will include species productivity, identifica-
tion of constraints and potentials on fruit marketing and product devel-
opment potentials.
(iv) Fruit quality and fruit processing
Fruit quality concerns nutritional value, taste, etc of the fruits as well as
their conservation properties and processing potentials
(v) People’s knowledge, perception and behavior
In depth assessments on people’s knowledge, perception and behavior
will provide a better understanding on constraints and potentials of indige-
nous fruit tree domestication. The information gathered will be impor-
tant for developing sustainable natural resource use strategies and for
the integration of indigenous fruit species into agroforestry systems
Acknowledgements
This research was funded by the PASN Project (Projet d’Appui au Silo
National), Intercooperation, Switzerland and the National Tree Seed Center
of Madagascar, SNGF (Silo National des Graines Forestières). We wish to
express our gratitude to the responsibles of these two institutions Philippe
Vaneberg, Guy Rakotondranony and Lolona Ramamonjisoa for their generous
support. We wish also to thank the Vita Project in Andasibe, National Park
Project in Ranomafana and the Masoala Project in Masoala who provided
excellent assistance during our fieldwork. The botanical species determina-
308

tions were possible with the generous cooperation of the National Herbarium
at the Department of Forestry and Fishery Research of FOFIFA, Antananarivo.
We are grateful to all the staff of the above mentioned projects who have
actively contributed to the success of this research.
References
Abadie MC (1953) Inventaire des espèces fruitières comestibles à Madagascar. Bulletin de
l’Académie Malgache 1951–1952 30: 185–204
Davis SD, Heywood VH and Hamilton AC (1994) Centers of Plant Diversity. A Guide and
Strategy for their Conservation. Volume 1: Europe, Africa, South West Asia, and the Middle
East. WWF and IUCN
Déjardin J, Guillaumet JL and Mangenot G (1973) Contribution à la connaissance de l’élément
non endémique de la flore malgache (végétaux vasculaires). Candollea 28: 325–391
FAO (1983) Food and Fruit-bearing Forest Species. 1: Examples from Eastern Africa. FAO
Forestry Paper 44/1. FAO Forestry Department, Rome, Italy
Ferraro PJ (1994) Natural resource use in the Southeastern rain forests of Madagascar and the
local impacts of establishing the Ranomafana National Park. MS Thesis. Duke University,
Durham, NC, 249 pp
Green GM and Sussman RW (1990) Deforestation history of the eastern rain forests of
Madagascar from satellite images. Science 248: 212–215
Gueye B and Schoonmaker Freudenberger K (1991) Introduction à la Méthode Accélerée de
Recherche Participative (MARP). Rapid Rural Appraisal. Quelques notes pour appuyer une
formation pratique. Deuxième édition. IIED (International Institute for Environment and
Development), London, UK, 70 pp
Hardenbergh SHB (1997) Why are boys so small? Child growth, diet and gender near
Ranomafana, Madagascar. Soc Sci Med 44(11): 1725–1738
Humbert H (1927) La déstruction d’une flore insulaire par le feu; Principaux aspects de la
végétation à Madagascar. Mémoires de l’Académie Malgache, Fascicule V, Antananarivo,
Madagascar, 79 pp
Humbert H (1959) Origines présumées et affinités de la flore de Madagascar. Mémoire d’Institut
Science Madagascar, Série b (Biologie et Végétation) 9: 149–187
Humbert H (1965) Description des types de végétation. In: Humbert H and Cours Darne G
(eds) Notice de la carte de Madagascar, Vol 6. Travaux de la Section Scientifique et
Technique de l’Institut Français de Pondichéry (hors série)
Isaia R (1995) Rapport de mission dans le cadre du programme sur l’utilisation de la forêt à
Ambanizana. In: Projet Masoala (ed) Proposition des limites du Parc National Masoala.
Soumise à la Direction des Eaux et Forêts par: Care International Madagascar, Wildlife
Conservation Society et Perigrine Fund. Projet Masoala, Antananarivo, Madagascar, Annexe
No. 3, 10 pp
Iwano T, Randalana R and Rakotoarisoa G (1991) Ecology of the aye-aye (Daubentonia mada-
gascariensis) I. Distribution. In: Akiyoshi E et al. (eds) Primatology Today, pp 41–42.
Elsevier Science Publishers B.V. (Biomedical Division)
Koechlin J, Guillaumet JL and Morat P (1974) Flore et végétation de Madagascar. J Cramer
Verlag, Vaduz, Liechtenstein, 687 pp
Kress WJ (1993) Coevolution of plants and animals: Pollination of flowers by primates in
Madagascar. Current Science 65(3): 253–257
Langrand O (1990) Guide to the Birds of Madagascar. Yale University Press, New Haven,
Connecticut, USA
Leakey RRB (1999) Potential for novel food products from agroforestry trees: a review. Food
Chemistry 66: 1–14
309

Leakey RRB and Newton AC (1994) Domestication of ‘Cinderella’ species as the star of a
woody-plant revolution. Preface. In: Leakey RRB and Newton AC (eds) Tropical Trees:
The Potential for Domestication and the Rebuilding of Forest Resources. Proceedings of a
Conference organized by ECTF at Edinburgh UK on 23–28 August 1992. ITE Symposium
No. 29, ECTF Symposium No. 1, pp 3–6. IUFRO, Edinburgh Centre for Tropical Forests
(ECTF) and Institute of Terrestrial Ecology (ITE), London, UK
Lowry II PP, Schatz GE and Phillipson PB (1997) The classification of natural and anthro-
pogenic vegetation in Madagascar. In: Goodman SM and Patterson PD (eds) Natural Change
and Human Impact in Madagascar, pp 93–123. Smithsonian Institution Press, Washington
and London
Maghembe JA, Kwesiga F, Ngulube M, Prins H and Malaya FM (1994) Domestication poten-
tial of indigenous fruit trees of the miombo woodlands of southern Africa. In: Leakey RRB
and Newton AC (eds) Tropical Trees: The Potential for Domestication and the Rebuilding
of Forest Resources. Proceedings of a Conference organised by ECTF at Edingburgh UK
on 23–28 August 1992. ITE Symposium No. 29, ECTF Symposium No. 1, pp 220–229.
IUFRO, Edinburgh Centre for Tropical Forests (ECTF) and Institute of Terrestrial Ecology
(ITE), London, UK
Martin GJ (1995) Ethnobotany. A Methods Manual. ‘People and Plants’ Conservation Manuals:
Volume 1. WWF International, UNESCO, Royal Botanic Gardens (Kew, UK), Chapman &
Hall, London, UK, 268 pp
Ministère de l’Economie et du Plan (1990) Charte de l’Environnement. Ministère de l’Economie
et du Plan, République Démocratique de Madagascar, Antananarivo, Madagascar
Perrier de la Bathie H (1936) Biogéographie des plantes de Madagascar. Paris, France, 156 pp
Razanaka SJ (1989) Délimitation des régions de provenances pour les semences forestières à
Madagascar. Mémoire de Diplôme d’Etudes Approfondies de Sciences Biologiques
Appliquées. Option Ecologie Forestière. Ecole supérieure des Sciences Agronomiques,
Département des Eaux et Forêts, Université d’Antananarivo, Antananarivo, Madagascar,
63 pp
Rocheleau DE (1988) Gender, resource management and the rural landscape: Implications for
agroforestry and farming systems research. In: Poats SV, Schmink M and Spring A (eds)
Gender Issues in Farming Systems Research and Extensions, pp 149–169. Westview Press,
Boulder
Scharfe F and Schlund W (1989) Seed dispersal of two tropical trees through lemurs in a
deciduous dry forest of western Madagascar. Primate Report 25: 82–83
Schatz GE, Lowry II PP, Lescot M, Wolf AE, Andriambololonera S, Raharimalala V and
Rahrimampionona J (1996) Conspectus of the vascular plants of Madagscar: a taxonomic
and conservation electronic database. In: van der Maesen LJG et al. (eds) The Biodiversity
of African Plants, pp 10–17. Kluwer Academic Publishers, the Netherlands
Smith AP and Ganzhorn JU (1996) Convergence in community structure and dietary adapta-
tion in Australian possums and gliders and Malagasy lemurs. Sixth International Theriological
Congress. Australian Journal of Ecology 21: 31–46
Sussman RW and Raven PH (1978) Pollination by lemurs and marsupials: An archaic coevo-
lutionary system. Science 200: 731–736
Turk D (1995) A Guide to Trees of Ranomafana National Park and Central Eastern Madagascar.
USAID/ANGAP, Antananarivo, Madagascar, 330 pp
Verheij EWM and Coronel RE (1992) Edible Fruits and Nuts. Plant Resources of South-East
Asia. No. 2. Prosea, Bogor, Indonesia, 446 pp
Villachica H (1996) Frutales y hortalizas promisorios de la Amazonia. Tratado de Cooperacion
Amazonica, Lima, Peru, 365 pp
310