Content uploaded by Tina Kyndt
Author content
All content in this area was uploaded by Tina Kyndt
Content may be subject to copyright.
Genetic fingerprinting using AFLP cannot distinguish
traditionally classified baobab morphotypes
A. E. Assogbadjo ÆT. Kyndt ÆF. J. Chadare ÆB. Sinsin Æ
G. Gheysen ÆO. Eyog-Matig ÆP. Van Damme
Received: 30 November 2007 / Accepted: 22 May 2008 / Published online: 6 June 2008
ÓSpringer Science+Business Media B.V. 2008
Abstract Baobab (Adansonia digitata L.) is one of
the predominant tree species in West African agrofor-
estry systems. A local morphological classification
system is used by farmers, identifying trees with
desired or undesired combinations of traits. This study
evaluates the genetic significance of these morpho-
types by comparing local identification with AFLP
marker information. Eight morphotypes were recog-
nized by seven ethnic groups from Benin, Ghana and
Senegal, among 182 sampled baobab trees. Five primer
pairs were used for DNA fingerprinting, resulting in a
total of 254 scored bands, of which between 94.1% and
100% was polymorphic within morphotypes. Gener-
ally, genetic fingerprinting did not correlate with the
traditional morphological identification of Adansonia
digitata. Probably, AFLP markers are not directly
linked to the differences in phenotype or the traits used
for the traditional classification are largely dependent
on environmental factors. Since no genetic differenti-
ation is found between the morphotypes, a
morphotype-based approach in the collection of
genetic variation for conservation programs is not
advisable.
Keywords Adansonia digitata Local
classification AFLP West Africa
Introduction
The multipurpose baobab tree (Adansonia digitata
L.) is expected to play a major role in future crop
diversification programs and in the development of
West-African agroforestry systems (IPGRI 1999).
This was also expressed by the rural people in
A. E. Assogbadjo and T. Kyndt equally contributed to this
work.
A. E. Assogbadjo F. J. Chadare B. Sinsin
Laboratory of Applied Ecology, Faculty of Agronomic
Sciences, University of Abomey-Calavi, 05 BP 1752
Cotonou, Benin
e-mail: assogbadjo@yahoo.fr
T. Kyndt (&)G. Gheysen
Department of Molecular Biotechnology, Ghent
University (UGent), Coupure Links 653, 9000 Ghent,
Belgium
e-mail: tina.kyndt@ugent.be
T. Kyndt G. Gheysen
Institute for Plant Biotechnology for Developing
Countries (IPBO), Ghent University, K.L.
Ledeganckstraat 35, 9000 Ghent, Belgium
O. Eyog-Matig
Bioversity International, c/o IITA, 08 BP 0932, Cotonou,
Benin
P. Van Damme
Laboratory of Tropical
and Subtropical Agriculture and Ethnobotany, Department
of Plant Production, Ghent University (UGent), Coupure
links 653, 9000 Ghent, Belgium
123
Agroforest Syst (2009) 75:157–165
DOI 10.1007/s10457-008-9157-y
semi-arid West Africa, who selected baobab as the
number one species that would merit to receive more
attention in future domestication programs by ICRAF
(Leakey and Simons 1998). Indeed, baobab tree
(Adansonia digitata L.) is a key economic species
used daily in the diet of rural communities in West
Africa (Assogbadjo et al. 2005a,b,2006; Codjia
et al. 2001,2003; Sidibe
´and Williams 2002). The
species contributes to rural incomes (Diop et al.
2005) and has various important medicinal and food
uses (Assogbadjo et al. 2005a,b; Delisle et al. 1997;
Diop et al. 2005; Sena et al. 1998; Sidibe
´and
Williams 2002; Yazzie et al. 1994). The participatory
domestication of indigenous fruits has been proposed
as an appropriate means to alleviate poverty (Poulton
and Poole 2001), and could also have positive
benefits on the environment since new plantings of
baobab would help to restore the declining resources
of this important tree.
Adansonia digitata L. (Bombacaceae family) is a
majestic tree from the African savannas. The
drought-tolerant baobab is a good fodder tree,
especially for wild browsing animals. Although local
traditions prohibit communities from cutting down
baobab trees in some regions, the natural reproduc-
tion and regeneration cycles of the baobab are
threatened due to damage caused by domestic
animals and seedlings clearance for other land use.
Within the species, there is evidence indicating the
existence of a number of local forms differing in
habit, vigor, size, quality of the fruits and foliar
vitamin content (Assogbadjo et al. 2005a; Gebauer
et al. 2002; Sidibe
´and Williams 2002; Maranz et al.
2008). In previous studies, focusing on Benin (Asso-
gbadjo et al. 2005a,2006), we observed a link
between morphological diversity of baobab and both
(1) abiotic, environmental factors and (2) genetic
basis of some specific traits.
Recently, an ethnobotanical survey of the percep-
tions and human/cultural meaning of morphological
variation, use forms, preferences (desirable/undesir-
able traits) and links between traits has been
undertaken by our research group (Assogbadjo et al.
2008). The results showed that local people apply a
morphological classification system for baobab trees
and are able to guide in selecting and collecting
germplasm from trees with preferred combinations of
traits. However, at present, nothing is known about
the genetic basis of the desired or undesired traits.
Because human interference in traditional agrofor-
estry systems is already known to have an important
impact on the genetic structure of tree species
through their effect on seed and pollen dispersal,
density, fragmentation and selection (Young and
Merriam 1994; Aldrich et al. 1998; Kelly et al.
2004), increased selection of the preferred morpho-
types might in the long term alter the population
genetic structure of this multipurpose tree.
The main objective of the present study is to
combine modern molecular tools and large-scale
ethnobotanical surveys to assess the relevance of folk
classification of baobab in three countries of West
Africa: Benin, Ghana and Senegal, in order to define
whether the locally recognised morphotypes are
genetically defined.
To this aim, the genetic diversity and differentiation
within and between locally characterized baobab
morphotypes were studied. Amplified fragment length
polymorphism (AFLP) analysis (Vos et al. 1995) was
applied to assess the intra-specific genetic diversity in
the context of those locally recognised morphotypes.
Materials and methods
Sampling and ethnobotanical survey
In this study baobab individuals from three countries
have been sampled in the Sudanian and Sudano-
Sahelian regions of West Africa: Benin, Ghana and
Senegal. Within each country, sampling of localities
has been done in the areas where local ethnic groups
use baobabs on daily basis and have been shown to
have an excellent knowledge on baobabs. This has
been assessed through both literature review and
surveys among local populations, with the help of the
forestry and agricultural department in each selected
country. Local informants participated and provided
information on a voluntary basis. Surveys have been
conducted among women and men randomly drawn
from seven ethnic groups. There were the Ditamari
ethnic group (Benin), the Grune, Dagbale, Kaseem and
Wale ethnic groups (Ghana); and the Wolof and Se
´re
´re
ethnic groups (Senegal) (Fig. 1). In each ethnic group
interviews have been done with men and women of
different ages. Ethnobotanical surveys were carried
out between October 2006 and January 2007. Inter-
views included questions on perception and human/
158 Agroforest Syst (2009) 75:157–165
123
cultural meaning of morphological baobab variation,
use forms, preferences (desirables/non desirable
traits), and links between traits according to local
people of different countries. For detailed results on
this survey we refer to Assogbadjo et al. (2008).
The survey revealed that local people are able to
recognize eight morphotypes, a morphotype being
defined as a group of baobabs sharing some linked
traits identified by the ethnic groups. The persons most
familiar with the traditionally recognized morphotypes
became our key people in each district and were asked
to participate in the selection of baobab individuals to
be sampled for DNA fingerprinting. In total, 18 key
people provided us their help, their number varying
between 2 and 3 per ethnic group. The sampled trees
were collected in local traditional agroforestry systems
and their total number did not reflect the population
densities of the species in the selected locality.
Eight different morphotypes of baobab were sampled
(Table 1), but not all morphotypes were distinguished
by all ethnic groups (Table 2). In total 182 individ-
ual baobab trees were sampled for this study.
Molecular analyses
DNA isolation was performed as previously described
in Assogbadjo et al. (2006). For AFLP analysis, two
different sets of pre-amplification products were
generated using an EcoRI-primer carrying zero selec-
tive nucleotides (E-0) in combination with a MseI-
primer with two selective nucleotides (either M-AC or
M-GC). For the final selective amplification, five
primer pairs (E-GT/M-ACGG; E-GT/M-ACGA;
E-GA/M-ACGC; E-TC/M-GCGA; E-AT/M-GCGG)
were chosen based on an initial screening for poly-
morphism among a limited number of samples and on
band consistency and repeatability, in the course of
previous work on the genetic diversity of baobab in
Benin (Assogbadjo et al. 2006). For each individual,
the DNA fingerprints were scored by visual inspection
Fig. 1 Studied countries
(shown in white) and ethnic
groups. The number in
brackets (n) represents the
total number of the sampled
baobab trees per ethnic
group
Table 1 Matrix of identified baobab morphotypes
Type Pulp characteristics Leaves
characteristics
Fertility of tree
Sweet Slimy Acidic Tasty Bitter Unfertile Fertile
A+-- + - - +
B+-- - + - +
C--+ + - - +
D-+- + - - +
E--- - + + -
F--+ - + - +
G--- + - - +
H--- + - + -
Characteristics of the locally recognised morphotypes are
indicated as present (+) or absent (-)
Agroforest Syst (2009) 75:157–165 159
123
for presence (1) or absence (0) of specific AFLP-bands.
Only distinct, major bands were scored.
Data matrices were analyzed using Treecon 1.3b
(Van de Peer and De Wachter 1994). Genetic similar-
ities were calculated using Jaccard’s coefficient
(Jaccard 1908) and the resulting matrices were ana-
lyzed using the UPGMA method. Allele-frequency
based analyses of genetic diversity and structure were
performed using AFLPsurv version 1.0. (Vekemans
2002) which is based on the methods described by
Lynch and Milligan (1994). Allelic frequencies at
AFLP loci were estimated from the binary presence/
absence matrix, under the assumption of Hardy–
Weinberg equilibrium, from the observed frequencies
of fragments using the Bayesian approach proposed by
Zhivotovsky (1999). A non-uniform prior distribution
of allelic frequencies was assumed with its parameters
derived from the observed distribution of fragment
frequencies among loci (see note 4 in Zhivotovsky
1999). Nei’s (1973) gene diversity (also known as
expected heterozygosity) as well as global and pairwise
genetic differentiation (F
ST
) values were computed.
Significance of the genetic differentiation between
groups was tested by comparison of the observed F
ST
with a distribution of F
ST
under a hypothesis of no
genetic structure, obtained by means of 1,000 random
permutations of individuals among groups.
A model-based (Bayesian) clustering method was
applied on the presence/absence matrix to infer
genetic structure in the dataset, using the software
Structure version 2.0. (Pritchard et al. 2000). Apply-
ing a ‘no admixture’ model (250,000 iterations) and
without using prior information of the number of
populations (USEPOPINFO =0), different K-values
(2–16) were evaluated, in order to estimate the number
of genepools present in the dataset. Individuals of the
eight morphotypes were then assigned probabilisti-
cally to the inferred gene pools.
Results
Indigenous characterization of baobab in the
parklands systems of West Africa
Adansonia digitata is a multipurpose tree species daily
used by local people of West Africa for food,
medicine, cultural, artistic, agronomic and commer-
cial purposes. All baobab products have at least one
utilization in rural areas of West Africa. The leaves,
pulp, kernel are used as food whereas all these organs
as well as capsule, sap, bark, branches and roots are
incorporated into the traditional pharmacopoeia.
Apart from its therapeutic and food uses, baobab is
considered as a fetish tree, sacred, deified and full of
mysteriousness.
In West Africa, local perceptions of baobab
differentiation vary from one country to another.
Local people recognized 21 traits, which can be used
as criteria to distinguish baobab individuals in the
parklands systems. For a more detailed description
and analysis of the results of the ethnobotanical
surveys, we refer to Assogbadjo et al. (2008).
The most commonly used criteria to differentiate
among baobab individuals were: leaf taste, pulp taste,
the sliminess of fruit pulp and the fertility of baobab
trees. Based on locally recognized variants, eight
different morphotypes of baobab were distinguished
Table 2 Number of
baobab individuals per
morphotype (A–H) as
sampled by seven ethnic
groups from three West
African countries
Morphotypes Benin Ghana Senegal Total
Ditamari Dagbale Grune Kaseem Wale Wolof Se
´re
`re
A 45 23 8 0 11 5 8 100
B60000006
C1940446845
D40000239
E10000012
F10020104
G20000024
H010203612
28 8 8 15 17 28
Total 78 59 45 182
160 Agroforest Syst (2009) 75:157–165
123
among the 182 individuals sampled in the traditional
agroforestry systems of West Africa (Tables 1and 2).
Figure 1shows the geographical location of the
studied ethnic groups and the number of baobabs
which they have morphotyped.
For instance, morphotype A groups ‘‘female’’
baobabs (fertile) always producing sweet pulp and
tasty leaves whereas morphotype H groups ‘‘male’’
baobabs (infertile) producing tasty leaves (Table 1).
It is important to notice that baobabs are hermaph-
rodites (Wickens 1982) and the so-called ‘male’ tree
(a term used by local people) is based only on local
perception since this kind of baobab never produces
(mature) fruits. In reality, they are not really biolog-
ically male. The fact that they do not produce fruits
may be due to the incompatibility of the reproduction
system or genetic inbreeding.
For some further statistical analyses, only morpho-
types with at least four individuals were considered.
AFLP-based genetic variation within and between
morphotypes
Considering AFLP fingerprints for all analysed
individuals, the five primer combinations resulted in
a total of 254 scored bands of which 83.85% was
polymorphic. UPGMA-clustering based on Jaccard’s
dissimilarity coefficient showed no clear grouping
according to morphotype classification but rather
correlated with the geographical distribution of the
samples (results not shown).
Estimates of within-morphotype genetic diversity
were calculated using AFLPsurv and the results are
summarized in Table 3. Generally, Nei’s (1973) gene
diversity within morphotypes ranged between 0.31
and 0.37, indicating a substantial amount of variation
within morphotype. The values of gene diversity
within morphotypes in each country ranged from 0.29
to 0.37. Levels of polymorphism within morphotypes
varied between 94.1% and 100% reflecting a high
level of polymorphism within morphotypes.
Analysis of population structure with allele-fre-
quency based F-statistics revealed that morphotypes
are not significantly differentiated from each other at
genetic level. A non-significant F
ST
value was
observed within all countries and between all distin-
guished morphotypes (Table 4). Recently, a spatial
population genetic structure analysis of baobab was
performed by our research group (Assogbadjo et al.
submitted), revealing spatial autocorrelation at both
large and regional geographical scale. To rule out bias
coming from country-scale spatial genetic structure,
the ethnic group that distinguished the largest number
of different morphotypes was selected for each country
and the differentiation between the baobab individuals
identified by that specific ethnic group was evaluated.
Again, no differentiation between morphotypes iden-
tified by the Ditamari (Benin), the Dagbale (Ghana) or
the Se
´re
`re (Senegal) was found (P[0.05) (Table 4).
Also, pairwise F
ST
-values confirmed that none of the
analysed morphotypes are genetically differentiated
(P[0.05) (data not shown).
A model-based clustering method (Pritchard et al.
2000) using a no admixture model with corre-
lated allele frequencies was performed in Structure
2.0, in order to detect genetic structuring in the analyzed
Table 3 Genetic diversity within baobab morphotypes identified by local people in West Africa, expressed as polymorphism rate
and Nei’s gene diversity, based on AFLP data
Morphotypes Polymorphism (%) Nei’s gene diversity ±SD
Benin Ghana Senegal Global
A 94.1 0.29 ±0.01 0.33 ±0.01 0.29 ±0.01 0.35 ±0.01
B 94.5 0.37 ±0.01 na na 0.37 ±0.01
C 98.8 0.35 ±0.01 0.33 ±0.01 0.30 ±0.01 0.36 ±0.01
D 100.0 0.37 ±0.01 na 0.33 ±0.01 0.36 ±0.01
E 100.0 na na na 0.31 ±0.01
F 100.0 na na na 0.37 ±0.01
G 100.0 na na na 0.35 ±0.01
H 98.0 na na 0.30 ±0.01 0.36 ±0.01
na, Not applicable (low size class)
Agroforest Syst (2009) 75:157–165 161
123
sample set. The results showed that clustering all
genotypes into 11 gene pools correlates with a maxi-
mum estimate of the likelihood of the data (data not
shown) and indicate that morphotype classification does
not correlate with molecular marker-based gene pool
assignment of the individual baobab trees. For instance,
the 100 individuals identified as morphotype A, are
scattered across all inferred gene pools in the sample set.
But also for morphotypes with a low number of
representatives, like E, F and G, individuals show
maximum diversity in gene pool assignment. In
conclusion, our results show that local morphotype
classification is not correlated with AFLP fingerprinting
in West African baobab.
Discussion
The use of traditional classification occurs generally in
those instances where a species has attained a high
degree of cultural significance (Bye 1993). Indeed,
A. digitata played an important nutritional and cultural
role in West Africa over centuries (Wickens 1982;
Sidibe
´and Williams 2002). Generally, considerable
variation in local knowledge was found of both
utilization and classification of A. digitata in the
parkland systems of West Africa (Assogbadjo et al.
2008). Recognition of the different forms of A. digitata
is believed to be of ancient origin, and knowledge of its
use, names, and classification schemes has been passed
on from older members of the society to younger ones.
Although the large genome of Adansonia digitata
(1C equals 3,773 Mbp according tothe RBG Kew DNA
C-values database) cannot be completely sampled by
AFLP (or any other molecular marker system), AFLP
markers are known to map throughout the genome of
any particular species analyzed so far, and therefore this
high-throughput DNA fingerprinting techniques gives
fast and efficient measurements of genome-wide diver-
sity. We used this technique to investigate if the
traditional classifications of A. digitata are confirmed
by genome-level genetic differentiation. This compar-
ison between an ethnobotanical survey and a genetic
analysis of baobab individuals, shows that AFLP
fingerprinting cannot distinguish the traditionally clas-
sified forms of A. digitata, although our studies have
shown before that AFLP is a very useful technique to
distinguish baobab genotypes (Assogbadjo et al. 2006).
The locally recognized morphotypes seem to include a
substantial amount of genetic variation and are not
differentiated from each other on genome-wide scale.
This means that the traditional selection of morphotypes
with desired traits will not directly alter the natural
population genetic structure of baobab.
High polymorphism and genetic diversity rates
within morphotypes and lack of differentiation between
them can be due to the fact that AFLPs are neutral
markers that are not correlated or directly linked to
differences in phenotype. In addition, the lack of
observed genetic differences between morphotypes
may be a result of strong non-genic effects underlying
these traditional classifications. Phenotypic differences
observed between locally recognized morphotypes
might be plastic responses to differences in environ-
ment and habitat. This phenomenon of phenotypic
plasticity was found in several species of tropical and
temperate trees for many traits, usually in response to
changes in climate (Kramer 1995; Heaton et al. 1999).
A link between the morphometric diversity of baobab
and abiotic/environmental factors (rainfall, relative
Table 4 Genetic diversity and differentiation of baobab morphotypes identified by local people in West Africa, as calculated with
the Bayesian method using AFLPsurv 1.0
Country Ethnic groups N Ht Hw Hb F
ST
Probability (F
ST
)
Benin Ditamari 4 0.34 0.34 -0.002 -0.01 P=0.83
Ghana Dagbale, Grune, Kaseem and Wale 2 0.33 0.33 -0.003 -0.01 P=0.77
Dagbale 2 0.33 0.32 0.003 0.01 P=0.53
Senegal Wolof and Se
´re
`re 4 0.30 0.30 -0.005 -0.02 P=0.77
Se
´re
`re 4 0.30 0.30 -0.002 -0.01 P=0.53
Global Ditamari, Dagbale, Grune, Kaseem,
Wale, Wolof and Se
´re
`re
8 0.36 0.35 0.009 0.02 P=0.38
Ht, total genetic diversity; Hw, genetic diversity within morphotypes; Hb, genetic diversity between morphotypes; F
ST
, genetic
differentiation between morphotypes
162 Agroforest Syst (2009) 75:157–165
123
humidity, potential evaporation, type of soil, etc.) has
already been demonstrated(Assogbadjo et al. 2005a,b).
Baobab characteristics such as pulp and leaf taste,
fertility of the tree (which are the traits used for the
traditional classification in this study) might thus be to a
large extent dependent on environmental factors. It has
to be noted that the limited number of analyzed
perceived traits might not adequately reflect the total
morphometric diversity of the species, but previous
studies on baobab in Benin as well reported that, except
for three specific morphological characteristics (height
of the trees, number of branches and thickness of the
capsules), no significant correlation between the general
morphometric diversity and genetic dissimilarity was
observed (Assogbadjo et al. 2006). On the other hand,
genetic differentiation between populations from dif-
ferent climatic zones in Benin was relatively high
(Assogbadjo et al. 2006). Recently, our research group
has performed a detailed study on spatial genetic
structuring of baobab populations across four West
African countries, showing isolation-by-distance at
large and regional geographical scale (Assogbadjo et al.
submitted).
A similar lack of correlation with AFLP fingerprint-
ing was also observed for traditional morphological
classification of Opuntia pilifera (Nilsen et al. 2005).
Other studies involving genetic markers and morpho-
logical differences have found that there is generally no
correlation between phenotypic differences and varia-
tion in genetic markers (Borghetti et al. 1993;Katajima
et al. 1997; Venable et al. 1998; Heaton et al. 1999;
Geleta et al. 2006). On the other hand, environmental
effects on the morphological variables have been
observed in several edible trees in Africa. Maranz and
Wiesman (2003) showed for the shea tree (Vitellatria
paradoxa) a significant relationship between trait values
(fruit size and shape, pulp sweetness, and kernel content
of the species) and abiotic variables (temperature and
rainfall) in sub-Saharian Africa north of the equator.
Also, Soloviev et al. (2004)showedforBalanites
aegyptiaca and Tamarindus indica (savanna trees) the
significant influence of different climatic zones of
Senegal on fruit pulp production.
Since no genetic differentiation is found between
the locally recognized morphotypes, a morphotype-
based approach in the collection of genetic variation
for conservation programs is not advisable. As such,
conservation strategies in West African baobab
should target not only the various morphotypes but
should also incorporate data on morphometric diver-
sity and molecular information on genetic diversity.
The amount of genetic variation in terms of genetic
diversity and differentiation among populations
should be considered to estimate both the number
of individuals and populations to be sampled to
capture a sufficient amount of the species’ genetic
variability, and to select the most threatened popu-
lations for conservation programmes. Next to AFLP
fingerprinting, the application of codominant molec-
ular marker methods (like SSR, which is at present
not available for baobab) would provide useful
information on ploidy, heterozygosity and levels of
inbreeding for baobab populations. To avoid the
negative impact that can result from artificial selec-
tion in traditional agroforestry systems, it is
recommended to conserve the non-desirable baobabs
ex situ in gene banks. At the same time, desirable
baobabs can be preferably conserved in situ as natural
gene pools in traditional agroforestry systems.
Farmers are able to guide breeders in collecting
germplasm from trees since they have knowledge to
distinguish types of baobab. This can allow selecting
the ‘‘best’’ candidate trees for propagation, and plan-
ning a domestication programme, combining
indigenous knowledge and genetic findings. In order
to assess the genetic base of economically interesting
traits for baobab breeding a genetic map of the species
should be constructed. A linked marker analysis (e.g.
QTL) will be useful to aid in the identification of
specific genomic regions involved in the variation of
interesting traits for breeding. Most morphological
traits are polygenic, and the information obtained from
molecular markers can be used to facilitate the strategic
planning of new breeding approaches. However, since
baobab is a long generation species, mapping studies
are a challenging goal for the future.
Acknowledgements This work was supported by Bioversity
International and Pioneer Hi-Bred International Inc., a Dupont
Company, through the receipt of a Vavilov-Frankel Fellowship.
We also thank DADOBAT-Project (EU-Funding) and The
Rufford Maurice Laing Foundation for its additional financial
support through The Rufford Small Grant for Nature
Conservation as well as The King Leopold III Fund for Nature
Conservation and Exploration for its financial support for the
fieldwork in West Africa. We thank all these institutions and
their donors. Our acknowledgements also go to local people of
Benin, Ghana and Senegal and to Ir. Hugues Akpona (LEA-
FSA-Benin), Dr. Dogo Seck (CERRAS-Senegal), Dr.
Macoumba Diouf (ISRA-Senegal) and Mr. Joseph Mireku
(FORIG-Ghana).
Agroforest Syst (2009) 75:157–165 163
123
References
Aldrich PR, Hamrick JL, Chavarriaga P, Kochert G (1998)
Microsatellite analysis of demographic genetic structure
in fragmented populations of the tropical tree Symphonia
globulifera. Mol Ecol 7:933–944. doi:10.1046/j.1365-
294x.1998.00396.x
Assogbadjo AE, Sinsin B, Van Damme P (2005a) Caracte
`res
morphologiques et production des capsules de baobab
(Adansonia digitata L.)auBe
´nin. Fruits 60(5):327–340.
doi:10.1051/fruits:2005039
Assogbadjo AE, Sinsin B, Codjia JTC, Van Damme P (2005b)
Ecological diversity and pulp, seed and kernel production
of the baobab (Adansonia digitata) in Benin. Belg J Bot
138(1):47–56
Assogbadjo AE, Kyndt T, Sinsin B, Gheysen G, Van Damme P
(2006) Patterns of genetic and morphometric diversity in
baobab (Adansonia digitata L.) populations across dif-
ferent climatic zones of Benin (West Africa). Ann Bot
(Lond) 97:819–830. doi:10.1093/aob/mcl043
Assogbadjo AE, Gle
`le
`Kakaı
¨R, Chadare FJ, Thomson L,
Kyndt T, Sinsin B, et al (2008) Folk classification, per-
ception and preferences of baobab products in West
Africa: consequences for species conservation and
improvement. Econ Bot 62(1):74–84. doi:10.1007/
s12231-007-9003-6
Bye R (1993) The role of humans in the diversification of
plants in Mexico. In: Ramamoorthy TP, Bye R, Lot A, Fa
JE (eds) Biological diversity of Mexico: origins and dis-
tribution. Oxford University Press, New York
Borghetti M, Leonardi S, Raschi A, Synderman D, Tognetti R
(1993) Ecotypic variation of xylem embolism, phenolog-
ical traits, growth parameters and allozyme characteristics
in Fagus sylvatica. Funct Ecol 7:713–720. doi:10.2307/
2390193
Codjia JTC, Fonton-Kiki B, Assogbadjo AE, Ekue MRM
(2001) Le baobab (Adansonia digitata), une espe
`ce a
`
usage multiple au Be
´nin. CECODI/CBDD/Veco/SNV/
FSA, Cotonou
Codjia JTC, Assogbadjo AE, Ekue
´MRM (2003) Diversite
´et
valorisation au niveau local des ressources forestie
`res al-
imentaires ve
´ge
´tales du Be
´nin. Cah Agric 12:321–331
Delisle H, Bakari S, Gevry G, Picard C, Ferland G (1997)
Provitamin A content of traditional green leaves from
Niger. Original title: Teneur en provitamine A de feuilles
vertes traditionnelles du Niger. Cah Agric 6(8):553–560
Diop AG, Sakho M, Dornier M, Cisse M, Reynes M (2005) Le
baobab africain (Adansonia digitata L.): principales ca-
racte
´ristiques et utilisations. Fruits 61(1):55–69. doi:
10.1051/fruits:2006005
Gebauer J, El-Siddig K, Ebert G (2002) Baobab (Adansonia
digitata L.): a review on a multipurpose tree with prom-
ising future in the Sudan. Gartenbauwissenschaft
67(4):155–160
Geleta N, Labuschagne MT, Viljoen CD (2006) Genetic
diversity analysis in sorghum germplasm as estimated by
AFLP, SSR and morpho-agronomical markers. Biodivers
Conserv 15:3251–3265. doi:10.1007/s10531-005-0313-7
Heaton HJ, Whitkus R, Gomez-Pompa A (1999) Extreme
ecological and phenotypic differences in the tropical tree
chicozapote (Manilkara zapota (L.) P. Royen) are not
matched by genetic divergence: a random amplified
polymorphic DNA (RAPD) analysis. Mol Ecol 8:627–
632. doi:10.1046/j.1365-294x.1999.00616.x
IPGRI (1999) Vers une approche re
´gionale des ressources
ge
´ne
´tiques forestie
`res en Afrique sub-saharienne. Actes
du premier atelier re
´gional de formation sur la conserva-
tion et l’utilisation durable des ressources ge
´ne
´tiques
forestie
`res en Afrique de l’Ouest. Afrique Centrale et de
Madagascar, Ouagadougou, Burkina-Faso
Jaccard P (1908) Nouvelles recherches sur la distribution flo-
rale. Bull Soc Vaud Sc Natur 44:223–270
Katajima K, Mulkey SS, Wright SJ (1997) Seasonal leaf phe-
notypes in the canopy of a tropical dry forest:
photosynthetic characteristics and associated traits. Oec-
ologia 109:490–498. doi:10.1007/s004420050109
Kelly BA, Hardy OJ, Bouvet J-M (2004) Temporal and spatial
genetic structure in Vitellaria paradoxa (shea tree) in an
agroforestry system in southern Mali. Mol Ecol 13:1231–
1240. doi:10.1111/j.1365-294X.2004.02144.x
Kramer K (1995) Phenotypic plasticity of the phenology of
seven European tree species in relation to climatic
warming. Plant Cell Environ 18:93–104. doi:10.1111/
j.1365-3040.1995.tb00356.x
Leakey RRB, Simons AJ (1998) The domestication and com-
mercialization of indigenous trees in agroforestry for the
alleviation of poverty. Agrofor Syst 38:165–176. doi:
10.1023/A:1005912729225
Lynch M, Milligan BG (1994) Analysis of population genetic
structure with RAPD markers. Mol Ecol 3:91–99. doi:
10.1111/j.1365-294X.1994.tb00109.x
Maranz S, Wiesman Z (2003) Evidence for indigenous selec-
tion and distribution of the shea tree, Vitellaria paradoxa,
and its potential significance to prevailing parkland
savanna tree patterns in sub-Saharan Africa north of the
equator. J Biogeogr 30:1505–1516. doi:10.1046/j.1365-
2699.2003.00892.x
Maranz S, Niang A, Kalinganire A, Konate
´D, Kaya B (2008)
Potential to harness superior nutritional qualities of exotic
baobabs if local adaptation can be conferred through
grafting. Agrofor Syst 72:231–239. doi:10.1007/s10457-
007-9093-2
Nei M (1973) Analysis of gene diversity in subdivided popu-
lations. Proc Natl Acad Sci USA 70:3321–3323. doi:
10.1073/pnas.70.12.3321
Nilsen LB, Shivcharn SD, Sara LACR, N-Aguilar BR, Heun M
(2005) Traditional knowledge and genetic diversity of
Opuntia pilifera (cactaceae) in the Tehuaca
´n-Cuicatla
´n
Valley, Mexico. Econ Bot 59(4):366–376. doi:10.1663/
0013-0001(2005)059[0366:TKAGDO]2.0.CO;2
Poulton C, Poole N (2001) Poverty and fruit tree research:
issues and options paper. DFID Forestry Research Pro-
gramme, Wye College, Ashford
Pritchard JK, Stephens M, Donnelly P (2000) Inference of
population structure using multilocus genotype data.
Genetics 155:945–959
Sidibe
´M, Williams JT (2002) Baobab Adansonia digitata.
International Centre for Underutilised Crops, Southamp-
ton, UK P 100, ISBN 0854327762
Sena LP, Vanderjagt DJ, Rivera C, Tsin ATC, Muhamadu I,
Mahamadou O et al (1998) Analysis of nutritional
164 Agroforest Syst (2009) 75:157–165
123
components of eight famine foods of the Republic of
Niger. Plant Foods Hum Nutr 52(1):17–30. doi:10.1023/
A:1008010009170
Soloviev P, Niang TD, Gaye A, Totte A (2004) Variabilite
´des
caracte
`res physico-chimiques des fruits de trois espe
`ces
ligneuses de cueillette, re
´colte
´sauSe
´ne
´gal: Adansonia
digitata, Balanites aegyptiaca et Tamarindus indica. Fruits
59:109–119. doi:10.1051/fruits:2004011
Van de Peer Y, De Wachter R (1994) TREECON for windows:
a software package for the construction and drawing of
evolutionary trees for the Microsoft Windows environ-
ment. Comput Appl Biosci 10:569–570
Vekemans X (2002) AFLP-surv version 1.0. Distributed by the
author. Laboratoire de Ge
´ne
´tique et Ecologie Ve
´ge
´tale,
Universite
´Libre de Bruxelles, Belgium
Venable Dl, Dyreson E, Pinero D, Becarra JX (1998) Seed
morphometrics and adaptive geographic differentiation.
Evolution 52:344–354. doi:10.2307/2411072
Vos P, Hogers R, Bleeker M et al (1995) AFLP—a new
technique for DNA fingerprinting. Nucleic Acids Res
23:319–332
Wickens GE (1982) The baobab—Africa’s upside-down tree.
Kew Bull 37:173–209. doi:10.2307/4109961
Yazzie D, Vanderjagt DJ, Pastuszyn A, Okolo A, Glew RH
(1994) The amino acid and mineral content of baobab
(Adansonia digitata L.) leaves. J Food Compost Anal
7(3):189–193. doi:10.1006/jfca.1994.1018
Young AC, Merriam HG (1994) Effect of forest fragmentation
on the spatial genetic structure of Acer saccharum Marsh.
(sugar maple) populations. Heredity 72:201–208. doi:
10.1038/hdy.1994.27
Zhivotovsky LA (1999) Estimating population structure in
diploids with multilocus dominant markers. Mol Ecol
8:907–913. doi:10.1046/j.1365-294x.1999.00620.x
Agroforest Syst (2009) 75:157–165 165
123
