Content uploaded by Enoch Gbenato Achigan-Dako
Author content
All content in this area was uploaded by Enoch Gbenato Achigan-Dako on Nov 25, 2020
Content may be subject to copyright.
ORIGINAL ARTICLE
Flow cytometric analysis in Lagenaria siceraria (Cucurbitaceae)
indicates correlation of genome size with usage types
and growing elevation
Enoch G. Achigan-Dako ÆJo
¨rg Fuchs Æ
Adam Ahanchede ÆFrank R. Blattner
Received: 14 March 2008 / Accepted: 21 July 2008
ÓSpringer-Verlag 2008
Abstract The occurrence and extent of genome size
variation within species is controversially discussed and
thorough analyses are rare. Given the large morphological
variation in Lagenaria siceraria (bottle gourds) and its
wide distribution in Africa we here analysed (1) the gen-
ome size variation within cultivars of L. siceraria, (2) the
correlation between genome size and morphological traits,
and (3) the geographical patterns of DNA content within
the species. We measured 2C-values of 366 individuals
from 117 accessions of L. siceraria (2n=22) from Africa,
America and Asia via flow cytometry with propidium
iodide as DNA stain. We found that 2C-value in L. sice-
raria (0.683–0.776 pg/2C) is about two times lower than
previously reported and varies by about 12% among all
accessions. Moreover, our results indicated a clear corre-
lation of genome size with two different seed or usage
types and with growing elevations in West Africa. Within
the seed types genome size varies by 6.6 and 7.5%,
respectively. The genome size differences in seed types of
L. siceraria might indicate differences in their evolutionary
history and necessitates a re-evaluation of the phylogenetic
relationships within L. siceraria while the correlation
between 2C-value and the elevation of the collecting sites
might indicate an adaptation of genome size to an unknown
ecological parameter connected to altitude.
Keywords C-value Cucurbitaceae Flow cytometry
Genome size Gourd Lagenaria siceraria Taxonomy
Introduction
The genus Lagenaria Ser. contains five wild and one
cultivated species (Jeffrey 1967,1995; Keraudren 1967;
Adjakidje
`2006). All wild taxa thrive in Africa. Lagenaria
siceraria (Molina) Standley, commonly known as the
bottle gourd, is the only cultivated species. This species
originated from Africa or could have been independently
domesticated in Africa and Asia (Heiser 1979; Decker-
Walters et al. 2001; Decker-Walters et al. 2004; Erickson
et al. 2005) and is cultivated today all over the tropics.
Lagenaria siceraria is diploid with 2n=22 chromo-
somes (Beevy and Kuriachan 1996). It accounts for
tremendous morphological variation particularly with
respect to fruit and seed size, shape, and colour, rind
hardness, etc. (Heiser 1979; Decker-Walters et al. 2001;
Morimoto and Mvere 2004; Morimoto et al. 2005; Achigan
Dako et al. 2008). The species is monoecious and a strong
climber or trailer. The fruit is a berry, often globular,
bottle- or club-shaped, white-yellow to dark green when
young, with a hard and durable rind. The seeds are oblong,
generally up to 2 cm long, emarginated at the base, with
two facial ridges, mostly smooth, whitish to brownish.
Electronic supplementary material The online version of this
article (doi:10.1007/s00606-008-0075-2) contains supplementary
material, which is available to authorized users.
E. G. Achigan-Dako (&)F. R. Blattner
Department of Taxonomy and Evolutionary Biology,
Leibniz Institute of Plant Genetics and Crop Plant Research
(IPK), Corrensstr. 3, 06466 Gatersleben, Germany
e-mail: eachigan@ipk-gatersleben.de; dachigan@gmail.com
J. Fuchs
Cytogenetics and Genome Analysis, Leibniz Institute of Plant
Genetics and Crop Plant Research (IPK), Corrensstr. 3,
06466 Gatersleben, Germany
E. G. Achigan-Dako A. Ahanchede
Plant Science Laboratory, Faculty of Agronomic Sciences,
University of Abomey Calavi, 01 BP 526 Cotonou, Benin
123
Plant Syst Evol
DOI 10.1007/s00606-008-0075-2
L. siceraria has been most widely used as bottles or con-
tainers for both liquid and dry materials. The fruits are also
used for floats, music instruments, medicine, artistic
endeavours, and as an almost indispensable item in
human’s attire (Heiser 1979). Furthermore, young, tender
and non-bitter fruits are partly used as vegetables
(Morimoto and Mvere 2004; Schippers 2004), and in West
Africa so-called Egusi types exist, which are grown for the
consumption of their seeds instead of the fruits (Zoro Bi
et al. 2003; Schippers 2004; Achigan Dako et al. 2006,
2008).
Although L. siceraria is known for its high morpho-
logical variability of seeds and fruits, no systematic
investigation of the genome size in different accessions
has been performed yet. A single genome size estimation
was published for L. siceraria (Ingle et al. 1975) and no
intraspecific or intrapopulation variation of C-values has
been analysed for this or other Lagenaria species. Pro-
nounced differences in genome size within plant species
were often reported (e.g. Price et al. 1981; Rayburn et al.
1989; Graham et al. 1994; Bennett and Leitch 1995;
Poggio et al. 1998). However, Obermayer and Greilhuber
(2005) argue that intraspecific variation in genome size
must be considered with due scepticism, as many older
studies seemingly were hindered by methodological flaws
and often reported differences due to measuring artefacts.
Careful analyses arrived mostly at relatively uniform
values within single species, while differences between
closely related species might be much higher (Jakob
et al. 2004; Mahelka et al. 2005). Although intraspecific
variation was found in samples from distant populations
(Jakob et al. 2004; Schmuths et al. 2004;S
ˇmarda and
Bures
ˇ2006) genome size differences are mostly corre-
lated with ecological differences of the habitat (e.g.
Poggio et al. 1998; Kalendar et al. 2000; Knight and
Ackerly 2002; Jakob et al. 2004) or differences in plant
phenotype (Knight et al. 2005; Murray 2005; Beaulieu
et al. 2007). Intraspecific genome size variation may thus
also indicate micro-evolutionary differentiation and could
be taxonomically significant. Unfortunately, very few
reports of intraspecific genome size variation were cor-
related with plant phenotype (Meagher and Costish 1996;
Meagher et al. 2005) or phylogeny (Jakob et al. 2004).
Given the large morphological variation in L. sice-
raria and its wide distribution in Africa we here
analysed (1) the genome size variation within cultivars
of L. siceraria, (2) the correlation between genome size
and morphological traits, and (3) the geographical pat-
terns of DNA content within the species. Although we
analysed mostly material from western Africa, we also
included samples from Asia and the New World to see
how representative the African material is for the entire
taxon.
Materials and methods
Plant material and seed characters measurement
We used 366 individuals from 117 accessions of L. sice-
raria from West Africa, Asia and America. Accessions are
made up of 74 bottle gourd types from Benin, Ghana, Niger
and Mali, 11 gourd types from China, Korea, Irak and
Georgia, one from Cuba, and 31 Egusi types from Benin
and Ghana (supplement online material Table S1). The
gourd types were described in detail by Heiser (1979: pp.
94–96). The Egusi type is a non-hard-coat form of L.
siceraria mainly cultivated for the use of its seeds. Egusi
(in Yoruba language) refers to a group of cucurbit species
that produce protein and oil rich seeds (Burkill 1985;
Norman 1992). They consist mainly of species such as
Citrullus lanatus subsp. mucosospermus Fursa, known as
Egusi watermelon, and Cucumeropsis edulis (Hook. f.)
Cogn., considered as the true Egusi (Burkill 1985). The
Egusi cultivars of L. siceraria are known as ‘Aklampa’ or
‘Accra-kakoun’ in Benin, ‘Apantra kesa’ in Ghana, or
‘Se
ˆre
ˆgbe
ˆ’inCo
ˆte d’Ivoire (Zoro Bi et al. 2003; Schippers
2004; Achigan-Dako et al. 2006,2008). Accessions from
Africa were collected in 2005, 2006 and 2007 by the first
author in the indicated countries covering four phytogeo-
graphic regions of West Africa: the shrubs and woodland
savannah in the Sudanian phytoregion, the woodland
savannah in the Guineo-Sudanian region, the mosaic of
forests and savannahs in the so-called Dahomey gap
(White 1983; Salzmann and Hoelzmann 2005), and the
evergreen forests and semi-deciduous forests of the Gui-
neo-Congolian region. Accessions from Asia and America
were provided by the Genbank of the Leibniz-Institute of
Plant Genetics and Crop Research (IPK, Gatersleben,
Germany).
Seed measurements were carried out using the Digital
Seed Analyser/Counter (GTA Sensorik GmbH, Neubran-
denburg, Germany), which allowed a high counting
accuracy (error \0.01). A sample of 15–30 seeds were
measured per accession and the following characters were
recorded: surface area (mm
2
), length (mm), width (mm)
and thousand-seed weight (g). Data were recorded using
Marvin 4.0 software (GTA Sensorik GmbH, Neubranden-
burg, Germany).
Nuclear genome size estimation
For flow cytometric measurements of the genome size,
plants were grown from seeds in the green house at 26°C.
Specimen vouchers of all accessions are kept in the her-
barium of IPK, Gatersleben (GAT). Samples of fresh
young leaves were co-chopped together with Raphanus
sativus cv. ‘Voran’ (Genebank Gatersleben accession
E. G. Achigan-Dako et al.
123
number: RA 34; 2C=1.11 pg, Schmidt-Lebuhn et al.
2008) as internal standard with a sharp razor blade in a
Petri dish containing 1 ml nuclei isolation buffer (Galbraith
et al. 1983) supplemented with DNAse-free RNase (50 lg/
ml) and propidium iodide (50 lg/ml) and filtrated through
a35lm mesh. After at least 10 min of incubation on ice
relative fluorescence intensities of nuclei were measured at
a FACStar
PLUS
(BD Biosciences) flow sorter equipped with
an argon ion laser INNOVA 90C (Coherent). Usually,
10,000 nuclei per sample were analysed using the software
CELL Quest 3.3 (BD Biosciences). Depending on the
availability and the germination rate of seeds, each acces-
sion was measured on average more than three times
(Table 1) on samples of the same plant and/or accession on
different days. Accessions with the most contrasting values
underwent further measurements. The absolute DNA
amounts of samples were calculated based on the values of
the G1 peak means.
Data analysis
Every variable was examined graphically to check out the
frequency distribution pattern and normal distribution of
group frequencies with the Kolmogorov–Smirnov test. To
compare the variation of the 2C-values among groups (for
instance gourd type vs. Egusi type or African gourd vs.
Asian gourd) we used the univariate general linear model
(GLM). Levene test was used to check for homogeneity of
variances. Post hoc comparison was performed using
Hochberg’s GT2. Welch and Brown–Forsythe statistics
were complementarily used to test the equality of means in
case of variance heterogeneity (Hill and Lewicki 2006). To
describe allometric relationships for seed quantitative traits
and between seed traits and genome size, all variables were
log-transformed. The allometric relationship is expressed
as log Y¼log aþblog Xwhere Yand Xstand for vari-
ables and aand brepresent regression coefficients. We
used ordinary least square simple or multiple regressions to
test for relationship between 2C-value, geographical origin
(altitude, latitude and longitude) and seed quantitative
traits. We rejected the null hypothesis at P\0.05. In case
of significant relationship between two variables we
examined the frequency distribution and the q–q plot of the
standardized residuals to check for normality and we
checked the goodness-of-fit of the model (Sokal and Rohlf
1995; Crawley 2007). The coefficient of correlation (r)
indicates the strength of the relationship and r
2
indicates
the proportion of the explained variation. To estimate the
line that best describes the bivariate scatter of the thousand-
seed weight and the seed surface area or the seed length
and the 2C-value for example we used the standardized
major axis (SMA) methods with SMATR (Warton et al.
2006). The SMA method tests for equality of slopes of the
bivariate scatter of groups. Others statistical analyses were
carried out using R (version 2.5.1, 2007) and SPSS (version
12, 2003). The single accession from the New World
(Cuba) was not considered for inference analysis but just
for checking its 2C-value.
Results
Allometric relationships between seed
quantitative traits
The thousand seed weight of L. siceraria was on average
180.24 g (SD =99.16) (Table 2). The gourd type had a
slightly higher thousand seed weight (182 g ±113.62,
coefficient of variation CV =62.2) than the Egusi type
(174.2 g ±46.5, CV =26.6). Seed surface area varies
between 50 ±5.4 and 385 ±26.3 mm
2
in gourd types
(mean =132.46 mm
2
±79.31, CV =59.8) and between
78.8 ±6.3 and 166.1 ±12.7 mm
2
for Egusi types
(mean =130.87 mm
2
±19.47, CV =14.8). Seed width
reaches on average 8.89 ±2.65 mm for gourd types and
8.38 mm ±0.68 for Egusi types. For thousand-seed weight,
seed surface area and seed width, no difference was noticed
between the two types according to the Welch and Brown–
Forsythe test (P[0.05). In contrast, the seed length was
significantly different between the two types of cultivars
(Welch and Brown–Forsythe =14.9, P\0.001). Seed
length varied between 10.9 ±0.9 and 35.2 ±1.78 mm for
gourd type (mean =17.27 mm ±5.24, CV =30.3) and
between 14.4 ±1.37 and 23.9 ±1.85 mm for the Egusi
type (mean =19.82 mm ±1.75, CV =8.8).
Table 3presents the results of SMA analysis of pairwise
combinations of thousand-seed weight (T
w
), seed surface
area (S
a
), seed length (S
l
) and seed width (S
w
) for 74
accessions of gourd type and 31 accessions of Egusi type.
The relation between seed surface area (S
a
) and thousand-
seed weight (T
w
) is highly significant for both gourd and
Egusi types (P\0.001), although the coefficient of
determination was stronger in the gourd types (r
2
=0.93)
than in Egusi types (r
2
=0.66). The test for common slope
of the regression lines is significantly different between the
two types (Test statistic =10.6, P=0.004). Detailed
examination of the relationship using seed length and
seed width as independent variables showed similar
results. The allometric equation between seed width and
thousand-seed weight is log Tw¼0:211 þ2:12 log Sw
(r
2
=0.84, P\0.01) for gourd types and log Tw¼
0:69 þ3:16 log Sw(r
2
=0.57, P\0.01) for Egusi types.
The test for common slope between types was significant
(Test statistic =9.0, P=0.003). The allometric equation
between seed length and thousand-seed weight was
log Tw¼0:56 þ2:11 log Sl(r
2
=0.91, P\0.01) for
Genome size variation in Lagenaria siceraria
123
Table 1 List of the 117 accessions of Lagenaria siceraria used and their 2C-values
Accession Type M P 2C-value SD Origin Accession Type M P 2C-value SD Origin
06NIA013 Gourd 6 3 0.741 0.025 Ghana 916IS11 Gourd 1 1 0.741 0 Benin
06NIA131 Gourd 4 4 0.702 0.006 Ghana 919IS12 Gourd 1 1 0.727 0 Benin
06NIA214a Gourd 2 2 0.753 0.008 Mali 920IS12 Gourd 1 1 0.720 0 Benin
06NIA214b Gourd 6 3 0.731 0.017 Mali 923IS2 Gourd 4 4 0.743 0.021 Benin
06NIA215 Gourd 2 2 0.751 0.014 Mali 924IS2 Gourd 4 4 0.742 0.018 Benin
06NIA216 Gourd 2 2 0.757 0.002 Mali 941KO-019 Gourd 4 4 0.735 0.022 Mali
06NIA217 Gourd 2 2 0.756 0.002 Mali 942KO_020 Gourd 1 1 0.705 0 Mali
06NIA218 Gourd 3 3 0.730 0.002 Mali 943KO_021 Gourd 1 1 0.722 0 Mali
06NIA220 Gourd 2 2 0.740 0.001 Mali 946KO_023 Gourd 1 1 0.735 0 Mali
06NIA227 Gourd 9 3 0.715 0.012 Mali 947KO_024 Gourd 1 1 0.712 0 Mali
06NIA435 Gourd 1 1 0.739 0 Mali 956KU0130 Gourd 3 2 0.715 0.023 Benin
06NIA505 Gourd 1 1 0.741 0 Benin 957KU0131 Gourd 1 1 0.742 0 Benin
1001SC19 Gourd 1 1 0.730 0 Benin 974LAF005 Gourd 4 3 0.733 0.027 Benin
1004SE031 Gourd 1 1 0.710 0 Mali 981SC003 Gourd 4 4 0.708 0.018 Benin
1009SE035 Gourd 1 1 0.720 0 Mali 995SC018 Gourd 4 3 0.719 0.027 Benin
1011SN001 Gourd 7 6 0.708 0.020 Benin LAG24 Gourd 4 3 0.723 0.012 China
1016SN007 Gourd 1 1 0.705 0 Benin LAG36 Gourd 3 3 0.734 0.005 Georgia
1017SN008 Gourd 1 1 0.733 0 Benin LAG48 Gourd 3 3 0.724 0.014 Georgia
1018SN009 Gourd 4 3 0.733 0.029 Benin LAG50 Gourd 3 3 0.712 0.001 Korea
1019SN011 Gourd 3 3 0.715 0.018 Benin LAG51 Gourd 3 3 0.710 0.009 Korea
1024SN014 Gourd 2 1 0.735 0.046 Benin LAG52 Gourd 3 3 0.749 0.004 Korea
1025SN014 Gourd 4 2 0.731 0.031 Benin LAG55 Gourd 3 3 0.740 0.005 Irak
1026SN014 Gourd 1 1 0.721 0 Benin LAG56 Gourd 3 3 0.746 0.025 Irak
1028SN016 Gourd 4 4 0.708 0.015 Benin LAG60 Gourd 3 2 0.742 0.004 Korea
1029SN018 Gourd 1 1 0.717 0 Benin LAG61 Gourd 3 3 0.725 0.006 Korea
1036SN025a Gourd 1 1 0.710 0 Benin LAG62 Gourd 3 3 0.726 0.009 Korea
1043SN029a Gourd 4 3 0.706 0.022 Benin LAG78 Gourd 4 3 0.725 0.013 Cuba
1045SN030a Gourd 4 3 0.727 0.028 Benin 06NIA124 Egusi 4 4 0.742 0.022 Ghana
1046SN030a Gourd 2 2 0.733 0.067 Benin 1012SN002 Egusi 1 1 0.748 0 Benin
1047SN030b Gourd 4 4 0.730 0.035 Benin 1013SN003 Egusi 4 3 0.734 0.011 Benin
1048SN031 Gourd 4 4 0.735 0.025 Benin 1014SN004 Egusi 1 1 0.741 0 Benin
1067SN044 Gourd 4 4 0.717 0.014 Benin 1015SN006 Egusi 4 3 0.745 0.019 Benin
1071SN19 Gourd 1 1 0.728 0 Benin 1021SN012 Egusi 1 1 0.742 0 Benin
1073SNX1 Gourd 4 4 0.715 0.020 Benin 1066SN043 Egusi 1 1 0.737 0 Benin
1079SC009 Gourd 4 4 0.730 0.014 Benin 1072SNX005 Egusi 1 1 0.744 0 Benin
113KO_022 Gourd 8 8 0.731 0.016 Mali 120SN10 Egusi 4 4 0.750 0.010 Benin
114KO_026 Gourd 4 4 0.750 0.012 Mali 121AA609 Egusi 4 4 0.759 0.007 Benin
115KO_019 Gourd 7 7 0.731 0.010 Mali 803AA436a Egusi 6 6 0.756 0.032 Benin
117SE030 Gourd 4 4 0.737 0.023 Mali 803AA436b Egusi 4 3 0.760 0.012 Benin
806AA503 Gourd 4 4 0.735 0.013 Benin 805AA502 Egusi 6 6 0.767 0.018 Benin
815AAF320 Gourd 5 2 0.746 0.148 Benin 820AAGF33 Egusi 1 1 0.741 0 Benin
816AAF336 Gourd 3 3 0.746 0.002 Benin 858BSN005 Egusi 4 4 0.764 0.033 Benin
817AAF33 Gourd 4 4 0.737 0.031 Benin 871BSN011 Egusi 1 1 0.725 0 Benin
830AE0106 Gourd 4 4 0.741 0.011 Niger 872BSN011 Egusi 2 2 0.730 0.017 Benin
843BA006 Gourd 4 4 0.757 0.018 Mali 882BSN016 Egusi 13 1 0.683 0.005 Benin
845BAX02 Gourd 14 9 0.703 0.017 Mali 888BSNX1 Egusi 3 3 0.746 0.023 Benin
848BSNX3 Gourd 4 4 0.724 0.025 Benin 895DAF11 Egusi 3 3 0.754 0.040 Benin
865BSN008 Gourd 3 3 0.727 0.026 Benin 905IS001 M Egusi 16 9 0.767 0.024 Benin
E. G. Achigan-Dako et al.
123
gourd and log Tw¼2:53 þ2:9logSl(r
2
=0.51, P\0.01)
for Egusi types. The test for common slope between
types was significant as well (Test statistic =5.49, P=
0.02).
Intraspecific variation of the genome size
Based on the 2C-value of 1.11 pg for Raphanus sativus, the
average DNA content of the 117 investigated accessions of
L. siceraria was 0.734 ±0.017 pg/2C. We did not observe
significant differences between the accessions with regard
to their geographical origin. Accessions of gourd types
from Africa showed similar 2C-values as accessions
from Asia (0.729 ±0.014 vs. 0.730 ±0.013 pg DNA,
Hochberg GT2 post hoc test) and also the single accession
from Cuba had a 2C-value within the same range (0.725 ±
0.013 pg DNA). Genome size values of accessions measured
only once fitted well into the range of their particular group.
However, genome size differs within the taxon by about
12%, and if we exclude the surprisingly low estimate of the
Egusi accession 882BSN016 we found intraspecific genome
size differences of 9.5% in L. siceraria.
We got a significant difference (F=28.44, P\0.001)
in the genome size between plants with different seed
types. The Egusi type had on average a 2C-value of
0.747 ±0.017 pg DNA and the gourd type of 0.729 ±
0.014 pg DNA (Fig. 1). The genome size varied from
0.702 to 0.759 pg/2Cin gourd types and from 0.683 to
0.776 pg/2Cin Egusi types. The magnitude of the variation
was about 7.5 and 12%, respectively. If we excluded the
very low estimate of the Egusi type 882BSN016 the vari-
ation dropped to 6.6 for all other Egusi accessions.
Table 2 Descriptive statistics
of the gourd and Egusi types of
Lagenaria siceraria (values in
bracket represent SD)
Coefficients of variation (CV)
are expressed in percentages
Mean Coefficient of variation Minimum Maximum
Thousand-seed weight (g) Gourd type 182.6 (113.6) 62.2 47.9 542.5
Egusi type 174.2 (46.5) 26.6 92.2 285.3
Total 180.2 (99.2) 47.9 542.5
Seed surface area (mm
2
) Gourd type 132.5 (79.3) 59.8 50.0 385.3
Egusi type 130.9 (19.5) 14.8 78.8 166.1
Total 132.0 (67.8) 50.0 385.3
Seed length (mm) Gourd type 17.3 (5.2) 30.3 10.9 35.2
Egusi type 19.8 (1.7) 8.8 14.4 23.9
Total 17.9 (4.5) 10.9 35.2
Seed width (mm) Gourd type 8.9 (2.6) 29.8 5.4 16.1
Egusi type 8.4 (0.7) 8.1 6.6 9.9
Total 8.7 (2.3) 5.4 16.1
2C-value (pg) Gourd type 0.729 (0.014) 1.9 0.702 0.759
Egusi type 0.747 (0.017) 2.2 0.683 0.776
Total 0.734 (0.017) 0.683 0.776
Table 1 continued
Accession Type M P 2C-value SD Origin Accession Type M P 2C-value SD Origin
886BSN018 Gourd 6 5 0.750 0.023 Benin 912IS010 Egusi 4 4 0.759 0.032 Benin
887BSN018 Gourd 4 4 0.759 0.030 Benin 929IS6 Egusi 8 7 0.761 0.020 Benin
890BSNX2 Gourd 7 7 0.722 0.008 Benin 930IS6 Egusi 3 3 0.760 0.018 Benin
892DAF008 Gourd 9 7 0.709 0.023 Benin 931IS7 Egusi 3 3 0.754 0.028 Benin
894DAF010 Gourd 4 4 0.741 0.029 Benin 932IS7 Egusi 2 2 0.761 0.020 Benin
897DAF014 Gourd 7 7 0.736 0.017 Benin 933IS8 Egusi 16 6 0.776 0.032 Benin
898DAF12 Gourd 4 3 0.718 0.014 Benin 934IS8 Egusi 1 1 0.739 0 Benin
904GFAX Gourd 7 7 0.720 0.033 Benin 973LAF001 Egusi 4 3 0.738 0.029 Benin
906IS002 M Gourd 4 4 0.740 0.028 Benin 983SC006 Egusi 4 4 0.724 0.018 Benin
911IS007 M Gourd 4 4 0.742 0.011 Benin 985SC007 Egusi 1 1 0.761 0 Benin
914IS1 Gourd 7 7 0.749 0.028 Benin
Mnumber of measurements, Pnumber of plants, SD standard deviation
Genome size variation in Lagenaria siceraria
123
Moreover, within each seed type reproducible differences
in the genome size were detectable. In order to demonstrate
that the variation observed reflect genuine differences in
the genome size, we co-chopped and co-measured acces-
sions from different as well as from the same seed types.
Depending on the difference or similarity in genome size
these experiments resulted either in single peak, indicating
that the genome size of the two is almost similar, or clearly
distinguishable individual peaks (Fig. 2). In addition, we
increased the number of individual measurements for the
accessions with the most contrasting 2C-values. These data
show that the genome size of L. siceraria varies between
Egusi and gourd type as well as within each seed type. The
gourd type cultivars have a lower average DNA content
than the Egusi type independent of their geographical
origin.
Relationship between seed quantitative traits,
geographical coordinates and genome size
As the DNA contents were different between the gourd and
the Egusi types, we analysed the relationships between the
genome size, the geographical coordinates and the seed
traits separately for both types.
For the gourd type, the relationship between 2C-value
(as dependant variable) and independent variables such as
thousand-seed weight (T
w
), seed surface area (S
a
), seed
length (S
l
), seed width (S
w
), altitude (A
l
), latitude (L
t
) and
longitude (L
g
) as revealed by the multiple linear regression
is not significant (P\0.218). However, among the set of
regression variable coefficients only growing elevation (A
l
)
showed the smallest Pvalue (P=0.143). The simple
regression analysis between the 2C-value and elevation
indicated highly significant relation (r
2
=0.126, F=9.74,
P=0.002, residual standard error =0.013 on 71 degrees
of freedom). The normal q–q plot indicated a normal dis-
tribution of errors and supports the goodness of fit of the
model. Figure 3a shows the scatter plot of the 2C-value of
gourds against the elevation.
For Egusi types, regression analyses were performed
without the accession 882BSN016. For the remaining acces-
sions, the multiple regression analysis did not indicate any
relationship between 2C-value and variables such as thou-
sand-seed weight, seed surface area, seed length, seed width,
altitude, latitude and longitude (F=1.077, P=0.410).
Bivariate correlation analysis indicates, however, medium
significant correlation between 2C-value and growing eleva-
tion (r=0.374, P=0.041). The simple regression analysis
using elevation as independent variable found a significant
correlation (r
2
=0.114, F=4.567, P=0.041, residual
standard error =0.012 on 28 degrees of freedom). Figure 3b
shows the scatter plot of the 2C-value of Egusi types against
the elevation.
Fig. 1 Variation of 2C-values in accessions of African gourd
(n=74), Asian gourd (n=11) and Egusi type (n=31). In all
box-and-whisker plots, the horizontal line is the median, the rectangle
is one quartile and the vertical line is the range
Table 3 Results of standardized major axis regression (SMA) analysis of pairwise combinations of thousand-seed weight (T
w
), seed surface area
(S
a
), seed length (S
l
) and seed width (S
w
) for 74 African accessions of gourd type and 31 accessions of Egusi type of Lagenaria siceraria
Trait pair Yand XSeed type nr
2
PSlope Intercept Slope homogeneity
T
w
and S
a
Gourd 74 0.928 \0.001 1.135 -0.152 P=0.004
Egusi 31 0.658 \0.001 1.672 -0.305
T
w
and S
w
Gourd 74 0.843 \0.001 2.120 0.211 P=0.003
Egusi 31 0.577 \0.001 3.165 -0.690
T
w
and S
l
Gourd 74 0.914 \0.001 2.114 -0.562 P=0.02
Egusi 31 0.508 \0.001 2.914 -2.536
E. G. Achigan-Dako et al.
123
Discussion
DNA content estimation
Genome size data are known for about 1.8% of angio-
sperms, a sample not representative of the global flora
(Bennett and Leitch 2005a, Leitch and Bennett 2007).
Furthermore, many values were just first approximations;
most are unverifiable, and some are for misidentified spe-
cies and the need to generate new data on plants C-values
has become self-evident (Bennett 1998; Hanson et al. 2003;
Dolezel and Bartos
ˇ2005). A first estimate has yet to be
Fig. 2 Flow-cytometric
genome size measurements in
Lagenaria siceraria. Examples
of histograms of relative
fluorescence intensity obtained
after analysis of propidium-
iodide stained nuclei of the
Egusi-type accession
882BSN016 (a) and 933IS8 (c),
respectively, together with the
internal standard Raphanus
sativus (Rs). Co-processing of
accessions with deviating
genome sizes resulted in
histograms with clearly
distinguishable individual peaks
in the Egusi- (b) as well as the
gourd-type (d)
Fig. 3 Relationship between
genome size and growing
elevations for (a) gourd types
and (b) Egusi types in
Lagenaria siceraria from
West Africa
Genome size variation in Lagenaria siceraria
123
reported for most families including many large tropical
groups. For Cucurbitaceae with about 800 described spe-
cies (Jeffrey 2005), genome size data are only available for
some species of the genera Citrullus,Cucumis,Cucurbita,
Lagenaria and Bryonia (Barlow 1975; Ingle et al. 1975;
Ramachandran and Narayan 1985; Arumuganathan and
Earle 1991; Chattopadhyay and Sharma 1991). The only
data available for L. siceraria suggest a 2C-value of 1.4 pg
DNA (Ingle et al. 1975). This value is about two times
higher than the average value we obtained in our mea-
surements (0.734 ±0.017 DNA/2C) using 117 accessions
from West Africa, Asia and the New World. Although we
did not find any indication of the incidence of polyploidy in
our L. siceraria samples, it cannot be completely excluded,
especially because the origin of L. siceraria used in the
study of Ingle et al. (1975) is not mentioned. However, the
discrepancy between the results might also be due to
methodological errors, since Ingle et al. (1975) estimated
the DNA content using comparative Feulgen photometry
based on a protocol that included hot hydrolysis (McLeish
and Sunderland 1961), which is discussed as a potential
error in genome size estimation (Greilhuber 1998,2005;
Kron et al. 2007). The use of flow cytometry in our study
and the high number of individual plants screened
(Table 1) led to more reliable results (Dolezel et al. 1998,
2007) on the 2C-value of L. siceraria. Moreover, it should
be recognised that most of our samples were repeatedly
measured at different days to minimise the error introduced
by the instrumental set-up.
Intraspecific variation of genome size
The occurrence and extent of genome size variation below
the species level is controversially discussed and is con-
sidered to be less frequent than previously thought
(Greilhuber 1998,2005; Dolezel et al. 2007; Kron et al.
2007). Dolezel and Bartos
ˇ(2005) considered various
methodological aspects of DNA flow cytometry and con-
cluded that reliable detection of intraspecific variation in
genome size is not a trivial task. However, genome size
variation has been observed in Hordeum marinum subsp.
marinum (Poaceae: Jakob et al. 2004), in Arabidopsis
thaliana (Brassicaceae: Schmuths et al. 2004), and in
Festuca pallens (Poaceae: S
ˇmarda and Bures
ˇ2006). Our
results indicate that in L. siceraria intraspecific variation of
genome size is partially related to differences among gourd
and Egusi types. The gourd type has a smaller genome size
(mean 2C±SD =0.729 pg DNA ±0.014) in compari-
son to the Egusi type (2C=0.747 pg DNA ±0.017). We
found also significant differences within each of the two
seed types. The co-processing runs, which revealed in a
couple of different combinations a clear separation of the
individual peaks representing different accessions, is an
additional argument that the observed variation is not just
due to technical errors (Greilhuber et al. 2007). The gen-
ome size variation observed in L. siceraria seems to bear
taxonomic or evolutionary significance. Clearly, Egusi type
accessions possess discriminant characters in comparison
with gourd types. Conspicuous are for example the rind
consistency, the seed coat hardness, the seed colour and
ornamentation (Achigan-Dako et al. 2008), and they are
used for different purposes by human communities in West
Africa. Seeds of the Egusi type have been primarily used as
protein and oil source (Schippers 2004) while the primary
utilization of the gourd type is as containers (Heiser 1979).
The species has surely been under cultivation and human
selection for a long time (Erickson et al. 2005). The gen-
ome size differences between the Egusi and gourd type
represent additional evidences of possible reproductive
isolation between both forms. Thus, it may be convenient
to split L. siceraria taxonomically into subspecies or
varieties according to their fruit or seed types.
The processes of variation in DNA content include (1)
polyploidy, (2) aneuploidy, (3) variation in the amount of
heterochromatin, (4) occurrence of supernumerary chro-
mosomes (B chromosomes), (5) deletion or duplication of
chromosomes segments, and (6) variation in the copy
number of repeated sequences (Poggio et al. 1998). In our
study no polyploidy events have been detected. Counting
of the extremely small chromosomes of L. siceraria indi-
cated identical numbers (2n=22) for accessions with
low or high DNA content (e.g. 882BSN016, 933IS8,
905IS001M, 845BAX02, 06NIA216, data not shown).
Genome size variation could also be associated to the
presence of supernumerary chromosomes (B chromo-
somes). However, to our knowledge, B chromosomes were
not documented for L. siceraria. Studying genome size
variation in maize populations at various altitudes Poggio
et al. (1998) concluded that the described variation corre-
sponds to differences in DNA content of the regular
A-chromosome complements irrespective of the presence
of B-chromosomes although the numerical polymorphism
of B chromosomes was one of the main factors contribut-
ing to intraspecific genome size variation (Rosato et al.
1998; Lia et al. 2007). S
ˇmarda and Bures
ˇ(2006) also
observed that the intrapopulational variation in Festuca
pallens was not exclusively caused by the presence of B
chromosomes.
Relation between genome size, geographical
occurrence and seed traits
Allometric relationships as described for seed quantitative
traits indicate differences in slope between the gourd and
Egusi types. Although the seed size (seed surface, seed
length and width) and the thousand-seed weight are highly
E. G. Achigan-Dako et al.
123
correlated, the contribution of the seed size to the thousand
seed weight is different for both cultivar types. However, it
was noticed that neither gourd nor Egusi types showed any
association between seed quantitative traits and 2C-values.
Comparative studies of angiosperms have played a leading
role in showing that DNA C-value is correlated with a wide
range of phenotypic characters at the cellular level (Bennett
1998). However, Jakob et al. (2004) and Greilhuber (2005)
strongly argue that many correlations indicated in the
literature are either not reproducible or influenced by a
multitude of parameters, and should therefore be consid-
ered with great care.
Ecology may have also an important impact on genome
size. Altitude and latitude were used as proxies for abiotic
influences putatively acting on genome size. Positive,
negative or polynomial relationships have been variously
described (Knight et al. 2005). In L. siceraria, we observed
a positive relationship between the genome size and
growing elevation for gourd (P=0.002, n=74) and
Egusi types (P=0.041, n=30), although the magnitude
of the variation is low and the proportion of the variation
explained is small (r
2
=12 and 14%, respectively). Similar
results observed in populations of the grass Dasypyrum
villosum (Ca
´ceres et al. 1998) were critically discussed by
Greilhuber (2005). However, other data support a geo-
graphic gradient of DNA content in cultivated maize
(Poaceae: Zea mays) populations at different altitudes
(Bennett 1985; Poggio et al. 1998), in populations of
Festuca pallens within its distribution area (S
ˇmarda and
Bures
ˇ2006) and of Arabidopsis thaliana (Schmuths et al.
2004) from western to eastern Eurasia. Rosato et al. (1998)
further demonstrated that mean number of supernumerary
chromosomes in maize populations from northern Argen-
tina were positively correlated with altitude. Although the
DNA content variation in L. siceraria follows also a
South–North gradient this is superimposed by growing
elevation differences changing in the same direction. To
capture the full relationship between genome size variation
and ecology in L. siceraria, a more detailed analysis
including parameters such as rainfall, soil types and aver-
age temperatures would be necessary.
Microevolution in L. siceraria
Genome size diversification is an important process prob-
ably occurring during or shortly after speciation in plants
(Jakob et al. 2004; Lowe et al. 2004, Eilam et al. 2007), and
thus, might be a useful parameter for estimating cryptic
taxonomic differentiation within species (Greilhuber 1998;
Ohri 1998). Previous studies have indicated that L. siceraria
Egusi types show morphologically divergent characters in
comparison to gourd types (Achigan Dako et al. 2008).
However, no taxonomical conclusion was drawn about the
type’s taxonomic rank. The genome size variation in L.
siceraria may be an adaptation to farmers’ selection or
could suggest incipient speciation (Murray 2005). A func-
tional and adaptive role of DNA amount is envisaged in the
nucleotypic theory which shows that non-informative DNA
can produce various phenotypic effects at nuclear, cellular,
tissue and organism levels (Bennett 1972,1973,1985;
Knight et al. 2005). Furthermore, Bennett and Leitch
(2005b) argued that the possession of a large genome may
itself act as an evolutionary force. The Egusi type may
therefore represent a nucleotype of L. siceraria. Analysis of
the nuclear rDNA internal transcribed spacer (ITS) region
in the two types did, however, not indicate any nucleotide
substitution (own unpublished data), suggesting that the two
types still belong to the same species. From a taxonomic
standpoint, intraspecific C-value variation is probably an
important indicator, showing that there may be more than
one entity within a species (Murray 2005), also in L. sice-
raria. This has, however, to be proven by phylogenetic
analyses as suggested by Bennett and Leitch (2005b).
Morphometric analysis between seed quantitative traits and
the genome size found no significant association. However,
there is a significant correlation between 2C-value and the
elevation of the collecting sites, which might indicate an
adaptation of genome size to an unknown ecological
parameter connected to altitude.
Acknowledgments E.G.A-D. is supported by the Deutscher Aka-
demischer Austauschdienst (DAAD). We acknowledge funding of
this work by the Deutsche Forschungsgemeinschaft (DFG) to F.R.B
and the International Foundation for Science (IFS)’s grant T/3709 to
E.G.A.-D. We thank the Genbank of the IPK for providing accessions
from Asia and America, Raymond S. Vodouhe
`for facilitating col-
lecting missions, Christina Koch (IPK) for greenhouse sample
management and help with the laboratory works and O.G. Gaoue for
contribution to statistical analysis with R.
References
Achigan-Dako GE, Fanou N, Kouke A, Avohou H, Vodouhe SR,
Ahanchede A (2006) Evaluation agronomique de trois espe
`ces
de Egusi (Cucurbitaceae) utilise
´es dans l’alimentation au Be
´nin
et e
´laboration d’un mode
`le de pre
´diction du rendement.
Biotechnol Agron Soc Environ 10:121–129
Achigan-Dako GE, Vodouhe SR, Sangare A (2008) Caracte
´risation
morphologique des cultivars locaux de Lagenaria siceraria
(Cucurbitaceae) collecte
´sauBe
´nin et au Togo. Belg J Bot 141(1)
Adjakidje
`V (2006) Cucurbitaceae. In: Akoe
`gninou A, van der Burg
WJ, van der Maesen LJG, Adjakidje
`V, Essou JP, Sinsin B,
Ye
´domonhan H (eds) Flore du Be
´nin. Backhuys Publishers,
Cotonou & Wageningen, pp 520–534
Arumuganathan K, Earle ED (1991) Nuclear DNA content of some
important plant species. Pl Molec Biol Rep 9:208–218
Barlow PW (1975) Polytene nucleus of giant hair cell of Bryonia
anthers. Protoplasma 83:339–349
Beaulieu JM, Moles TA, Leitch IJ, Bennett MD, Dickie JB, Knight
CA (2007) Correlated evolution of genome size and seed mass.
New Phytol 173:422–437
Genome size variation in Lagenaria siceraria
123
Beevy SS, Kuriachan P (1996) Chromosome numbers of south Indian
Cucurbitaceae and a note on the cytological evolution in the
family. J Cytol Genet 31:65–71
Bennett MD (1972) Nuclear DNA content and minimum generation
time in herbaceous plants. Proc Roy Soc Lond B 181:109–135
Bennett MD (1973) Nuclear characters in plants. In: Basic mecha-
nisms in plant morphogenesis. Brookhaven Sym Biol 25:344–
366
Bennett MD (1985) Intraspecific variation in DNA amount and the
nucleotypic dimension in plant genetics. In: Freeling M (ed)
Plant genetics, UCLA symposium on molecular and cellular
biology, Alan R Liss, New York, pp 283–302
Bennett MD (1998) Plant genome values: How much do we know?
Proc Natl Acad Sci USA 95:2011–2016
Bennett MD, Leitch IJ (1995) Nuclear DNA amounts in angiosperms.
Ann Bot 76:113–176
Bennett MD, Leitch IJ (2005a) Nuclear DNA amounts in angio-
sperms—progress, problems and prospects. Ann Bot 95:45–90
Bennett MD, Leitch IJ (2005b) Plant genome size research: a field in
focus. Ann Bot 95:1–6
Burkill HM (1985) The useful plants of West Tropical Africa, vol 1,
2nd edn. Families A-D. Royal Botanical Gardens, Kew
Caceres ME, De Pace C, Scarascia Mugnozza GT, Kotsonis P,
Ceccarelli M, Cionini PG (1998) Genome size variation within
Dasypyrum villosum: correlations with chromosomal traits,
environmental factors and plant phenotypic characteristics and
behaviour in reproduction. Theor Appl Genet 96:559–567
Chattopadhyay D, Sharma AK (1991) Chromosome studies and
nuclear DNA in relation to sex difference and plant habit in two
species of Cucurbitaceae. Cytologia 56:409–417
Crawley JM (2007) The R book. Wiley, West Sussex
Decker-Walters D, Staub J, Lopez-Sese A, Nakata E (2001) Diversity
in landraces and cultivars of bottle gourd (Lagenaria siceraria;
Cucurbitaceae) as assessed by random amplified polymorphic
DNA. Genet Resour Crop Evol 48:369–380
Decker-Walters D, Wilkins-Ellert M, Chung S-M, Staub EJ (2004)
Discovery and genetic assessment of wild bottle gourd [Lagena-
ria siceraria (Mol.) Standley; Cucurbitaceae] from Zimbabwe.
Econ Bot 58:501–508
Dolezel J, Greilhuber J, Lucretti S, Meister A, Lysak MA, Nardi L,
Obermayer R (1998) Plant genome size estimation by flow
cytometry: inter-laboratory comparison. Ann Bot 82 (suppl
A):17–26
Dolezel J, Bartos
ˇJ (2005) Plant DNA flow cytometry and estimation
of nuclear genome size. Ann Bot 95:99–110
Dolezel J, Greilhuber J, Suda J (2007) Flow cytometry with plants: an
overview. In: Dolezel J, Greilhuber J, Suda J (eds) Flow
cytometry with plant cells. Wiley-VCH, Weinheim, pp 41–65
Eilam T, Anikster Y, Millet E, Manisterski J, Sagi-Assif O, Feldman
M (2007) Genome size and genome evolution in diploid
Triticeae species. Genome 50:1029–1035
Erickson LD, Smith DB, Clarke CA, Sandweiss HD, Tuross N (2005)
An Asian origin for a 10, 000-year-old domesticated plant in the
Americas. Proc Natl Acad Sci USA 102:18315–18320
Galbraith DW, Harkins RK, Maddox MJ, Ayres MN, Sharma PD,
Firoozabady E (1983) Rapid flow cytometry analysis of the cell
cycle in intact plant tissues. Science 220:1049–1051
Graham MJ, Nickell CD, Rayburn AL (1994) Relationship between
genome size and maturity group in soybean. Theor Appl Genet
88:429–432
Greilhuber J (1998) Intraspecific variation in genome size: a critical
reassessment. Ann Bot 82 (suppl A):27–35
Greilhuber J (2005) Intraspecific variation in genome size in
angiosperms: identifying its existence. Ann Bot 95:91–98
Greilhuber J, Temsch EM, Loureiro JCM (2007) Nuclear DNA
content measurement. In: Dolezel J, Greilhuber J, Suda J (eds)
Flow cytometry with plant cells. Wiley-VCH, Weinheim, pp 67–
101
Hanson L, Brown RL, Boyd A, Johnson MAT, Bennett MD (2003)
First nuclear DNA C-values for 28 Angiosperm genera. Ann Bot
91:31–38
Heiser CB (1979) The gourd book. University of Oklahoma Press,
Norman
Hill T, Lewicki P (2006) Statistics: methods and applications. A
comprehensive reference for science, industry, and data mining,
1st edn. StatSoft, Tulsa
Ingle J, Timmis JN, Sinclair J (1975) The relationship between
satellite deoxyribonucleic acid, ribosomal ribonucleic acid gene
redundancy and genome size in plants. Pl Physiol 55:496–501
Jakob SS, Meister A, Blattner FR (2004) The considerable genome
size variation of Hordeum species (Poaceae) is linked to
phylogeny, life form, ecology, and speciation rates. Molec Biol
Evol 21:860–869
Jeffrey C (1967) On the classification of the Cucurbitaceae. Kew Bull
20:417–426
Jeffrey C (1995) Cucurbitaceae. In: Edwards S, Tadesse M, Hedberg I
(eds) Flora of Ethiopia and Eritrea. Canellaceae to Euphorbia-
ceae, vol 2, Part 2. The National Herbarium/Uppsala University,
pp 17–59
Jeffrey C (2005) A new system of Cucurbitaceae. Bot Zhurn 90:332–
335
Kalendar R, Tanskanen J, Immonen S, Nevo E, Schulman AH (2000)
Genome evolution of wild barley (Hordeum spontaneum)by
BARE-1 retrotransposon dynamics in response to sharp micro-
climatic divergence. Proc Natl Acad Sci USA 97:6603–6607
Keraudren M (1967) Cucurbitaceae. In: Raynal J (ed) Flore du
Cameroun, vol 6. Museum National d’Histoire Naturelle, Paris,
pp 154–185
Knight CA, Ackerly DD (2002) Variation in nuclear DNA content
across environmental gradients: a quantile regression analysis.
Ecol Lett 5:66–76
Knight CA, Molinari NA, Petrov DA (2005) The large genome
constraint hypothesis: evolution, ecology and phenotype. Ann
Bot 95:177–190
Kron P, Suda J, Husband BC (2007) Applications of flow cytometry
to evolutionary and population biology. Annual Rev Ecol Evol
Syst 38:847–876
Lia VV, Confalonieri AV, Poggio L (2007) B chromosome
polymorphism in maize landraces: adaptive vs demographic
hypothesis of clinal variation. Genetics 177:895–904
Leitch IJ, Bennett MD (2007) Genome size and its uses: the impact of
flow cytometry. In: Dolezel J, Greilhuber J, Suda J (eds) Flow
cytometry with plant cells. WILEY-VCH, Weinheim, pp 153–176
Lowe A, Harris S, Ashton P (2004) Ecological genetics. Design,
analysis and application. Blackwell Publishing, TJ International,
Cornwall
Mahelka V, Suda J, Jarolı
´mova
´V, Tra
´vnı
´cek P, Krahulec F (2005)
Genome size discriminates between closely related taxa Elytri-
gia repens and E. intermdia (Poaceae: Triticeae) and their
hybrid. Folia Geobot 40:367–384
McLeish S, Sunderland N (1961) Measurements of DNA in higher
plants by feulgen photometry and chemical methods. Exp Cell
Res 24:527–540
Meagher TR, Costish DE (1996) Nuclear DNA content and floral
evolution in Silene latifolia. Proc Roy Soc Lond B 263:1455–
1460
Meagher TR, Gillies ACM, Costish DE (2005) Genome size,
quantitative genetics and the genome basis for flower size
evolution in Silene latifolia. Ann Bot 95:247–254
Morimoto Y, Mvere B (2004) Lagenaria siceraria. In: Grubben GJH,
Denton OA (eds) Vegetables. Plant resources of Tropical Africa
2. Backhuys Publishers/CTA, Wageningen/Leiden, pp 353–358
E. G. Achigan-Dako et al.
123
Morimoto Y, Maundu P, Fujimaki H, Morishima H (2005) Diversity
of the white-flowered gourd (Lagenaria siceraria) and its
relatives in Kenya: fruit and seed morphology. Genet Resour
Crop Evol 52:737–747
Murray GB (2005) When does intraspecific C-value variation become
taxonomically significant? Ann Bot 95:119–125
Norman JC (1992) Tropical vegetable crops Cucurbitaceous crops.
Arthur Stockwell, Great Britain, pp 107–119
Obermayer R, Greilhuber J (2005) Does genome size in Dasypyrum
villosum vary with fruit colour? Heredity 95:91–95
Ohri D (1998) Genome size variation and plant systematics. Ann Bot
82 (suppl A) 75–83
Poggio L, Rosato M, Chiavarino AM, Naranjo AC (1998) Genome
size and environmental correlations in Maize (Zea mays spp.
mays, Poaceae). Ann Bot 82:107–115
Price HJ, Chambers KL, Bachmann K (1981) Genome size variation in
diploid Microseris bigelovii (Asteraceae). Bot Gaz 142:156–159
Ramachandran C, Narayan RKJ (1985) Chromosomal DNA variation
in Cucumis. Theor Appl Genet 69:497–502
Rayburn AL, Auger JA, Benzinger ES, Hepburn AG (1989) Detection
of intraspecific DNA content variation in Zea mays L. by flow
cytometry. J Exp Bot 40:1179–1183
R Development Core Team (2007) R: A language and environment
for statistical computing. R Foundation for Statistical Comput-
ing, Vienna, Austria. URL http://www.R-project.org
Rosato M, Chiavarino AM, Naranjo CA, Hernandez CJ, Poggio L
(1998) Genome size and numerical polymorphism for the B
chromosome in races of maize (Zea mays spp. mays, Poaceae).
Amer J Bot 85:168–174
Salzmann U, Hoelzmann P (2005) The Dahomey Gap: an abrupt
climatically induced rain forest fragmentation in West Africa
during the late Holocene. The Holocene 15:190–199
Schippers RR (2004) Le
´gumes Africains Indige
`nes: pre
´sentation des
espe
`ces cultive
´es. Cucurbitaceae. Margraf Publishers, CTA,
113–182
Schmidt-Lebuhn AN, Fuchs J, Kessler M (2008) Flow cytometric
measurements do not reveal different ploidy levels in Mintho-
stachys (Lamiaceae). Pl Syst Evol 271:123–128
Schmuths H, Meister A, Horres R, Bachmann K (2004) Genome size
variation among accessions of Arabidopsis thaliana. Ann Bot
93:317–321
S
ˇmarda P, Bures
ˇP (2006) Intraspecific DNA content variability in
Festuca pallens on different geographical scales and ploidy
levels. Ann Bot 98:665–678
Sokal RR, Rohlf FJ (1995) Biometry, 3rd edn. WH Freeman and
Company, New York
SPSS for Windows, Rel. 12.0.0. 2003. SPSS, Chicago
Warton DI, Wright IJ, Falster DS, Westoby M (2006) Bivariate line
fitting methods for allometry. Biol Rev 81:259–291
White F (1983) The vegetation of Africa. A descriptive memoir to
accompany the Unesco AETFAT/UNSO vegetation map of
Africa. UNESCO, Paris
Zoro Bi I, Koffi KK, Dje Y (2003) Caracte
´risation botanique et
agronomique de trois espe
`ces de cucurbits consomme
´es en sauce
en Afrique de l’Ouest: Citrullus sp., Cucumeropsis manii,
Lagenaria siceraria. Biotechnol Agron Soc Environ 7:187–199
Genome size variation in Lagenaria siceraria
123
