Dear Jeremias Maerki,
Yes, i am using batik version with fop.
i attached my fo file. please look into this.
Regards,
H. Balakrishnan
--- Jeremias Maerki <[EMAIL PROTECTED]> wrote:
> I've downloaded Custard and tried two of their
> generated SVG files. No
> problems with the way you specified the SVG. There
> must be something
> else. Please send in that stacktrace I was talking
> about in my last mail.
> It would help me track down the problem. Your last
> mail didn't contain
> any helpful information.
>
> Are you sure you're using the Batik version that
> comes with FOP? Another
> version may be incompatible.
>
> On 11.02.2003 10:58:46 H. krishna wrote:
> > my fop bat file is :
> >
> > java -Xmx250m -cp %LOCALCLASSPATH%
> > org.apache.fop.apps.Fop %1 %2 %3 %4 %5 %6 %7 %8 -d
> -c
> > conf\userconfig.xml
> >
> > my svg is:
> >
> > <fo:external-graphic alignment-baseline="baseline"
>
> > src="m001.svg"/>
>
>
> Jeremias Maerki
>
>
>
---------------------------------------------------------------------
> To unsubscribe, e-mail:
> [EMAIL PROTECTED]
> For additional commands, e-mail:
> [EMAIL PROTECTED]
>
__________________________________________________
Do you Yahoo!?
Yahoo! Shopping - Send Flowers for Valentine's Day
http://shopping.yahoo.com
<?xml version="1.0" encoding="iso-8859-1"?>
<fo:root xmlns:fo="http://www.w3.org/1999/XSL/Format";>
<fo:layout-master-set>
<fo:simple-page-master master-name="firstPage"
page-height="540pt"
page-width="450pt"
margin-top="45pt"
margin-bottom="56pt"
margin-left="50pt"
margin-right="50pt">
<fo:region-before extent="20mm" region-name="firstPage-before"/>
<fo:region-body margin-bottom="20mm" margin-top="20mm"/>
</fo:simple-page-master>
<fo:simple-page-master master-name="otherPageOdd"
page-height="540pt"
page-width="450pt"
margin-top="45pt"
margin-bottom="56pt"
margin-left="50pt"
margin-right="50pt">
<fo:region-before extent="20mm" region-name="otherPageOdd-before"/>
<fo:region-body margin-bottom="20mm" margin-top="20mm"/>
</fo:simple-page-master>
<fo:simple-page-master master-name="otherPageEven"
page-height="540pt"
page-width="450pt"
margin-top="45pt"
margin-bottom="56pt"
margin-left="50pt"
margin-right="50pt">
<fo:region-before extent="20mm" region-name="otherPageEven-before"/>
<fo:region-body margin-bottom="20mm" margin-top="20mm"/>
</fo:simple-page-master>
<fo:page-sequence-master master-name="contents">
<fo:repeatable-page-master-alternatives>
<fo:conditional-page-master-reference
master-reference="firstPage"
page-position="first"/>
<fo:conditional-page-master-reference
master-reference="otherPageOdd"
odd-or-even="odd"
page-position="any"/>
<fo:conditional-page-master-reference
master-reference="otherPageEven"
odd-or-even="even"
page-position="any"/>
</fo:repeatable-page-master-alternatives>
</fo:page-sequence-master>
</fo:layout-master-set>
<fo:page-sequence master-reference="contents">
<fo:static-content flow-name="firstPage-before">
<fo:block>
</fo:block>
</fo:static-content>
<fo:static-content flow-name="otherPageOdd-before">
<fo:block text-align="right">
<fo:page-number/>
</fo:block>
</fo:static-content>
<fo:static-content flow-name="otherPageEven-before">
<fo:block text-align="left">
<fo:page-number/>
</fo:block>
</fo:static-content>
<fo:flow flow-name="xsl-region-body">
<fo:table table-layout="fixed" width="100%">
<fo:table-column column-width="proportional-column-width(1)"/>
<fo:table-body>
<fo:table-row height="50mm">
<fo:table-cell display-align="before">
<fo:block font-size="8pt" span="all" line-height="10pt" font-style="italic">
<fo:external-graphic width="20pt" alignment-baseline="baseline" height="20pt" src="logo.jpg"/>Biology and Philosophy 00: 1-0, 2002.<fo:block font-size="8pt" space-before.optimum="-12pt" line-height="10pt" font-style="italic" start-indent="20pt">2002 Academic Publishers. Printed in the Netherlands.</fo:block>
</fo:block></fo:table-cell>
</fo:table-row>
</fo:table-body></fo:table>
<fo:block span="all" font-size="14pt" font-family="serif" line-height="16pt" space-after.optimum="12pt" space-before.optimum="-48pt" text-align="justify"><fo:inline font-weight="bold">Comparison between phenetic characterisation using RAPD and ISSR markers and phenotypic data of cultivated chestnut <fo:inline font-style="italic">(Castanea sativa Mill.)</fo:inline></fo:inline> This paper was first presented at a congress in Bilbao</fo:block>
<fo:block span="all" font-size="12pt" font-family="serif" line-height="14pt" space-after.optimum="0pt" text-align="start"><fo:inline font-weight="bold">LUÍS GOULAÕ, TERESA VALDIVIESSO, CARLOS SANTANA and CRISTINA MONIZ OLIVEIRA</fo:inline></fo:block>
<fo:block span="all" font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="10pt" text-align="start" font-style="italic">Instituto Superior de Agronomia,Tapada da Ajuda, 1349-018, Lisboa, Portugal; Estaão Nacional de Fruticultura Vieira Natividade, Estrada de Leiria, 2460-059, Alcobaa, Portugal; International Institute for Tropical Agriculture, ESARC, P.O. Box 7878, Kampala, Uganda; Kawanda Agricultural Research Institute, P.O. Box 7065, Kampala, Uganda (e-mail: [EMAIL PROTECTED]; phone: +3512136023053; fax: +351213635031).</fo:block>
<fo:block span="all" font-size="8pt" font-family="serif" line-height="10pt" space-after.optimum="24pt" text-align="start">Received 29 November 1999; accepted in revised form 3 March 2000</fo:block>
<fo:block span="all" font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="24pt" text-align="left"><fo:inline font-weight="bold">Key words:</fo:inline> <fo:inline font-style="italic">Castanea sativa</fo:inline>, ISSR markers, Marker comparison, Morphological traits, Phenetic similarities, RAPD markers</fo:block>
<fo:block span="all" font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="24pt" text-align="justify"><fo:inline font-weight="bold">Abstract</fo:inline> Patterns of phenotypic and phenetic variability in six
Portuguese cultivars of chestnut (<fo:inline font-style="italic">Castanea sativa
</fo:inline> Mill.) are evaluated. Morphological characterisation
was based on the quantification of seventeen traits. Variance
analysis showed significant differences among cultivars, and cultivar
× year for all the traits studied, and trees within cultivars
showed also some significant differences for some of the
morphological variables. A significant correlation was obtained
between length of the leaf blade and the percentage of unisexual and
androgynic inflorescence with the effective thermal index,
accumulated rainfall from April to October and from July to October,
or the accumulated temperature below seven during the dormant period.
Principal Component and cluster analysis were performed to group the
cultivars, according to their similarity coefficients. For molecular
characterisation, 125 RAPD and 157 ISSR polymorphic markers were
amplified using 28 and 7 primers respectively. High level of
congruence among the two marker systems (<fo:inline font-style="italic">r</fo:inline> = 90.5%) was obtained from comparison of phenetic
similarities based on the percentage of shared fragments. ISSR
markers revealed important advantages over RAPDs, due to a high
effective multiplex ratio (12.5 for ISSR compared with 2.2 for
RAPD analysis) and reproducibility. Although morphological and
molecular results are comparable, slight differences are showed in
cluster analysis UPGMA dendrograms. Molecular analysis explained
homonym situations among `Martainha' and
`Longal' cultivars in Portugal.</fo:block>
<fo:block span="all" font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="24pt" text-align="justify"><fo:inline font-weight="bold">Abbreviations:</fo:inline> APOD; ascorbate peroxidase, BP; 2,2-bipyridine, DNA; 2,2-bipyridine, GR; glutathione reductase, HQ; 8-hydroxylquinoline, MDA; malondialdehyde</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="12pt" text-align="start" font-weight="bold">Introduction</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">In Europe there are two main areas of particular biological
value for sweet chestnut (<fo:inline font-style="italic">Castanea sativa</fo:inline>
Mill.) genetic resources, Turkey and Iberian Peninsula, each of
them showing peculiar characteristics (Villani et al. 1999). These two areas, together with Italy, are the
leading European chestnut producers and collectively account for
55.3% of the 240000 tons world's annual
production. In Portugal, chestnut has a relevant place at the
socioeconomic level, reaching an annual fruit production of 20000
tons
(INE 1997).</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">The genotype identification is traditionally based on the
observation of morphological characters whose expression is largely
influenced by development, environmental and cultivation factors.
Furthermore it is necessary to find morphological descriptors that
are able to distinguish among different cultivars, and make
observations in different years and regions in order to reduce the
environmental interactions (for chestnut, see
Pereira-Lorenzo et al., 1996a and Oraguzie et al., 1998). Isozymes
(Pereira-Lorenzo et al., 1996b; Pereira et al. 1999) and molecular markers as RAPDs (Fineschi et al., 1993, Galderisi et al., 1998, Santana et al., 1999) or
SSRs (Botta et al., 1999) have been used to classify cultivars of chestnut.
The choice of appropriate markers for different aspects of germplasm
evaluation is of the most interest, and different results
(phenetic dendrograms) of each approach have been reported
for several species, as the basis for detecting each of these markers
is different (Rafalski et al., 1996).</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">In this work, we aimed to compare RAPDs, ISSRs and
morphological data, in order to evaluate the suitability and
congruence of these different markers for estimating overall phenetic
similarity between a group of important Portuguese chestnut
cultivars. RAPD (Welsh and McClelland, 1990; Williams et al., 1990) relies on the amplification of the genome, using a
single randomly chosen 10-mer primer, under low annealing
temperatures. ISSR (Inter simple sequence repeats) is based
on the amplification of inter repeat regions using
microsatellite-anchored primers (Zietkiewicz et al., 1994). Both RAPD and ISSR are dominant markers, present
low start up costs, especially if the ISSR amplification fragments
are detected by silver staining, and automation is possible. Whereas
RAPD analysis can present reproducibility problems, ISSR combines two
important features, high reproducibility due to the use of longer
primers, and high multiplex ratio, since microsatellites are
ubiquitous and abundant in eukariotic genomes.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">In fruit crops, several reports have compared different
molecular markers, as in <fo:inline font-style="italic">Citrus</fo:inline>
(Fang et al., 1997)
<fo:inline font-style="italic">Vitis</fo:inline>
(Bowers and Meredith, 1998), <fo:inline font-style="italic">Ribes</fo:inline>
(Lanham and Brennan, 1999) and <fo:inline font-style="italic">Pyrus</fo:inline>
(monte-Corvo et al., 2000). Few studies have, however, related
morphological data and molecular markers for estimating diversity
between genotypes and, although in chestnut, studies that aimed to
quantify the level of correlation between genetic, morphometric or
physiological data have been published (Villani et al., 1992, Pereira-Lorenzo et al., 1996a, 1996b),
they were based on isozyme markers. Comparison between different
molecular markers and morphological characters were not accessed.
</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">Molecular technique comparisons are important because,
depending on the question being addressed, one technique is more
appropriate than another. Furthermore, different techniques are
informative at different taxonomic levels. With the increasing
development and generalised use of a large number of different
molecular markers during the last years, the debate about adequacy of
morphological as opposed to molecular data for the reconstruction of
phylogenies arouse (Petersen and Seberg, 1998). It is
important to point out that in this work we aim to estimate
indicators of overall phenetic similarity and this approach is
inappropriate for phylogenetic analysis. Accurate identification of
the cultivars and the assessment of intra and intercultivar
variability are important in clonal selection programs. Since
Portuguese cultivars are not well characterised and are classified
according to their origin, clarifying synonymy and homonymy
situations, is important for management of germplasm collections.
</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="12pt" text-align="start" font-weight="bold">Materials and methods</fo:block>
<fo:block font-size="9pt" font-family="serif" line-height="11pt" space-after.optimum="12pt" text-align="start" font-weight="bold">Plant materials</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">The six accessions studied belong to a collection of ENFVN
(Estação Nacional de Fruticultura Vieira
Natividade), Alcobaça, Portugal
(Table 1).
`Martainha1' and `Martainha2' as well as
`Longal5' and `Longal6' were purposely
selected since they are suspected to be homonyms. For morphological
characterisation, four trees of each cultivar were analysed during a
five years (1993–1997) period, for the morphological
characters listed in Table 2. For RAPD and ISSR analysis one tree per cultivar was
selected for DNA extraction, since in a previous assay, no
differences between RAPD and ISSR patterns of different trees of each
cultivar were detected (unpublished data).</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="12pt" text-align="start" font-style="italic">Genomic DNA extraction</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">DNA was extracted from fresh newly expanded leaves, using a
hexadecyltrimethylammonium bromide (CTAB) protocol adapted
from Doyle and Doyle (1990) as described in
Santana et al. (1999).</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="12pt" text-align="start" font-style="italic">RAPD assays</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">The reaction mixtures had a total volume of 25 μl
. The mixture contained 1.0 unit of <fo:inline font-style="italic">Taq</fo:inline>
DNA polymerase (Promega), 0.4 μM primer,
0.16 mM of each dNTP (Promega), 2.5 mM
MgCl<fo:inline vertical-align="sub" font-size="7pt">2</fo:inline>, 1 × reaction buffer (100 mM
NaCl, 50 mM Tris-HCl pH 8.0, 0.1 mM
EDTA, 1 mM DTT, 50% glycerol, 0.5%
Tween<fo:inline vertical-align="super" font-size="7pt">™</fo:inline>20, 0.5% Nonidet<fo:inline vertical-align="super" font-size="7pt">™</fo:inline>P40), and
50 ng
of template DNA. Reactions were performed in a
UNO-Thermoblock thermal cycler (Biometra version 2.72,
Göttingen, Germany), programmed as follows: 5 min
at 94 °C for initial denaturation, 45 cycles of
5 s
at 94 °C (denaturation), 30 s at 36 °C
(annealing), and 1 min at 72°C
(extension). A final extension step at 72 °C
for 5
min followed. A total of twenty-eight
10-mer primers of arbitrary sequence (Operon
Technologies, Alameda Calif.) were selected for PCR
amplification (OPA01, OPA02, OPA03, OPA04, OPA07, OPA09, OPA10,
OPA13, OPA15, OPA18, OPC05, OPC06, OPC08, OPC12, OPE01, OPE03, OPE04,
OPE06, OPE07, OPE08, OPE11, OPE12, OPE14, OPE15, OPE16, OPE18, OPE19,
OPE20). The amplification products were visualised on
1.5% agarose gels stained with ethidium bromide, using
standard methods (Sambrook et al., 1989). Three replications of each reaction
for each selected primer were carried out and only reproducible bands
were scored.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="12pt" text-align="start" font-style="italic">ISSR assays</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">Amplification reactions were carried out in volumes of 20
μl containing 30 ng template DNA, 1.0
unit <fo:inline font-style="italic">Taq</fo:inline> DNA polymerase (Pharmacia,
Biotech), 0.25 mM each dNTP (Gibco BRL)
and 1 μM primer (Gibco BRL), in 1 ×
reaction buffer (50 mM KCl, 1.5 mM
MgCl<fo:inline vertical-align="sub" font-size="7pt">2</fo:inline>, 10 mM Tris HCl pH 9.0). PCR
reactions were performed under the following conditions: 4 min
at 94 °C for initial denaturation, 27 cycles of
30 s
at 94° (denaturation), 45 s at
52 °C (annealing) and 120 s at
72 °C
(extension), followed by 7
min at 72 °C for
final extension of the single strands. Thirteen primers were screened
using a bulked DNA sample and seven primers that produced a complex
band pattern in agarose gels were selected for final amplifications
(Table 3 ). ISSR
amplified fragments were mixed with equal volume of formamide dye
(98% formamide, 10 mM EDTA, 0.05%
xylene cyanol), loaded onto pre-warmed denaturing
6% polyacrilamide gels 7.5 M urea, and
electrophoresed in 1XTBE buffer, at 50 W constant power
until the dye front reached the end of the gels. The gels were silver
stained as described in Bassam et al. (1991).
</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="12pt" text-align="start" font-style="italic">Data analysis</fo:block>
<fo:block font-size="8pt" font-family="serif" line-height="10pt" space-after.optimum="12pt" text-align="start" font-style="italic">Morphological traits</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify"> Data analysis followed
three steps: analysis of variance, multivariate analysis and
regression analysis between morphological and climate variables.
</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">The effect of interaction cultivar × year and tree within
cultivar were determined according to the following model:
<fo:external-graphic alignment-baseline="baseline" src="file:e001.svg"/>
where <fo:external-graphic width="20pt" alignment-baseline="baseline" height="20pt" src="m001.svg"/> is the observation of the tree <fo:external-graphic alignment-baseline="baseline" src="m002.svg"/>) within the cultivar <fo:external-graphic alignment-baseline="baseline" src="m003.svg"/>) in the year <fo:external-graphic alignment-baseline="baseline" src="m004.svg"/>); μ is the mean value of all
observations; <fo:external-graphic alignment-baseline="baseline" src="m005.svg"/> are the effects of the cultivar <fo:inline font-style="italic">m</fo:inline>, the tree <fo:inline font-style="italic">i</fo:inline> within cultivar <fo:inline font-style="italic">m</fo:inline>, the year <fo:inline font-style="italic">j</fo:inline>, interaction cultivar × year,
and the error associated to the interaction tree within cultivar
× year.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">Cluster and Principal Component Analysis (PCA) were
performed using the NTsys-pc ver.1.8 (Rohlf, 1993) software. For both analyses, a correlation matrix was
computed from the standardised data to determine the similarity among
cultivars. For cluster analysis, the UPGMA (unweighted
pair-group method with arithmetic averages) was used and
dendrograms were constructed. In PCA, eigenvectors were calculated to
determine the contribution of each variable for the separation of the
cultivars. A Minimum Spanning Tree (MST) was also included.
</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">Environmental variables were computed from the meteorological
station nearby (7 km) for the period
1992–1997, in order to perform regression analysis between the
morphological data and climate variables. These variables were the
same as reported by Pereira-Lorenzo et al (1996a), plus the accumulated temperature
(°C)
below seven degrees during the dormant period, between October of the
previous year and April of the current year (COA). The
effective thermal index (ETI) is defined as the accumulated
temperature degrees (°C) above seven between
budbreak
and harvesting time. The accumulated rainfall from April to October
(RAO) and from July to October (RJO) were the
other variables computed (Table 4).
</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="12pt" text-align="start" font-style="italic">Molecular markers</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">The information provided by each marker system was determined
according to Powell et al. (1996) indices: Effective Multiplex Ratio
(number of polymorphic products from a single amplification
reaction), Expected Heterozygosity (<fo:external-graphic alignment-baseline="baseline" src="m006.svg"/>, where <fo:external-graphic alignment-baseline="baseline" src="m007.svg"/> is the allele frequency for the <fo:inline font-style="italic">m</fo:inline>
<fo:inline vertical-align="super" font-size="7pt">th</fo:inline> allele) and Marker Index (the product
of Effective Multiplex Ratio and Expected Heterozygosity).
Similarity values were estimated based on the fraction of bands
common to each pair of cultivars, according to
Nei and Li's (1979) coefficient. Cluster
analyses were performed to construct dendrograms, using the
unweighted pair-group method with arithmetic averages
(UPGMA) from the similarity data matrices. The Numerical
Taxonomy and Multivariate Analysis System program package for
personal computer (NTSYS-pc version 1.8;
Rohlf, 1993))
was used in statistical calculations.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="12pt" text-align="start" font-weight="bold">Results</fo:block>
<fo:block font-size="12pt" font-family="serif" line-height="12pt" space-after.optimum="12pt" text-align="start" font-weight="bold">Morphological traits</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">The F values resulting from the conducted analysis of variance
are displayed in Table 5 .
The results of interaction cultivar × year revealed significant
differences for all the morphological characters studied except for
percentage of nuts with diameter 30–35 mm
(PN30-35). Significant differences among trees
within cultivars were found only for length of the leaf blade
(LLB), width of the leaf blade (WLB), length of
the petiole (LPE), number of veins (NVE) and
length of inflorescence (LIN).</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">Dendrograms based on the correlation dissimilarity coefficients
from cluster analysis of the six cultivars, calculated for each of
the five years separately, revealed different phenotypic
classifications (Figure 1). This aspect was
particularly true for the years
1993 and 1994, reinforcing the importance of collecting morphological
data during several years or in different regions.
</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">Principal Component Analysis associated with the Minimum
Spanning Tree, based on the five-year averages, provided
complementary information to cluster analysis, as it allows a
graphical presentation of the distribution of the cultivars in a
three-dimensional plot
(Figure 2). In
this representation `Martainha1',
`Martainha2' and `Verdeal' are clearly
separated from `Longal5', `Longal6' and
`Amarelal'. `Longal5' and
`Amarelal' are the most similar cultivars.
</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">The eigenvalues indicate that three components provide a very
good description of the data, as account for 80.7% of the
standardised variance (Table 6). The analysis of eigenvectors provides information
about the traits responsible for the separations along the first
three Principal Components. PC1 had 45.6% of the total
variation. Percentage of nuts with diameter <30 cm
(PN30) contributed positively to PC1. In contrast, number
of veins in 10 cm (NVE), percentage of nuts
with diameter 35 to 40 cm (PN35-40)
contributed negatively. This Principal Component is responsible for
the separation of `Martainha2' from `Longal5'
and `Amarelal'. PC2 exhibited 20.8% of the total
morphological variability mainly caused by differences in the length
of the leaf blade (LLB) and length of the petiole
(LPE), and is responsible for the individualisation of
`Martainha1' apart from the other cultivars. PC3 had
14.2% of the total variation and was positively associated
with the number of feminine inflorescences (NFI), whereas
percentage of brachystaminate inflorescences (PBI) was
negatively associated. PC3 is responsible for the separation of
`Longal5' from `Amarelal'.
</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">No significant correlations were found between morphological
data and climatic variables except for length of the leaf blade, and
percentage of unisexual and androgynic inflorescence. Length of
the leaf blade (LLB) was correlated (<fo:inline font-style="italic">p</fo:inline> ≤ 0.05)
with the two independent variables RAO/RJO, percentage of
unisexual inflorescences (PUI) was highly correlated
(<fo:inline font-style="italic">p</fo:inline> ≤ 0.001) with IET and correlated (<fo:inline font-style="italic">p</fo:inline> ≤ 0.01)
with any combination of IET and other variables, while percentage of
androgynic inflorescences (PAI) was correlated
(<fo:inline font-style="italic">p</fo:inline> ≤ 0.01) with COA and its combination with any of the
other three variables.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="12pt" text-align="start" font-style="italic">Molecular markers</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">The twenty-eight selected primers yielded a total of 224
reproducible, well defined RAPDs, of which 125 (56%)
were polymorphic. An average of 8.2 bands, ranging from 7 to 12, was
observed for each primer.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">A total of thirteen ISSR primers were tested. Primers
(AT)8YC, (AG)8YC, (GA)8YC,
(GT)8YC, (AGC)4YR and (TCC)5RY failed
to amplify a sufficient number of fragments visible in agarose gels
and were not used in the analysis. The seven primers used detected 157
polymorphic ISSR fragments, in a total of 291 bands resolved,
accounting for 54% of polymorphisms. The average number of
scored bands per primer was 41.6, ranging from 31 to 68, about
5-fold superior to RAPD analysis.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">Both RAPD and ISSR fragments ranged from about 250 to 1750
bp in the scored region. Although a small percentage of
RAPD markers were not reproducible among the three replication
reactions, ISSR analysis revealed perfect reproducibility, even if
performed by different operators, in different days and in different
PCR reactions and electrophoresis.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">A summary of the effectiveness of the different markers is
given in Table 8. ISSR
markers proved to be more useful than RAPD markers, since it present
a marker index of 8.12, while this indice is 1.54 for RAPDs. This
aspect is due to the superior multiplex and effective multiplex
ratios provided by ISSR analysis, since the calculated percentage of
polymorphic bands and Expected heterozygosity were similar for both
marker systems.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">Similarity coefficients calculated according to
Nei and Li (1979) coefficient revealed
high mean similarities among cultivars of 82% and 83%
for RAPD and ISSR analysis, respectively. The phenetic classification
obtained using UPGMA as a clustering method is represented in the
dendrograms of Figure 3. High congruence was obtained among the two markers. The
similarity matrices calculated for RAPD and ISSR data were compared
and a significant correlation (<fo:inline font-style="italic">r</fo:inline> = 90.5%) was
obtained. In both dendrograms the cultivars are grouped in two
clusters: The first cluster is composed of `Amarelal' and
`Longal5', while in the second cluster,
`Longal6' and `Martainha2' clustered
together, as well as `Verdeal' and
`Martainha1'. These two cultivars showed very high
similarity in both analysis. Supposed homonym cultivars appeared
clearly separated.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" space-after.optimum="12pt" text-align="start" font-weight="bold">Discussion</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">The results of this work point out the advantages of molecular
markers over morphological data for diversity studies among European
chestnut. Since RAPD and ISSR markers assess differences among the
cultivars at the DNA level, no interaction with the environment is
expected. On the contrary, morphological traits were affected by the
particular conditions of each individual year. Considerable
differences were obtained between cluster analysis performed with
data from the years 1993 and 1994 and the remaining ones. This is not
fully unexpected, since these years had particular climatic
conditions, especially with respect to COA. The year 1993 showed also
significantly higher rainfall, compared to the other years
(Table 4). Correlation
of individual morphological traits with climate variables showed
significant correlation of variables related with temperatures
(IET and COA) with the percentage of unisexual and
androgynic inflorescences, respectively. The effect of temperature in
flower physiology and sex expression is well known for many species
(Sedgley and Griffin, 1989). This correlation was also
observed by Pereira-Lorenzo et al. (1996a).
The length of the leaf blade was correlated with RAO/RJO. These
correlations may have been responsible for the differences in the
phenotypic classification observed, since when this three variables
were removed from the analysis, a more congruent classification was
obtained among the different years (data not shown). These
aspect strengths the need to include only morphological characters
strongly genetically determined in the descriptor lists for germplasm
classification and evaluation.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">The existence of significant differences among trees within
cultivars was reported for some morphological traits and isozyme
systems (Pereira Lorenzo et al., 1996a, 1996b). Although morphological and isozyme
characterisation are environmental-dependent, according to
these authors, the genetic effects are expected to be more important
than the environmental effects, so this suggests the existence of
genetic variability within cultivars. In our study, however, although
significant differences among trees were also observed for some
traits, the intercultivar differences were higher than intracultivar
variation, and the results of molecular characterisation provided
indication that intracultivar variability should be minimal.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">Both morphological and molecular data proved that the
similarities observed among supposed homonym cultivars,
`Martainha1' and `Martainha2', as well as
`Longal5' and `Longal6', are sufficiently low
to be considered the same cultivar.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">High congruence was obtained among phenetic classification
using RAPD and ISSR data. This can be explain by the similar way that
this two marker systems detect polymorphisms, since both are dominant
markers, the primers anneal arbitrarily and the fragments detected
present about the same size. ISSR markers proved, however, to be more
useful, due to the high effective multiplex ratio and
reproducibility. Even using polyacrylamide gels to resolve the ISSR
amplification products, this technique can be less time and money
consuming due to these two features.
Yang et al. (1996) found lower relative costs for ISSR, compared to RAPD,
since RAPD present extra costs due to the relatively low frequency of
reproducible polymorphisms.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">Despite the disadvantages of morphological classification, a
significant correlation was obtained among similarity matrices of
RAPD and ISSR with binarized morphological data, without the three
variables correlated with climate variables (<fo:inline font-style="italic">r</fo:inline> =
71.8% and 71.3%, respectively). This aspect reveal
that, for the cultivars employed in this study, morphological
classification is comparable to molecular classification, if
environmental-dependent variables were excluded from the
descriptor lists. In many situations, the significance of the derived
variables (principal components) may be useful in giving a
general and biological description of the main traits of the
cultivars with efficiency of discrimination greater than would be in
the case if only non-mapped molecular markers were used.
</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">In this study, the main difference among phenotypic and
phenetic dendrograms, was about `Longal6'. This cultivar
was, with morphological characters, clustered together with
`Longal5' and `Amarelal'. This is not fully
unexpected, since both cultivars are named `Longal', in
different regions of Portugal, according to a morphological
classification. This aspect points out other disadvantage of
morphological data, since it only reveals information about a limited
part of the genome.</fo:block>
<fo:block font-size="10pt" font-family="serif" line-height="12pt" text-align="justify">In conclusion, characterisation based on molecular markers is
faster, less expensive and more reliable that the one based on
morphological characters. However, morphological classification can
be useful in cases when it is necessary to obtain an agronomic
description of the germplasm used. In these cases, only variables
with a strong genetic control should be quantified. Mapped molecular
markers closely associated to these characteristics (both single
traits and Quantitative Trait Loci) can provide, however, the
best solution for characterisation studies. Therefore, the
development of mapping programs for all species, including chestnut,
for which this information is still not available, is necessary.
</fo:block>
<fo:external-graphic display-align="center" src="f001.jpg"/>
<fo:block font-size="10pt" space-before.optimum="12pt" text-align="start"><fo:inline font-weight="bold">Figure 1.</fo:inline> UPGMA dendrograms obtained by Cluster Analysis based on the
morphological data collected by individual year.</fo:block>
<fo:external-graphic display-align="center" src="f001.jpg"/>
<fo:block font-size="10pt" space-before.optimum="12pt" text-align="start"><fo:inline font-weight="bold">Figure 2</fo:inline> Projection of the cultivars and Minimum Spanning Trees on the
first three principal components based on the five-year mean
values of the morphological observations.</fo:block>
<fo:external-graphic display-align="center" src="f002.jpg"/>
<fo:external-graphic display-align="center" src="f001.jpg"/>
<fo:block font-size="10pt" space-before.optimum="12pt" text-align="start"><fo:inline font-weight="bold">Figure 3</fo:inline> UPGMA dendrogram obtained by Cluster Analysis based on a)
RAPD and B) ISSR amplified fragments, for the six cultivars
studies.</fo:block>
<fo:external-graphic display-align="center" src="f003.jpg"/>
</fo:flow>
</fo:page-sequence>
</fo:root>
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
