Basic HTML Version
Molecular Plant Breeding 2011, Vol.2, No.12, 83
-
91
http://mpb.sophiapublisher.com
86
were randomly assigned to one of them. Hybrids
formed a distinct and homogenous group. Among the
42 local Tall, 19 of them were assigned to their own
accessions, showing that there are some differences
between accessions. The expected value for the null
hypothesis of no differences between accessions was
5.7 (χ
2
=208
***
).
Table 2 Average numbers of alleles amplified and observed
(
Ho
) and expected (
He
) heterozygosities for each accession
Hybrid
Average alleles number
Ho
He
WY
2.759 0
0.515 4 0.600 9
Tall
HK-GT
1.846 2
0.153 8 0.475 3
SJ-GT
2.204 3
0.336 5 0.153 8
WC-GT
2.637 2
0.192 3 0.329 9
WC-YT
1.576 9
0.423 1 0.423 1
WC-RT
2.336 6
0.307 7 0.417 2
CP-GT
2.333 6
0.236 3 0.425 2
QH-GT
2.508 6
0.211 5 0.364 0
XL-GT
2.226 0
0.326 9 0.432 7
XL-RT
2.603 8
0.205 1 0.294 9
Average
2.252 6
0.265 9 0.368 5
Table 3
F
statistics in the local Tall accessions
Parameter
Estimate
Confidence interval (α=0.95)
F
IT
0.157
0.089~0.237
F
ST
0.078
0.046~0.117
F
IS
0.054
0.018~0.106
Note: Confidence intervals are estimated by bootstrapping
In spite of the deviations from the sampling protocol,
members from the same accession still showed closer
relationships than those from the different accessions.
Three plantations (Wenchang, Changpo and Qionghai)
were represented by two accessions with a total of 26
individuals (except WY78F1). Among them, 11 are
assigned to their own accessions. Among the other
remaining 15 individuals, four were assigned to the
other accession of the same plantation. This result
tends to weak the pattern suggested by the overall FST
value (see above): A small part of the differences
among accessions was due to the differences among
the plantations.
1.3 Cluster analysis
The dendrogram was constructed from the genetic
distance matrix by NTSYS program (Rohlf, 2000)
(Figure 1). The 45 coconut individuals of HNT were
divided into two main groups (i.e., Group
Ⅰ
and
Group
Ⅱ
). Furthermore, Group
Ⅰ
was subdivided
into two subgroups. This dendrogram shows that the
members of the same plantation were not assigned to
the same group as expected: E.g., both groups
included Qionghai Green Tall (QH-GT), Xinglong
Green Tall (XL-GT) and Wengchang Yellow Tall
(WC-YT). Some accessions from different plantations
were put into one group (e.g., CP-GT (Changpo Green
Tall) 1, 5, 7 and QH-GT (Qionghai Green Tall) 1, 2, 3).
This result is consistent with that from the Bayesin
assignment tests.
2 Discussion
Genetic diversity is necessary to sustain the
productivity of a crop since it furnishes new genes for
yield, adaptation, disease resistance, high-value uses
and characters (Frankel and Soulé, 1981). Rich
diversity of coconuts exists in the field of farmers, and
it is tremendously necessary to use this diversity to
formulate strategies, thereby solving the problems of
coconut farmers. In the present study, we
characterized 10 accessions of coconuts by using the
molecular tools in six coconut plantations in Hainan
province. Moreover, we quantified the extent of
genetic diversity and determined the population
structure and its relationship with other coconut
populations. Furthermore, our results also showed the
relatively high genetic diversity of coconuts in Hainan
province.
Our study has practical implications in farmer
participatory evaluation and conservation of coconut
genetic resources. Coconut accessions, which are
adapted to the local conditions, high-yielding and
possessing valuable characteristics, are under a threat
of genetic erosion. In order to maintain coconut
genetic resources, it is necessary to evaluate the extent
of genetic diversity in the accessions. This process
would make a significant contribution in promoting
conservation of coconut germplasm in farmer′s fields
through in situ and on-farm conservation (Batugal and