Reproductive biology of the Magenta Lilly Pilly (Syzygium
paniculatum) and its implications for conservation
Katie A. G. Thurlby
, William B. Sherwin
, Maurizio Rossetto
, and Peter G.
National Herbarium of NSW, Botanic Gardens Trust, Mrs Macquaries Rd Sydney, NSW
School of Biotechnology and BioMolecular Sciences, University of New South Wales,
TABLE OF CONTENTS
Syzygium paniculatum: a brief description
Threatened species: population size and diversity
Polyembryony, clonality and Syzygium paniculatum
Fitness, clonality and Syzygium paniculatum
Ploidy in Syzygium paniculatum
Applicability of nuclear microsatellite
2. Materials and Methods
Obtaining and analyzing diversity data
Obtaining and analyzing germination and reproduction data
Conservation and horticultural implications
The Magenta Lilly Pilly (Syzygium paniculatum), endemic to a narrow strip along the
New South Wales coast, is currently listed as vulnerable at both state and national
levels. At present management of the species focuses on minimizing currently known
threats, such as weed invasion, while little is known about the reproductive biology of
the species. S. paniculatum is the only recorded polyembryonic Australian species of
Syzygium; polyembryony being the development of multiple (and often asexual)
embryos in one seed. Nuclear microsatellite markers were used to investigate the
genetic outcome of polyembryony on the reproductive and population biology of the
species focusing particularly on the population located on The Entrance Peninsular.
Low within-population diversity was found, with low heterozygosity levels and a low
level diversity indices when compared to other rare or rainforest species. Multiple
embryos from single seeds were found to be identical to the mother. Multiple embryos
germinated and survived but one seedling was always significantly taller than all
others in the seed but was not considered sexual. It was concluded that the rare S.
paniculatum is an apomictic clonal species with extremely low genetic diversity.
Syzygium paniculatum: a brief description
Gaertn., commonly known as Magenta Lilly Pilly, is a small
to medium tree in the family Myrtaceae. S. paniculatum is endemic to New South Wales,
occurring along a narrow, linear, coastal strip in five separate geographical areas: Jervis Bay,
Coalcliff, Botany Bay, Central Coast and Seal Rocks (Figure 1). At Jervis Bay and Coalcliff,
grows in remnant stands of littoral rainforest in grey soils over sandstone. At
Botany Bay, the Central Coast, and Seal Rocks, the species can be found growing in
remnant littoral rainforest patches and riverside gallery rainforests on sandy soils, stabilized
dunes, gravels, silts or clays. Plants produce clusters of white flowers at the ends of
branches, usually between November and February. The fruits are deep magenta, maturing
from March to June, and containing a single seed, which is often polyembryonic (containing
more than one embryo, see Figures 2, 3, 4 & 5).
S. paniculatum is of conservation significance as it is listed as endangered under
NSW Threatened Species Conservation Act 1995 (TSC Act) and vulnerable under Federal
legislation. If appropriate conservation of S. paniculatum is to take place, to increase
population numbers and protect the species from extinction, natural populations need to be
studied to determine genetic diversity, breeding strategy and fitness potential of offspring.
An understanding of the reproductive biology of this species is likely to be crucial because
the polyembryonic seeds may have arisen through asexual apomixis.
If polyembryony in Syzygium paniculatum is a product of apomixis, it is likely to
have substantial implications for any recovery plan because apomixis may reduce genetic
diversity within each population and increase genetic divergence between populations.
Although uncommon in native vegetation communities, the species is widely
cultivated and a number of cultivars are on sale commercially. Genetic distinctiveness and
reproductive biology could also have important implications for potential interactions
between cultivars and natural populations.
Distribution of S. paniculatum
along the NSW coast and indicates the five
geographically separated areas in which Syzygium paniculatum occurs; Seal Rocks
(orange), Central Coast (blue), Botany Bay (red), Coalcliff (pink) and Jervis Bay
Figure 2. Botanical illustration of S. paniculatum. a. fruiting branch – scale bar 3cm,
leaf venation – scale bar 2.5cm, c.
opening bud – scale bar 1cm, d.
– scale bar 1cm, e. transverse section of fruit showing seed containing multiple
embryos – scale bar 1.5cm. (L. Elkan 2008 © Botanic Gardens Trust)
Threatened species: population size and diversity
The fragmentation and isolation of habitats, due to human disturbance or geological
change, has resulted in the reduction in population size of many species of plants. Small
population size is often a characteristic of threatened species (Frankham, 1996) and has been
found to be positively correlated with genetic variation (Leimu et al., 2006). Reduced
population size can lead to reduced evolutionary potential due to genetic drift and inbreeding
(Frankham, 1999). Genetic drift and inbreeding may result in marked genetic effects for
populations such as reduced genetic diversity through changes in allele frequencies and loss
of heterozygosity (Allendorf & Luikart, 2007; Frankham, 1996; Frankham, 1999). Thus the
potential for rapid erosion of genetic variation may make small populations poor candidates
for conservation efforts (Lesica & Allendorf, 1992). However,
it is not always the case that
rare species with reduced population sizes have low diversity (Gitzendanner & Soltis, 2000;
Lesica & Allendorf, 1992; Lewis & Crawford, 1995; Ranker, 1994). Because genetic
diversity is more likely to depend on the life history and evolutionary history of species, it
cannot be assumed that all rare species will maintain low levels of genetic diversity
(Gitzendanner & Soltis, 2000)
Polyembryony, clonality and Syzygium paniculatum
There are a number of different processes by which multiple embryos (polyembryony) can
arise in plant seeds (Stebbins, 1941; Naumova, 1992; Ozias-Askins, 2006; Richards, 2003).
Polyembryony can occur through the division of a fertilized diploid zygotic embryo
(cleavage polyembryony), resulting in the development of multiple yet identical sexually
derived embryos. It can also arise through various forms of apomixis, where additional
asexual embryos develop from diploid cells, usually alongside the fertilized diploid zygotic
embryo. The latter may result in the development of both sexual and asexual embryos (if a
zygote develops and survives), or only asexually derived embryos (if the zygote fails to
develop or develops and subsequently degenerates or is out-competed). Apomixis can
therefore, in some cases, result in the production of high levels of clonal individuals in a
population, who have inherited their entire genome from their mother. This has the potential
to drastically reduce the genetic diversity in the population or across the species if the clonal
offspring of one mother happen to predominate. Also, if sexual reproduction is reduced (in
facultative apomicts) or virtually non-existent (in obligate apomicts), the rate at which new
combinations of alleles are made will also be reduced, in effect, hindering potential increase
in genetic diversity. Apomicts can acquire genetic variability through means such as somatic
mutation or somatic recombination; however most of the genetic variability found within
apomictic species is a product of past sexuality (Richards, 2003).
Polyembryony has not been recorded for any other Australian species of Syzygium,
but has been observed in a number of Asian species: S. malaccense (L.) Merr. & L.M.Perry,
S. jambos (L.) Alston, and S. cumini (L.) Skeels. In S. jambos, Roy (1953) described the
embryos as deriving from diploid cells in the nucellus, the maternal tissue surrounding the
embryo sac. In Myrtaceae, formation of triploid endosperm (after pollination) is a source of
nutrition to developing sexual embryos and this is also the case for nucellar embryos (Roy,
1961). This pollination requirement suggests that polyembryony in Syzygium paniculatum is
not brought on by male sterility or lack of pollination. Sahai & Roy (1962) report the
formation of a zygote that degenerates soon after fertilization so that only asexually derived
Understanding the immediate genetic outcome of polyembryony in S. paniculatum,
will help in gaining an understanding of the effect that polyembryony has on population
biology; both in S. paniculatum, as well as in other species. To determine which type of cells
asexual embryos are derived from in S. paniculatum, would require a detailed embryological
study and this was beyond the scope of this study. Regardless which type of maternal tissue
gives rise to asexual embryos, they do not undergo meiosis and hence there is no
recombination to facilitate chromosomal crossover and embryos will be genetically identical
to the maternal plant, except in rare cases of somatic mutation or mitotic crossover. By using
suitable molecular techniques to study the population genetics of the species and the
genotypes of offspring, it may be possible to define the reproductive biology of S.
paniculatum as either sexual or asexual, or both. If embryos or offspring show genetic
evidence of recombination, reproduction in S. paniculatum may be defined as sexual. If all
offspring are found to be identical to the mother, reproduction in the species may be defined
as asexual and the species could be regarded as clonal.
Fruit, seed and embryos of S. paniculatum
. From left to right: the magenta
coloured fruit; a single seed with seed coat left on; three embryos all dissected from a
Figure 4. Five embryos of varying size, dissected from a single S. paniculatum seed,
collected from Abrahams Bosom. Each square in the scale bar represent 1 cm.
Polyembryonic seed of S. paniculatum
with seed coat removed (left) and the
same seed dissected to reveal five embryos and an under-developed embryo (right).
Clonality, fitness and Syzygium paniculatum
The process of clonality can be seen as having a fitness trade-off between
proliferation of provisionally fit clonal genotypes and the ability to generate potentially fitter
sexually derived genotypes. Plants that maintain purely clonal reproductive habits have the
potential for high genetic costs and disadvantages in changing environments (Callaghan et
al., 1992). In some species, such as Syzygium jambos, polyembryony has been suggested to
be so prolific that sexual reproduction is suppressed altogether (Sahai & Roy, 1962). If
polyembryony in S. paniculatum is also found to be a completely clonal process, with all
embryos identical to the parent plant, the number of potential new genotypic combinations
will be limited. This could make the species even more vulnerable to future environmental
change than originally thought. Even if some sexual reproduction is occurring alongside
clonal reproduction, the consequence could still be a dramatic reduction in genetic diversity
(Richards, 2003; Silvertown, 2008).
Ploidy in Syzygium paniculatum
Polyploidy is known to be linked to polyembryony and apomixis. It is thought that
polyploidy may serve as a barrier for sexual reproduction (Andrew et al., 2003; Lo et al.,
2009; Ozias-Askins, 2006; Van Der Hulst et al., 2003; Van Dijk & Van Damme, 2000)
whilst duplicate genes may be expressed asynchronously causing multiple embryos to arise
(Carman, 1997). Alternatively, it has also been suggested that the occurrence of apomixis
can actually promote polyploidy (Richards, 2003).
It is possible that polyembryonic S. paniculatum may also be a polyploid. In the
Myrtaceae, diploid (2n=22) is the most common ploidy level but polyploidy has been
documented, mainly in fleshy-fruited species (Da Costa & Forni-Martins, 2006 & 2007).
The polyembryonic species, Syzygium jambos, has been found to be tetraploid (2n=44)
(Oginuma et al., 1993; Roy, 1953; Singhal et al., 1985). However, diploid apomicts do exist
(Bicknell, 1997) so polyploidy in S. paniculatum is speculative at best and needs to be
Applicability of nuclear microsatellite markers
The most suitable molecular markers for population studies are nuclear microsatellite
markers (nSSRs) since they show co-dominance, simple Mendelian inheritance, high rate of
polymorphism and mutation, abundance throughout the genome, reproducibility and relative
ease of screening (Rossetto, 2001). They are also suitable
for studying parentage, when the
reproductive biology of a species is essentially unknown, but clonality is considered a
possibility (Jones et al., 2009).
This research project will focus on the largest and better documented population
(Payne 1991) located on the North Entrance Peninsula (c. 33º 18' 30″S, 151º 31' 10″E). The
population will be sampled at three hierarchical levels in order to obtain information on the
occurrence of apomixis, and on its potential effect on native populations as well as on the
We will test mature trees as a representation of genetic diversity across the
population. We will test embryos to determine the likely origin (whether sexual, asexual
reproduction or both, occur in S. paniculatum) We will also test embryos to assess
germinability and fitness and the likely survival of apomictic vs. zygotic embryos within
controlled conditions (i.e. information relevant to the horticultural trade) and within natural
conditions (i.e. within the representative individuals sampled; information relevant to the
conservation of this rare species).
Highly informative molecular tools (microsatellite loci) will be used to obtain the
relevant genetic information without the need for complex morphological and
developmental investigations. Co-dominant and highly variable molecular tools such as
microsatellites (SSR) are particularly suited to these investigations as changes in allelic
frequencies and heterozygosity measures can differentiate between outcrossed, selfed and
This is a unique approach, not been attempted before on a rare tree and likely to
provide relevant information for conservation and management. For instance, cultivated
material could include greater levels of apomixis than natural populations, and since S.
is a popular garden plant, dispersal of cultivated material onto native
populations could potentially result in significant losses of natural diversity.
This project will provide enhanced understanding of genetic diversity at the population
level, indication as to whether apomictically derived embryos can be used in propagating
individuals with a particular genotype rather than vegetative propagation, and results
relevant for the development of conservation and management strategies for wild
populations and cultivated plants
2 Materials and Methods
Obtaining and analyzing diversity data
Leaf sample collections were made at the main study site, The Entrance, with the
aim of collecting over 30 individuals across the site.
In addition to leaf collections from The Entrance, collections were also made for ten
other populations of Syzygium paniculatum (Figure 6). At two populations, Captain Cook
Drive and Wamberal Lagoon this included wild juveniles. . Additional fruit samples were
also collected at Abrahams Bosom, Towra Point and Cams Wharf, to supplement those
collected at the Entrance (as fruit set was generally low in the two years the project was
running and the proposed targets could not be achieved). Collections at additional
populations was work extra to the AFF funded project, and was conducted as part of an
Honours project (Katie Thurlby, UNSW) which is currently in preparation for publication.
Total genomic DNA was extracted from dried leaf tissue and frozen embryo tissue
using the Qiagen® DNeasy® 96 Plant Kit protocol. Nine nuclear microsatellite markers
were developed for S. paniculatum and were used to determine genotypes of adults in wild
populations as a measure of diversity and to study the parentage or breeding system through
offspring genotypes. Preliminary genotyping results showed extremely low variation, so
primers were also tested in eight diverse species of the tribe Syzygieae, including the
polyploid, Syzygium jambos.
Captain Cook Drive
Conjola National Park
Figure 6. Populations sampled are grouped according to the geographical area by
coloured symbols: Seal Rocks (orange triangle), Central Coast (blue circle), Botany
Bay (red square), Jervis Bay (green star).
Genotyping traces were analysed in Genemapper 4.0 (Applied Biosystems) to look
for similarities between species in microsatellite peak patterns, which may serve as an
indicator for ploidy level. Genotyping traces of the known tetraploid Syzygium jambos and
assumed diploid species, S. corynanthum and S. francisii, were compared with genotyping
traces of S. paniculatum. Chromosome counts were also attempted, however time and
resources did not allow for optimization of the necessary techniques and as such,
chromosome counts are yet to be completed.
Analysis of diversity data
The preliminary results for ploidy identification indicated that the species might be
tetraploid; having four copies of the genome rather than the usual two, making a co-
dominant analysis approach difficult. Standard analysis packages such as GenAlEx (Peakall
& Smouse, 2005) require determination of microsatellite allele copy number. In diploid
species, determining allele copy number is simple because a heterozygote will yield two
peaks (two different microsatellite lengths) in a genotyping trace, whilst a homozygote will
yield one peak (two copies of the one microsatellite length). For tetraploid species, with two
copies of a microsatellite from each parent, determining allele copy number is not always
possible. For this reason, diversity statistics were calculated using the program ATETRA
(Van Puyvelde et al., 2009) which is designed specifically for analysing tetraploid data. Use
of ATETRA required an assumption of ploidy level. After preliminary testing, tetraploidy
was inferred for Syzygium paniculatum however, ploidy should ideally be confirmed by
Also, following other studies of clonal polyploid species (Andrew et al., 2003;
Samadi et al., 1999) and to confirm results obtained through ATETRA, a dominant binary
approach was also taken. This approach uses all the alleles present in the population to
construct a genetic barcode for each individual by scoring alleles across all microsatellite
loci as present or absent. Data could then be used in the program GenAlEx (Peakall &
Smouse, 2005) to perform various data analyses.
Obtaining and analyzing germination and reproduction data
Fruit samples were collected from some individuals at The Entrance and were
dissected and used for both the reproduction and germination trials. Fruit samples were also
collected from Abrahams Bosom, Towra Point and Cams Wharf, for use in the Honours
project (Katie Thurlby, UNSW) to supplement the data obtained at The Entrance.
The germination and reproduction trials were designed to investigate the fitness and
parentage of embryos. Both seeds with single and multiple embryos were germinated. The
first embryo to germinate from each polyembryonic seed was recorded and measurements of
seedling leaf number and height were taken every three to six weeks. Statistical analysis was
Leaf samples were then taken from germinated seedlings, DNA extracted and
genotyping conducted. Genotypes of each individual seedling were compared to other
individual seedlings originally belonging to the same seed (family groups) to look for allelic
differences which may suggest the occurrence of either sexual reproduction or
recombination. Embryo material was also collected and DNA extracted however, the DNA
obtained from embryos was not of satisfactory quality for use in genotyping and as such,
only leaf material from the germinated embryos could be successfully genotyped.
The genotyping data for Syzygium paniculatum showed four allele peaks at the SP33
microsatellite locus and three allele peaks at the SP38, SP54 and SP86 loci, suggesting
polyploidy (possibly tetraploid). Syzygium jambos (known tetraploid) also showed four
alleles at marker SP33. The two other Syzygium species, S. corynanthum and S. francisii,
showed normal diploid patterns of one (homozygous) and two (heterozygous) peaks
respectively (Figure 7).
Figure 7. Genotyping traces for one microsatellite locus (SP33). Syzygium
(complete homozygous tetraploid= 4 peaks), Syzygium francisii
heterozygous diploid =2 peaks), Syzygium jambos (complete heterozygous tetraploid=
4 peaks) and Syzygium corynanthum (assumed homozygous diploid=1 peak).
Both low heterozygoisty and extremely low genetic diversity were found at The
Entrance. Of the 31 individuals genotyped across nine microsatellite loci, 30 possessed an
identical genotype with the single remaining individual possessing a genotype that showed
allelic differences at four of the nine loci. These differences were three instances of
homozygosity (in loci that were heterozygous in all other samples) and one instance of a 2bp
change. These allelic differences may be attributable to a rare sexual or selfing event, a
remnant of past sexual reproduction or a genotyping error.
Between population data, obtained as part of the honours project (Katie Thurlby et
al., unpubl), showed that genotypes were distributed along a geographic gradient. It also
showed that the main genotype found at The Entrance is the most common genotype across
the entire species, being present in 70.89% of individuals. The majority of divergences from
the most common genotype occurred in northern populations. All statistical tests showed the
southern populations (most of which contain only the most common genotype) grouped
together with northern populations as outliers. Almost all variation between populations is
explained by this southern grouping. Variation was not found to be related to current
population size, as the largest populations (The Entrance and Wamberal Lagoon) exhibited
proportionally less variation than smaller populations.
Embryo dissection confirmed polyembryony, with embryo number ranging from
one to nine per seed. In most seeds there is one large embryo, with each subsequent
embryo diminishing in size.
Leaves were taken from seedlings germinated ex-situ and family groups were
genotyped. Embryos proved extremely difficult to genotype, probably due to the high
carbohydrate levels in seeds. As fruit set was low, data was supplemented by including
individuals from other populations collected as part of the aforementioned honours
One seedling (single embryo in a seed) from The Entrance showed evidence of
sexual origin (two loci heterozygous in the parent were homozygous in the offspring
and two new alleles were present). Four seedlings germinated from two seeds (two
embryos per seed) showed evidence for the sexual origin of one embryo in the seed
(multiple new instances of homozygosity) and the asexual origin of the other (identical
genotype to parent). This suggests that both sexual reproduction and asexual
reproduction is occurring at The Entrance. In other populations tested there was also
evidence for the sexual origin of embryos (single embryos or one embryo from a
polyembryonic seed) as well as extensive evidence for asexual origin of embryos
(including polyembryonic seeds showing no evidence of sexual origin).
Additionally, juveniles from two populations were sampled and genotyped. Of
the eight wild juvenile leaf samples only one had a different genotype to the adults
sampled at the same location. This individual had a 1bp change at one allele in one
microsatellite locus. This single allele change most likely provides evidence of somatic
mutation (rather than reproduction) as no changes in heterozygosity were found in the
genotype as would be expected if recombination had occurred. It is possible that the
parent plant was not sampled, and as such, that the juvenile is also a single surviving
germinant from an apomictic seed.
From polyembryonic seeds, multiple embryos germinated and multiple seedlings
survived (Figure 8 & 9). Embryos germinated successfully whether they were kept
within the bounds of the seed coat or whether embryos were separated. One seedling
was always taller than other seedlings arising from the same seed suggesting a possible
fitness advantage and this was found to be statistically significant. Seedling height was
highly correlated to embryo weight suggesting that embryo size is a determining factor
in embryo height. When genotype was taken into account and all sexual embryos
compared to all asexual embryos, it was found that sexual seedlings were significantly
taller than asexual seedlings; however the test was no longer significant when the
smaller embryos from polyembryonic seeds were removed from the test (leaving the
largest embryo from a purely asexual seed and the sexual embryo). Seedling height was
not correlated to embryo weight in sexual embryos but was found to be highly
significant in polyembryonic asexual seedlings. Leaf number and branch number were
also recorded but no statistically significant differences were found.
Multiple seedlings arising from a single seed.
Figure 9. Survival of the multiple seedlings (Figure 8) over time. Left to right:
Weeks 12, 18, 21 and 27, (photographs not to same scale).
The results suggest that because Syzygium paniculatum displays similar genotypic
patterns to the tetraploid S. jambos, S. paniculatum is likely to be tetraploid; however, this will
need confirmation through chromosome counts. As the multiple alleles were found at four loci, it
is unlikely that the multiple alleles are caused by localized gene duplication and more likely that
it is a result of whole genome duplication. Although our results do not exclude other higher
ploidy levels, tetraploidy should be assumed to be most likely as it would require the least
number of events to arise. It is not yet known whether S. paniculatum is an allopolyploid, or an
There is a strongly documented link between polyploidy and apomixis (Bicknell &
Koltunow, 2004; Roche et al., 2001) and it is thought that polyembryony might be caused by
duplicate genes being expressed asynchronously (Carman, 1997). It is possible that the
polyembryony found in S. paniculatum is caused by the polyploid nature of the species; however
this would need further investigation.
The present study has shown that S. paniculatum has extremely low overall genetic
diversity. Only nine genotypes were found among eleven populations across the entire
geographic range of the species (Katie Thurlby unpubl. obs., 2011). S. paniculatum also showed
rather low heterozygosity and genetic diversity when compared to other rare species (Katie
Thurlby, unpubl. obs., 2011).
There are a number of explanations as to why The Entrance population possesses such
extremely low diversity and shares an identical genotype with almost all southern populations of
S. paniculatum. It is possible that The Entrance population represents one part of a large clonal
stand of S. paniculatum existing in the southern distribution of the species. This clonal stand is
likely to have been broken into smaller populations due to the occupation of S. paniculatum in
specific costal habitats of littoral rainforest which has been largely cleared or modified for coastal
development, agriculture and sand mining (Floyd, 1990). This clearing has reduced population
numbers substantially and reduced habitat to small, highly vulnerable fragments (Payne, 1991).
Conversely, low levels of heterozygosity, extremely low within population diversity and
almost no diversity, may be an indication of a founder effect. Clonal reproduction, through
dispersible apomictic seed, would facilitate the fast spread of genetically identical individuals to
colonize and expand populations with any new genetic diversity arising via somatic mutation. It
is likely that The Entrance population, along with other southern populations of S. paniculatum,
was colonized in such a way.
It is possible that The Entrance population of S. paniculatum arose as one of multiple
hybridization events. Variability across populations of an apomictic species can be described by
the multiple origins of asexual polyploids by hybridization (Soltis & Soltis, 2000). S.
paniculatum may have arisen in such a manner at a number of different sites where the two
parental species co-occur (or did co-occur). This would result in the pattern of unique genotypes
which we see across the range of S. paniculatum (Katie Thurlby, pers. obs. 2011) and could also
explain the existence of polyembryony and polyploidy in the species. However, none of the
results obtained from this study are able to directly substantiate this idea and further studies are
4.3 Reproduction and Fitness
Reproduction in Syzygium paniculatum appears to be asexual or, clonal. This study has
shown that embryos produced in seeds of S. paniculatum are of both sexual and asexual origin.
This supports the view that S. paniculatum is a facultative apomictic species i.e. able to reproduce
both sexually and asexually, which is more common than obligate apomixis (Asker & Jerling,
1992; Bicknell & Koltunow, 2004). Sexual embryos occurred both in monoembryonic seeds and
polyembryonic seeds however most embryos from polyembryonic seeds showed no evidence of
sexual origin. It is possible that a sexual embryo develops first in all seeds but is sometimes out-
competed by the proliferation of multiple asexual embryos. If sexual reproduction occurs and the
sexual embryo is not out-competed, a sexual individual can survive either along in a seed or
alongside asexual embryos. This study found limited evidence for recent recombination in adult
populations, suggesting that if there is a sexual embryo it never (or hardly ever) survives in-situ.
Also, as a sexual event may occur in as little as 2% of progeny of apomictic species (Bicknell &
Koltunow, 2004) the small samples sizes in this study may have resulted in a gross overestimate
of the sexual reproduction in the species as a whole.
Apomixis can be influenced by environmental factors such as climate, nutritional supply
or competition (Asker & Jerling, 1992) which suggest that environmental influences during
particular seasons may result in a higher rate of sexual reproduction than others. Apomicts are
often found in highly disturbed areas or where individuals are widely dispersed (Asker & Jerling,
1992), so it is possible that habitat disruption and habitat fragmentation may be an influencing
factor for sexuality in S. paniculatum.
The results of this study did not find a significant fitness advantage for sexual individuals
in terms of height; rather, they suggest that embryo size is a significant contributor to seedling
height while embryo origin is not. Sexual reproduction does involve some advantages however,
because through sexual reproduction, species are equipped for relatively fast genetic changes
should they be required by a changing environment conversely, the genotype of one sexual seed
has one chance to survive whereas the identical genotype of multiple asexual seeds has multiple
chances to survive. So it is possible that there underlying are trade-offs between the advantages
and disadvantages of different reproductive modes.
4.4 Conservation Implications
The results of this study imply that Syzygium paniculatum may have some difficulty
adapting to future environmental change. Until now, the species has survived with moderately
low overall genetic diversity and extremely low (if any) variation between individuals. This
survival is likely to be due to the ability of the species to reproduce prolifically by way of
polyembryony and the theoretical fitness of the persistent genotypes in the current environment.
It should however be noted, that whilst there is low between-individual variation in the species,
the nature of apparent tetraploidy in S. paniculatum means that there is high within-individual
variation. Nevertheless, the lack of variation between individuals and in the species as a whole,
and the apparent asexual reproduction, means that the development of new variation is restricted
to the slow process of somatic mutation.
The detection of some sexual reproduction within the species is positive as it means there
is potential within the species to adapt genetically to future environmental change however the
distinct lack of variation particularly at The Entrance suggests that such sexual individuals rarely
It is unlikely that the reduced population sizes and low genetic variation found in S.
are completely due to human interference. However, human activity is likely to have
caused significant reduction in the range and population sizes of the species. From a conservation
perspective, there is a need to improve environmental conditions and habitats where practicable,
not only to preserve current genetic diversity but to give the species the best possible chance for
the accumulation of variation in the future, be that by somatic mutation or sexual reproduction.
This particularly applies to the northern populations over southern populations, where the
majority of known genetic diversity has been found (Katie Thurlby, pers. comm. 2011).
Appropriate activities would include habitat protection and management, weed management, and
fire management plus on-going monitoring and surveying as well as more extensive research.
In addition to preserving habitats, in the interest of preserving allelic variation, it may be
advisable to introduce each of the various different genotypes found across the species into
cultivation to make sure that the variation found within the species is not diluted further by the
cultivation of only the most common genotype.
Propagation of S. paniculatum via seed may be a viable mode of cultivating genetically
identical plants as multiple embryos in each seed are easily germinated more reliably than
through methods of vegetative cultivation and a high proportion of individuals survive beyond
the seedling and sapling stages. Further studies would need to determine the rate of possible
sexual events and the practicality of cultivating via seed for large scale production.
Threatened Species Conservation Act 1995. Department of Environment, Climate Change and
Water, NSW Government, Sydney.
Allendorf FW, Luikart G (2007) Conservation and the Genetics of Populations. Blackwell
Andrew RL, Miller JT, Peakall R, Crisp MD, Bayer RJ (2003) Genetic, cytogenetic and
morphological patterns in a mixed mulga populations: evidence for apomixis. Australian
Systematic Botany 16, 69-80.
Asker S, Jerling L (1992) Apomixis in plants CRC Press, Boca Raton, Florida.
Bicknell R (1997) Isolation of a diploid, apomictic plant of Hieracium aurantiacum. Sexual Plant
Reproduction 10, 168-172.
Bicknell RA, Koltunow AM (2004) Understanding Apomixis:Recent Advances and Remaining
Conundrums. The Plant Cell 16, S228-S245.
Callaghan TV, Carlsson BA, Jonsdottir IS, Svensson BM, Jonasson S (1992) Clonal plants and
environmental change: introduction to the proceedings and summary Oikos 63, 341-347.
Carman JG (1997) Asynchronous expression of duplicate genes in angiosperms may cause
apomixis, bispory, tetraspory, and polyembryony. Biological Journal of the Linnean
Society 61, 51-94.
Da Costa I, Forni-Martins E (2007) Karyotype analysis in South American species of Myrtaceae.
Botanical Journal of the Linnean Society 155
Da Costa IR, Forni-Martins ER (2006) Chromosome studies in species of Eugenia, Myrciaria
and Plinia (Myrtaceae) from south-eastern Brazil. Australian Journal of Botany 54, 409-
Floyd A (1990) Australian Rainforests in NSW
Surrey Beatty, Chipping Norton.
Frankham R (1996) Relationship of Genetic Variation to Population Size in Wildlife.
Conservation Biology 10
Frankham R (1999) Quantitative genetics in conservation biology. Genetics Research 74, 237-
Gitzendanner M, Soltis P (2000) Patterns of genetic variation in rare and widespread plant
congeners. American Journal of Botany 87
Jones AG, Small CM, Paczolt KA, Ratterman NL (2009) A practical guide to methods of
parentage analysis. Molecular Ecology Resources 10, 6-30.
Leimu R, Mutikainen P, Koricheva J, Fischer M (2006) How general are positive relationships
between plant population size, fitness and genetic variation? Ecology 94, 942-952.
Lesica P, Allendorf FW (1992) Are Small Populations of Plants Worth Preserving? Conservation
Lewis P, Crawford D (1995) Pleistocene refugium endemics exhibit greater allozymic diversity
than widespread congeners in the genus Polygonella (Polygonaceae). American Journal
of Botany 82, 141-149.
Lo EYY, Stefanovic S, Dickinson TA (2009) Population genetic structure of dipload sexual and
polyploid apomictic hawthorns (Crataegus; Rosaceae) in the Pacific Northwest.
Molecular Ecology 18, 1145-1160.
Naumova T (1992) Apomixis in angiosperms: nucellar and integumentary embryony CRC.
Nygren A (1948) Further studies in spontaneous and synthetic Calamagrostis purpurea.
Hereditas 34, 113–134.
Oginuma K, Kato A, Tobe H, Mathenge SG, Juma FD (1993) Chromosomes of some woody
plants in Kenya. Acta Phytotaxonomica et Geobotanica 44, 53-58.
Ozias-Askins P (2006) Apomixis: Developmental Characteristics and Genetics. Critical Reviews
in Plant Sciences 25
Payne R (1991) New findings of the rare tree Syzygium paniculatum in the Wyong area, New
South Wales. Cunninghamia 2, 495-498.
Peakall R, Smouse P (2005) GENALEX 6: genetic analysis in Excel. Population genetic software
for teaching and research. Molecular Ecology Notes 6, 288-295.
Quarin C, Espinoza F, Martinez E, Pessino S, Bovo O (2001) A rise of ploidy level induces the
expression of apomixis in Paspalum notatum. Sexual Plant Reproduction 13, 243-249.
Ranker T (1994) Evolution of high genetic variability in the rare Hawaiian fern Adenophorus
and implications for conservation management. Biological Conservation 70
Richards AJ (2003) Apomixis in Flowering Plants: An Overview. Philosophical Transactions:
Biological Sciences 358
Roche D, Hanna W, Ozias-Akins P (2001) Is supernumerary chromatin involved in gametophytic
apomixis of polyploid plants? Sexual Plant Reproduction 13, 343-349.
Rossetto M (2001) Sourcing of SSR markers from related plant species. Plant genotyping: the
DNA fingerprinting of plants
Roy SK (1953) Embryology of Eugenia jambos L. Current Science 8, 249-250.
Roy SK (1961) Embryology of Eugenia fruticosa L. Proceedings of the National Academy of
Sciences, India Section B 31, 80-87.
Sahai R, Roy SK (1962) Polyembryony in Eugenia heyneana
Duthie. Science and Culture 28
Samadi S, Mavarez J, Pointier JP, Delay B, Jarne P (1999) Microsatellite and morphological
analysis of population structure in the parthenogenetic freshwater snail Melanoides
tuberculata: insights into the creation of clonal variability. Molecular Ecology 8, 1141-
Silvertown J (2008) The Evolutionary Maintenance of Sexual Reproduction: Evidence from the
Ecological Distribution of Asexual Reproduction in Clonal Plants. International Journal
of Plant Sciences 169
Singhal V, Gill B, Bir S (1985) Cytology of woody species. Proceedings of the Indian Academy
: Plant Sciences 94
Soltis PS & Soltis DE (2000) The role of genetic and genomic attributes in the success of
polyploids. Proceedings of the National Academy of Sciences 97(13), 7051-7057.
Stebbins GL, Jr. (1941) Apomixis in the Angiosperms. The Botanical Review 7, 507-542.
Van Der Hulst RGM, Mes THM, Falque M, et al.
(2003) Genetic structure of a population
sample of apomictic dandelions. Heredity 90, 326-355.
Van Dijk P, Van Damme J (2000) Apomixis technology and the paradox of sex. Trends in Plant
Science 5, 81-84.
Van Puyvelde K, Van Geert A, Triest L (2009) ATETRA, a new software program to analyse
tetraploid microsatellite data: comparison with tetra and tetrasat. Molecular Ecology
- 1.2. Threatened species: population size and diversity 5
- 2. Materials and Methods 12
- The Magenta Lilly Pilly (Syzygium paniculatum), endemic to a narrow strip along the New South Wales coast, is currently listed as vulnerable at both state and national levels. At present management of the species focuses on minimizing currently known threats, such as weed invasion, while little is known about the reproductive biology of the species. S. paniculatum is the only recorded polyembryonic Australian species of Syzygium; polyembryony being the development of multiple (and often asexual) embryos in one seed. Nuclear microsatellite markers were used to investigate the genetic outcome of polyembryony on the reproductive and population biology of the species focusing particularly on the population located on The Entrance Peninsular. Low within-population diversity was found, with low heterozygosity levels and a low level diversity indices when compared to other rare or rainforest species. Multiple embryos from single seeds were found to be identical to the mother. Multiple embryos germinated and survived but one seedling was always significantly taller than all others in the seed but was not considered sexual. It was concluded that the rare S. paniculatum is an apomictic clonal species with extremely low genetic diversity.
- 1 Introduction
- 1.1. Syzygium paniculatum: a brief description
- 1.2 Threatened species: population size and diversity
- 1.3 Polyembryony, clonality and Syzygium paniculatum
- 1.4 Clonality, fitness and Syzygium paniculatum
- 1.5 Ploidy in Syzygium paniculatum
- 1.6 Applicability of nuclear microsatellite markers
- 1.7 Project outline
- 2.1 Obtaining and analyzing diversity data
- 2.1.3 Ploidy identification
- 2.2 Obtaining and analyzing germination and reproduction data
- 3.1. Ploidy
- 3.2. Genetic Diversity
- 3.3. Reproduction
- 3.4. Germination Trial
- 4 Discussion
- 4.1. Ploidy
- 4.3 Reproduction and Fitness
- 4.4 Conservation Implications