Journal of the Royal Society of Western Australia, 84:103-110, 2001
© Royal Society of Western Australia 2001
Variation in seed production and germination in 22 rare and threatened
Western Australian Verticordia (Myrtaceae)
, K Brown
, S Cunneen
& A Kelly
Threatened Flora Seed Centre, Department of Conservation and Land Management, Locked Bag 104,
Environmental Weeds Action Network, 108 Adelaide Terrace, East Perth WA 6000
CSIRO Centre for Mediterranean Agricultural Research, Floreat WA 6014
24 Carnarvon St, East Victoria Park WA 6100
This study investigates the reproductive potential of 22 rare and threatened Western Australian
taxa in the genus Verticordia (Myrtaceae) over a 5-year period. Considerable inter- and intra-specific
variation in both seed production and germinability was demonstrated for the majority of taxa. The
seed to flower ratio, or “seed set”, ranged from 0% to 68% with an overall mean of 21% in 82
accessions representing seed from 48 populations of the 22 taxa. Percentage germination ranged
from 7% to 100% with an average of 49% for 68 accessions. The precariously low annual
reproductive capacity of some of the more restricted and critically endangered taxa threatens their
survival and unexpected disturbance events may result in population decline or even localised
extinction. Mitigation measures such as the reintroduction of plant material into new sites and the
enhancement of existing populations through additional plantings may be warranted for many of
Western Australia’s rare and threatened Verticordia.
Keywords: Verticordia, seed production, germination
Verticordia (family Myrtaceae, sub-family
Leptospermoideae) consists of woody perennials largely
endemic to the South West Botanical Province of Western
Australia. This genus occupies a prominent place in
many shrub and heathland communities along with other
myrtaceaous genera such as Beaufortia, Calytrix, Agonis,
There has been a considerable increase in our taxonomic
understanding of the genus in the past decade, and a
revision in 1991 identified 3 subgenera, 24 sections, 97
species, and 62 subspecies and varieties (George 1991).
Material was later re-examined and 100 species and 43
infraspecific taxa were described for the genus (George &
George 1994). Many taxa in the genus are considered of
high conservation value. There were 76 taxa of Verticordia
on the Western Australian Department of Conservation
and Land Management’s Declared Rare and Priority
Flora List (Atkins 1998). Seventeen of these were declared
rare (WA Government Gazette 1999), and five of these
were ranked under IUCN criteria as Critically
Endangered by CALM’s Western Australian Threatened
Species and Communities Unit (Anon 1998b). Many
populations are at risk of local extinction in the near future
due to a range of threatening processes. These include
disease, weed invasion, salinity, small population sizes,
habitat fragmentation and/or continued land clearing.
The genus possesses many species with great potential
for ornamental horticulture (George 1990). The often
prominently displayed feathery flowers are borne singly
but appear as heads or spikes and are generally brightly
coloured, ranging from yellow to red to purple. The
flowers are long lasting, and can be picked for the cut
flower market. Between 1996 and 1998 over 1 million
flowering stems of 23 taxa of Verticordia were bush picked
from private, crown lands and cultivated stands for the
cut flower industry (L Rohl, CALM, unpublished data).
Over-picking of flowers from the wild has been
impacting on wild populations of a number of species
(e.g. V. eryocephala) with detrimental effects (McEvoy &
Although there have been a number of studies on
pollen germination and tissue culture propagation in
has been limited research conducted on the reproductive
biology of particular species in the genus. Seed appears
to be of the dependent embryo type (small embryos
relative to the endosperm) according to classification by
Atwater & Vivrette (1987). Tyagi et al. (1991) reported
that in Verticordia only a single seed is set despite ovule
numbers of up to 13 in some taxa. Our experience has
shown that on rare occasions 2 and 3 seed per flower
may form in taxa with multiple ovules (A Cochrane,
personal observations). Houston et al. (1993) report seed
set (based on seed to flower ratio) for V. nitens averaging
27% with a range from 0.6% to 54% and 36% for V. aurea.
Tyagi et al. (1991) also reported low seed set for a range
of taxa in the genus.
Holm (1988) considered that pollinators for this genus
were likely to be unspecialised insects, but postulated
Journal of the Royal Society of Western Australia, 84(3), September 2001
that V. grandis may be bird pollinated due to the flower
structure. Houston et al. (1993) reported apparent
pollinator mutualism in V. nitens and possibly in V. aurea.
Profuse flowering in some species of Verticordia pointed
towards intense competition for pollinators (Holm 1988),
and given the great range of floral morphology, scent,
colour and flowering times, it is possible that many
pollinators will be found to be highly specific.
contain a single seed and are shed annually. They are
never discharged but the entire flower dries and breaks
off below the receptacle. It is thought that dormancy in
some members of the Myrtaceae (for example in
Chamelaucium, Verticorida and Darwinia) is controlled by
the seed coat inside the fruit that breaks down in time
due to weathering and soil disturbance (Beardsell et al.
1993a). Verticordia plants for the nursery industry have
traditionally been propagated vegetatively due to
inadequate knowledge of seed collection and germination
techniques (Watkins & Shepherd 1984). Relatively few
studies have been undertaken on seed germination in this
genus, although Ashby (1961) reported some success with
V. picta, V. chrysantha and V. brownii. Over the past 5
years considerable developments have been made in the
techniques required for successful germination of a range
of Western Australian species (see review by Bell et al.
1993; Cochrane & Kelly 1996). Smoke responsiveness has
been demonstrated in the genus (Dixon et al. 1995) and
after-ripening requirements can be overcome with the
addition of the growth hormone gibberellic acid
(Cochrane et al. unpublished observations).
Over the past 5 years, seeds of a large range of
threatened taxa in the genus Verticordia have been
collected for conservation in Western Australia’s ex situ
program. The aim of this program is to conserve the
genetic diversity of threatened taxa under low moisture
and low temperature conditions for long periods of time
(> 50 years) until material is required for recovery
purposes (Cochrane & Coates 1994).
This present study assessed the reproductive potential
of a range of rare and threatened taxa in the genus
Verticordia through an analysis of seed set and
germination data collected over a 5 year period. These
data are useful as a basis for recommendations for
conservation and management of the populations. Sound
knowledge of germination mechanisms will enable
adequate monitoring of seed viability in storage and
enhances the opportunity to provide whole plants for
All Verticordia seeds used in this study were collected
from wild populations between January 1994 and
December 1998. Site names have been abbreviated due to
the confidentiality of locational information for
conservation flora. Seed stocks are held ex situ at the
Department of Conservation and Land Management’s
Threatened Flora Seed Centre, a seed-based genebank for
the conservation of genetic material from rare and
threatened taxa. Seeds were tested for germinability
freshly collected and, in many cases, after moisture
content reduction and storage at –20
C for periods of up
Seed set was assessed by sectioning 3 replicate
samples of 100 old flowers (fruits) through the
hypanthium to establish the presence or absence of a
healthy seed (swollen, moist, white embryo). Calculation
of seed set by the cut test method was based on the
proportion of seeds to old flowers rather than the seed to
ovule ratio. Seed to flower ratio was considered to be a
more useful indicator of reproductive potential than seed
to ovule ratio as rarely did more than 1 ovule per flower
set. “Seed set” was therefore defined as the number of
intact and healthy seeds for a given number of flowers.
Some predation of developing ovules was observed
during cut tests, although the level of predation was not
Seed germination trials were conducted under
laboratory conditions. Seed sample sizes were dependent
on the number of old flowers (fruits) collected, as well as
the number of seeds obtained by the cut test (seed set)
and ranged from 5 to over 1000 seeds (x– 61). To aid
germination, seeds were completely excised from the old
flowers with a scalpel under a dissecting microscope.
Prior to seed coat removal, flowers were soaked in
distilled water for a minimum of 2 hours to soften the
seed coat. Seeds were germinated in 90 mm glass Petrie
dishes on a 0.75% (w/v) agar solution in temperature
and light controlled incubation cabinets, using a 12-hour
photoperiod. Cabinets were set at a constant 15
C. A 2%
solution to inhibit fungal growth. Seeds were not surface
sterilised prior to incubation. Petrie dishes were checked
twice weekly and germination was determined by radicle
Previous research (A Cochrane, unpublished data) has
indicated that many species of Verticordia are smoke
responsive, and the seeds of those species requiring
aqueous smoke treatment for optimum germination were
soaked for 24 hours in a smoke solution obtained from
Perth’s Kings Park and Botanic Gardens and produced
according to Dixon et al. (1995). After soaking, seeds were
rinsed with distilled water prior to incubation. Growth
promoters have been found to be necessary to cue
germination in fresh seed of Verticordia, and the growth
hormone gibberellic acid (GA
) was added to the agar
or 10 mg L
The number of ovules per flower in the genus
Verticordia varies between species and ranges from 1 to 8
for the 22 taxa studied (Table 1). Rarely did more than 1
seed per flower reach maturity and as such the seed to
ovule ratio ranged from 0.125 to 1 (Table 1). There was a
broad range of inter- and intra-specific variation in seed
set and germination within the genus (Table 2). The mean
seed set for all collections was 21% (± 1.85) with a range
from 0 to 68%. Population seed set was never greater
than 68%, although seed set for a single individual within
a taxon did reach 90% in V. staminosa subsp cylindraceae
var erecta (see Fig 1). The mean percentage germination
was 49% (± 2.96), range 7-100%, for 68 accessions.
Fourteen taxa are known to be obligate seeders; 4 are
respouters, with 3 considered to have the potential to
both resprout and seed. Information on the reproductive
A Cochrane et al.: Seed production and germination in Verticordia
Fire response, number of ovules per flower, seed to ovule ratio, and percentage seed set and germination for 22 rare and threatened taxa
in the genus Verticordia.
% seed set
% seed germination
(± se, range)
V. endlicheriana Schauer var angustifolia AS George (sect Chrysoma)
79.4 ± 11.6 (57-96)
84.5 ± 3.5 (81-88)
76.7 ± 12.0 (60-100)
V. albida AS George (sect Pennuligera)
43.8 ± 5.02 (20-68)
30.4 ± 8.6 (10-87)
37.5 ± 5.9 (26-54)
34.7 ± 11.9 (11-48)
63.0 ± 6.0 (57-69)
8.6 ± 2.1 (0-19)
43.0 ± 8.6 (7-86)
V. densiflora Lindley var caespitosa Turcz (sect Corymbiformis)
56.7 ± 21.5 (15-87)
53.8 ± 6.5 (25-86)
47.3 ± 9.940 (29-63)
V. plumosa var ananeotes (Desf) Druce (sect Verticordia)
26.0 ± 7.0 (11-42)
46.0 ± 8.7 (25-75)
strategy for one taxon (V. dasystylis subsp oestopoia) was
Seed production and germinability for the 82
collections of Verticordia exhibited year to year (Fig 2A-
C), population to population (Fig 3), plant to plant (Fig 4)
and seasonal (Figs 5, 6) variation. The mean seed set and
germination for each taxon was calculated (Table 1);
however, the wide spatial and temporal variation make
these figures somewhat misleading. Nonetheless these
data provide a reference point to illustrate the variation
between taxa. There were no apparent trends evident for
seed set and germination or for reproductive strategy
within the different subgenera or sections (Table 1). There
was also no correlation between condition of the
population as determined by location (road verge versus
reserve or bushland) and health of population (degraded
or healthy) and reproductive potential (Table 2). In
addition, there was no correlation between intra-specific
population size and levels of seed set (Table 2).
It would appear from this study that rare and
threatened taxa within the genus Verticordia exhibit
excess flower production and a corresponding low seed
to flower ratio, in keeping with previous results for
common taxa in the genus (Tyagi et al. 1991; Houston et
beneficial for plants to produce excess flowers and only
allocate energy for seed production to those flowers with
a minimal level of surviving embryos. Flower production
is less energy draining than the necessary proteins and
lipids required for seed production. Surplus flower
production may enable plants to exploit favourable
conditions such as increased resources or pollinator
activity that occur unpredictably. Seasonal and yearly
changes in seed set may be due to changes in flower
density during the flowering period that may in turn
affect pollinator assemblages, abundance and behaviour.
It may also give a plant the opportunity to increase its
Differences in seed set between individual plants for V.
Site location, condition (D=degraded H=healthy RV=road verge P=reserve, park, remnant bush), population size, time of collection,
number of plants sampled, percentage seed set, and germination for 22 rare and threatened taxa in the genus Verticordia.
V. dasystylis ssp oestopoia
V. densiflora var pedunculata
V. fimbrilepis ssp australis
V. fimbrilepis ssp fimbrilepis
V. plumosa var pleiobotrya
offspring vigour through selective abortion. Surplus
flower production may also provide an ovary reserve in
case of mortality of flowers, or may provide a buffer
during adverse weather conditions or during competition
that may reduce pollen flow (Lee & Bazzaz 1982). The
interacting factors of pollination failure, resource
deficiency, predation and genetic defects causing
developmental failure may cause pre-dispersal seed and
ovule mortality (Fenner 1985; Wallace & O’Dowd 1989).
A range of biological constraints can also lead to
reductions in seed production (Owens 1995). These
include (1) periodic or inadequate floral initiation, (2)
asynchronous development and flowering, (3) floral
abortion, (4) ovule abortion, (5) embryo abortion, and (6)
failure of seeds and fruits to mature and our inability to
determine maturity. Constraints and their importance
will vary among species, sites and years and occur
throughout all stages of development. Stephenson (1981)
reports that whole fruit abortion is common among
outcrossing perennials and that a low fruit to flower ratio
(i.e. seed set) is observed in many species. Similarly,
Wiens et al. (1987) noted that reproductive success as
measured by seed to ovule ratio in outcrossing plants
was considerably lower (22%) than in inbreeding plants
V. plumosa var vassensis
V. plumosa var vassensis
V. spicata ssp squamosa
V. spicata ssp squamosa
V. staminosa ssp staminosa
A number of researchers have reported that self-
incompatibility in the family Myrtaceae contributed to
low seed set (e.g. Briggs 1962, 1964; Prakash 1969; Rye
1980; Barlow & Forrestor 1984; Griffin et al. 1987;
Beardsell et al. 1993b, Sedgley & Granger 1996). Despite
widespread self-incompatibility within the family, recent
research into the mating system of the rare Verticordia
rates are very high (J Samson, CALM, personal
communication). Rye (1980) and Tyagi et al. (1991) also
noted self-compatibility in a number of species in the
genus. Tyagi et al. (1991) considered that low seed set in
field populations of some species of Verticordia was
determined by factors such as efficiency of pollen
transfer, genetic diversity within populations and
physiological constraints rather than loss of pollen
fertility. They were unable to demonstrate that seed
produced by inbreeding was viable, although recent
studies (J Samson, CALM, personal communication) have
now shown that 78% of seeds of V. fimbrilepis subsp
fimbrilepis produced as a result of self-fertilisation were
viable. McEvoy (1995) also reported that seed viability
was not reduced by selfing in the common V. eriocephala.
These data suggest that inbreeding depression is unlikely
to be a cause of low seed set in this genus.
Environmental stress and demographic structure of
the population have also been known to contribute to
low seed production (Jordano 1992) and can account for
some of the wide variation in seed set noted between
individuals, populations and years. However, this study
demonstrated that smaller populations of Verticordia
located on degraded road verges did not necessarily
exhibit lower seed set than larger populations occurring
in remnant bush or in reserves or bushland (Table 2).
There appeared to be no difference in seed set between
healthy and degraded sites indicating that environmental
Seasonal differences in seed set and germination for V
Difference in seed set and germination between years
for V. carinata Site NIT, V. spicata subsp squamosa Site C, and V.
staminosa subsp staminosa Site MH.
Differences in seed set and germination between sites
of collection for V albida 1997 collections.
Differences in seed set and germination between
individual plants for V fimbrilepis subsp fimbrilepis Site A, 1998.
stress at a gross level may not be impacting on the
reproductive capacity of those taxa studied. It is possible
that reproductive success may be a function of plant
maturity, although at the time of this study the age
structure of each population was unknown. Another
significant factor affecting seed set is the presence and
health of pollinators, which may in part be related to the
health of the site and opportunities for supporting
Various studies have reported a relationship between
seed set and life history (Wiens 1984; Ehrlen 1991;
Time During Season
Richmond & Chinnock 1994; Meney et al. 1997). It
appears that reproductive success in some genera may in
part be determined by the internal allocation of resources
as dictated by the plants’ regenerative mode (Carpenter
& Recher 1979; Hansen et al. 1991; Ladd & Wooller 1997;
Meney et al. 1997). In this study, there was no evidence to
suggest that seed set in Verticordia varied between
obligate seeding and resprouting species. The wide intra-
specific variation in seed set makes it difficult to
determine relationship between levels of seed set and
regenerative mode and further investigations are
The breakdown of seed dormancy in Verticordia
appears to require not only the removal of the seed coat,
which acts as a barrier to water uptake, but also the
addition of growth hormones to overcome an after-
ripening requirement. It is possible that the hypanthium
and perianth might help protect the seed from
weathering, thus maintaining dormancy. This has been
observed in Thrytomene calycina (Beardsell et al. 1993c).
Recent work on other genera in the family Myrtaceae
(Darwinia and Chamelaucium) has also established the
need for hypanthium removal and the application of
growth hormones to aid germination (A Cochrane,
unpublished data). Despite considerable inroads into
understanding the germination requirements of the
genus Verticordia, we are not achieving maximum
germination in most cases from what appears to be
healthy, mature seed. It is obvious that there is still a
great deal more to be understood about the germination
requirements of particular taxa. Further research is
required to determine whether this incomplete
germination is due to after-ripening requirements, to
maturity of seed, to inadequate dormancy breaking
treatments and/or to genetically related defects.
Given that sampling of material for ex situ
conservation occurs on a random basis to enable the
range of genetic characters and reproductive potential to
be represented, it is not unexpected to find such intra-
and inter-specific variability in germination.
Germinability of seed will be affected by the
environmental conditions under which the seed
developed. It will also be affected by the timing of
collection (see Figs 5, 6) and post-harvest conditions prior
to germination. Germination differences among
individuals have important fitness consequences and
germination differences between populations could well
be reflecting inbreeding depression (Menges 1991). There
have been suggestions that there is a relationship
between population size and germinability in species,
with larger populations exhibiting higher germination
than smaller populations (Menges 1991). Our data on
Verticordia indicate no such relationship (Figure 3).
This study has demonstrated considerable variation in
seed set and germinability in a range of taxa in the genus
Verticordia. The precariously low annual reproductive
success of some of the taxa studied indicates a need for
considerable monitoring of the health and biology of
these taxa over the long-term. Continued disturbance
such as clearing, disease or fire may result in population
decline or even localised extinction of some of the more
critically endangered taxa. Changes in population size,
degree of isolation and fitness are warning signs that
populations may be vulnerable (Ellstrand & Elam 1993)
and should cause concern to conservation managers.
Regeneration plays a major role in the composition and
floral diversity of plant communities. The regeneration
potential of a population depends on the proportions of
germinable seed successfully growing, maturing and
attaining reproductive status, as well as the reproductive
potential determined by seed set and germination. For
most taxa the effective size of the soil seedbank remains
unknown. Seedling survival through summer months is
also unknown. Further stresses due to post-dispersal
factors (seed predation and seedling mortality) in species
that already exhibit low pre-dispersal reproductive
success will critically restrict the ability of the taxon to
reconstitute populations from seed, or to maintain levels
of plants in the face of disturbance and senescence.
Seed set in plants occurring in intact vegetation is no
greater than in plants located in degraded sites, but the
ability of populations to attract and support pollinators
may be dependent on the condition of the site and its
associated vegetation. Despite the ability of plants to self-
pollinate, the impact of reducing population size and
health on the survival of plants and their reproductive
capacity is unknown. Mitigation measures such as the
reintroduction of material into new sites and the
enhancement of existing populations may be warranted
for many of Western Australia’s rare and threatened
Verticordia. The observed year to year and site to site
variations in seed set and germination suggest a cycle of
alternating high and low reproductive activity which
may be affected by seasonal influences on fruit survival
and maturation. Reproductive success in terms of seed to
flower ratio may be a function of size, age, condition and
genetic make-up of the plants, as well as seasonal factors
and pollinator activity. Research into spatial and
temporal factors (e.g. climatic data, nutrient status of sites,
pollinator visitors, pollen loads, age structure and genetic
variability within each population) affecting seed
production and germinability in Verticordia is considered
necessary to ensure local extinction of small isolated
populations does not occur.
The authors wish to acknowledge the assistance of E
George for identification of specimens and for the provision of
information on ovule numbers and regenerative mode for the taxa utilised
in this present study. Between 1992 and 1998 the Threatened Flora Seed
Centre was funded by the Threatened Species and Communities Section,
Biodiversity Group, Environment Australia.
Ashby E 1961 Notes on the germination of the seeds of
Australian native flora. Australian Plants June:17-18.
Atkins, K 1998 Declared Rare and Priority Flora List. Department
of Conservation and Land Management, Perth.
Atwater BR & Vivrette NJ 1987 Natural protective blocks in the
germination of seeds. Acta Horticulturae 202:57-68.
Barlow BA & Forrester J 1984 Pollen tube interactions in
University of Melbourne, Melbourne, 154-160.
Beardsell DV, Knox RB & Williams EG 1993a Germination of
seeds from the fruits of Thryptomene calycina (Myrtaceae).
Australian Journal of Botany 41:263-73.
Beardsell DV, Knox RB & Williams EG 1993b Breeding system
and reproductive success of Thryptomene calycina (Myrtaceae).
Australian Journal of Botany 41:333-353.
Beardsell DV, Knox RB & Williams EG 1993c Fruit and seed
structure of Thryptomene calycina (Myrtaceae). Australian
Journal of Botany 41:183-193.
Bell DT, Plummer JA & Taylor SK 1993 Seed germination
ecology in southwestern Western Australia. The Botanical
Briggs BG 1962 The New South Wales species of Darwinia.
Contributions to the NSW National Herbarium 3:129-150.
Briggs BG 1964 The control of interspecific hybridisation in
Darwinia. Evolution 18:292-303.
Carpenter FL & Recher HF 1979 Pollination, reproduction, and
fire. The American Naturalist 116:871-879.
Cochrane A & Coates DJ 1994 Western Australia’s Threatened
Flora Seed Centre and its role in the conservation of genetic
diversity. Danthonia 3:4-7.
Cochrane A & Kelly A 1996 Germination and storage of seed
from rare and threatened plants of the SouthWest of Western
Australia. In: Proceedings of the Second Workshop on Native
Seed Biology for Revegetation (eds SM Bellairs & JM
Osborne). Australian Centre for Minesite Rehabilitation
Research, Brisbane, 101-106.
Dixon KW, Roche S & Pate JS 1995 The promotive effect of
smoke derived from burnt native vegetation on seed
germination of Western Australian plants. Oecologia 101:185-
reserve-ovary model. The American Naturalist 138:918-933.
Ellstrand NC & Elam DR 1993 Population genetic consequences
of small population size: implications for plant conservation.
Annual Review of Ecology and Systematics 24:217-242.
Fenner M 1985 Seed Ecology. Chapman & Hall, London.
George AS 1991 New taxa, combinations and typifications in
Verticordia (Myrtaceae: Chamelaucieae). Nuytsia 7:231-294.
George EA & George AS 1994 New taxa of Verticordia
(Myrtaceae: Chamelaucieae) from Western Australia. Nuytsia
George EA 1990 Unravelling the mysteries of Verticordia.
Griffin AR, Moran GF & Fripp YJ 1987 Preferential outcrossing
in Eucalyptus regnans F. Muell. Australian Journal of Botany
Hansen A, Pate J & Hansen AP 1991 Growth and reproductive
performance of a seeder and a reprouter species of Bossiaea as
a function of plant age after fire. Annals of Botany 67:497-509.
Holm E 1988 On Pollination and Pollinators in Western
Australia. Eigil Holm, Denmark.
Houston TF, Lamont BB, Radford S & Errington SG 1993
Apparent mutualism between Verticordia nitens and V. aurea
(Myrtaceae) and their oil-ingesting bee pollinators
(Hymenoptera: Colletidae). Australian Journal of Botany
Ladd PG & Wooller SJ 1997 Explaining variation in pollination
and seed set in an andromonoecious genus of the Proteaceae.
Acta Horticulturae 437:115-119.
Jordano P 1992 Fruits and frugivory. In: Seeds. The Ecology of
Regeneration in Plant Communities (ed M Fenner). Cab
International, Oxon, 105-156.
Lee TD & Bazzaz FA 1982 Regulation of fruit and seed
production in an annual legume, Cassia fasciculata. Ecology
McComb JA, Newell C & Arthur N 1986 Micropropagation of
Verticordia grandis and other Verticordia species. Plant
McEvoy S & True D 1995 The Conservation and Cultivation of
Verticordia eriocephala and related species. Department of
Conservation and Land Management, Perth.
Meney KA, Dixon KD & Pate JS 1997 Reproductive potential of
obligate seeder and resprouter herbaceous perennial
monocots (Restionaceae, Anarthriaceae, Ecdeiocoleaceae)
from southwestern Western Australia. Australian Journal of
Menges ES 1991 Seed germination percentage increase with
population size in a fragmented prairie species. Conservation
Owens JN 1995 Constraints to seed production: temperate and
tropical forest trees. Tree Physiology 15:477-484.
Prakash N 1969 Reproductive development in two species of
Darwinia Rudge (Myrtaceae). Australian Journal of Botany
Richmond GS & Chinnock RJ 1994 Seed germination of the
Australian desert shrub Eremophila (Myoporaceae). The
Botanical Review 60:483-503.
Rye BL 1980 Chromosome numbers, reproductive biology and
evolution in the Myrtaceae. PhD Thesis, University of
Western Australia, Perth.
Sedgely M & Granger L 1996 Embryology of Eucalyptus
spathulata and E. platypus (Myrtaceae) following selfing,
crossing and reciprocal interspecific pollination. Australian
Journal of Botany 44:661-671.
Stephenson AG 1981 Flower and fruit abortion: proximate causes
and ultimate functions. Annual Review of Ecology and
Turnbull JW & Doran C 1987 Seed development and
germination in the Myrtaceae. In: Germination of Australian
Native Plant Seed (ed P J Langkamp). Inkata Press,
Tyagi AP, McComb J & Considine J 1991 Cytogenetic and
pollination studies in the genus Verticordia DC. Australian
Journal of Botany 39:261-272.
Tyagi A, Considine J & McComb J 1992 Germination of
Australian Journal of Botany 40:151-5.
Wallace DD & O’Dowd DJ 1989 The effect of nutrients and
inflorescence damage by insects on fruit-set by Banksia
Watkins PA & Shepherd RW 1984 Studies in propagation of
Wiens D 1984 Ovule survivorship, brood size, life history,
breeding systems, and reproductive success in plants.
Wiens D, Calvin CL, Wilson CA, Davern CI, Frank D & Seavey
SR 1987 Reproductive success, spontaneous embryo abortion,