Regeneration mechanisms in Swamp Paperbark (Melaleuca ericifolia Sm.) and their implications for wetland rehabilitation

3.4.2 Implications for revegetation 
 
While the benefits of extensive physiologically integrated ramets and expansive clonality 
are many, there are several distinct disadvantages in the longer-term. Reduction or loss of 
genetic diversity or sexual reproduction in populations of clonal plants is well 
documented (Widen et al. 1994; Jelinski and Cheliak 1992; Kreher et al. 2000; Eckert 

 
78
2001). The evolution of the clonal growth form is believed to be a consequence of limited 
opportunity for sexual recruitment (Mogori et al. 2003). Habitat change, including 
alteration to water regimes and salinity levels can move the environment beyond the 
ecological tolerances of the species forcing clonal plants to rely on asexual reproduction 
for continued survival (Honnay and Bossuyt 2005; Wesche et al. 2005) and on sexual 
reproduction for colonisation of new sites (Rea and Ganf 1994). The existing conditions 
of the wetlands of the Gippsland Lakes have, through permanent opening of the 
intermittent opening to the sea and human induced alteration to the water regime through 
the installation of levee banks, severely impacted the ecological conditions of the 
wetlands, increasing salinity levels and varying from historical wetting and drying 
patterns (Bird 1962; Greyson 2003).  
 
In the short-term, altered water regimes and salinity level changes seems to have 
benefited  M. ericifolia in the wetlands of the Gippsland Lakes, allowing expansion of 
existing genets and recruitment of new genets (Raulings et al. 2006). However, longer-
term effects of these altered regimes are becoming apparent with decline of existing 
genets and reduced sexual and asexual reproductive capacity (Raulings et al. 2006). The 
requirements for sexual recruitment need to be understood to allow rehabilitation of these 
now highly managed and degraded wetlands. Some studies investigating the hydrology, 
microtopography and planting technique of M. ericifolia in the Gippsland Lakes wetlands 
have been carried out (Raulings et al. 2006) although the planting techniques trialled 
have been largely unsuccessful. Specific germination and recruitment condition 
information is needed to allow planned intervention and site manipulation to achieve 
maximum success. Knowledge of expansion rates and interactions of individual genets 
has major implications for spatial and temporal appropriate planting techniques. Present 

 
79
planting techniques that focus on dense plantings of individual plants (genets), usually on 
2 x 2 to 3 x 3m centres, primarily to achieve rapid canopy cover, may be particularly 
inappropriate for the long-term survival of M. ericifolia in brackish-water wetlands. If 
hand-planting of seedlings is to be used, it may be more appropriate to plant far fewer 
plants under ideal establishment conditions and allow the rapid expansion rates of the 
species to come into play. Counts of numbers of genets per hectare, based on the work 
carried out in this study, would indicate that there are as few as three genets per hectare 
and that on average there are 7 per hectare. Present planting densities utilising 3 m 
centres between planted seedlings utilises approximately 1,110 plants to revegetate one 
hectare. Fewer, well-placed plants coupled with time represents both a significant cost 
saving and probably a superior ecological outcome.   
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Chapter 4 
Comparison of two contrasting life forms of Melaleuca in 
south-eastern Australia. 
 
 
Abstract 
 
The viability of seed and the trade-off between sexual and asexual reproduction can 
vary widely between plants of the same genus occupying different aspects of an 
environmental gradient in wetlands. Viability can vary among plants of the same 
species across the range of that species in response to different environmental 
parameters. A comparative study was carried out between two co-occurring members 
of the genus Melaleuca (M. ericifolia and M. parvistaminea) with contrasting growth 
type (rootstock regenerator vs. seed-only regenerator). Additional viability testing was 
carried out on 23 populations across the southern part of the range of M. ericifolia in 
Victoria and Tasmania in order to test the hypothesis that a range of factors including 
population size, distance to nearest population and alterations to habitat may play a 
role in viability over and above inherent viability. Seed weight and viability of the 
rootstock regenerating M. ericifolia was consistently lower (<30 µg, < 38 %) than the 
seed-only regenerating M. parvistaminea (>30 µg, > 70 %). Across the range of M. 
ericifolia, viability varied from 0 – 38%, with large, relatively undisturbed 
populations having higher percentage germination. Populations of M. ericifolia 
affected by isolation, limited gene flow, disturbance and secondary salination had 
markedly reduced viability regardless of area covered by the population.  

 
81
 
4.1 Introduction 
 
Melaleuca ericifolia (Swamp Paperbark) and Melaleuca parvistaminea are two 
members of the large serotinous genus Melaleuca (Myrtaceae) that occurs primarily in 
Australia (Spencer 1996). There is considerable overlap in the distribution of the two 
species and until recently both were considered varieties of M. ericifolia (Albrecht 
1987). Differences in regeneration and reproductive ability are two of the main 
distinguishing features between these taxa: Melaleuca parvistaminea is a seed-only 
regenerator while M. ericifolia forms extensive colonies of vegetatively reproduced 
stems through root suckering (Albrecht 1987).  Both species occur in wetlands, but 
are usually spatially and ecologically segregated:  M. parvistaminea occurs in areas in 
south-east Australia with only intermittent, short-term inundation in primarily 
freshwater swamps (Figure 4.1), whereas M. ericifolia occurs in wetter habitats that 
may be flooded for many months, ranging from fresh to brackish swamps (Figure 
4.2). Accordingly, it is not uncommon to find both species growing in the same 
wetland but spatially separated along an elevational gradient. The co-occurrence of 
these two species in many wetlands throughout south-eastern Australia provides an 
ideal opportunity to investigate the hypothesis that there is a trade-off between sexual 
and asexual reproduction and that this may be expressed as reduced seed viability in 
the clonal species. 
 

 
82
 
Figure 4.1 Distribution of Melaleuca ericifolia in Australia (Australian Virtual 
Herbarium)  
 
Figure 4.2 Distribution of Melaleuca parvistaminea in Australia (Australian Virtual 
Herbarium). 
 

 
83
It is commonly assumed that there is a trade-off between sexual and asexual 
reproduction in plants due to the high costs (energetic and nutrient) of sexual 
reproduction (Lovett Doust 1989; Reekie 1999; Eckert 2001). There are, however, 
conflicting findings regarding resource allocation to sexual reproduction in clonal 
plants, with some authors finding reduced commitment to sexual reproduction (in 
Mimulus: Sutherland and Vickery 1988) and others no reduction in sexual 
reproductive outputs (Rea and Ganf 1994).  
 
While there are some seed viability studies of several populations of M. ericifolia in 
South Gippsland (Ladiges et al. 1981; Robinson et al. 2006) there are no 
comprehensive studies of this species over the majority of its range from northern 
New South Wales through to western Victoria and Tasmania. Such studies are 
essential if generalisations about trade-offs between sexual and asexual reproduction 
are to be made.  No references to the seed viability of any populations of M. 
parvistaminea were found.  
 
The aims of this component of the thesis are to: 
 
1.  Compare the seed viability in the clonal M. ericifolia to the non-clonal M. 
parvistaminea, 
2.  Determine the importance of site isolation and population size on the seed 
viability of M. ericifolia and M. parvistaminea, and  
3.  Determine the importance of site degradation on the seed viability of M. 
ericifolia and M. parvistaminea.  
 
 

 
84
4.2 Methods 
 
 
4.2.1 Seed collection 
 
Seed capsules were collected between April 2004 and November 2005 from 12 – 25 
adult trees (different clones) scattered throughout various populations of M. ericifolia 
across its range in Victoria and Tasmania, including several Bass Strait Islands (Table 
4.1). Seed was collected from seven populations of M. parvistaminea in South 
Gippsland that were known to be natural occurrences (Table 4.2) were the range of 
this species overlaps with M. ericifolia.  The characteristic phalanx (dense advancing 
front) growth form and dome shape was used to determine individual plants of M. 
ericifolia; seed was collected from widely separated plants that were clearly not 
related clonally. Approximately 12 individual plants were sampled scattered 
throughout each population of M. parvistaminea. Approximately 500 seed capsules 
were collected from each putative individual of each species. Seed loses viability after 
approximately 1 year, the short period of storage, mostly over the cooler winter 
months was not viewed as materially altering overall viability.  
 
Information was collected for each of the populations of M. ericifolia, including 
locality (latitude and longitude), adjacent disturbance and an estimate of area covered 
by the species. Area of population for M. ericifolia was visually assessed and classed 
into categories shown in Table 4.3. Site salinisation and distance to nearest population 
was calculated using site records from the Waterwatch and the National Herbarium 
Melbourne  

 
85
Table 4.1 Populations and location of seed collection sites for M. ericifolia across 
southern Australia. Localities marked with an * are potentially introduced 
populations.  
Population location 
Latitude 
Longitude 
Population 
size (ha) 
Little River (Victoria) 
37
o
 56' 17"  144
o
 28' 33"  >1 ha 
Cape Nelson (Victoria) 
38
o
 24' 39 '
141
o
 23' 41 
1-5 ha 
*Mount Wellington (Tasmania) 
42
o
 53' 09"  147
o
 15' 32"  >1 ha 
Heritage Golf Course (Victoria) 
37
o
 40' 55"  145
o
 21' 05"  >1 ha 
Brushy Creek (Victoria) 
37
o
 46' 00"  145
o
 17' 00"  1-5 ha 
Yanakie Peninsula (Victoria) 
38
o
 48' 52"  146
o
 14' 09"  1 ha 
Dowd Morass (Victoria) 
37
o
 28' 58"  149
o
 40' 38"  > 1000 ha 
*Kemps Marsh (Lake Sorell) (Tasmania) 
42
o
 47' 31"  147
o
 33' 26"  1-5 ha 
Spadoni Reserve (Victoria) 
37
o
 28' 58"  149
o
 40' 38"  1-5 ha 
Boggy Creek (Victoria) 
37
o
 42' 00"  145
o
 33' 12"  1-5 ha 
Cades Road (Victoria) 
37
o
 33' 38"  149
o
 07' 13"  10-50 ha 
Apsley Marshes (Tasmania) 
41
o
 59' 59"  148
o
 15' 53"  10-50 ha 
Lightwood Creek, Point Nepean Nation Park (Victoria)
38
o
 25' 5" 
144
o
 56' 19"  5-10 ha 
Garfield (Victoria) 
38
o
 05' 06"  145
o
 36' 38"  5-10 ha 
Cann River (Victoria) 
37
o
 32' 35"  149
o
 07' 07"  10-50 ha 
Gypsy Point (Victoria) 
37
o
 28' 58"  149
o
 40' 38"  10-50 ha 
Yarram (Victoria) 
38
o
 37' 55"  146
o
 40' 05"  10-50 ha 
Cape Barren Island Population 1 (Tasmania) 
40
o
 24' 43"  148
o
 01' 42"  10-50 ha 
Cape Barren Island Population 2 (Tasmania) 
40
o
26' 28" 
148
o
 08' 14"  10-50 ha 
Flinders Island Population 1 (Tasmania) 
39
o
 57' 10"  148
o
 10' 49"  50-100 ha 
Flinders Island Population 2 (Tasmania) 
40
o
 13' 17"  148
o
 17' 40"  50-100 ha 
Gladstone (Tasmania) 
40
o
 55' 01"  147
o
 54' 30"  50-100 ha 
Narawntapu National Park (Tasmania) 
41
o
 10' 72"  146
o
 36' 70"  100-1000 ha
 
Table 4.2 Populations and locations of seed collection sites for M. parvistaminea in 
South Gippsland, Victoria 
Population location 
Latitude 
Longitude
Population  size (ha) 
Rosedale 38
o
10'4" 146
o
57'24"
1-5 ha 
Maffra 37
o
56'2" 146
o
48'41"
1-5 ha 
Fernbank 37
o
52'31"
147
o
16'16"
1-5 ha 
Providence Ponds 
37
o
57'46"
147
o
19
'
18
"
 5-10 
ha 
Heyfield 37
o
56'33"
146
o
43'34"
5-10 ha 
Briagalong 37
o
49'9" 146
o
58'39"
5-10 ha 
Sale Common  
38
o 
8'2" 147
o
5'0" 5-10 
ha 
 
 

 
86
Table 4.3 Classification of population size of M. ericifolia across southern Australia.  
Classification Population 
size 
Very small 
Under 1 ha 
Small 
1-5 ha 
Medium 
5-10 ha 
Medium/Large 
10-50 ha 
Large 
50-100 ha 
Very Large 
100-1000 ha 
Expansive 
Greater than 1000 ha 
 
 
Capsules were stored in paper bags at 20
0
C for one week.  The bags were lightly 
shaken to release seed from the capsules and the contents sieved to remove the empty 
capsules and other detritus.  Seed was placed in clean paper bags for a further three 
days to remove excess moisture, then transferred to sealed glass containers and stored 
at 20
0
C in darkness until used, usually within 2 months of harvest.  Preliminary 
germination trials began in late April 2004, with the main trials carried out in July 
2004 and December 2005.  
 
4.2.2 Seed viability 
 
The number of seeds per gram was determined by taking five samples of known 
weight (0.003-0.007 g), which were counted by eye. The average number of seeds 
was calculated from the mean of the aggregated totals of the five samples. The weight 
of individual seeds was determined by dividing the number of seeds per sample by the 
weight of the sample (Association of Official Seed Analysts 1990). 

 
87
 
Viability testing followed procedures outlined by the Association of Official Seed 
Analysts (1990), except that the number of seeds per replicate run was increased from 
25 to 100 to improve statistical rigour (Robinson et al. 2006).  A small pilot study was 
carried out before the main trials to determine statistical power and the number of 
replicates needed to detect significant responses to environmental variables (Zar 
1999).  These trials indicated that four replicates, each using 100 seeds, yielded a 
power of >0.99.   
 
Seeds were surface sterilized by placing them in small sealed muslin bags and 
plunging the bags in 10 % w/v sodium hypochlorite solution for 20 seconds and then 
rinsing them three times in distilled water.  For each viability-trial replicate, 100 seeds 
were evenly spaced in a grid pattern on a disc of Whatmans #3 filter paper in a 9-cm 
diameter petri dish.  Each paper disc was wetted with 8 mL of distilled water and the 
dish sealed with Labfilm to reduce moisture loss.  A total of four replicate petri dishes 
(i.e., 400 seeds) were used for each viability test. Seeds were germinated and then 
incubated in growth cabinets with day-time temperatures of 20
o
C and night-time 
temperatures of 10
o
C.  A 12:12 hour light:dark cycle was used.  A bank of fluorescent 
tubes designed for plant growth provided light, emitting a PAR of 40 
μ
E m
-2 
s
-1
 at the 
level of the seeds.   All replicates were randomly shuffled daily within the cabinet to 
randomise spatial variability, and germination was measured after 7, 14 and 21 days.  
The trial was terminated at 21 days because no additional germination was recorded 
after Day 14. Germination was judged by the emergence of the base of the hypocotyl 
from the testa.   
 

 
88
 
4.3 Results 
 
4.3.1 Melaleuca ericifolia 
 
Viability of seed from various population of M. ericifolia varied widely, from 0 % for 
two isolated populations in western Victoria and one from southern Tasmania through 
32 % for populations from East Gippsland and 33–38 % for the majority of 
Tasmanian and Bass Strait Island populations (Table 4.4). Germination percentage 
was closely correlated with seed weight (R
2
 = 0.933, see Figure 4.3) with little 
deviation from a linear relationship.  
 
Populations with lower viability had a large percentage of unfilled seed. The three 
populations with zero germination were excluded from data analysis (rows 1-3 Table 
4.4). There was little relationship between viability and population size with three 
larger populations, Apsley Marshes, Cades Road and Dowd Morass (10-50 ha and > 
1000 ha, respectively) exhibiting a clear deviation from the mean (Figure 4.4). 
Populations under 5 ha also exhibited wide divergence from the mean (Table 4.4 and 
Figure 4.4). Features in common with all sites with very low viability were the degree 
of adjacent human disturbance, secondary salinisation, and isolation from the nearest 
population.   
 
 
 
 

 
89
 
Table 4.4 Population size, seed weight and viability of various populations of 
Melaleuca ericifolia in Victoria and Tasmania, Australia 
 
 
Population location 
 
Population 
size (ha) 
 
Number 
of 
seeds 
per mg 
Mean 
Seed 
weight 
(µg) 
 
Percent 
germination 
(%) 
 
Little River (Victoria) 
> 1 ha 
0 
0 
0 
Cape Nelson (Victoria) 
1-5 ha 
0 
0 
0 
Mount Wellington (Tasmania) 
> 1 ha 
0 
0 
0 
Heritage Golf Course (Victoria) 
> 1 ha 
61.5 
16 
0.3 
Brushy Creek (Victoria) 
1-5 ha 
66.4 
15 
0.5 
Yanakie Peninsula (Victoria) 
1 ha 
55.3  
18 
5.5 
Dowd Morass (Victoria) 
> 1000 ha 
55.6 
18 
6.0 
Kemps Marsh (Lake Sorell) 
(Tasmania) 1-5 
ha 
53.6 
19 
6.0 
Spadoni Reserve (Victoria) 
1-5 ha 
53.4 
19 
6.3 
Boggy Creek (Victoria) 
1-5 ha 
49.9 
20 
9.3 
Cades Road (Victoria) 
10-50 ha 
49.2 
20 
12.8 
Apsley Marshes (Tasmania) 
10-50 ha 
47.3 
21 
14.5 
Lightwood Creek Point Nepean 
National Park (Victoria) 
5-10 ha 
42.5  
23 
21.5 
Garfield (Victoria) 
5-10 ha 
45.2 
22 
22.8 
Cann River (Victoria) 
10-50 ha 
43 
23 
23.3 
Gypsy Point (Victoria) 
10-50 ha 
41.4 
24 
27.0 
Yarram (Victoria) 
10-50 ha 
39.2 
26 
32.0 
Cape Barren Island Population 
1 (Tasmania) 
10-50 ha 
39 
26 
32.3 
Cape Barren Island Population 
2 (Tasmania) 
10-50 ha 
38.7 
26 
33.0 
Flinders Island Population 1 
(Tasmania) 
50-100 ha 
34.2 
29 
34.0 
Flinders Island Population 2 
(Tasmania) 50-100 
ha 
34.4 
29 
34.5 
Gladstone (Tasmania) 
50-100ha 
37.9 
26 
35.8 
Narawntapu National Park 
(Tasmania) 
100-1000 
ha 
37.7 26  38.0 
 
 

 
90
y = 3.0579x - 48.442
R
2
 = 0.933
0
5
10
15
20
25
30
35
40
45
0
5
10
15
20
25
30
35
Mean Seed weight (µg) 
G
e
rm
inat
ion (%
)
 
Figure 4.3 Linear regression of seed weight versus germination rate for Melaleuca 
ericifolia. 
 
 
 
0
5
10
15
20
25
30
35
40
Population size (ha)
G
e
rminat
ion (
%
)
< 1
1
1 - 
5
5 -1
0
10

Kataloq: 1469
1469 -> HƏSƏn məCİdzadə saVALAN (1947) Hətatı
1469 -> Transurethral Prostatectomy (turp) & Open