ArticlePDF Available

The biology of Australian weeds 65.'Tradescantia fluminensis' Vell.

116 Plant Protection Quarterly Vol.30(4) 2015
Botanical name
Tradescantia fluminensis Vell. (Figure 1).
Synonym: T. albiflora Kunth (misap-
plied; APNI 2014).
The genus name is derived from John
Tradescant (1608–1662), gardener to King
Charles I of England, while the species is
named for the Rio de Janeiro from the
Latin meaning “from the river” (CABI
Common names
There are many common names for
T. fluminensis. Historically, wandering Jew
was probably most common; however,
because of its pejorative origins its use is
now less common and accepted. Trad and
tradescantia are probably the most used
colloquially, but there are others listed in
the literature: wandering Jew (Auld and
Medd 1992, Harden et al. 2004, Wilson
2015), trad (Pallin 2000, Blood 2001,
Harden et al. 2004), wandering creeper
(Blood 20 01, Richardson et al. 2006, VRO
2011), creeping Christian (Blood 2001),
wandering trad or tradescantia (Blood
2001, AG DE 2015), water spiderwort
(Blood 20 01, Jessop and Conran 2011),
and inch plant (Jessop and Conran 2011).
Names applied outside Australia
include trapoeraba (Brazil, Pereira et al.
2008), Nohakata karakusa (Japan), small-
leaf spiderwort, spiderwort, Vandrande
Jude (Germany), wandering creeper,
wandering Jew, wandering Willie, white
flowered wandering Jew (GISD 2014),
river spiderwort, and wandering Gypsy
(Wikipedia 2015).
Tradescantia fluminensis belongs to the
New World tribe Tradescantieae within
the subtribe Tradescantiinae in the
Commelinaceae, a pantropical family of
monocotyledonous herbs containing 650
species in 41 genera (Burns et al. 2 011).
Combining the naturalized
Commelinaceae species and the species
used in horticulture, but not naturalized,
there is a total of 59 species (plus at
least 19 varieties) in 14 genera present in
Australia (APNI 2014). Fourteen of these
species are naturalized alien species and
36 are native. A further nine are alien, but
are not known to be naturalized. There
are 18 species, plus at least 17 varieties,
cultivated for horticultural purposes
(ornamental pot or garden plants). All of
these horticultural taxa are alien, except
Aneilema biflorum R.B r., A. acuminatum
R.Br., Commelina cyanea R.Br., and Pollia
macrophylla (R.Br.) Benth.
Phylogenetic relationships
The intergeneric phylogenetic rela-
tionships of the Commelinaceae have
been examined by Evans et al. (2003)
by combining rbcL and morphological
datasets, where one (to three) species
were used to represent 30 genera. Burns
et al. (2011) and Wade et al. (2006) also
investigated the phylogenetic relation-
ships of Commelinaceae. While some
of their findings differed from those of
Evans et al. (2003), none of these changed
the apparent relationships between the
genera present in Australia. Therefore
the relationships of the Australian genera
described below have been based on the
work of Evans et al.
Although there was incongruence
between the morphology and rbcL data-
sets, the Australian genus Car tonema
R.Br. was placed consistently as sister
to the rest of the family as the tribe
Cartonemateae, while the monophyletic
tribe Commelineae, which contains five
genera present in Australia (Commelina
L., Aneilema R.B r., Floscopa Lou r.,
Murdannia Royle and Pollia Thunb.), is the
next most distantly related lineage.
The remainder of the Commelinaceae
fall within the tribe Tradescantieae. This
includes the genus Palisota Rchb. ex
Endl., which does not occur in Australia
and is in its own subtribe (Palisotinae).
There are six additional subtribes within
the Tradescantieae. The most distantly
related to Tradescantia L. is Palisontinae,
of which there are no representatives in
Australia, and then Dichorisandrinae,
which contains members of two genera
cultivated in Australia (Geogenanthus
Ule and Dichorisandra J.C.Mikan). The
subtribes Coleotrypinae and Cyanotinae
together are the next most closely
related, with two species represented in
Australia (both occurring in Cyanotinae).
Thyrsantheminae is most closely related
to Tradescantia, but the re are no members
present in Australia.
Within the subtribe Tradescantiinae
there are 12 species from three
genera (Callisia Loef l., Gibasis Raf. and
Tradescantia) present in Australia. Of
these, Callisia is the most dis tantly related,
while Gibasis is very closely related. The
relationship of the recently described
Australian species Tapheocarpa calan-
drinioides (F.Muell.) Conran to the rest
of Commelinaceae is yet to be deter-
mined, but preliminary morphological
and molecular data suggest that its
affinities are with Commelineae, close to
Commelina (J. Conran unpublished data).
While its polyploidy status is unknown
(AG DE 2015), T. fluminensis has a chro-
mosome number of 2n=40, 67+B (Owens
1981), although Scaramuzzi et al. (2000)
report a stable chromosome number
of 2n=72 for the horticultural cultivar
T. fluminensis var. foliis variegatis.
As part of a joint biological control
project, a microsatellite genetic analysis
of T. fluminensis populations collected
from 15 populations sourced from
Australia, New Zealand and South Africa
has been undertaken by Gar y Houliston
(Manaaki Whenua Landcare Research,
New Zealand). One group of Australian
accessions clustered with accessions
from South Africa and New Zealand,
while another group from Australia was
genetically distinct. Interestingly, a large-
leaved Australian Tradescantia accession
from Browns Reserve, Victoria did not
cluster with any of the other populations
investigated and supports the conclusion
from the author (JGC) that this accession
is a species different to T. fluminensis.
It was resistant to the leaf smut path-
ogen, further supporting this conclusion
(Sheppard 2015).
Tradescantia fluminensis is an invasive
perennial (or annual), procumbent clonal
herbaceous weed of the spiderwort
family (Commelinaceae). It forms dense
swards (of ten referred to as mats) that
consist of many interwoven stems (Figure
2). Stems form adventitious roots and
small stem fragments with at least one
leaf node readily develop into new plants.
The leaves are sub-sessile, stem-
sheathing, spirally alternate becoming
distichous by twisting; the sheath ciliate
apically; the lamina glabrous, lanceolate
The Biology of Australian Weeds 65. Tradescantia
fluminensis Ve ll.
Tony M. DugdaleA, David A. McLarenA,B and John G. ConranC
ADepar tment of Economic Dev elopment, Jobs, Transport a nd Resources , AgriBio, 5 R ing
Road, Bundoora 3083, Melbourne, Australia.
BLa Trobe Universit y, Bundoora 3083, Melbourne, Australia.
CACEBB and SGC (Australian Centre for Evolutionary Biology and Biodiversity and
The Sprigg Geobiology Centre), School of Biological Sciences, Benham Building, North
Terrace, The Universit y of Adelaide, South Australia 5005, Australia.
Plant Protection Quarterly Vol.30(4) 2015 117
to ovate, 25–70 mm long, 10 –30 mm wide,
acute and green. The inflorescence is a
terminal, pedunculate, sub-umbellate
cyme, wi th 2 spathes , boat-shaped, g reen
and acute, with the inner spathe slightly
smaller. There are up to 20 flowers. The
sepals are 5– 8 mm long, 2–3 mm wide
and green. The petals are broadly ovate,
7–10 mm long, 4–6 mm wide and white.
The stamen filaments are white, bearded
with long white hairs; the anthers are
yellow. The capsule is c. 2 mm long. The
seeds are c. 1.5 mm long; reticulate and
dark brown (Conran 2005, Figure 1).
Similar species
Native species that look similar to
T. fluminensis include Commelina cyanea
(native wandering creeper; scurvy weed),
C. diffusa, Burm.f. (native wandering
Jew, scurvy weed), Aneilema biflorum,
Oplismenus spp. (native grasses), and
Cheirostylis ovata (F.M.Bailey) Schltr. and
Zeuxine oblonga R.S.Rogers & C.T.White
(a native orchid) (Blood 2001, AG DE 2015).
C. cyanea differs from T. fluminensis in
that the former has blue flowers, three
fertile stamens, smaller and narrower
leaves, less dense foliage and thicker,
fleshier roots (Muyt 2001, AG DE 2015).
C. diffusa differs because it has blue
flowers (Biosecurity Queensland 2014).
The white flowers of Aneilema biflorum
also have only three fer tile stamens (AG
DE 2015). When not in flower the native
grasses of the genus Oplismenus can
be similar but their leaves are hairy. In
Victoria, members of this genus are not
usually luxuriant enough to be confused
with T. fluminensis.
Other native species such as Aneilema
acuminatum, A. biflorum. and C. cyanea
can also be mistaken for this taxon, espe-
cially the creeping A. biflorum, but the
absence of cincinnus (lateral branches
arising alternately on opposite sides of
the false axis), bracts or stamen filament
hairs distinguishes them from the weed.
Introduced species that look similar are
the weedy (undesirable or troublesome
plant) Tradescantia zebrina Bosse (Blood
20 01), T. pallida (Rose) D.R.Hunt (which
has a very restricted distribution) and
T. cerinthoides Kunth (al so with a rest ricted
distribution, in Adelaide and Sydney;
CHAH 2013, AG DE 2015). T. zebrina
is more widespread and is recorded
from New South Wales (including Lord
Howe Island), Queensland, and Western
Australia (CHAH 2013), but can be differ-
entiated by t he silvery- white stripe s on the
upper leaf surfaces, purple undersides
and pink to purplish flowers (Richardson
et al. 2006). T. pallida and T. cerinthoides
both have purplish colouration of the
undersides of their leaves (AG DE 2015).
Vinca major L. (blue periwinkle) can also
be confused when not in flower, but this
has oppositely arranged leaves (T. flumin -
ensis has alternately arranged leaves) and
exudes milky sap when damaged.
Althou gh not similar enoug h to confuse
with T. fluminensis, the weed species
Tradescantia spathacea Sw., Commelina
Figure 1. Tradescantia fluminensis Vell. A. Habit showing creeping stems with leaves twisted to sit more or less horizontally; B.
Close-up of sheathing leaf base showing ciliate apex; C. and D. Cincinnate inflorescence with two enclosing bracts and f lower
with densely hairy anther filaments. (Cincinnate refers to lateral branches arising alternately on opposite sides of a false axis). A.,
B. and D. by J.G. Conran; C. with permission from PlantNET (2015).
118 Plant Protection Quarterly Vol.30(4) 2015
benghalensis L., and C. africana L. are
closely related (Blood 2001).
Tradescantia fluminensis is a very
common houseplant in many countries
around the world, including Ireland,
Germany, the United Kingdom and
United States (CABI 2014). It was probably
introduced to Australia as a garden plant
(AG DE 2015). The earliest herbarium
record from Australia is dated 1924, when
it was collected from the Sydney suburb
of Ashfield (CHAH 2013).
Tradescantia fluminensis is recorded
from all states and territories except the
Northern Territory and the Australian
Capital Territory (Figure 3), as well as
being present on Lord Howe Island and
Norfolk Island (AG DE 2015). It is more
common in coastal areas and population
centres. In Victoria it is widespread, with
medium to large populations (Carr et al.
1992). In Tasmania it is a weed of peri-
urban areas and suburban waterways (M.
Baker Tasmanian Herbarium, personal
communication 2014).
Tradescantia fluminensis is endemic to
the tropical rainforests of south-eastern
Brazil and neighbouring areas of Uruguay
and Argentina (CABI 2014). It is particu-
larly abundant along the coast in south-
eastern and southern Brazil. It does not
form extensive or dense populations in
its native range and is not regarded as
a weed (Pereira et al. 2008, Fowler et al.
2013), although it does occur along road-
sides and in gardens (CABI 2014 – citing
Barreto 1997, in Spanish).
T. fluminensis is recorded as weedy
from New Zealand, South East Asia,
South Africa, and the United States of
America (Blood 2001). The species has
also become naturalized in Argentina,
Bermuda, Italy, Japan, Kenya, Portugal,
Puerto Rico, Russian Federation, Saint
Lucia, and Swaziland (GISD 2014).
Climatic requirements
In its native range (southern Brazil)
T. fluminensis occurs in tropical rainforest
and other damp places (CABI 2014), but in
its introduced range it occupies a broad
range of climates from cool temperate to
tropical. Climate-based modelling of its
potential distribution in Australia (Figure
4) indicates that it will grow best in south-
eastern and south-western Australia,
with poor climate matches in central and
northern Australia, except for parts of
northern Queensland. It should be noted
that this modelling does not take into
account microclimate factors (such as
shade and moisture) that are important
for T. fluminensis.
Tradescantia fluminesis is frost sensi-
tive (Bannister 1984, Blood 2001). In New
Zealand it is found in frost-free areas,
either in the north of the country, or
protected under forest cover as far south
as Dunedin on the lower south-western
coast of the South Island, where frosts are
severe (Butcher and Kelly 2011, J. Conran
personal observation 2014).
In southern Brazil, T fluminesis is mostly
found in coastal humid, damp, rocky
areas, particularly in riparian habitats
(Pereira et al. 2008). It is also reported in
natural and managed forests, roadsides,
urban and peri-urban areas, disturbed
areas, riverbanks, and wetlands (locations
not specified; CABI 2014). In Australia,
T. fluminensis is found mainly in damp,
shady areas, where it tolerates periodic
inundation and poorly drained soils
(Blood 20 01, Muyt 2001, Richardson et
al. 2006) and commonly grows on silt y
alluvial soils (VRO 2011). Dense infesta-
tions form on damp fertile soil and its
growth is sparse on rocky substrates in
Australia (GISD 2014). It responds rapidly
to nitrogen and growth is greatest at
nitrogen concentrations likely to occur
only on disturbed sites (Maule et al. 19 95)
or rich alluvial deposits. It is also capable
of accumulating nitrogen in its stem
tissues, which it can then utilize when soil
nitrogen becomes depleted (Maule et al.
Plant associations
In Australia, T. fluminensis invades damp
or seasonally moist sites, including wood-
lands, heathlands, rainforest, damp and
wet sclerophyll forest, riparian vegeta-
tion, warm and cool temperate rainforest,
dry sclerophyll forest and woodland (Carr
et al. 1992, Blood 2001, Muyt 2001, AG
DE 2015), and subtropical rainforest
(Stockard 1996, Harden et al. 2004). It is
also weedy in gardens and greenhouses
(Blood 20 01, Richardson et al. 2006).
Growth and development
Data from Australia on the growth and
development of T. fluminensis are lacking.
Nevertheless some general statements
can be made. Growth is very rapid, with
stems able to spread several metres in a
single year (Blood 2001). There are two
distinct growth phases: prostrate vegeta-
tive growth, rooting readily at the nodes,
and a flowering stage with shoots that are
more or less erect but having a reduced
capacity to form roots (Blood 20 01).
Although the plant produces flowers,
viable seed has not been identified in
either New Zealand or Australia (Kelly
and Skipworth 1984a, Blood 2001). Frost
cuts back plants, but they can quickly
regenerate from stems protected by
leaf litter (Blood 2001). Regardless, in its
typical understory habitat T. fluminensis
is generally protected from frost.
Swards of T. fluminensis ty pically
consist of three zones: 1) the top of the
sward contains erect soft green stems
bearing 8 –15 fleshy leaves; 2) the middle
Figure 2. Sward of Tradescantia fluminensis below riparian forest overstorey, Yarra
River, Melbourne, Victoria. (Trevor D. Hunt).
Plant Protection Quarterly Vol.30(4) 2015 119
contains stems that are chlorotic, with few
leaves and many adventitious roots; and
3) the bottom contains the oldest stems
where the tissues are dying (Kelly and
Skipworth 1984a). The dynamics of indi-
vidual plants of T. fluminensis have been
studied by Kelly and Skipworth (1984a)
and Maule et al. (1995). These studies
showed that plants consist of one major
vertical stem of about 0.6 m, with little
occurrence of branching. Growth occurs
at the apex and the plant consists of a
vertical portion and a basal, horizontal
portion of 0.3–1.5 m which decays from
its end. The horizontal portion has roots
at the nodes, penetrating the soil or
humus only a short distance. Aerial roots
are also formed at nodes within the sward
(Standish et al. 2004).
The growth form of T. fluminensis
(sward or mat) can be considered as
typical of the phalanx type. That is, the
plant develops dense clusters of ramets
arising because of highly branched,
rooting stems with short internodes, that
consolidate occupancy of a site (cf. guer-
rilla species which have long internodes
and a wandering habit) (Thomas and Hay
2010). Note the highly branched rooting
stems described here contrast with the
minimal branching described by Maule
et al. (1995) above. However, both plant
forms result in the formation of dense
swards. Continued outward growth of
stems is not possible without the growth
of adventitious roots at their nodes, which
then stimulate bud activation. This may be
necessary because the species is unable
to transport water adequately from its
basal roots to stem tips (Thomas and Hay
2010). The adventitious root primordia
activate only under moist conditions, so
lateral spread of this species is limited by
dry barriers.
The most thorough survey of T. flumin-
ensis found mean sward dry biomass
in New Zealand to be 455 g m-2 (range
of 116 –1201 g m-2; Fowler et al. 2013),
with other studies reporting swards of
T. fluminensis to contain ~500 g m-2 (or
~300 stems m-2, Maule et al. 1995); ~1400
g m-2 (or 900 m of stem m-2
, equating
to 300 plants m-2, Kelly and Skipwor th
1984a), and a productivity of ~3 t ha-1 yr-1
(Kelly and Skipwor th 1984b) and 0.64 –1.28
t ha-1 yr-1 (Standish et al. 2004). by compar-
ison, in its country of origin (Brazil), large
swards of T. fluminensis are difficult to
find, where mean biomass was 164 g m-2
(range of 54–334 g m-2; Fowler et al. 2 013).
The species is often considered an obli-
gate shade plant, but recent research
shows that it has an unusual photosyn-
thetic acclimation profile that allows it
to grow in full sunlight through to deep
Figure 3. Location of herbarium records for Tradescantia fluminensis (and
T. albiflora) in Australia (CHAH 2013).
Figure 4. Potential distribution of Tradescantia fluminensis in Australia based on
climatic conditions. Colours represent the possibility of T. fluminensis infesting
these areas: red = very high, yellow = high, orange = medium and green = likely.
The CLIMATE model (Pheloung 1996) was used, based on locations where this
plant is known to be native or naturalized, sourced from GBIF (2007), Missouri
Botanical Gardens (2007) and Australia’s Virtual Herbarium (CHAH 2013).
120 Plant Protection Quarterly Vol.30(4) 2015
shade, where it can persist in light as low
as 1.0–1.4% of full sunlight (Adamson et
al. 1991, Maule et al. 1995), with a light
compensation point of 1.3% (Standish
et al. 2001). In New Zealand forests,
T. fluminensis occurs in canopy gaps
and edges when light levels are above
1% of ambient light, but standing crop
increases logis tically with increasing light,
up to 10–15% of ambient light (Kelly and
Skipworth 1984a, Standish et al. 2001).
Maule et al. (1995) reported three
mechanisms that allow T. fluminensis
to flourish in moderate shade: 1) it has
an unusually high stem-to-root ratio; 2)
it increases its specific leaf area when
grown in shade, without a major change
in chlorophyll content per unit area; and
3) it can decrease its leaf protein content
under low light (without reducing chloro-
phyll or carotenoid content), presumably
to reduce metabolic costs. Such mech-
anisms provide the plant an advantage
in low light, but in open areas they may
lead to excessive water loss, which may
be a factor limiting the species to shaded
sites. Interestingly, this evidence supports
the Australian observation of Muyt (2001),
who reports that T. fluminensis tends to
occur in forested areas and its growth in
open sunny positions is slow unless soil
moisture is high.
Tradescantia fluminensis growth (shoot
extension) shows a seasonal pattern
that matches the seasonal trends in air
temperature and day length. Annual shoot
extension rates were 0.60.7 m, with 2–3
mm day-1 in summer and 0.5 mm day-1 in
winter (Maule et al. 1995, Standish 2002).
Despite this annual shoot extension rate,
and the seasonal cycle of grow th rate,
total shoot length and dry weight did
not change through the two year study.
This is because apical growth is balanced
by basal decay and the expansion of the
swards was constrained by a stream on
one side and a dense forest on the other
(Maule et al. 1995). In contrast, during a
12 month study, above-ground growth
exceeded decay over the duration of the
study (indicating that biomass was accu-
mulating), but decay exceeded growth
during summer at one site and autumn at
the other (Standish et al. 2004).
In Victoria, T. fluminensis swards typi-
cally persist throughout the year (R. Dabal
(Melbourne Water) and F. Ede (Melbourne
University) personal communications
2014), although the plant behaves as a
geophytic (resprouts from underground
buds) annual in hot dry climates, dying
back to the rootstocks and also persists
overwinter as rootstocks in cold areas
(Conran 2005). T. fluminensis flowers from
spring to summer (Wilson 2015).
Mycorrhizae are absent from the order
Commelinales, to which T. fluminensis
belongs (Stevens 2001).
Floral biology
Although many members of
Tradescantiinae are self-incompatible,
T. fluminensis is self-compatible and a
high degree of self-fertilization is thought
to occur (Owens 1981, but see below).
Flowers in the Commelinaceae are visited
by a wide range of insect pollinators and
development of large showy anthers may
be a deceptive mechanism to attract
pollinators in the absence of a nectar
reward (Evans et al. 2003).
Seed production and dispersal
Seedlings are very rarely recorded
(Blood 20 01, Jessop and Conran 2013),
or not recorded (Muy t 2001, AG DE 2015)
in Australia, or New Zealand (Kelly and
Skipworth 1984a) and it is believed that
the seeds may not be viable. Because
of this, dispersal is limited to areas with
streams, people or other animals to
disperse its vegetative propagules.
Vegetative reproduction and
Stem fragments that are at least 1 cm
long and contain a node, with or without
leaves, sprout and form new plants (Kelly
and Skipworth 1984a). Subsequently,
T. fluminensis is very easily dispersed
by stems transported in water currents
(Stockard 1996, Peel 2010), dumped
garden waste, soil, and on vehicles and
mowing equipment (Blood 20 01, Muyt
2001). The presence of T. fluminensis was
found to be predicted best by anthro-
pogenic factors (e.g. distance to road,
distance to house, road surface, road
lanes) that facilitated the dumping of
garden waste, rather than physical factors
such as fo rest compo sition, cano py height,
proximity to streams, and slope (Butcher
and Kelly 2011). T. fluminensis can also be
dispersed in cattle hooves and probably
chicken feet (Standish 2001). It establishes
along animal paths tha t lead from infested
watercourses in the Dandenong Ranges
National Park, Victoria, suggesting that it
is also dispersed by the large fauna there,
including deer and wallabies (B. Incoll,
Friends of Sherbrooke Forest personal
communication 2014).
Stem fragments are desiccation-tol-
erant and can sur vive for one year without
roots or contact with the soil (Blood
2001). Further, they can survive for up to
48 hours in seawater, so dispersal from
riparian infestations via river mouths to
nearby coastal areas is likely (Hurrell and
Lusk 2012).
Given that T. fluminensis is traded on
the internet, is desiccation-tolerant and
can be dispersed by water (including
seawater), and dumping of garden
waste occurs widely in Australia, it is very
likely that T. fluminensis will continue to
increase its distribution.
Tradescantia fluminensis is classified as a
moderately to highly invasive weed in the
Victorian Weed Risk Assessment (Weiss
and Iaconis 2002). It is sometimes consid-
ered a “symptomatic invader” because
it requires disturbance (increased light
and soil nitrogen) for establishment
(GISD 2014). Indeed, invasion after
canopy disturbance has been reported
in Australia by Pallin (2000) and Harden et
al. (2004). However, Peel (2010) describes
invasion of intact littoral Australian rain-
forest by T. fluminesis, indicating that it
does not necessarily require disturbance.
Maule et al. (1995) propose an invasion
strategy in New Zealand in which breaks
in the forest canopy (by tree fall) accom-
panied by disturbance of the ground
cover (by tree fall or animal browsing)
result in greater light and soil nitrogen,
which favour T. fluminensis invasion from
its neighbouring forest edge populations.
They also showed that once established
in these environments T. fluminensis
can alter its root and shoot biomass and
nutrient flow to dominate low light forest
Detrimental – Environment
Tradescantia fluminesis is a serious envi-
ronmental weed (Carr et al. 1992, GISD
2014). In Australia it is under-reported,
but has been reported in 23 published
and online articles that describe it as a
weed in specific locations, or describe
control programs for it in Victoria, New
South Wales, South Australia, Tasmania,
Western Australia, and Queensland
(authors’ unpublished data). It is regarded
as a very high risk to the following
Victorian bioregions: 1) Coastal Plains
and Heathy Forest; 2) Inland Plains; and
3) Ranges (Adair et al. 2008a, b, c). In New
South Wales it is reported as a threat to
five endangered ecological communities
(Illawarra subtropical rainforest, Littoral
rainforest, Riverflat eucalypt forest on
coastal floodplains, Sub-tropical coastal
floodplain forest and Swamp-oak flood-
plain forest; Coutts-Smith and Downey
The phalanx growth form of T. f lu m i n -
ensis leads to the formation of smoth-
ering swards, consisting of interwoven
Plant Protection Quarterly Vol.30(4) 2015 121
stems to 60 cm deep (or 1 m in northern
Queensland (Biosecurity Queensland
2014)), which dominate the ground-layer,
excluding all other ground cover species
and preventing recruitment of most seed-
lings of shrubs and t rees in bushland areas
(Figure 2). These effects interfere with (or
prevent) forest succession (Dunphy 1988,
Stockard 1996, Pallin 2000, Blood 2001,
Muyt 2001, Harden et al. 2004). Growth
is so dense that T. fluminensis excludes
all other plants, resulting in bare ground
when the swards or mats are removed
during manual control operations (Muyt
2001). Further, the swards can be so
dense that mature rainforest trees are
killed, apparently through competition
for nutrients and water (Peel 2010). This
has led to T. fluminensis being described
as a transforming weed species, or a
‘water-hogging’ weed (Peel 2010).
In New Zealand, T. fluminensis is rated
as one of the most threatening environ-
mental weeds in several areas of the
North Island (Hurrell et al. 2009) and is
listed in the National Pest Plant Accord
(MPI NZG 2015), meaning it cannot be
sold, propagated, or distributed. There, it
has established in native forest and flour-
ishes where canopy damage increases
light levels on the forest floor, preventing
regeneration of native species (Kelly and
Skipworth 1984a).
Its impacts have been well-studied in
New Zealand, where it alters forest floor
habitat in at least three ways, summarised
by Standish et al. (2004) as: 1) the forma-
tion of dense swards more than 6 0 cm tall,
in contrast to other ground covers, which
are typically smaller and sparsely distrib-
uted; 2) the production of leaf litter that
decomposes faster and alters nutrient
availability and cycling; and 3) increased
soil moisture under swards. The impacts
of these habitat alterations have been
investigated by several authors (Maule
et al. 1995, Standish et al. 2001, Standish
2002, McAlpine et al. 2015). Dense swards
of T. fluminensis are associated with an
exponential decrease in tree seedling
abundance and species richness, owing
to reduced light under the swards rather
than reduced seed availability (Standish
et al. 2001). Survival of native seedlings
and saplings is greatly enhanced when
individuals are taller than the surrounding
sward (Standish et al. 2001). Removal of
swards results in seedling recruitment of
native species (Standish 2002).
Research in New Zealand has indi-
cated that forest seedling recruitment
is compromised when stems within
T. fluminensis swards are longer than
21–23 cm (Mc Alpine et al. 2015). This stem
height correlates with a sward biomass of
200 g m-2 (or 70–90% cover) (derived from
Standish et al. 2001) or a sward biovolume
of 0.9 m3 per 4-m2 plot (McAlpine et al.
2015). This threshold is used as a target
for the New Zealand biocontrol program
(i.e. if the agents reduce sward biomass
to <200 g m-2 they will be considered
successful) (Fowler et al. 2013). Th is
compares with dense infestations of up
to 1200 g m-2 identified in New Zealand
(Fowler et al. 2013). We are not aware of
any measures of biomass in Australia,
nor of relationships of sapling or seed-
ling survival in relation to T. fluminensis
density. However, it is clear that natural
forest regeneration and active forest
restoration programs are both impeded
by dense mats of T. fluminensis.
T. fluminensis swards increase forest
leaf litter decomposition rates relative
to areas of forest floor where the ground
cover is mostly leaf litter (probably
because of the microclimate provided
within the swards) and they alter nutrient
availability (most notably increasing
N availability). However, it is not clear
whether these changes result in any
impact on the forest ecosystems in which
they were measured (Standish et al. 20 04).
T. fluminensis impacts on epigaeic
(living above ground) invertebrates living
beneath swards by decreasing their
species richness and abundance, while
changing the invertebrate community
composition, compared to uninfested
areas (Standish 2004). Such differences
were not detected for invertebrates
above the weed, where T. fluminensis
cover was found to be a poor predictor
of beetle and fungus gnat species rich-
ness or abundance (Toft et al. 2001). The
strongest predictor of beetle and fungus
gnat species richness and abundance
was species richness of the forest vege-
tation, in a positive relationship (Toft et al.
2001). Given that T. fluminensis is known
to reduce regeneration of forests, it is
likely that in the long term these inverte-
brate communities will decline in associ-
ation with T. fluminensis-driven decline in
forest vegetation species richness.
The impact of T. fluminensis on soil
microfauna generally varies with site,
but it was shown to cause a change in
species assemblage of plant-feeding
nematodes (Yeates and Williams 2001).
Seven taxa were only associated with soil
below T. fluminensis mats (or swards),
while eight taxa were absent from soil
below T. fluminensis, relative to paired
areas containing native New Zealand
forest understory. These changes were
attributed to the greater root resources
available in T. fluminensis infestations.
T. fluminensis swards also threaten the
survival of endangered species. In New
South Wales it is reported as a threat to
the plants Allocasuarina portuensis L.A.S.
Johnson and Myrsine richmondensis
Jackes (Syn. Rapanea sp. A Richmond
River) and the parrot Cyclopsitta dioph-
thalma coxeni Gould (Coutts-Smith and
Downey 2006). It impairs access of tuatara
(reptile, Sphenodon punctatus Gray) and
fairy prions (seabird, Pachyptila turtur
Kuhl) to their burrows in New Zealand
(Brown and Brown 2015).
Tradescantia fluminensis as an
ecosystem engineer
Some of the ecological impac ts of
T. fluminensis described in the preceding
section are consistent with those of an
autogenic ecosystem engineer, based
on the definition of Jones et al. (1994).
Autogenic engineers change the envi-
ronment by their own physical structures,
which modulate the supply of other
resources for other species. Invasion by
ecosystem engineers can have profound
effects on the invaded system (Crooks
2002), as can be seen with T. f l u m i n -
ensis. The most important way in which
T. fluminensis modulates the environ-
ment is by reducing light levels near the
soil, which prevents seedling recruitment.
Over time this will change the composi-
tion of the forest where it grows, as trees
and shrubs die without replacement. It
is unlikely that this will continue to the
ultimate point where only T. fluminensis
remains because T. fluminensis does not
occur in full sunlight (unless soil moisture
is high), so this process will cease where
the sward density is reduced enough to
allow seedling recruitment. The impact
of T. fluminensis on forest succession is
likely to be greater in Australia than New
Zealand because it is reported to invade
intact forests in Australia, but only forest
remnants and margins in New Zealand. In
addition, T. fluminensis alters soil nutrient
availability and cycling and increases soil
water, while the sward growth form can
also exclude other ground layer species
by occupying living space, thus reducing
diversity. These can all be considered
forms of autogenic engineering.
The importance of T. fluminensis in
altering ecosystems through changes
in trophic struc ture is less clear. Its high
production of leaf litter that decom-
poses faster than other species is likely
to increase the short-term abundance
of particular herbivores, detritivores and
decomposers, which may explain the
changes in soil microfauna and epigaeic
invertebrates observed in some studies
(see above).
Detrimental – Agriculture,
animal and human health
Tradescantia fluminensis does not
appear to be a weed of crops (GISD
2014), although it has been identified as a
122 Plant Protection Quarterly Vol.30(4) 2015
serious urban garden weed (Blood 20 01,
Richardson et al. 2006).
T. fluminensis can cause skin aller-
gies in people, although this seems rare
(Paulsen and Thormann 2010, Wüthrich
and Johansson 1997). It also very
commonly causes allergic reactions on
the skin of dogs (AG DE 2015). It usually
causes contac t dermatitis in the muzzle,
axillae, groin, interdigital areas, legs,
and ventral abdomen (McBarron 1977,
Marsella et al. 1997, Lee and Mason
2006), but can also result in mange-like
hair loss, localized scabbing, and skin
eruptions over affected areas, including
the lower back (Figure 5). There are an
estimated 4.2 million dogs in Australia
(RSPCA 2015) and T. fluminensis occu-
pies many areas frequented by pet dogs,
including riparian areas and urban back
yards of many high population cities. A
typical vet consultation plus medication
costs for T. fluminensis is in excess of $200
(K. Wallace, Prahran Vet Hospital personal
communication 2014), suggesting that
economic impacts to urban communi-
ties and health issues to pets could be
Purportedly, T. fluminensis can cause
nitrate poisoning in cattle, with rapid
death (even within 1–2 h of access to the
plant) when cattle are hungry (McBarron
1991). However, the evidence for its
toxicity is poor or inconsistent (McKenzie
2012) and it is not recorded in three
volumes dealing with poisonous plants
in Australia (Everist 1981, Covacevich et
al. 1987, Shepherd 2010). A senior veter-
inary pathologist has not come across
any cases of poisoning suspected to
be caused by this species (G. Rawlin,
Department of Economic Development,
personal communication 2014). Also, its
purported toxicity to cattle is inconsistent
with anecdotal observations that high
cattle stocking rates can destroy large
infestations, by trampling of the fleshy
stems (Borchard and Eldridge 2012), or
possibly direct grazing (Peel 2010), and
with observations that it is found under
forest in New Zealand since domestic
stock are excluded from these areas and
would not usually be able to graze it
(Butcher and Kelly 2011).
Tradescantia fluminensis is grown widely
as a house plant (CABI 2014, GISD
2014), particularly variegated varieties
with pronounced silver and/or purple
colouration in their leaves. Variegation is
caused by a genetic mutation, but varie-
gated cultivars can rever t to green when
grown in the shade (CABI 2014). There are
several forms cultivated as ornamentals
in Australia that differ from the wild type
T. fluminensis (Conran 2005). Although
T. fluminensis was widely available in
Australian nurseries (Carr et al. 1992), its
trade now appears limited.
Tradescantia fluminensis is currently not
listed as a noxious weed under legisla-
tion of any state or territory of Australia.
However, up until February 2014, it was
declared in New South Wales as a class 4
weed in seven council areas (Ku-ring-gai,
Lane Cove, Manly, North Sydney, Ryde,
Hornsby and Willoughby). A review of
New South Wales noxious weeds prior to
2014 determined that T. fluminensis was
beyond the capacity of people to control,
meaning that its control could no longer
be legally enforced (M. Michelmore,
NSW Department of Primary Industries
personal communication 2015).
In Western Australia T. fluminensis
is listed under the Western Australia
Organism List (Biosecurity and
Agricultural Management Act 2007) as a
Permitted (s11) organism, which means it
is allowed entry into Western Australia.
Weed Management
Herbicide application is considered the
only practical way to control large infes-
tations of T. fluminensis (Standish 2002,
Lusk et al. 2012). Because leaves and
stems of T. fluminensis are succulent and
very water repellent, an adjuvant is gener-
ally required for herbicides to be effec-
tive. The active ingredients fluroxypyr
and picloram are contained in products
registered for control of T. fluminensis
in Australia. Off-label permits existed
for four further herbicides that contain
the following active ingredients: glypho-
sate, and metsulfuron-methyl (APVMA
PER9907), triclopyr + picloram, and tric-
lopyr + picloram + aminopyralid (both
APVMA PER12367, now expired) (Ensbey
et al. 2011). Wick wipers with glyphosate
(33% solution) have been used to control
T. fluminensis (Carr et al. 1992).
Of the control programs documented
in Australia, glyphosate (2–5%) has been
used successfully in a rainforest remnant.
It was applied after the canopy had partly
reformed (in the later stages of restora-
tion) and was effec tive in late autumn–
winter (Stockard 1996, Harden et al. 20 04).
Fluroxypyr is the most effective herbicide
in rainforest situations in south-eastern
Australia and glyphosate (2%) is also
recommended, but requires repeat appli-
cations at 4– 6 week intervals (Peel 2010).
A problem with using herbicide on
T. fluminensis is that some of the herbi-
cides reg istered for use against it are toxic
to aquatic life, so they must be prevented
from entering water ways. This is difficult
when most T. fluminensis infestations
occur along stream margins. Despite
these problems, approved herbicides are
used by Melbourne Water for T. f l u m i n -
ensis control in some riparian situations.
In order to restore T fluminensis-domi-
nated areas, the infestations may need
to be treated with a herbicide 5 6 times
over a two year period. The aim is to get
bare ground, free from weed competition
(particularly T. fluminensis), then plant
tube stock or sow seed. After this, two
years of spot spraying is required, and
then ongoing maintenance as necessary
to prevent reinvasion (particularly after
floods), aiming for 90–95% reduction in
T. fluminensis cover (R. Dabal, Melbourne
Water personal communication 2014).
The herbicide triclopyr (butoxyethyl
ester) has been shown to be effective
against T. fluminensis in several New
Zealand studies and is used widely
(Brown and Rees 1995, Standish 2002,
Hurrell et al. 2008, Hurrell et al. 20 0 9,
Hurrell et al. 2012, Lusk et al. 2012); use of
this herbicide against T. fluminensis is not
permitted in Australia. Because triclopyr
causes damage to many native seed-
lings, Hurrell et al. (2008, 2009) screened
19 herbicides and herbicide combina-
tions against T. fluminensis to determine
their efficacy and off-target impac ts to
surrounding New Zealand vegetation.
They concluded that picloram, flurox ypyr,
Figure 5. Contact dermatitis and hair loss on the lower back (A) and abdomen
(B) of a 2 year-old female Bernese mountain dog exposed to T. fluminensis. (J.G.
Plant Protection Quarterly Vol.30(4) 2015 123
metsulfuron-methyl and amitrole
provided useful control (along with three
mixtures glyphosate + fluroxypyr, metsul-
furon-methyl + triclopyr, and picloram +
triclopyr), but concluded that there was
little evidence that these herbicides
would be tolerated better by native
plants than triclopyr (which performed
the best against T. fluminensis). Although
the research of Hurrell et al. (2008) shows
that glyophosate is much less effective
than other herbicides, almost complete
kill of thick mats of T. fluminensis has
been achieved with glyphosate when it is
applied three times over a single growing
season (McCluggage 1998). Therefore,
in areas where there is risk of off-target
damage, such as near water bodies, a
regime based on some glyphosate prod-
ucts (that can be used near water) may
provide an option.
Detailed descriptions of T. fluminensis
herbicide-based management programs,
including methodology employed,
cost, duration, off-target impacts and
outcomes are provided by Brown and
Rees (1995), Hurrell et al. (2012), Lusk et
al. (2012) and Brown and Brown (2015).
Other control methods
Careful manual removal can be done
where T. fluminensis swards (or mats)
are raked and/or hand pulled and rolled
up and taken off-site for disposal. Care
must be taken to remove every stem and
stem fragment (Muyt 2001). A variation
on this method has been employed
where T. fluminensis mats are rolled up,
then sprayed periodically with herbi-
cide and re-rolled, and/or wrapped in
black plastic for 18 months, to facilitate
on-site composting (Pallin 2000). This
method has been employed by Friends of
Sherbooke Forest, where an estimate of
the volunteer labour was kept, and it was
calculated to cost $100 000 ha-1 (at $30
h-1) to clear an area adjacent to Monbulk
Creek. Even after this intensive control,
reinvasion of T. fluminensis occurred over
the next few years (B. Incoll, Friends of
Sherbrooke Forest personal communica-
tion 2014).
Solarization is another effective way to
control T. fluminensis where it is growing
in positions that allow direct sunlight.
The plants should be covered with
plastic sheets and left in place for several
weeks over the warmer months. Piles
of T. fluminensis removed during weed
control ac tivities can also be killed with
solarization, negating the need to trans-
port the material off-site (Muy t 2001).
Although regrow th occurs, co ntrol with
these methods is generally effective, with
regrowth being limited to missed stems
or fragments. This good control may be
due to its shallow, fibrous roots that do
not develop strongly (Lamp and Collet
1989), providing little resistance to hand-
weeding and no source of below-ground
propagules from which it can regrow.
The use of shade has also been
trialled in New Zealand, by constructing
shade houses in the field that reduced
light to 2–5% of full light versus 15–27%
in unshaded areas of the study site
(Standish 2002). T. fluminensis biomass
was reduced to 81 g m-2 in shaded plots,
compared to 598 g m-2 in unshaded plots.
This provided more sustained suppres-
sion of T. fluminensis than either manual
removal or herbicide, and it also reduced
the invasion by other weeds.
Chickens and ducks eat T. fluminensis
in gardens and have been very effec tive
in removing it from small areas (Anon.
2002, AG DE 2015). Also, it has been fed
to chickens to dispose of large quanti-
ties generated from manual removal
programs (Peel 2010).
In summary, large infestations of
T. fluminensis can be controlled with
herbicide but several applications will be
required. Smaller infestations can also
be easily hand pulled, raked and rolled
up for manual removal but will readily
regrow from any remaining stem or stem
fragments. Solarization using plastic can
also be employed to control infestations
in direct sunlight. Ongoing monitoring
and follow up treatments will be required.
Natural enemies
Potyvirus has been detected causing
damage on Tradescantia species in
green houses in United States of America
(Lockhart et al. 1981) and causing leaf
distor tion and mild mosaic in T. f lumin -
ensis in Italy (Ciuffo et al. 2006).
T. fluminensis has no known significant
associations with invertebrates or patho-
gens in New Zealand or eastern Australia
(Standish 2001), although no-one has
looked systematically in Australia. The
lack of invertebrate or pathogen asso-
ciations could be because the species
contains flavonoids, which may deter
generalist insect feeders, but promote
host-specific relationships with inverte-
brates from its native range.
Reproduction of T. fluminensis in
Australia and New Zealand, and probably
other areas of naturalization, appears to
be wholly vegetative (Kelly and Skipworth
1984a). By virtue of their genetic
uniformity, plant species that reproduce
vegetatively are excellent candidates for
biological control (Burdon and Marshall
1981). New Zealand has initiated a biolog-
ical control project and has completed
an overseas exploration for natural
enemies of T. fluminensis in its country
of origin, Brazil (Fowler et al. 2 013).
Arthropods and fungal pathogens were
recorded on T. fluminensis in its native
range and these resulted in plants that
had much greater damage than plants in
New Zealand. Several biological control
agents have been identified, with three
Chrysomelid beetles (Neolema ogloblini
(Monrós), N. abbreviata (Lacordaire),
1845 and Lema basicostata Monrós) now
released (causing damage to leaves,
shoot-tips and mature stems, respec-
tively) and the Brazilian yellow leaf smut
fungus (Kordyana sp.) is expected to be
released in New Zealand soon (Auckland
Council 2012). The beetle species have all
established on T. fluminensis infestations
in New Zealand and their impacts to date
have been impressive (S. Fowler, Manaaki
Whenua Landcare Research personal
communication 2014).
A biological control project on
T. fluminensis has recently commenced
in Australia. Preliminary host specificity
testin g suggests t hat the leaf smut f ungus,
Kordyana sp. and the Tradescantia leaf
beetle, N. ogloblini have potential as
host specific biological control agents
for Australia (Lefoe 2015, Morin 2015a, b).
This project was funded by Melbourne
Water. John Weiss and Jackie Steel
(Department of Economic Development)
contributed Figure 4. We thank Andy
Sheppard (CSIRO) for reviewing a
draft manuscript and an anonymous
reviewer and the series editor for helpful
Adair, R., Cheal, D. and White, M. (2008a).
Advisory list of environmental weeds in
coastal plains and heathy forest biore-
gions of Victoria. Victorian Government
Department of Sustainability and
Environment, Melbourne. 42 pp.
Adair, R., Cheal, D. and White, M. (2008b).
Advisory list of environmental weeds in
the Inland Plains bioregions of Vic toria.
Victorian Government Department
of Sustainability and Environment,
Melbourne. 44 pp.
Adair, R., Cheal, D. and White, M. (2008c).
Advisory list of environmental weeds in the
Ranges bioregions of Vic toria. Victorian
Government Department of Sustainability
and Environment, Melbourne. 42 pp.
Adamson, H.Y., Chow, W.S., Anderson, J.M.,
Vesk, M. and Sutherland, M.W. (1991).
Photosynthetic acclimation of Tradescantia
albiflora to growth irradiance: morpholog-
ical, ultrastruc tural and growth responses.
Physiologia Plantarum 82, 353-9.
AG DE, Australian Government Department
of the Environment (2015). Tradescantia
au/cgi-bin/biodiversity/invasive/weeds/ (accessed
25 November 2015).
Anonymous (2002). Tradescantia (trad) it’s
chicken feed! Weed and Bushcare News.
124 Plant Protection Quarterly Vol.30(4) 2015
Gecko Newslet ter – Weeds of the Blue
Mountains. Article posted 5 December
2002, Website viewed 14 February 2014.
APNI, Australian Plant Name Index (2014).
Australian Plant Name Index. Integrated
Botanical Information System http://www. (accessed 2
January 2014).
Auckland Council (2012). Application to import
for release or to release from containment
new organisms under the Hazardous
Substances and New Organisms Act
1996 (Application number APP201362).
Environmental Protection Authority,
Wellington, New Zealand. http://www.epa.
tion.pdf (accessed 25 November 2015).
Auld, B.A. and Medd, R.W. (1992). ‘Weeds-An
illustrated botanical guide to weeds
of Australia.’ (Inkata Press Melbourne,
Sydney). 255 pp.
Bannister, P. (1984). Winter frost resistance of
leaves of some plants from the Three Kings
Islands, grown outdoors in Dunedin, New
Zealand. New Zealand Journal of Botany
22, 303-6.
Biosecurity Queensland (2014). Wandering Jew
or trad Tradescantia albiflora. Fac t Sheet
Not Declared Pest Plant. PP97 February
2014. Depar tment of Agriculture Fisheries
and Forestry, Biosecurity Queensland. 2
Blood, K. (2001). ‘Environmental weeds-A
field guide for SE Australia.’ (C.H. Jerram-
Scientific Publishers, Mount Waverly,
Victoria). 232 pp.
Borchard, P. and Eldridge, D.J. (2012).
Vegetation changes associated with cattle
(Bos taurus) and wombat (Vombatus ursi-
nus) activity in a riparian forest. Applied
Vegetation Science 15, 62-70.
Brown, D. and Rees, D. (1995). Control of
Tradescantia on Stephens I sland. Ecological
Management 3, 6 -9.
Brown, K. and Brown, D. (2015). Control to
eradication of Tradescantia fluminensis on
Stephens Island (Takapourewa): the impor-
tance of systematic and persistent effort.
Department of Conservation Research and
Development Series 346. Department of
Conservation, Wellington, New Zealand.
.15 pp.
Burdon, J.J. and Marshall, D.R. (1981).
Biological control and the reproduction
mode of we eds. Journal of A pplied Ecolog y
18, 649-58.
Burns , J.H., Faden, R.B. and St eppan, S.J. (2 011).
Phylogenetic studies in the Commelinaceae
subfamily Commelinoideae inferred from
nuclear ribosomal and chloroplast DNA
sequences. Systematic Botany 36, 268-76.
Butcher, E.R. and Kelly, D. (2011). Physical and
anthropogenic factors predict distribution
of the invasive weed Tradescantia flumin-
ensis. Austral Ecology 36, 621-7.
CABI, Centre for Agriculture and Bioscience
International (2014). Tradescantia flumin-
ensis (wandering Jew). Invasive Species
datasheet/54389 (accessed 25 November
2015 ).
Carr, G.W., Yugovic, J.V. and Robinson, K.E.
(1992). ‘Environmental weed invasions in
Victoria: Conservation and management
implications.’ (Department of Conservation
and Environment and Ecological
Horticulture Pt y Ltd, Melbourne, Victoria).
78 pp.
CHAH, The Council of Heads of Australasian
Herbaria (2013). Australia’s Virtual
Herbarium. (accessed
14 November 2013).
Ciuffo, M., Masenga, V. and Turina M. (20 06).
Charac terization of a potyvirus isolated
from Tradescantia fluminensis in northern
Italy. Archives of Virology 151, 123 5 -41.
Conran, J.G. (2005). Family Commelinaceae
spiderwort and wandering Jew family.
In ‘Horticultural flora of south-eastern
Australia. The identification of garden
and cultivated plants. Volume 5. Flowering
Plants. Monocotyledons.’ ed. R. Spencer,
pp. 499-506. (University of New South
Wales Press, Sydney, New South Wales).
Coutts-Smith, A.J. and Downey, P.O. (2006).
Impac ts of weeds on threatened biodiver-
sity in New South Wales. Technical series
no. 11. (Cooperative Research Centre for
Australian Weed Management, Adelaide,
South Australia).
Covacevich, J., Davie, P. and Pearin, J. (1987).
‘Toxic plants and animals: a guide for
Australia.’ (Queensland Museum, Brisbane).
504 pp.
Crooks, J.A. (2002). Characterizing ecosys-
tem-level consequences of biological
invasions: the role of ecosystem engineers.
Oikos 9 7, 15 3- 66 .
Dunphy, M. (1988). Rainforest weeds of
the Big Scrub. In ‘Rainforest remnants’,
Proceedings of a Workshop on Rainforest
Rehabilitation held at the North Coast
Agricultural Institute, 17-18th November,
1988. ed S. Phillips, pp. 85-93. (New South
Wales National Parks and Wildlife Service,
Hurstville, Sydney).
Ensbey, R., Cook, T., Scott, M., Johnson, S. and
van Oosterhout E. (2011). ‘Noxious and
environmental weed control handbook – A
guide to weed control in non-crop, aquatic
and bushland situations’. 5th Edition.
(New South Wales Department of Primary
Industries, Orange, New South Wales).
Evans, T.M., Sytsma, K.J., Faden, R.B. and
Givnish, T.J. (2003). Phylogenetic relation-
ships in the Commelinaceae: II. A cladistic
analysis of rbcL sequences and morphol-
ogy. Systematic Botany 28, 270-92.
Everist, S.L. (1981). ‘Poisonous plants of
Australia’. Revised Edition. (Angus and
Robertson, Sydney). 966 pp.
Fowler, S.V., Barreto, R., Dodd, S., Macedo,
D.M., Paynter, Q., Pedrosa-Macedo,
J.H., Pereira, O.L., Peterson, P., Smith, L.,
Waipara, N., Winks, C.J. and Forrester, G.
(2 013 ). Tradescantia fluminensis, an exotic
weed affecting native forest regeneration
in New Zealand: Ecological surveys, safety
tests and releases of fo ur biocontrol ag ents
from Brazil. Biological Control 64, 323 -9.
GBIF, Global Biodiversity Information Facility
(2007). (accessed 16
August 2007).
GISD, Global Invasive Species Database
(2014). Tradescantia fluminensis. ht t p: //
ogy.a sp?si= 49 7&fr=1&sts=ss s&lan g=EN
(accessed 8 January 2014).
Harden, G.J., Fox, M.D. and Fox, B.J. (2004).
Monitoring and assessment of restoration
of a rainforest remnant at Wingham Brush,
NSW. Austral Ecology 29, 489-507.
Hurrell, G.A., Belton, T., Lusk,
C.S. and Lamoureaux, S.L. (2012).
Tradescantia managemen t in a New Zealan d
National Park. Ecological Management and
Restoration 13, 311-4.
Hurrell, G.A., James, T.K., Lamoureaux, S.L.,
Lusk, C.S. and Trolove, M.R. (2009). Ef fects
of rate of application of triclopyr on
wandering Jew (Tradescantia fluminensis
Vell.). New Zealand Plant Protection 62,
3 6 3 -7.
Hurrell, G.A., James, T.K., Lusk, C.S. and
Trolove, M. (2008). Herbicide selec tion for
wandering Jew (Tradescantia fluminensis)
control. New Zealand Plant Protection 61,
Hurrell, G.A. and Lusk, C.S. (2012). Survival
of Tradescantia fluminensis in sea water.
Ecological Management and Restoration
13, 314-6.
Jessop, J.P and Conran, J.G. (2011).
Commelinaceae (version 1). In ‘Flora of
South Australia’, ed. J. Kellermann. 5th
(accessed 26 September 2011).
Jessop, J.P. and Conran, J.G. (2013).
Commelinaceae (version 2). In ‘Flora of
South Australia’, ed. J. Kellermann. (State
Herbarium of South Australia, Adelaide). (accessed
25 November 2015).
Jones, C.G., Lawton, J.H. and Shachak, M.
(1994). Organisms as ecosystem engineers.
Oikos 69, 373-86.
Kelly, D. and Skipworth, J.P. (1984a).
Tradescantia fluminensis in a Manawatu
(New Zealand) forest: I. Growth and effects
on regeneration. New Zealand Journal of
Botany 22, 393-7.
Kelly, D. and Skipworth, J.P. (1984b).
Tradescantia fluminensis in a Manawatu
(New Zealand) forest: II. Management by
herbicides. New Zealand Journal of Botany
22, 399-402.
Lamp, C. and Collet, F. (1989). ‘Field guide
to weeds in Australia’. 3rd Edition. (Inkata
Press, Melbourne). 332 pp.
Lee, S.E. and Mason, K.V. (2006). Immediate
hypersensitivity to leaf extracts of Callisia
fragrans (inch plant) in a dog. Veterinary
Dermatology 17, 70 -8 0.
Lefoe, G. (2015). Tradescantia leaf beetle
Neolema ogloblini (Coleoptera:
Chrysomelidae): a potential biologi-
cal control for Tradescantia fluminen-
sis (Commelinales: Commelinaceae).
Unpublished report to National
Environmental Research Program. Victoria
Department of Economic Development,
Jobs, Transport and Resources, Melbourne.
33 pp.
Lockhart, B.E., Betzold, J.A. and Pfleger, F.L.
(1981). Character ization of a pot yvirus caus-
ing a leaf di stortion disease of Tradescantia
and Zebrina species. Phytopathology 71,
602- 4.
Lusk, C.S., Hurrell, G.A. and Lamoureaux, S.L.
(2012). Native plant and weed diversity in
forest fragments affected by Tradescantia
fluminensis management. New Zealand
Plant Protection 65, 74-80.
Marsella, R., Kunkle, G.A., and Lewis, D.T.
(1997). Use of pe ntoxifylli ne in the treatm ent
of allergic contact reactions to plants of the
Commelinaceae family in dogs. Veterinary
Dermatology 8, 121-6.
Plant Protection Quarterly Vol.30(4) 2015 125
Maule, H.G., Andrews, M., Morton, J.D.,
Jones, A.V. and Daly, G.T. (1995). Sun/
shade acclimation and nitrogen nutrition of
Tradescantia fluminensis, a problem weed
in New Zealand native forest remnants.
New Zealand Journal of Ecology 19, 3 5 -46.
McAlpine, K.G., Lamoureaux, S.L. and
Westbrooke, I. (2015). Ecological impacts
of ground cover weeds in New Zealand
lowland forests. New Zealand Journal of
Ecology 39, 50- 60.
McBarron, E.J. (1977). ‘Medical and veterinary
aspec ts of plant poisons in New South
Wales’. (Department of Agriculture, New
South Wales, Sydney). 243 pp.
McBarron, E.J. (1991). ‘Poisonous plants:
handbook for farmers and graziers’. (Inkata
Press, Melbourne). 150 pp.
McCluggage, T. (1998). Herbicide trials on
Tradescantia fluminensis. Conservation
Advisory Science Notes No. 180.
Department of Conservation, Wellington,
New Zealand. 12 pp.
McKenzie, R. (2012). ‘Australia’s poisonous
plants, fungi and cyanobacteria: A guide
to species of medical and veterinary impor-
tance’. (CSIRO (Commonwealth Scientific
and Industrial Research Organisation),
Melbourne). 950 pp.
Missouri Botanical Gardens (2007). w3TROPI-
COS, Miss ouri Botanic al Gardens Data base
vast.html (accessed 16 August 2007).
Morin, L. (2015a). Information package to
suppo rt application to release th e leaf smut
fungus Kordyana brasiliensis for the biolog -
ical control of wa ndering trad (Tradescantia
fluminensis) in Australia. Unpublished
repor t to National Environmental Research
Progra m. CSIRO (Commonw ealth Scientific
and Industrial Research Organisation),
Canberra). 37 pp.
Morin, L. (2015b). Using pathogens to
biologically control environmental weeds
– updates. Plant Protection Quarterly. 30,
MPI NZG, Ministry for Primary Industries, New
Zealand Government (2015). National Pest
Plant Accord. http://www.biosecurity.govt.
nz/nppa (accessed 25 November 2015).
Muyt, A. (2001). ‘Bush invaders of south-east
Australia: A guide to the identification
and control of environmental weeds
found in south-east Australia’. (R.G and F.J
Richardson, Meredith, Victoria). 30 4 pp.
Owens, S.J. (1981). Self-incompatibility in
Commelinaceae. Annals of Botany 47,
Pallin, B.N. (200 0). Ku-ring-gai flying-fox
Reser ve. Ecological Management and
Restoration 1, 10-20.
Paulsen, E. and Thormann, H. (2010).
Occupational sensitization from the
Tradescantia albiflora cultivar ‘Pink Joy’.
Contact Dermatitis 62, 186-8.
Peel, B. (2010). ‘Rainforest restoration
manual for south-east Australia’. (CSIRO
(Commonwealth Scientific and Industrial
Research Organisation), Melbourne). 336
Pereira, O.L., Barreto, R.W. and Waipara, N.
(2008). Pathogens from Brazil for classical
biocontrol of Tradescantia fluminensis.
Proceedings of the 12th International
Symposium on Biological Control of
Weeds. (CAB (Centre for Agriculture and
Bioscience) International, Wallingford,
United Kingdom).
Pheloung, P.C. (1996). ‘CLIMATE: A system
to predic t the distribution of an organism
based on climate preferences’. (Agriculture
Western Australia, Per th).
PlantNET (2015). The Plant Information
Network System. Royal Botanic Gardens
and Doma in Tr ust, Sydney, Australia. h ttp:// (accessed 24
November 2015).
Richardson, F.J., Richardson, R.G. and
Shepherd, R.C.H. (2006). ‘Weeds of the
south-east: An identification guide for
Australia’. (R.G and F.J Richardson,
Meredith, Vic toria). 438 pp.
RSPCA, Royal Society for the Prevention of
Cruelty to Animals (2015). How many pet s
are there in Australia? http://kb.rspca.
Australia_58.html (accessed 25 November
2015 ).
Scara muzzi, F., Apollonio, G. and D’emerico, S.
(2000). In vitro propagation of two species
of Commelinaceae from bud cultures and
repeated subcultures on a grow th regu-
lator-free medium. In Vitro Cellular and
Developmental Biology-Plant 36, 379-82.
Shepherd, R.C.H. (2010). ‘Is that plant poison-
ous? An Australian field guide for livestock,
pets and people’. (R.G. and F.J. Richardson,
Meredith, Vic toria). 250 pp.
Sheppard, A. (2015). Biological control of
wandering tradescantia, Tradescantia
fluminensis-Phase 1-Emerging Priorities.
Final Report-National Environmental
Researc h Program. CSIRO (C ommonwealth
Scientific and Industrial Research
Organisation), Canberra). 7 pp. Standish,
R.J. (2001). Prospects for biological
control of Tradescantia fluminensis
Vell. (Commelinaceae). Department of
Conservation Science Internal series 9.
Department of Conservation, Wellington,
New Zealand. 25 pp.
Standish, R.J. (2002). Experimenting with
methods to control Tradescantia flumin-
ensis, an invasive weed of native forest
remnants in New Zealand. New Zealand
Journal of Ecology 26 , 161-70.
Standish, R.J. (200 4). Impact of an invasive
clonal herb on epigaeic invertebrates in
forest remnant s in New Zealand. Biological
Conservation 116, 49-58.
Standish, R.J., Robertson, A.W. and Williams,
P.A. (2001). The impact of an invasive weed
Tradescantia fluminensis on native forest
regeneration. Journal of Applied Ecology
38 , 12 53- 63.
Standish, R.J., Williams, P.A., Robert son, A.W.,
Scott, N.A. and Hedderley, D.I. (2004).
Invasion by a perennial herb increases
decomposition rate and alters nutrient
availability in warm temperate lowland
forest remnant s. Biological Invasions 6,
Stevens, P.F. (2001). Angiosperm phylog-
eny website. Version 12, July 2012 [and
more or less continuously updated since.
APweb/ (accessed 15 April 2014).
Stockard, J.D. (1996). Restoration of Wingham
brush 1980-1996. Proceedings of the 11th
Australian Weeds Conference, ed R.C.H
Shepherd, pp. 432-6 (Weed Science Society
of Victoria, Melbourne).
Thomas, R.G. and Hay, M.J.M. (2010). The role
of nodal roots in prostrate clonal herbs:
‘phalanx’ versus ‘guerrilla’. Evolutionary
Ecology 24, 1489-504.
Toft, R.J., Harris, R. J. and Williams, P.A.
(2001). Impacts of the weed Tradescantia
fluminensis on insect communities in frag-
mented forests in New Zealand. Biological
Conservation 102, 31-46.
VRO, Victorian Resources Online (2011).
Invasiveness assessment–wandering
creeper (Tradescantia fluminensis) in
ing_creeper (accessed 15 April 2014)
Wade, D.J., Evans, T.M. and Faden, R.B.
(2006). Subtribal relationships in tribe
Tradescantieae (Commelinaceae) based on
molecular and morphological data. Aliso
22, 520 -6.
Weiss, J.E.R. and Iaconis, L.J. (20 02). Pest plant
invasiveness assessment. Keith Turnbull
Research Institute, Department of Natural
Resources and Environment, Frankston,
Wikipedia (2015). Tradescantia flumin-
Tradescantia_fluminensis (accessed 25
November 2015).
Wilson, P.G. (2015). Tradescantia fluminensis
Vell. In PlantNET The Plant Information
Network System. Royal Botanic Gardens
and Domain Trust, Sydney, Australia.
(accessed 25 November 2015).
Wüthrich, B. and Johansson S.G.O. (1997).
Allergy to the ornamental indoor green
plant Tradescantia (Albifloxia). Allergy:
European Journal of Allergy and Clinical
Immunology 52, 5 5 6 -9.
Yeates, G.W. and Williams, P.A. (2001).
Influence of three invasive weeds and site
factors on soil microfauna in New Zealand.
Pedobiologia 45, 367-83.
... To conserve these species, the "action threshold" (i.e., the target abundance of the invader at which management intervention is triggered) for the control of T. fluminensis would likely be lower than the impact threshold assessed in terms of less sensitive community metrics, such as native species richness, relative abundance, and diversity. Herbicide application is considered the only practical way to control large infestations of T. fluminensis 43 . However, in Australia some of the herbicides registered for its control cannot be used close to waterways. ...
... Tradescantia fluminensis Vell. (family Commelinaceae, commonly known as small-leaf spiderwort) is a sprawling herb native to rainforests of south-eastern Brazil 43 . Adventitious roots forming along stem nodes enable rapid clonal spread, including from fragmented stems 43 . ...
... (family Commelinaceae, commonly known as small-leaf spiderwort) is a sprawling herb native to rainforests of south-eastern Brazil 43 . Adventitious roots forming along stem nodes enable rapid clonal spread, including from fragmented stems 43 . Tradescantia fluminensis is considered a significant invasive plant of subtropical to cool temperate rainforest ecosystems worldwide, especially in New Zealand and Australia 46 , where it was introduced as a popular ornamental house and garden plant 43 . ...
Full-text available
Impacts of invasive species are often difficult to quantify, meaning that many invaders are prioritised for management without robust, contextual evidence of impact. Most impact studies for invasive plants compare heavily invaded with non-invaded sites, revealing little about abundance–impact relationships. We examined effects of increasing cover and volume of the non-native herbaceous groundcover Tradescantia fluminensis on a temperate rainforest community of southern Australia. We hypothesised that there would be critical thresholds in T. fluminensis abundance, below which the native plant community would not be significantly impacted, but above which the community’s condition would degrade markedly. We modelled the abundance–impact relationship from 83 plots that varied in T. fluminensis abundance and landscape context and found the responses of almost all native plant indicators to invasion were non-linear. Native species richness, abundance and diversity exhibited negative exponential relationships with increasing T. fluminensis volume, but negative threshold relationships with increasing T. fluminensis cover. In the latter case, all metrics were relatively stable until cover reached between 20 and 30%, after which each decreased linearly, with a 50% decline occurring at 75–80% invader cover. Few growth forms (notably shrubs and climbers) exhibited such thresholds, with most exhibiting negative exponential relationships. Tradescantia fluminensis biomass increased dramatically at > 80% cover, with few native species able to persist at such high levels of invasion. Landscape context had almost no influence on native communities, or the abundance–impact relationships between T. fluminensis and the plant community metrics. Our results suggest that the diversity of native rainforest community can be maintained where T. fluminensis is present at moderate-to-low cover levels.
... Austrotradescantia comprising thirteen species restricted to South America (Pellegrini 2017(Pellegrini , 2018. One species of this subgenus, Tradescantia fluminensis Vell., is a dense, ground-smothering perennial herb that is widely naturalised in New Zealand and elsewhere, including Australia (Dunphy 1991;Burns 2004;Dugdale et al. 2015;Morin 2018), Hawaii (Staples et al. 2006), South Africa (Obbermeyer and Faden 1985) and North America (Wunderlin 1998;Faden 2000). It occurs naturally in the rainforests of south-east Brazil (Pellegrini et al. 2015;Pellegrini 2018). ...
Tradescantia fluminensis Vell. is an invasive species in Australia, New Zealand and South Africa. To assist biocontrol initiatives and management of the species we examine genetic variation in these countries and compare this to samples collected from its natural range in Brazil. Tradescantia fluminensis comprises two genetic groupings in New Zealand, both of which are shared with Australia and South Africa. One of these genotypes is relatively common in New Zealand and this is also shared with a Brazilian population. Populations of T. fluminensis in Australia and South Africa are genetically more variable than in New Zealand. Two other entities, T. mundula Kunth (syn. T. albiflora hort.) and T. umbraculifera Hand-Mazz. (syn. T. aff. fluminensis “Big”), new names for naturalised species in New Zealand, also comprise distinct genetic groups. These genetic data emphasise the importance of correct taxonomic identification of weed species being considered for biological control programmes. Tradescantia mundula and T. umbraculifera share a similar genome size and chromosome numbers (2n = 66, 68, 70 and 2C = 14.9 picograms), whereas T. fluminensis had lower values (2n = 56, 58; 2C = 11.7 picograms). Self-pollinations of T. fluminensis and T. umbraculifera failed to produce seed, confirming that these two taxa are self-incompatible. Tradescantia mundula is self-compatible as the majority (93%) of self-pollinations produced fruit. Tradescantia umbraculifera produced a low number of fruit and seeds per fruit when pollinated by T. mundula, but no fruit or seeds were formed when it was pollinated by T. fluminensis. Tradescantia fluminensis pollinated with T. mundula or T. umbraculifera failed to produce fruit or seeds. Self-incompatibility and failure to set seed when cross-pollinated with other species suggests the invasive T. fluminensis does not pose a threat of seedling establishment in indigenous ecosystems and vegetative spread remains the main method of reproduction and invasion.
... The species shows effective vegetative reproduction by stolons throughout the year. Similarly to the congeneric species Tradescantia fluminensis Vell., which is an aggressive invasive species in New Zealand (Standish et al., 2001) and Australia (Dugdale et al., 2015), T. zebrina is more common in forests that are degraded, disturbed, or undergoing natural regeneration (Mantoani et al., 2013;Chiba de Castro et al., 2019). ...
Full-text available
As a result of biodiversity and ecosystem service losses associated with biological invasions, there has been growing interest in basic and applied research on invasive species aiming to improve management strategies. Tradescantia zebrina is a herbaceous species increasingly reported as invasive in the understory of disturbed forest ecosystems. In this study, we assess the effect of spatial and seasonal variation on biological attributes of this species in the Atlantic Forest. To this end, we measured attributes of T. zebrina associated with plant growth and stress in the four seasons at the forest edge and in the forest interior of invaded sites in the Iguaçu National Park, Southern Brazil. The invasive plant had higher growth at the forest edge than in the forest interior and lower leaf asymmetry and herbivory in the winter than in the summer. Our findings suggest that the forest edge environment favours the growth of T. zebrina. This invasive species is highly competitive in the understory of semi-deciduous seasonal forests all over the year. Our study contributes to the management of T. zebrina by showing that the summer is the best season for controlling this species.
... Temperature and precipitation seemed to play a key role in its distribution as indicated by the bioclimatic variables that, in order of importance, mean annual temperature (Bio7), temperature annual range (Bio1) and precipitation in the driest quarter (Bio17) contributed the most on the model (Table 2). Previous reports have indicated that T. fluminensis thrives best in damp habitats in its tropical native region of Brazil (Fowler et al., 2013;Mbande et al., 2019Mbande et al., , 2020 even though it also occurs in both tropical and temperate areas in its introduced ranges (Dugdale et al., 2015). It is therefore not surprising that precipitation in the driest quarter was revealed as a limiting factor as the weed will not be able to withstand prolonged dry spells as also supported by contemporary empirical studies (Burns, 2004;Mbande et al., 2019Mbande et al., , 2020. ...
... Tradescantia zebrina is a 15 to 25 cm tall herbaceous perennial that has invaded disturbed tropical and subtropical areas around the world, including the United States of America (Burns 2004), South Africa (Foxcroft et al. 2008), China (Weber et al. 2008), Australia (Biosecurity Queensland 2016) and Brazil (Zenni and Ziller 2011). Similarly to its congeneric Tradescatia fluminensis Vell., which is an aggressive invader in New Zealand (Standish et al. 2001) and Australia (Dugdale et al. 2015), T. zebrina spreads vegetatively from stem and root fragments, often producing dense mats that prevent the regeneration of native plants by smothering recruits (Lorenzi and Souza 2008;Pellegrini et al. 2017). In Brazil, T. zebrina has invaded a wide range of environments, including semi-arid dry forests, savannas and the Atlantic Forest (Mantoani et al., 2013;Ribeiro et al., 2014;Zenni and Ziller 2011). ...
Aims Invasive plants modify the structure and functioning of natural environments and threat native plant communities. Invasive species are often favored by human interference such as the creation of artificial forest edges. Field removal experiments may clarify if invasive plants are detrimental to native plant regeneration and how this is related to other local factors. We assessed the joint effect of environment and competition with the invasive Tradescantia zebrina on tree species recruitment in an Atlantic forest fragment. Methods We carried out the experimental study in the Iguaçu National Park, located in southern Brazil, using 30 plots distributed across five invaded sites during six months. We counted T. zebrina leaves and recorded the abundance and height of tree recruits over time under contrasting environmental (forest edge versus forest interior) and removal (all aboveground biomass, only T. zebrina removal, and control) treatments. We analyzed the effects of environment and removal treatment using Generalized Linear Mixed Models. Important Findings The invasive species performed better at the forest edge than in the interior. The higher competitive pressure of T. zebrina led to higher mortality and lower height of tree recruits. Invader removal favored tree recruitment, especially in the forest interior. Our study shows that T. zebrina hampers woody species regeneration in tropical Atlantic forests, especially at the forest edge.
The herbaceous groundcover plant Tradescantia fluminensis (wandering trad) has become a significant invader of temperate and subtropical forest ecosystems in Australia. Classical biological control (biocontrol) offers a sustainable and broad-scale management strategy to reduce wandering trad populations. The leaf-smut fungus Kordyana brasiliensis from Brazil, investigated as part of the New Zealand biocontrol program for wandering trad, was identified as a promising option for biocontrol in Australia. Additional research, however, was required to investigate further the host range of K. brasiliensis and fully assess risks since several native species in the family Commelinaceae are present in Australia. Kordyana brasiliensis only developed lesions on the 14 different accessions of wandering trad tested. Of the 28 non-target plant taxa tested, 5 developed flecks following inoculation with K. brasiliensis. In 2019, initial releases were performed in two regions, the Shoalhaven in New South Wales and Dandenong Ranges in Victoria, to obtain contrasting information on the development of the fungus post-release. Kordyana brasiliensis lesions were detected on wandering trad at the 4 plots in the Shoalhaven within a couple of months of the release. In contrast, a few lesions were detected only 5 months after the release at a few of the 9 plots in the Dandenong Ranges. Cooler temperatures in the Dandenong Ranges may have hampered development of the fungus. After 26–32 months of those initial releases, wandering trad foliage cover had declined substantially in 3 of the 4 plots in the Shoalhaven but had remained stable in the Dandenong Ranges plots. While the reduction in wandering trad cover may have resulted from recurrent disease caused by K. brasiliensis at the Shoalhaven plots, a longer post-release monitoring period is required to support this conjecture with greater certainty.
Widely used throughout the world as traditional medicine for treating a variety of diseases ranging from cancer to microbial infections, members of the Tradescantia genus show promise as sources of desirable bioactive compounds. The bioactivity of several noteworthy species has been well-documented in scientific literature, but with nearly seventy-five species, there remains much to explore in this genus. This review aims to discuss all the bioactivity-related studies of Tradescantia plants and the compounds discovered, including their anticancer, antimicrobial, antioxidant, and antidiabetic activities. Gaps in knowledge will also be identified for future research opportunities.
Full-text available
Wandering Jew (Tradescantia fluminensis) prevents the regeneration of native forests in New Zealand. The herbicide triclopyr effectively controls this weed, but is damaging to many native plant species. To identify alternative herbicides, 16 active ingredients representing eight chemical groups were applied to container-grown wandering Jew plants of various ages in three experiments. In Experiment 1, triclopyr killed all plants (3 months old), while amitrole caused substantial damage to plants. In Experiment 2, amitrole, terbuthylazine, metsulfuron-methyl and triclopyr provided excellent control of 2 month-old plants. In Experiment 3, on 4 month-old plants, wandering Jew was highly susceptible to triclopyr, metsulfuron-methyl, fluroxypyr, glyphosate + fluroxypyr, metsulfuron-methyl + triclopyr and picloram + triclopyr. These herbicides were evaluated in a subsequent field trial and all except metsulfuron-methyl gave similar levels of control to Experiment 3. Further investigation of these chemicals is required to determine their optimal use rates and safety for native plants.
Full-text available
Ecological impacts of three weed species of similar life form, Asparagus scandens, Plectranthus ciliatus and Tradescantia fluminensis, were investigated in six lowland forest remnants in New Zealand. All three species form dense, ground-covering mats of vegetation, and are tolerant of a broad range of light environments. Relationships between canopy openness, weed volume, native plant abundance and native species richness were investigated. Volume of all three weed species increased as canopy openness increased. Tradescantia fluminensis appeared to be most detrimental to native vegetation, with both native abundance and native species richness decreasing sharply as weed volume increased. Plectranthus ciliatus and Asparagus scandens were also associated with declines in native abundance and native species richness, but the correlations were less pronounced and were inconsistent across sites. Regression tree analyses on data from individual sites suggested a potential threshold of weed volume for Tradescantia fluminensis, beyond which both native abundance and native species richness declined abruptly. A threshold was also evident when data from all sites were analysed together. Where native species richness did decline in association with increasing weed volume, there did not appear to be any particular native species that were more likely to be excluded than others. All three ground cover weed species are associated with declines in native plant abundance and native species richness, particularly under high light conditions where the weeds are most abundant.
Australia's Poisonous Plants, Fungi and Cyanobacteria is the first full-colour, comprehensive guide to the major natural threats to health in Australia affecting domestic and native animals and humans. The overriding aim of the book is to prevent poisoning, as there are few effective treatments available, particularly in domestic animals. The species have been chosen because of their capacity to threaten life or damage important organs, their relative abundance or wide distribution in native and naturalised Australian flora, or because of their extensive cultivation as crops, pastures or in gardens. These include flowering plants, ferns and cone-bearing plants, macrofungi, ergot fungi and cyanobacteria. The plant species are grouped by life form such as herbs, grasses and sedges, shrubs, trees, and for flowering plants by flower type and colour for ease of identification. Species described have colour photographs, distribution maps and notes on confusing species, habitats, toxins, animals affected, conditions of poisoning, clinical signs and symptoms, post mortem changes, therapy, prevention and control. Symbols are used for quick reference to poisoning duration and available ways of managing poisoning. As further aids to understanding, poisoning hot-spots are highlighted and the book lists plants under the headings of animals affected and organs affected. A Digest gives brief details for all poisonous species in Australia. This book is written in a straightforward style making it accessible to a wide audience including farmers, veterinarians, agricultural advisors, gardeners, horticulturists, botanists and park rangers, medical practitioners and paramedics, teachers, parents and pet owners.
The chloroplast-encoded gene rbcL was sequenced in 30 genera of Commelinaceae to evaluate intergeneric relationships within the family. The Australian Cartonema was consistently placed as sister to the rest of the family. The Commelineae is monophyletic, while the monophyly of Tradescantieae is in question, due to the position of Palisota as sister to all other Tradescantieae plus Commelineae. The phylogenly supports the most recent classification of the family with monophyletic tribes Tradescantieae (minus Palisota) and Commelineae, but is highly incongruent with a morphology-based phylogeny. This incongruence is attributed to convergent evolution of morphological characters associated with pollination strategies, especially those of the androecium and inflorescence. Analysis of the combined data sets produced a phylogeny similar to the rbcL phylogeny. The combined analysis differed from the molecular one, however, in supporting the monophyly of Dichorisandrinae The family appears to have arisen in the Old World, with one or possibly two movements to the New World in the Tradescantieae, and two (or possibly one) subsequent movements back to the Old World; the latter are required to account for the Old World distribution of Coleotrypinae Cyanotinae, which are nested within a New World clade.
This paper reports on a survey of self-incompatibility in 110 species of 22 genera in the family Commelinaceae. Genera from both tribes, Tradescantieae and Commelineae are included. Fifty-five species were found to be self-incompatible, 50 species self-compatible, and five species comprised individuals which were self-incompatible and individuals which were self-compatible. This variability and its possible evolutionary significance are discussed. Self-incompatible species had actinomorphic flowers and the majority of these were in the Tradescantieae. Species with zygomorphic flowers which were more commonly found in the Commelineae were self-compatible. The ubiquitous presence of binucleate pollen grains supports previous data that self-incompatibility is of the gametophytic type. The site of pollen tube arrest, however, was on the stigma at or near the base of the stigma papilla cells. There were two exceptions to this viz. an unnamed Dichorisandra species and Siderasis fuscata where pollen tube arrest was stylar. The significance of these data to taxonomy also receives comment.
This study aimed to determine the effects of different management practices for Tradescantia fluminensis in lowland podocarp/broadleaf forest remnants in the lower North Island. Fourteen 50 m line transects, across eight sites, were established in April 2009 and assessed annually until 2012. Management practices prior to and during the study period were documented. Over the four assessments, changes in the numbers of native plant seedlings and species differed greatly between management practices as did the percent cover of Tradescantia and other weeds. Native species diversity improved more and the abundance of Tradescantia and other weeds increased less, in forests that were less disturbed and where careful on-going control was carried out, than in forests with more disturbance prior to or during control operations. Effective monitoring of both weeds and native plants is essential to enable the outcome of weed management practices to be measured.
Interactions between organisms are a major determinant of the distribution and abundance of species. Ecology textbooks (e.g., Ricklefs 1984, Krebs 1985, Begon et al. 1990) summarise these important interactions as intra- and interspecific competition for abiotic and biotic resources, predation, parasitism and mutualism. Conspicuously lacking from the list of key processes in most text books is the role that many organisms play in the creation, modification and maintenance of habitats. These activities do not involve direct trophic interactions between species, but they are nevertheless important and common. The ecological literature is rich in examples of habitat modification by organisms, some of which have been extensively studied (e.g. Thayer 1979, Naiman et al. 1988).
(1) The genetic structure of agricultural plant populations nas long been recognized as an important factor in their vulnerability or resistance to disease and pest attack. However, in the biological control of weedy plants, the potential significance of the population genetic structure of the target species appears to have been severely underestimated. (2) An examination of the degree of control achieved in eighty-one different control attempts demonstrated a significant correlation between the degree of control achieved and the predominant mode of reproduction of the target plant: asexually reproducing species were effectively controlled significantly more often than sexually reproducing ones. (3) It is argued from this result that the genetic structure of the target species has important implications with respect to the selection of species to be controlled using biological agents.