Emex australis (Doublegee)
- Summary of Invasiveness
- Taxonomic Tree
- Notes on Taxonomy and Nomenclature
- Plant Type
- Distribution Table
- History of Introduction and Spread
- Risk of Introduction
- Habitat List
- Hosts/Species Affected
- Host Plants and Other Plants Affected
- Growth Stages
- Biology and Ecology
- Air Temperature
- Rainfall Regime
- Soil Tolerances
- Natural enemies
- Notes on Natural Enemies
- Means of Movement and Dispersal
- Pathway Vectors
- Plant Trade
- Impact Summary
- Environmental Impact
- Impact: Biodiversity
- Social Impact
- Risk and Impact Factors
- Similarities to Other Species/Conditions
- Prevention and Control
- Distribution Maps
Don't need the entire report?
Generate a print friendly version containing only the sections you need.Generate report
PicturesTop of page
IdentityTop of page
Preferred Scientific Name
- Emex australis Steinh.
Preferred Common Name
Other Scientific Names
- Emex centropodium Meisner
- Vibro australis Greene
International Common Names
- English: bindii; bull head; cape spinach; Cathead; cat's head; Devil's-thorn; Emex; giant bull head; goat head; prickly jack; southern three corner jack; Spiny emex; tanner's curse; three-cornered jack
Local Common Names
- Germany: Suedliche Emex
- South Africa: dubbeltiedorings; Duwweltjie; Emex-dubbeltjie; Inkunzane
- EMEAU (Emex australis)
Summary of InvasivenessTop of page
Taxonomic TreeTop of page
- Domain: Eukaryota
- Kingdom: Plantae
- Phylum: Spermatophyta
- Subphylum: Angiospermae
- Class: Dicotyledonae
- Order: Polygonales
- Family: Polygonaceae
- Genus: Emex
- Species: Emex australis
Notes on Taxonomy and NomenclatureTop of page
The common name most frequently used in Australia, doublegee, is derived from dubbeltjie, an Afrikaans word meaning devil's thorn (Gilbey, 1975; Gilbey and Weiss, 1980). It was also known as Cape spinach as the young leaves are palatable and can be eaten as spinach (Gardner, 1930); however, soon after being imported into Australia, the plant became troublesome and was more likely to be referred to as tanner's curse (Gilbey, 1975). Common names incorporating the word "jack" presumably refer to the children's game in which small 6-pointed metal pieces are picked up whilst bouncing a ball. Other common names such as cat's head, bull head, devil's thorn and goat head are presumably based upon the visual resemblance of the achene, when viewed from the side so that only two of the three spines are showing, to an animal/devil head with the spines on the achene being the horns/ears.
DescriptionTop of page
The achenes contain a single, trigonous seed, produced firstly in the rosette and then sequentially in leaf axils along the stems. The plant has indeterminate growth, with stem and seed production continuing as long as the season is favourable. They ripen in approximately the same order as conception so that green achenes are still developing near the growing apices of the stem whilst ripe achenes have been shed from the stem near the crown. Due to the continuous growing nature of the plant, the size of aerial achenes vary greatly even within a single plant, but achenes are commonly >10 mm wide (spine tip to spine tip). In E. australis the subterranean achenes rarely exceed 8 mm wide (spine tip to spine tip). Weiss and Simmons (1979) collected seed (achenes) from E. australis populations in its native South Africa as well as from Australia and Hawaii where it is an exotic. Two generations of plant were then grown from these seeds planted simultaneously and within the same glasshouse, so as to remove the effects of environmental variations. They found little difference between the Australian and South African populations of E. australis, but second-generation plants from the Hawaii population had fewer stems, nodes and leaves and produced 40% less seeds than the others. Scott (1990) studied the biology and population dynamics of E. australis growing in South Africa and compared this to similar data collected by Weiss (1981) for E. australis grown in Australia. Seed banks, seedling densities, survivorship data and achene production were all found to be similar.
Plant TypeTop of page
DistributionTop of page
Distribution TableTop of page
The distribution in this summary table is based on all the information available. When several references are cited, they may give conflicting information on the status. Further details may be available for individual references in the Distribution Table Details section which can be selected by going to Generate Report.Last updated: 30 Jun 2021
|Continent/Country/Region||Distribution||Last Reported||Origin||First Reported||Invasive||Reference||Notes|
|Madagascar||Present, Few occurrences||Introduced||First reported: before 1838|
|Malawi||Present, Few occurrences||Introduced|
|Mauritius||Present, Few occurrences||Introduced|
|Saint Helena||Present||Present based on regional distribution.|
|South Africa||Present, Localized||Native||Invasive|
|Tanzania||Present||Introduced||Original citation: Missouri Botanic Garden, 2003|
|India||Present||Introduced||First reported: before 1838|
|Pakistan||Present, Few occurrences||Introduced|
|Trinidad and Tobago||Present, Localized|
|United States||Present, Widespread|
|-New South Wales||Present, Widespread||Introduced||1883||Invasive|
|-Northern Territory||Present, Few occurrences||Introduced||1983|
|-South Australia||Present, Widespread||Introduced||1870||Invasive|
|-Western Australia||Present, Widespread||Introduced||1830||Invasive|
|New Zealand||Present, Localized||Introduced|
History of Introduction and SpreadTop of page
In 1830, passengers travelling to Western Australia via South Africa thought that E. australis would be a useful vegetable to take to their new settlement so seeds were collected from Cape Town for cultivation along the banks of the Swan River (Western Australia). In Australia, the plant quickly became an unwanted weed, its spiny fruits spread by stock and rubber tyres. It was confined to Western Australia until 1870 before appearing in South Australia and then shortly afterwards in Victoria, New South Wales and southern Queensland (Gardner, 1930; Gilbey, 1974a). Isozyme studies (Panetta, 1990a) found two genotypes of E. australis within the Australian populations, but only one of these occurs in Western Australia. Western Australia could not therefore have been the source of genetic material for all populations present in Australia. Both genotypes exist across South Africa. Kloot (1987) blames fodder imported from South Africa rather than seed from Western Australia as the initial source of E. australis found in South Australia.
In New Zealand, E. australis has been present since 1883 and is widespread but not common on North Island and, whereas Panetta and Mitchell (1991) predict that the climatic conditions should be ideal for E. australis in many areas, the plant remains uncommon there. The stability of the plant communities resulting from perennial pastures and lack of crop/pasture rotations are believed to be the reason for E. australis not becoming weedy in New Zealand (Panetta and Mitchell, 1991). In Hawaii, introduced E. australis is a weed of pasture land (Krauss, 1963) having spread alarmingly quickly with Emex (both species collectively referred to) affecting 11,700 ha on Hawaii, Maui, Molokai and Oahu by 1962 (Goeden, 1978). However, introductions of a biological control agent (Perapion antiquum) have given substantial to complete control at altitudes of 600-1200 m although had no impact below 150 m (Julien and Griffiths, 1998).
Risk of IntroductionTop of page
HabitatTop of page
Habitat ListTop of page
|Terrestrial||Managed||Cultivated / agricultural land||Present, no further details||Harmful (pest or invasive)|
|Terrestrial||Managed||Managed forests, plantations and orchards||Present, no further details||Harmful (pest or invasive)|
|Terrestrial||Managed||Managed grasslands (grazing systems)||Present, no further details||Harmful (pest or invasive)|
|Terrestrial||Managed||Disturbed areas||Present, no further details|
|Terrestrial||Managed||Rail / roadsides||Present, no further details||Harmful (pest or invasive)|
|Terrestrial||Managed||Urban / peri-urban areas||Present, no further details||Harmful (pest or invasive)|
|Terrestrial||Natural / Semi-natural||Natural forests||Present, no further details|
|Terrestrial||Natural / Semi-natural||Natural grasslands||Present, no further details||Harmful (pest or invasive)|
|Terrestrial||Natural / Semi-natural||Riverbanks||Present, no further details|
|Terrestrial||Natural / Semi-natural||Deserts||Present, no further details|
|Littoral||Coastal areas||Present, no further details||Harmful (pest or invasive)|
Hosts/Species AffectedTop of page
Host Plants and Other Plants AffectedTop of page
Growth StagesTop of page
Biology and EcologyTop of page
The chromosome number of E. australis (based upon Australian populations) is 2n=20 (Putievsky et al., 1980) and is the same as both Australian and Portuguese populations of E. spinosa (Putievsky et al., 1980; Queiros, 1983). In Australia, where both species co-exist, the plants readily hybridise. Both species are monoecious with flowers of both sexes forming near the apex of the stems as the plant grows. Due to the more erect morphology of E. spinosa and as pollen is more likely to fall down than go upwards, seeds collected from E. australis plants were commonly outcrossed with E. spinosa but seeds collected from E. spinosa plants were only rarely outcrossed with E. australis (Putievsky et al., 1980). Putievsky et al. (1980) found the hybrids grew more vigorously than either parent, but were completely sterile when self-pollinated. They were, however, able to backcross with the parental species giving the possibility of gene flow between the species. There is very little genetic variability within E. australis (Weiss and Simmons, 1979; Panetta, 1990a). Panetta (1990a) states "E. australis provides an example of a sexually reproducing species whose patterns of within and between population variation are generally similar in both source and colonial regions. Indeed, 16 of the 21 surveyed South African populations could not be distinguished from Australian populations."
Physiology and Phenology
In areas with a Mediterranean climate, the emergence of E. australis usually begins in autumn after sufficient rain or in years with earlier rainfall, multiple waves of germination are often observed, each associated with significant rainfall events. Weiss (1981) found that up to six distinct cohorts may sometimes occur within a single year and attributed these to the earlier rains arriving before the majority of viable seed had completed their required period of innate dormancy. Although mid-season accessions may contain the most individuals, it was the first accessions of each year that contributed the most to seed production. Flowers could be produced within 4 weeks of germination and seeds within 6 weeks; however, this was highly variable and in some years it took 18 weeks before flowers were even produced. Seeds are produced continuously at the growing apex of the stems throughout the growing season. Regardless of when they germinate, all Emex plants die at the end of the winter rains when soil moisture disappears (Weiss, 1981). The longevity of the growing season can vary immensely from site to site and from year to year with 20-32 weeks after the first accession, reported in Weiss (1981). Within a Mediterranean climatic region, irrigation during the summer will permit some E. australis seeds to germinate and these can successfully complete development (Panetta, 1990b) but germination rates are lower than normal due to a proportion of the seed being dormant. Seedling densities can be over 900 plants/m² (Gilbey and Weiss, 1980), though more commonly 20-500 plants/m² in areas where E. australis is a problem (Gilbey and Lightfoot, 1979; Weiss, 1981; Scott, 1990; Panetta and Randall, 1993b; Scott and Shivas, 1998; Yeoh et al., 2002). At seedling densities of approximately 200 plants/m², survival from emergence to seed production is 40-50% in both Australia and South Africa (Scott, 1990).
In sub-tropical climates (e.g. KwaZulu-Natal and Eastern Cape provinces, South Africa), E. australis can grow all year round, germinating at any time but remaining an annual (Scott and Way, 1990). E. australis has three distinct phases of growth rate (Gilbey and Weiss, 1980). Under glasshouse conditions, for the first 8 weeks after emergence, growth rates are rapid and most energy is diverted towards foliage production. At 8-15 weeks there is a slight decrease in growth with foliage production decreasing but with more stem elongation, branching and buds, flower and fruit formation and a corresponding increase in root biomass. From 15 weeks senescence occurs and whilst during this phase there is no net increase in plant weight, resources from the leaves, stems and roots are reabsorbed and translocated to the developing achene. Embryo development begins in the seed 3 weeks after flowering and seeds are mature 6 weeks after flowering. Gilbey and Weiss (1980) report the average weight of achenes as being 20 mg, 50 mg and 60 mg at 3, 6 and 8 weeks respectively after flowering. The weight of the achene remained stable after it was 8 weeks old. Weiss and Simmons (1977) found that the optimum temperature for production of seed, stems and leaves was 11.7°C and for roots 16.7°C. A delayed flowering and necrosis of stems was shown at the lowest (6.7°C) and highest (26.7°C) temperatures tested. Above 16.7°C, there were only a few fertile seeds produced with no seeds fully maturing within 15 weeks at 26.7°C. A very strong positive correlation exists between temperature and rate of development until the temperature reaches approximately 19°C, becoming a negative correlation above this.
Within an annual pasture system of a Mediterranean climatic region, early cohorts of E. australis seedlings can result in response to significant summer rainfall events. This is due to the ability for some E. australis seed to germinate in response to moisture, relatively independently of temperature requirements (Weiss and Simmons, 1977; Panetta and Randall, 1993c). In Australia, these result from events such as cyclonic activity which happens reasonably frequently within northern New South Wales and northern Western Australia. Out of season rainfall events are locally referred to as a "false break". Follow up rains may not occur until several months latter when the true autumn/winter rains begin. The long tap root of E. australis seedlings allows them to survive the long dry period following a false break far better than most other annual species and in years where significant E. australis individuals survive a false break they can dominate the pastures and produce massive seed banks if left uncontrolled (Gilbey, 1974a; Weiss, 1981). This can lead to dramatic increases in infestations of E. australis in the following year (Lemerle, 1996).
E. australis only reproduces by seed, is monoecious and self-compatible (Gardner, 1930). The spiny female flowers are sessile and in axillary clusters which form first on the crown, in the centre of the rosette, but subsequently in leaf axil nodes along the stems. The male flowers form in short axillary racemes, often emerging between the female achenes (Gilbey and Weiss, 1980). Stem growth, flower and subsequent seed production is continuous throughout the plant's lifetime. E. australis has a high degree of phenotypic plasticity with the extent of stem and node formation dependent upon environmental conditions and seed production being correlated with this stem and node production. Individuals growing under poor conditions (or in a short growing season) produce only a few seeds <10), but plants are capable of producing over 1100 seeds given favourable conditions (Weiss, 1978). E. australis invests a large proportion of its available resources into reproduction with highly stressed plants devoting 36%, and plants under more favourable conditions devoting 62% of all resources into achene and seed production (Weiss, 1978). Achene weight is variable from <10 mg to >90mg; however, Weiss (1978) found for Australian populations of E. australis the modal achene weight was 40-50 mg at the end of the growing season regardless of the plants' growing conditions. Weiss and Julien (1975) report mean achene lengths (capsule base to the top of the inner lobes of the female perianth) of 8.0 mm and mean widths (between two spine tips) of 9.5 mm. Massive seed banks can occur with up to 10,000 seeds/m² following the first year in pasture within a crop/pasture rotation, but over 17,000 seeds/m² is possible if E. australis plants are specifically protected from interspecific competition, pathogen and insect attack (Scott et al., 2000).
The persistence of E. australis and its ability to survive control measures is due to seed dormancy and longevity (Cheam, 1996). All seeds are dormant when freshly formed and require a 2-6 month period of after-ripening before they can germinate (Hagon and Simmons, 1978; Panetta and Randall, 1993c). Panetta and Randall (1993c), studying four populations of E. australis in Western Australian, found that if given the correct environmental conditions, most seeds (80-95%) were able to germinate at some stage during the following autumn. Whilst some seeds within the seed bank remained in this constant non-dormant state, the dormancy pattern in the majority (60-90%) followed the climatic patterns with germination being possible each winter but not possible each summer. A proportion of the seeds (5-20%) remained in a state of constant innate dormancy for the duration of the study (2 years). Seeds that remain in the constant non-dormant state give the species the flexibility to recruit opportunistically after summer rainfall events within the Mediterranean climate (Panetta and Randall, 1993c). Seeds that remain in the extended innate-dormant state ensure the continued recruitment of seedlings each year (Scott, 1990).
Despite most of the viable E. australis seeds being able to germinate each autumn, the proportion of seeds that actually germinates each season is usually low with only 15% of viable seeds reported germinating by Scott (1990) for E. australis in South Africa and 17% reported by Cheam (1996), 17.6% by Panetta and Randall (1993b), and 37% by Weiss (1981) for E. australis in Australia. Cultivation is known to increase germination (Gilbey and Weiss, 1980; Weiss, 1990). Another important factor is the ability of the seed to imbibe sufficient water for germination via its encapsulating, large, hard and spiny perianth (Cheam, 1987). As a consequence, unburied seeds have in general a low rate of germination. It would also explain Gilbey and Weiss's (1980) observation of lower emergence rates on sandy soils compared to heavier soils, which would retain higher moisture levels. Within a particular season, seeds buried 1-5 cm deep are most likely to become established (Scott, 1990; Cheam, 1996) whereas those on the soil surface or buried deeper than 10 cm are more likely to remain viable in future years thus acting as seed reservoirs in the event of future soil disturbances. Cheam (1996) studied germination patterns from E. australis seeds sown at 1-15 cm and over 4 years, there was 53.9% germination (79% of these within the first year) from seeds buried at 1 cm compared to 23.5% from seeds on the surface and 0% from seeds buried at 15cm. After 4 years, 21.0%, 10.5% and 18.3% of the initial seeds remained viable at 0, 1 and 15 cm depth, respectively. Some E. australis seeds can remain viable in the soil for more than 8 years (Gilbey, 1996).
A prerequisite to invading a new habitat is the ability to tolerate the new climatic conditions. Values estimated within the Climate table are derived from long-term climatic averages from sites known to be at the edge of the known distribution range for E. australis in the native range in South Africa (Schulze, 1986) and where introduced in Australia (Plumb, 1977). Models for predicting the potential distribution of both species, based upon both climate and plant development requirements, are given in Pheloung et al. (1996). For E. australis in Australia, the predicted range is approximately identical to the current range and this the plant is not expected to spread much further. A similar model based on climatic conditions for New Zealand (Panetta and Mitchell, 1991) predicts a range that is considerably larger than the actual range even following over 100 years of opportunity.
E. australis is a good host of the root lesion nematode, Pratylenchus neglectus, a pest of certain varieties of cereal crops. It can be managed by the inclusion of non-host species grown in rotation with the susceptible crops therefore the presence of E. australis in the pasture phase of the rotation will hinder any implemented nematode-management practices (Vanstone and Russ, 2001). There is no known mycorrhiza associated with E. australis (Gilbey and Weiss, 1980). In Australia, both within pastures and crops, E. australis it is usually associated with other plant species in the families Compositae (in particular Arctotheca calendula), Cruciferae, Geraniaceae (in particular Erodium spp.), Gramineae, Leguminosae (in particular Trifolium spp.) or Polygonaceae (Gilbey and Weiss, 1980).
Seed of E. australis is an important component of the diet for small mammals. In Australia, up to a quarter of the seeds on the soil surface are removed by mice (Weiss, 1981) and in South Africa, a third are removed by gerbils (Scott, 1990). In the northern grain-growing region of south-west Australia, the seeds of E. australis are a major part of the diet of the Major Mitchell and inland red-tailed black cockatoos, but the amount of seed removed is insignificant (Scott et al. 2000) and the seeds are a minor food source for galahs, little and long billed corellas (Keighery, 1996).
Air TemperatureTop of page
|Parameter||Lower limit||Upper limit|
|Mean annual temperature (ºC)||15||24|
|Mean maximum temperature of hottest month (ºC)||24||36|
RainfallTop of page
|Parameter||Lower limit||Upper limit||Description|
|Dry season duration||12||number of consecutive months with <40 mm rainfall|
Rainfall RegimeTop of page
Soil TolerancesTop of page
Natural enemiesTop of page
|Natural enemy||Type||Life stages||Specificity||References||Biological control in||Biological control on|
|Apion antiquum||Herbivore||Plants|Leaves; Plants|Stems|
|Brachycaudus rumexicolens||Herbivore||Plants|Leaves; Plants|Stems|
|Elegia inconspicuella||Herbivore||Plants|Leaves; Plants|Stems|
|Phomopsis emicis||Pathogen||Plants|Leaves; Plants|Stems|
|Rhinoncus australis||Herbivore||Plants|Leaves; Plants|Stems|
|Uromyces rumicis||Pathogen||Plants|Leaves; Plants|Stems|
Notes on Natural EnemiesTop of page
Means of Movement and DispersalTop of page
Emex achenes float in water. Water is an important mode of dispersal and new infestations are likely to originate along culverts, fence lines, contour banks or any place where water can be trapped and deposit the seeds (Gilbey, 1975). E. australis achenes are large and heavy and so are unlikely to be dispersed by wind.
Vector Transmission (Biotic)
E. australis has hard thorny achenes that lie on the ground so that one thorn is always pointing upwards. Livestock pick up the achene on their feet and move them short distances, although the achene's spines will eventually cause lameness, thereby limiting the distances seed are dispersed by this method (Gilbey, 1975; Gilbey and Lightfoot, 1979).
Attachment of the achenes to the tyres of vehicles, aircraft and machinery is the main method of spread, and allow the seeds to be transported long distances (Gilbey, 1975). Many farmers in Western Australia prevent the spread of the weed by picking out the seeds (by hand) from their tyres at paddock gates. Hay and grain purchased from infested farms can also be an important means of introduction (Gilbey, 1975; Lemerle, 1996) with hay fodder being more likely to contain E. australis seed than grain fodder in a study conducted within drought stricken areas of New South Wales, Australia (Thomas et al., 1984). Seeds of Emex are however sometimes found in feed wheat (grain) imported to Tasmania from mainland Australia (Parsons and Cuthbertson, 1992; DPIWE, 2003).
It is no longer regarded as a vegetable and has only limited beneficial characteristics (Gilbey and Weiss, 1980). It is not likely to be intentionally introduced anywhere.
Pathway VectorsTop of page
Plant TradeTop of page
|Plant parts liable to carry the pest in trade/transport||Pest stages||Borne internally||Borne externally||Visibility of pest or symptoms|
|Fruits (inc. pods)||seeds|
|Stems (above ground)/Shoots/Trunks/Branches||seeds|
|True seeds (inc. grain)||seeds|
|Plant parts not known to carry the pest in trade/transport|
|Growing medium accompanying plants|
Impact SummaryTop of page
|Fisheries / aquaculture||None|
ImpactTop of page
Economic impacts also occur post harvest. In Australia, the grain industry (cereals, pulses and oilseeds) has restrictions on levels of contamination with Emex seed, ranging from 0 (zero) Emex seed/500mls of malt barley in South Australia to 20 Emex seed/500ml of feed barley or lupins in Western Australia. For wheat, 8 Emex seed/500ml is the national maximum standard (Bowran, 1996). However, very few loads are declined due to Emex contamination (Fromm, 1996) as the comparatively large size of the E. australis seed facilitates its screening from the smaller sized grains (Weiss and Simmons, 1977). This, however, comes at a cost of A$12-$18 per tonne (Zaicou-Kunesch, 1996) or approximately 10% of the farm gate price that the farmer is likely to receive (Clark and Finlay, 1997). In the dried vine fruit industry, E. australis achenes attach to the bottom of plastic buckets and crates used during the picking and drying process. If dislodged, they contaminate the produce being difficult to remove because of their similar mass, size and colour. As the produce is used in breakfast cereals and other similar goods, there is an almost zero tolerance level (Pohlner, 1996). In 1993, the Australian penalty for having even one seed in a load was $173/tonne (Fletcher, 1993). Panagiotopoulos et al. (1987) summarises the tolerance for spiny fruits in this industry with his opening sentence "Dried fruit contaminated with caltrop, innocent weed or three corner jack (=E. australis) is of no commercial value."
As a weed of pastures, E. australis competes with beneficial pasture species and the sharp achenes can cause lameness of stock or introduce infection. In one example, 300 sheep from a flock size of 1800 contracted blackleg (Clostridium chauvoei) picked up through wounds inflicted whilst feeding on Emex plants, and was fatal for 60 of these (Rylands, 1966). In grazed pastures, reducing E. australis seedling densities from 150 plants/m to 30 plants/m resulted in a doubling of the available pasture for ewes and lambs, resulted in 73% heavier ewes, 21% higher values on lamb carcase value and 8% higher fleece value. The marked benefits gained from chemical control of E. australis in Australian pastures in 1979 (Gilbey and Lightfoot, 1979) disappeared during the 1980s and 1990s as meat and wool prices fell and chemical costs increased. Moderate direct short-term profits may still be made by using integrated control methods such as spray grazing (Peirce, 1993) but not always (Panetta and Randall, 1993b). Although Emex control within pasture may not result in immediate profits, it is important in the long term to reduce the seed bank (Peirce, 1993) allowing the legume component of the pasture to enrich soil nitrogen within a pasture/crop rotation farming system (Gilmour, 1996). Gilmour (1996) found controlling E. australis in the pasture phase of the rotation had benefits for the succeeding cereal phase, with a 16% (344 kg/ha) increase in wheat yield and a significant increase in the protein level of the grain (from 9.3% to 10.0%) resulting from the treatment.
Environmental ImpactTop of page
Impact: BiodiversityTop of page
Social ImpactTop of page
Risk and Impact FactorsTop of page
- Invasive in its native range
- Proved invasive outside its native range
- Highly adaptable to different environments
- Tolerates, or benefits from, cultivation, browsing pressure, mutilation, fire etc
- Highly mobile locally
- Has high reproductive potential
- Has propagules that can remain viable for more than one year
- Damaged ecosystem services
- Ecosystem change/ habitat alteration
- Negatively impacts agriculture
- Negatively impacts human health
- Negatively impacts animal health
- Negatively impacts tourism
- Reduced amenity values
- Reduced native biodiversity
- Competition - monopolizing resources
- Pest and disease transmission
- Produces spines, thorns or burrs
- Highly likely to be transported internationally accidentally
- Difficult/costly to control
UsesTop of page
Similarities to Other Species/ConditionsTop of page
Prevention and ControlTop of page
Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.Cutlural control
The success of E. australis in Australia has been attributed to the traditional use of a crop/pasture rotational farming system and, "Doublegee would probably become insignificant (in Australia) under continuous cropping or in permanent pasture" (Dodd and Panetta, 1993). Panetta and Mitchell (1991) considered that the lack of significant invasion of E. australis in New Zealand was due to the absence of rotational farming systems or other regular forms of soil disturbance and that E. australis is unable to successfully invade the stable perennial-pasture communities. Thus a change in land use or affected land management may be expected to have a significant effect on the spread of E. australis. In Australia, the country where E. australis has become most weedy, grain farmers have traditionally used cereal-legume rotational cropping systems so as to break crop disease cycles and to add nutrients to the soil. Prior to the 1980s, the rotational system typically incorporated legume based annual pastures that were grazed by sheep (Powles and Bowran, 2000). The chemical control of Emex within this pasture phase was difficult because reliable and selective herbicides were not available (Panetta and Randall, 1993b) and grazing was the main management tool to maintain pasture balance and prevent weed dominance (Peirce, 1993). E. australis has a relatively low competitiveness (Panetta and Randall, 1993a) that is exploited in vineyards by planting dense cover crops (cereals or legumes at twice their normal rates) between rows so as to minimize achene production from Emex plants that survive applications of pre-emergent herbicides. Cover crops are then slashed at the end of winter to form a mulch that not only suppresses further weed growth but also aids in retaining soil moisture (Code, 1990; Lang, 1990; MacGregor, 1990). Panetta and Randall (1993a) propose managing Emex by tolerating preferential weeds or by periodically resowing pasture legumes, a method which also proved successful in Hawaii prior to the development of biological control.
With small infestations, individual plants can be grubbed out and destroyed. Gardner (1930) reports good results being obtained by using a hoe to sever the taproot just below the crown, before seeds have formed. In broad acre farming, a shallow cultivation before germination encourages a quicker and thicker germination of Emex seedlings that can then be killed by a follow-up cultivation or by herbicides (Pearce, 1973). Several cultivations are effective in controlling most weeds, including Emex, but this can reduce the potential growing season of the crop by over 4 weeks reducing crop yield, and it is unsustainable in areas with light, fragile soils as cultivation promotes wind and water erosion (Powles and Bowran, 2000). In vineyards, undervine cultivation is useful in removing larger, hard-to-kill plants prior to herbicide usage. Carpet-covered 'prickle rollers' are used to manually collect and remove achenes from fruit drying areas and vehicle movement is restricted to reduce spreading the achenes.
As E. australis seeds can remain viable in the seed bank for over 8 years, chemical control will only affect those plants that have germinated in the current season and control measures must be repeated annually for many years to completely deplete the seed bank. The timing of the application of herbicides is critical for the control of E. australis, with post-emergence herbicides applied before the plants are at the 4-5 leaf stage when control becomes less reliable. Once runners have formed, most herbicides applied at the label rate will be ineffective and plants at this stage will also have already set seed (Gilbey, 1990). Gilbey (1990) and Commens (1997) are useful references for the use and costs of herbicide control of Emex in cereals, lupins, peas, clover and vines in parts of Australia. Herbicide-resistant weeds are an increasing problem promoted by the heavily reliance on post-emergent herbicides and in particularly by the persistent usage of the same type of chemical. Weed management strategies should aim to minimize herbicide use (e.g. by incorporating cultural methods), avoid the consecutive use of herbicides that belong to the same herbicide group, and understand that certain herbicide groups (e.g. A and B) are more prone to resistance than others (Preston, 2003). Although there are no current records of herbicide-resistant Emex plants (Heap, 2003), other members of the same family have developed resistant biotypes.
Members of the Polygonaceae (including Emex) are highly susceptible to dicamba and in Australia this has been used extensively for E. australis control within cereals from the 1970s until the mid 1980s when the sulfonylurea herbicides were used. The sulfonylurea herbicides are cheaper, have a broader target spectrum and can cope with a wider range in plant size and density, but have longer plant-back periods (Addenbrooke, 1996; Ralph, 1996). E. australis control in medic pasture in Australia was achieved with diuron + 2,4-DB, and a herbicide containing imazethapyr, with the former influencing pasture composition resulting in more grasses (Panetta and Randall, 1993b). Gilbey (1995) also warns that even low rates of diuron + 2,4-DB can damage clover. More recently there has been an increase in the availability of herbicides such as those containing flumetsulam or bromoxynil + diflufenican (Commens, 1997), but which must be used with caution on legume-based pastures and crops because herbicide tolerance varies widely between legume species/cultivars (Revell and Rose, 2001). Field chemical trials on Emex control within medic and sub-clover pastures (Gilmour, 1996) gave 97.8% reduction in E. australis seed production with flumetsulam + diuron and 77.2% seed reduction with diuron + 2,4-DB and neither treatments had any detrimental effects on legume seed production.
After the 1980s, decreasing values of meat and wool, the availability of non-selective knock-down herbicides such as glyphosate and the tendency towards minimum tillage systems so as to minimise erosion has resulted in more Australian farmers selling their livestock and turning to continuous cropping with cereals being rotated with lupins, pulses or oil seed crops. In the absence of the standard cultural control methods such as cultivation and grazing, weed control in these systems becomes heavily or totally dependent upon chemicals (Powles and Bowran, 2000). Broadleaf weed control in the previous cereal crop and the application of a total knock-down chemical (e.g. glyphosate or bipyridylium) before planting are essential to reduce in-crop weed control in the non-cereal phase. Simazine in conjunction with cultivation gives good control of Emex in Lupins (Gilbey, 1995); diuron can be used for control of Emex in lupins, peas and Lucerne; methabenzthiazuron can be used in peas and imazethapyr can be used in faba beans and peas (Commens, 1997). In conventional canola, broadleaf weed control in general is difficult due to the lack of suitable herbicides (Zaicou-Kunesch, 1996). Non-genetically modified triazine-tolerant (TT) and imidizalinone-tolerant (IT) varieties of canola were released in Australia in the late 1990s (OTGR, 2002). Simazine or atrazine are used to control Emex within TT Canola. IT Canola is widely planted in Emex-dominated areas because imazapic and imazapyr provide good control of E. australis; however, the emergence of resistance in other weeds (e.g. ryegrass) can curtail this (OTGR, 2002).
In Australian horticultural crops, ioxynil and methabenzthiazuron are registered for controlling Emex in onions, metribuzin for control in potatoes and in many other vegetables, chlorthal dimethyl or prometryn are used (Parsons and Cuthbertson, 1992). In vineyards and orchards, general knock-down herbicides such as glyphosate are used under the vines/trees to successfully control small Emex plants. The herbicides diuron, simazine and amitrole + atrazine (Gilbey, 1990) and bromacil are used for controlling Emex in citrus (Parsons and Cuthbertson, 1992). Moore and Moore (2003) provide information on the currently registered herbicides for control of E. australis plants at their various stages of development, listing 43 different active ingredients that are registered in Australia for the control of E. australis.
The early success of biological control against E. australis in Hawaii has led to a continuing effort to implement this technique in Australia where this weed is a major problem. Perapion antiquum, a weevil from South Africa, was released on four islands in Hawaii providing substantial to complete control at elevations of 600-1200 m (Julien and Griffiths, 1998). Other weevils, Perapion neofallax and Perapion violaceum were also released but did not establish on Hawaii. Perapion antiquum was released extensively in Australia during the 1970s and 1980s. Establishment was recorded at three sites but no control was achieved, and climatic reasons were suggested as the reason for the different degrees of success in Hawaii and Australia (Scott, 1992). Lixus cribricollis, a weevil from E. spinosa in Morocco, was also released for the control of E. australis in 1979, but did not establish (Julien and Griffiths, 1998). Recently, Apion miniatum, a weevil collected from E. spinosa in Israel has been released extensively in Western Australia but establishment is not confirmed (Yeoh et al., 2002).
The search for potential biological control agents against this weed also includes studies of the biology and host range of indigenous insects and pathogens attacking this weed. Species studied include: Microthrix inconspicuella (Harley et al., 1979; Shepherd, 1990); Perapion antiquum (Harley and Kassulke, 1975); Rhodometra sacraria (Scott and Way, 1989b; Shepherd, 1989); and Rhytirrhinus inaequalis (Scott and Way, 1989a). Indigenous pathogens that have been studied include Cercospora tripolitana (Morris, 1984); Uromyces rumicis (Morris, 1982), which has also been reported from greenhouse plants in Australia (Shivas, 1987), but not established in the field (Scott and Shivas, 1993); and Phomopsis emicis, which was subsequently discovered established in Australia (Shivas et al., 1994) where it damages up to 30% of seed production (Shivas et al., 1994; Scott and Shivas, 1998).
Emex is palatable to stock and is readily grazed until the formation of the spiny achenes. Defoliation delays stem and seed production as well as decreasing root growth (Weiss, 1976) which should result in the plant, which is by nature a poor competitor (Panetta and Randall, 1993a) becoming even less competitive. Goats preferentially graze on weeds over clovers but given a choice they prefer brassica weeds, annual rye or barley grass to Emex (Peirce, 1993). Indirect control of Emex can however be obtained by manipulating grazing pressures. The taller grasses are able to out-compete shorter broadleaf plants by shading them and therefore are favoured by no or light grazing. Emex and other similar early germinating broadleaved weeds are favoured by medium grazing; the highly competitive legumes such as subclover, which also produce more runners and below-ground seed in response to damage, are favoured by heavy grazing (Madin and Moore, 1993; Peirce, 1993). Excellent E. australis control in legume-based pastures can be obtained by integrating grazing with herbicides or the 'spray-graze' technique (Pearce, 1972). Sub-lethal low doses of herbicides such as 2,4-D, 2,4-DB or MCPA amine are used to induce wilting and elevated sugar levels in the Emex plants (and other broadleaved weeds). This increases their accessibility and palatability and under heavy grazing (four times the normal stocking rates) they are quickly either eaten out completely or at least made an unimportant species within the pasture. If grazing pressures are not high enough, stock may still avoid the Emex in preference for other broadleaved weeds and any plants not grazed within 2-3 weeks will fully recover from the sub-lethal dose of herbicide (Peirce, 1993). Although spray-grazing is the most common form of E. australis control in some regions, Panetta and Randall (1993b) state that it has not been widely adopted because of the significant effort and resources needed to obtain the required short-term elevated grazing pressures.
ReferencesTop of page
Agnew ADQ, 1974. Upland Kenya wild flowers. London, UK: Oxford University Press.
Binns B, 1968. A First Check List of the Herbaceous Flora of Malawi. Zomba, Malawi: The Government Printer.
Black I; Dyson C, 1993. Weed decide - a model to help in deciding whether to spray herbicides to reduce annual weed competition in wheat and barley. Primary Industries, South Australia Plant Protection Note, 31.
Clark N; Finlay J, 1997. Grain marketing handbook. Perth, Australia: The Kondinin Group.
Commens T, 1997. Prevent and control doublegee infestations. Farming Ahead, 49-51.
De Wit JL; Roberts BR; Vermeulen WJ, 1962. Preliminary trials on the use of 2,4-DB as a herbicide in newly-established lucerne. South African Journal of Agricultural Science, 5:341-342.
Dellow JJ; Milne BR; Markham N, 1984. Spiny Emex (Emex australis) control in wheat (late post emergent). Australian Weeds Research Newsletter, 32:25-26.
Dodd J; Panetta FD, 1993. Why weeds are difficult to control. Department of Agriculture Western Australia Bulletin, 4243:23-29.
DPIWE, 2003. Spiny Emex Draft Weed Management Plan. http://www.dpiwe.tas.gov.au/inter.nsf/WebPages/LBUN-5D96FA?open.
EPPO, 2014. PQR database. Paris, France: European and Mediterranean Plant Protection Organization. http://www.eppo.int/DATABASES/pqr/pqr.htm
Evenari M; Kadouri A; Gutterman Y, 1977. Eco- physiological investigations on the amphicarpy of Emex spinosa (L.) Campd. Flora: Morphologie, Geobotanik, Oekophysiologie, 166:223-238.
Fletcher G, 1993. Control of three cornered jack in the block. Australian Dried Fruits News, 20:15.
Fox FW; Young MEN, 1982. Food from the veld: edible wild plants of southern Africa botanically identified and described. Johannesburg, South Africa: Delta Books.
Gardner CA, 1930. The double gee (Emex australis, Steinh.). Journal of Agriculture, Western Australia, 7:504-506.
Gilbey D, 1995. Simple solution controls doublegee. Farming Ahead With the Kondinin Group, 41:24.
Gilmore JG, 1996. Broadstrike and Broadstrike + Diuron use in subterranean clover and medic pastures in Western Australia. In: Shepherd RCH, ed. Proceedings of the 11th Australian weeds conference, Frankston, Australia: Weed Science Society of Victoria Inc, 326-330.
Goeden RD, 1978. Biological control of weeds. Part II. United States Department of Agriculture Handbook, 480:357-414.
Graham RA, 1958. Polygonaceae. In: Turrill WB, Milne-Redhead E, eds. Flora of Tropical East Africa. London, UK: Crown agents for overseas governments and administrations, 1-40.
Guillarmod AJ, 1971. Flora of Lesotho. Cramer, LEHRE.
Harley KLS; Kassulke RC; Julien MH, 1979. Biology and host specificity of Microthrix inconspicuella Ragonot (Lepidoptera: Pyralidae), a natural enemy of Emex australis in South Africa. Journal of the Entomological Society of Southern Africa, 42(2):343-348
Hawkins HS; Black JN, 1958. Competition between wheat and three-cornered jack. Journal of the Australian Institute of Agricultural Science, 24:45-50.
Heap I, 2003. The International Survey of Herbicide Resistant Weeds. http://www.weedscience.com.
Holm L; Pancho JV; Herberger JP; Plucknett DL, 1979. A Geographical Atlas of World Weeds. Toronto, Canada: John Wiley and Sons Inc.
Julien MH; Matthews NC, 1980. The biology of Rhinoncus australis Oke (Coleoptera: Curculionidae), a weevil attacking the weed Emex australis Steinheil in eastern Australia. Australian Entomological Magazine, 6:87-92.
Kloot PM, 1987. The naturalised flora of South Australia 3. Its origin, introduction, distribution, growth forms and significance. Journal of the Adelaide Botanical Garden, 10:99-111.
Krauss NLH, 1963. Biological control investigations on Christmas Berry (Schinus terebinthifolius) and Emex (Emex spp.). Proceedings, Hawaiian Entomological Society, 18:281-287.
Lang DL, 1990. Soil management in South Australian vineyards. Plant Protection Quarterly, 5:114-115.
Lousley JE; Kent DH, 1981. Docks and Knotweeds of the British Isles. London, UK: Botanical Society of the British Isles.
Madin RW; Moore JH, 1993. Weeds in pastures. Department of Agriculture Western Australia Bulletin, 4243:121-122.
MAF, 2003. Ministry of Agriculture and Forestry, New Zealand. Appendix A: Plants listed as noxious or as pest plants in 1973, 1993 and 2000. http://www.maf.govt.nz/mafnet/rural-nz/sustainable-resource-use/land-management/emerging-weeds/appendices/appendix-a.htm.
Merxmuller H, 1969. Prodromus einer flora von Sudwestafrika. Germany: Verlag von J. Cramer.
Missouri Botanical Garden, 2003. VAScular Tropicos database. St. Louis, USA: Missouri Botanical Garden. http://mobot.mobot.org/W3T/Search/vast.html.
Moore CB; Moore JH, 2003. HerbiGuide - The Pesticide Expert on a Disk. Box 44, Albany, Western Australia, 6331.
OGTR, 2002. The biology and ecology of canola (Brassica napus). The office of the gene technology regulator. http://www.otgr.gov.au/pdf/ir/brassica.pdf.
Panagiotopoulos B; Simes RT; Dunstone KH; Jones JR, 1987. Control of caltrop, innocent weed and three cornered jack in vineyards and on drying greens. Australian Dried Fruits News, 15:7.
Panetta FD; Randall RP, 1993. Emex australis and the competitive hierarchy of a grazed annual pasture. Journal of Applied Ecology, 30:373-379.
Pearce GA, 1969. The control of weeds in cereals. Journal of Agriculture, Western Australia, 10:138-147.
Peirce JR, 1993. Control of broad-leaved weeds in pasture. Department of Agriculture Western Australia Bulletin, 4243:132-136.
Plumb TW, 1977. Atlas of Australian resources. Second series. Canberra: Division of National Mapping, Department of Natural Resources.
Preston C, 2003. GRDC research update: Latest developments in herbicide resistance. http://www.grdc.com.au/growers/res_upd/south/03/herbicide.htm.
Revell C; Rose I, 2001. Herbicide tolerance of some annual pasture legumes adapted to fine textured clay soils. http://www.agric.wa.gov.au/cropupdates/2001/Farm_systems/Revell_Rose2.htm.
Robbins WW; Bellue MK; Ball WS, 1951. Weeds of California. Sacramento, USA: California State Printing Office.
Rylands PF, 1966. Spiny Emex. The Agricultural Gazette of New South Wales, 77:610-615.
Schulze BR, 1986. Climate of South Africa part 8 general survey, edition 6. Pretoria: Weather Bureau, Department of Environmental Affairs, Republic of South Africa.
Scott JK, 1992. Biology and climatic requirements of Perapion antiquum (Coleoptera: Apionidae) in southern Africa: implications for the biological control of Emex spp. in Australia. Bulletin of Entomological Research, 82(3):399-406
Scott JK; Shivas RG, 1993. Occurrence of the rust fungus Uromyces rumicis, a biological control agent of fiddle dock (Rumex pulcher) in Western Australia. In: Proceedings of the 10th Australian and 14th Asian-Pacific Weed Conference, Brisbane: The Weed Society of Queensland, 145.
Scott JK; Way MJ, 1989. Biology of Rhytirrhinus inaequalis (F.), a weevil associated with Emex australis Steinh. (Polygonaceae) in southern Africa. Journal of the Entomological Society of Southern Africa, 52(1):1-9
Scott JK; Way MJ, 1989. Host plant specificity and biology of Rhodometra sacraria (L.) (Lepidoptera: Geometridae) in South Africa. Journal of the Entomological Society of Southern Africa, 52(2):245-251
Scott JK; Yeoh PB; Woodburn TL, 2000. Apion miniatum (Coleoptera: Apionidae) and the control of Emex australis (Polygonaceae): conflicts of interest and non target effects. In: Spencer NR, ed. Proceedings of the X international symposium on biological control of weeds, Bozeman: Montana State University, 473-485.
Shepherd RCH, 1990. Evaluation of Microthrix inconspicuella (Lepidoptera: Pyralidae), a potential biological control agent for Emex australis in Australia, carried out in apple orchards in South Africa. Entomophaga, 35(4):583-587
Shivas RG, 1995. Survey for pathogens of Emex australis in South Africa. In: Delfosse ES, Scott RR, eds. Proceedings of the 8th International Symposium on Biological Control of Weeds. Melbourne, Australia: DSIR/CSIRO, 351-354.
Shivas RG; Lewis JC; Groves RH, 1994. Distribution in Australia and host plant specificity of Phomopsis emicis, a stem blight pathogen of Emex australis. Australian Journal of Agricultural Research, 45(5):1025-1034
Siddiqi MA, 1973. New plant records for West Pakistan - 1. Pakistan Journal of Forestry, 23:128-132.
Smith CA, 1966. Common names of South African plants. Botanical Survey Memoir, 35:204-205.
Steinheil AD, 1838. Flore de Barbarie. Annales Des Sciences Naturelles Botanique, 9:193, 195-196.
USDA-ARS, 2003. Germplasm Resources Information Network (GRIN). Online Database. Beltsville, Maryland, USA: National Germplasm Resources Laboratory. https://npgsweb.ars-grin.gov/gringlobal/taxon/taxonomysearch.aspx
USDA-NRCS, 2002. The PLANTS Database, Version 3.5. National Plant Data Center, Baton Rouge, USA. http://plants.usda.gov.
Watt JM; Breyer-Brandwijk MG, 1962. The Medicinal and Poisonous Plants of Southern and Eastern Africa. Edinburgh and London, UK: E & S Livingstone Ltd.
Weiss PW, 1976. Effect of defoliation on growth of spiny emex (Emex australis). Field Station Record, CSIRO Division of Plant Industry, 15:27-33.
Weiss PW, 1978. Reproductive efficiency and growth of Emex australis in relation to stress. Australian Journal of Ecology, 3:57-65.
Weiss PW; Julien MH, 1975. A comparison of two species of spiny emex (Emex australis and E. spinosa) in northwestern Victoria. Journal of the Australian Institute of Agricultural Science, 41(3):211-213
Wells MJ; Balsinhas AA; Joffe H; Engelbrecht VM; Harding G; Stirton CH, 1986. A catalogue of problem plants in South Africa. Memoirs of the botanical survey of South Africa No 53. Pretoria, South Africa: Botanical Research Institute.
Wilding JL; Barnett AG; Amor RL, 1993. Crop weeds. Melbourne, Australia: Inkata Press.
Yeoh PB; Woodburn TL; Scott JK, 2002. Will the red apion (Apion miniatum), a potential biocontrol agent for doublegee (Emex australis), establish in Australia?. 13th Australian Weeds Conference: weeds "threats now and forever?", Sheraton Perth Hotel, Perth, Western Australia, 8-13 September 2002: papers and proceedings, 432-435; 9 ref.
Binns B, 1968. A First Check List of the Herbaceous Flora of Malawi., Zomba, Malawi: The Government Printer.
CABI, Undated. Compendium record. Wallingford, UK: CABI
CABI, Undated a. CABI Compendium: Status inferred from regional distribution. Wallingford, UK: CABI
CABI, Undated b. CABI Compendium: Status as determined by CABI editor. Wallingford, UK: CABI
DPIWE, 2003. Spiny Emex Draft Weed Management Plan., http://www.dpiwe.tas.gov.au/inter.nsf/WebPages/LBUN-5D96FA?open
Fox FW, Young MEN, 1982. Food from the veld: edible wild plants of southern Africa botanically identified and described., Johannesburg, South Africa: Delta Books.
Graham RA, 1958. Polygonaceae. In: Flora of Tropical East Africa, [ed. by Turrill WB, Milne-Redhead E]. London, UK: Crown agents for overseas governments and administrations. 1-40.
Kumar V, Kumari E, 2019. A report on Emex australis Steinh. - an emerging weed problem of wheat at Faridabad region of Haryana. Journal of Crop and Weed. 15 (2), 144-147. http://cwssbckv.org/Journal/default.aspx
Merxmuller H, 1969. (Prodromus einer flora von Sudwestafrika)., Germany: Verlag von J. Cramer.
Smith CA, 1966. Common names of South African plants. In: Botanical Survey Memoir, 35 204-205.
USDA-ARS, 2003. Hedychium flavescens. In: Germplasm Resources Information Network (GRIN). Online Database, Beltsville, USA: National Germplasm Resources Laboratory. http://www.ars-grin.gov/cgi-bin/npgs/html/tax_search.pl
Wells M J, Balsinhas A A, Joffe H, Engelbrecht V M, Harding G, Stirton C H, 1986. A catalogue of problem plants in southern Africa incorporating the national weed list of South Africa. Memoirs, Botanical Survey of South Africa. v + 658pp.
Distribution MapsTop of page
Select a dataset
CABI Summary Records
Unsupported Web Browser:
One or more of the features that are needed to show you the maps functionality are not available in the web browser that you are using.
Please consider upgrading your browser to the latest version or installing a new browser.
More information about modern web browsers can be found at http://browsehappy.com/