Tony M. Dugdale
Department of Primary Industries, AgriBiosciences Centre, Bundoora, Victoria, 3083, Australia.
Paul D. Champion
National Institute of Water and Atmospheric Research, PO Box 11-115, Hamilton, 3251, New Zealand.
Cite this article as:
Dugdale, T.M. and Champion, P.D. (2012). Control of alligator weed with herbicides: A review. Plant Protection Quarterly 27(2), 70-82.
Alligator weed (Alternanthera philoxeroides (Mart.) Griseb.) is a widespread weed that is difficult to control. Despite many published accounts relating to its control there is no published review on herbicide control of alligator weed. This paper describes the biology and impacts of alligator weed and the efficacy of herbicides and herbicide-based control programs used against it.
Alligator weed is native to the Parana River area of South America (Julien et al. 1995). It has subsequently spread to many countries outside of its indigenous range. It was first detected in the United States of America (USA) in 1897, in Mobile, Alabama, growing in ballast presumably sourced from its native range (Zeiger 1967). It was also collected from ballast in 1906, at Aratapu (Northland) in New Zealand (Cheeseman 1906). It was introduced as a forage crop to China in the 1930’s (Wang et al. 2005) and had established in India, Burma and Indonesia by the 1960’s (Sculthorpe 1967). It was first detected in Australia at Newcastle in 1946 and is likely to have been introduced with cargo during World War II (Julien and Bourne 1988). More recent introductions have occurred to Puerto Rico, Singapore, Vietnam and Thailand (Julien and Broadbent 1980). It was first found in Sri Lanka in 1999, established as a food plant possibly introduced from Australia (Tegan 2009), Italy (Garbari and Pedulla 2001) and France in 2002 (EPPO 2012).
Alligator weed is a perennial stoloniferous herb in the family Amaranthaceae. It requires a warm growing season but can tolerate a wide range of climate conditions, including winter frosts, which will kill exposed stem material. Although it flowers producing peduncled capitate inflorescences, seeds are not produced in its introduced range and it reproduces solely by clonal growth (Gangstad 1978, Julien and Broadbent 1980, Julien et al. 1992).
Alligator weed is an amphibious plant. It grows as an aquatic plant, either rooted into the hydrosoil and emerging above the water, or as a sprawling mat of entangled stems floating over the water, or in ephemerally damp or swampy soil (Julien and Broadbent 1980). It can also establish and become weedy in terrestrial habitats, including pasture (Julien and Broadbent 1980), urban areas (Gunasekera and Bonilla 2001) and arable crops (Shen et al. 2005).
Dispersal of alligator weed in water is aided by its hollow stems allowing dislodged stem fragments to float in water and drift with currents to lodge at new locations (Eggler 1953, Julien and Broadbent 1980, Dugdale et al. 2010). Any stem or taproot fragment that contains a bud, which breaks off the plant is capable of regeneration. Gangstad (1978) also reports that ~5% of internodes produce roots when floating in water but not when on soil. Julien et al. (1992) estimated that 9,800 to 17,300 stem nodes m-2 yr-1 were produced by alligator weed, representing its reproductive potential. It can also be spread by physical transport of stem or root fragments on machinery, boats, fishing nets or animals e.g. trapped within the cloven hoofs of livestock.
Alligator weed is also dispersed deliberately by humans. In Australia, it was cultivated as a food plant in home gardens by members of the Sri Lankan community in the mistaken belief that it was mukunuwenna or sessile joy weed (Alternanthera sessilis (L.) R.Br. ex DC.), which is very popular in Sri Lanka. In 2001, it was cultivated in at least 760 home gardens and there were 13 known naturalised sites in Victoria. However, deliberate cultivation in Australia has been reduced because of a very successful education campaign and provision of an alternative native plant (Alternanthera denticulata R.Br., Gunasekera and Bonilla 2001).
Alligator weed is more vigorous in aquatic situations, where it has taller, thicker stems and large amounts of above-soil biomass than when it is growing terrestrially (Julien and Broadbent 1980, Julien et al. 1992). Floating mats become very dense and have reached ~2 kg DW m-2 in Louisiana, USA (Lapham 1964). An aquatic infestation in Melbourne, Australia, spread outwards from bank-side patches to form floating mats that increased in area by 45 to 205% per annum over five years. The average biomass at the site was 4.3 kg DW m-2, consisting of emergent and submerged biomass (Clements et al. 2011). For terrestrial alligator weed, Julien and Bourne (1988) report an annual increase in biomass per unit area of 22% as it gradually displaced competing pasture species. Over 7.3 kg DW m-2 has been collected from a terrestrial site where it had been present for 20 years, with 10 fold higher biomass below ground than above (Schooler et al. 2008).
Alligator weed exhibits phenotypic plasticity with different growth forms occurring in aquatic and terrestrial environments, which are not driven by genetic differences (Li and Ye 2006, Geng et al. 2007). Terrestrial plants have a tough, fleshy, tap-root like structure that is lacking in the aquatic plants; their stems are thinner and more lignified; they have smaller internodal cavities; have shorter internodes; have less leaf area per stem; less relative chlorophyll content; and have lower growth rates than alligator weed growing at aquatic sites (Julien et al. 1992, Geng et al. 2007). Terrestrial plants also have a much greater biomass of root material than aquatic alligator weed (Julien and Broadbent 1980, Julien et al. 1992, Geng et al. 2007). This phenotypic plasticity is an important adaptive strategy for alligator weed because it allows a single invading genotype to adapt to a wide range of habitats in introduced areas, and therefore increases the chance of proliferation. Furthermore, Wilson et al. (2007) have shown that alligator weed develops a different morphology after physical control. Plants that were subject to shoot removal, just above soil level to mimic mowing, had higher below ground root biomass, a higher ratio of root to stem biomass and positioned its leaves much closer to the ground. In addition to this phenotypic plasticity, where genetically identical plants have different growth forms depending on their habitat, alligator weed may also have different growth forms that are due to a genetic basis (biotypes) (Kay and Haller 1982).
Differences in morphology or biotype are important because herbicide control can be affected. For example, Kay (1992) found that a slender stem biotype was more susceptible to quinclorac herbicide than a broad stem biotype. A difference in susceptibility of the terrestrial and aquatic forms of alligator weed to biological control has also been shown, where effective control of the aquatic form is achieved but not for the terrestrial form (Spencer and Coulson 1976, Julien and Broadbent 1980, Kay and Haller 1982, Li and Ye 2006). Such differences may be an additional cause of observed variable responses of alligator weed to herbicide treatment, which have been attributed to differences in water level and temperature (Steenis and McGilvery 1961).
Alligator weed grows rapidly over the surface of water bodies forming floating mats that cover the water surface, impeding access, navigation, water-use and drainage (Spencer and Coulson 1976, Gangstad 1978, Julien and Broadbent 1980, Clements et al. 2011). It alters the timing and magnitude of litter inputs into water along with having a faster decomposition rate than native riparian communities (Bassett et al. 2010). It has also been shown to reduce native invertebrate densities and alter community composition (Pan et al. 2010) and its evapotranspiration increases water loss from standing water relative to water without vegetation (Boyd 1987).
It displaces and competes with native plant communities, and pasture and cropping species in moist agricultural areas, including rice (Oryza sativa L.). It provides mosquito habitat, restricts light penetration into the underlying water and prevents gas exchange between the water and atmosphere (Gangstad 1978, Julien and Broadbent 1980, Julien and Bourne 1988, Shen et al. 2005). It is palatable to livestock, but the grazing of alligator weed has been associated with photosensitivity and resultant skin lesions, liver damage and death in cattle, calves and lambs (Bourke and Rayward 2003). Farming practices for cattle are affected by alligator weed in Kaipara District, New Zealand. Young livestock are initially raised in hill country, away from areas where this species occurs and black cattle are favoured (as white patches are most sensitive to damage). Crops such as kumara (Ipomoea batatas (L.) Lam.) and rice can be heavily impacted by alligator weed. Yield losses range up to 45% for rice (Shen et al. 2005) and some Northland kumara crops are lost when heavy infestations of alligator weed occur (P. Joynt, Northland Regional Council personal communication).
These impacts have led to alligator weed being regarded as one of the world’s worst weeds (Holm et al. 1997) and it is a legislated weed in most countries where it occurs. In Australia, it is one of the original 20 Weeds of National Significance (WoNS), which are managed nationally (van Oosterhout 2007). In New Zealand, it is designated an unwanted organism and is actively managed at all known sites outside of the Northland Region (Champion 2008). In China, US$72 million is spent each year to manage it (Liu and Diamond 2005) and there are occasional eradication efforts at local scales (Pan et al. 2010). In USA, there has been considerable effort to control it since the 1950’s (Gangstad 1978).
Efficacy of herbicides on alligator weed
Many publications exist that describe use of a large number of herbicides on alligator weed in various control programs, however much of the data presented are inadequate to allow efficacy of the herbicide to be determined and comparisons made. Sixteen publications were found that report on the efficacy of 32 herbicides in 49 combinations on alligator weed (Table 1). Short- (4 to 16 weeks), medium- (18 to 38 weeks), and long-term (~52 weeks) efficacy of each of the herbicides tested in these publications are shown (Table 1).
Although several herbicides achieved good to excellent control (80–89% and 90–100%, respectively) in the short- and medium-term, good to excellent long-term control was not as common and usually required repeated herbicide applications. For example, good to excellent long-term control was reported on eight occasions where herbicide was applied more than once: glyphosate three or four times within 12 months (Schooler et al. 2008); imazapyr once per year for two years (Langeland 1986a) and twice in 10 months (Hofstra and Champion 2010); metsulfuron-methyl, three or four times within 12 months (Schooler et al. 2008), five times over two years (Schooler et al. 2010) and twice in 10 months (Hofstra and Champion 2010); and triclopyr, three or four times within 12 months (Schooler et al. 2008) and twice in 10 months (Hofstra and Champion 2010).
Where herbicide was only applied once, good to excellent long-term control was only achieved on five occasions, twice with imazapyr (Allen et al. 2007, Hofstra and Champion 2010) and once each with dichlobenil, karbutilate (Blackburn and Durden 1974), and metsulfuron-methyl (Hofstra and Champion 2010). However, only a single application of metsulfuron-methyl and imazapyr to young plants (3-months old) resulted in excellent control, with older plants being less effected (Hofstra and Champion 2010). For established plants, Allen et al. (2007) reported that imazapyr was applied once to alligator weed in the spring and autumn, and only the autumn application resulted in effective long-term control. They suggested that this was because of better downward translocation of the herbicide at this time of year. The marshes that they did the experiment in were inundated after herbicide application, which offers an alternative explanation for the effective control after a single application because inundation after herbicide application has been observed anecdotally to improve control with herbicides (Steenis and McGilvery 1961, Langeland 1986a). Although the mechanism for this improved control is not known, it seems likely that alligator weed under water will be stressed through reduced gas exchange capacity and reduced light availability.
The need for multiple herbicide applications was also demonstrated by Bowmer et al. (1991), who tested at least 100 combinations of herbicides in ten experiments in replicated damp terrestrial plots over three years in New South Wales, Australia. They found that no single herbicide application eradicated alligator weed; repeat applications were always necessary. The best treatment regimes were 1) a single application of dichlobenil followed by metsulfuron or glyphosate nine months later or 2) three applications of metsulfuron or metsulfuron + glyphosate over 18 months.
The relative effectiveness of herbicides that achieved good to excellent control has been examined by a number of authors, where a selection of the herbicides glyphosate, imazapyr, metsulfuron-methyl and triclopyr triethylamine (TEA) have been tested against each other in long-term studies. Imazapyr and metsulfuron-methyl provide the best and most consistent control of alligator weed while triclopyr TEA and glyphosate are less effective.
Imazapyr and triclopyr amine were tested against alligator weed by Allen et al. (2007) in managed marshes in Alabama and Georgia, USA. Either herbicide was applied to 5 m by 5 m plots at one of three rates within the range recommended by the manufacturer in April (spring) or July (summer). In the autumn following application, both herbicides achieved almost 90% biomass reduction relative to controls when applied in summer and there was no difference in biomass with application rate. Only the high rate triclopyr and mid and high rate imazapyr still had significantly less biomass than controls by the second autumn after summer application (78, 88 and 99%, respectively). For the spring application, only the high rate of imazapyr resulted in a statistically significant biomass reduction by the first autumn (84%, relative to controls) and none of the treatments resulted in control by the second autumn after application.
Hofstra and Champion (2010) treated plants grown outdoors rooted in tanks with a single application of imazapyr, metsulfuron-methyl, triclopyr TEA or triclopyr TEA + picloram. For young plants there was regrowth observed for all treatments, beginning at 2.5 MAT (months after treatment) depending on treatment, but this was not substantial for imazapyr and triclopyr + picloram until 10-11 MAT (the spring/summer following summer application). Despite this apparent difference between treatments, reductions in biomass did not differ greatly between any of the treatments one year after application (ranging from a 75 to 90% reduction compared to controls). Two applications to mature plants of metsulfuron-methyl and imazapyr resulted in a 75 to 80% reduction in biomass, relative to controls, while biomass was not reduced for glyphosate and triclopyr TEA. Single applications of all four herbicides did not reduce biomass, relative to controls.
Schooler et al. (2008) investigated the relative effectiveness of broad spectrum (glyphosate) and selective (metsulfuron-methyl and triclopyr TEA) herbicides in moist terrestrial situations. Each herbicide was applied at two different frequency regimes (December, January, March and June or December, January and March). There was no difference in efficacy between metsulfuron-methyl and triclopyr TEA. Although, in the long-term there was no difference between the four- and three-application regimes, the broad spectrum herbicide reduced below ground biomass by ~80%, while the selective herbicides reduced below ground biomass by 95%. Above ground biomass was reduced by ~90% by the selective herbicides but not reduced by the broad spectrum herbicide, relative to the control. An important finding of this study is that even the poorest performing herbicide regime (3 applications of glyphosate) that was tested, which resulted in above ground biomass no different to controls, resulted in an 81% reduction of below ground biomass, 15 months after herbicide application. Reduction in alligator weed below ground biomass after herbicide application has also been demonstrated by Schooler et al. (2007) and Hofstra and Champion (2010).
The studies of Hofstra and Champion (2010) and Schooler et al. (2008) corroborate the recommendations of Champion (2008), van Oosterhout (2007), Sainty et al. (1998) and Bowmer et al. (1991) who all recommend multiple annual metsulfuron-methyl treatments of alligator weed in New Zealand and Australia. Hofstra and Champion (2010) found that the second application of glyphosate did not provide improved control of alligator weed; again this corroborates field experience in New Zealand (Champion 2002) and Australia (van Oosterhout 2007), where glyphosate does not provide adequate control in the long-term for rooted alligator weed.
The findings of Schooler et al. (2008) provide credibility to eradication programs in Australia and New Zealand that rely on a resource depletion strategy, whereby the below ground portions are gradually depleted with successive herbicide applications, ultimately aiming to eradicate the plant (see Control Programs). They also suggest that, despite the renowned poor translocation of herbicides to below ground parts of alligator weed (see Translocation of Herbicides), herbicide regimes can still be effective in reducing below ground biomass. Whether this is due to damage to the roots by the phytotoxicity of the herbicide itself or depletion of carbohydrates as the plant repeatedly re-establishes shoots and leaves is unknown. They also demonstrate the importance of determining below ground biomass, which is a key carbohydrate reserve and source of regenerating propagules. The importance of reducing below ground material was recognised by Bowmer et al. (1991) who determined the reductions in below ground biomass two months after treatment with sulfometuron-methyl, dichlobenil, glyphosate, glyphosate + clopyralid and imazapyr was 96, 91, 80, 29 and 16%, respectively, relative to controls. Remaining roots were also collected, cut at the internodes and then grown in potting mix. They found that all treatments reduced the viability of the underground biomass by 99, 94, 89, 88 and 78% relative to controls, respectively (which had 30,412 fragments m-2). These studies provide support for the gradual depletion approach to controlling or eradicating alligator weed. Given no other field studies have reported the impact on below ground biomass in relation to herbicide application and only two mesocosm studies do (Hofstra and Champion 2010, Schooler et al. 2007), we suggest more research would be valuable in determining the effectiveness of control programs in reducing below ground biomass.
Another factor that impacts on the usefulness of herbicides against alligator weed is stem fragmentation. Recent studies have confirmed that herbicide application creates viable alligator weed stem fragments (Dugdale et al. 2010, Clements et al. 2012). In aquatic situations, these fragments have the potential to disperse and establish new infestations. Approximately 40% of stem fragments produced after application of metsulfuron-methyl were viable compared to up to 2% after the application of glyphosate (Dugdale et al. 2010, Clements et al. 2012). Given each of these fragments has the potential to create a new infestation at an unknown downstream location, frequent applications of glyphosate on aquatic infestations may provide a better outcome than metsulfuron, in containment of infestations within catchments, even though the later kills the initial patch more effectively.
Translocation of herbicides
At least eight studies have been conducted on the absorption and translocation of radio-labelled herbicides in alligator weed (Earle et al. 1951, Funderburk and Lawrence 1963, Gangstad 1978, Bowmer et al. 1991, Bowmer and Eberbach 1993, Bowmer et al. 1993, Tucker et al. 1994, Eberbach and Bowmer 1995). These studies have demonstrated that herbicide translocation is poor within alligator weed, providing insight into why herbicide control of alligator weed is difficult with foliar applied herbicides.
The first step of herbicide translocation is absorption into the leaves. For alligator weed, absorption occurs at rates similar to other species for glyphosate and imazapyr (Tucker et al. 1994). In one study, 25 to 41% of foliar-applied glyphosate absorbed into alligator weed leaves (depending on spray droplet size, Bowmer et al. 1993), while in another study 42% was absorbed (Tucker et al. 1994). Absorption of imazapyr was much higher at 88% (Tucker et al. 1994). This occurs in alligator weed at a rate similar to other species for glyphosate and imazapyr (Tucker et al. 1994), with 25 to 41% of applied glyphosate absorbed into leaves (depending on spray droplet size, Bowmer et al. 1993) and 42 and 88% of glyphosate and imazapyr, respectively (Tucker et al. 1994). However, subsequent translocation out of the leaves to other plant parts is limited, with values ranging from 5 to 20% (Bowmer and Eberbach 1993, Bowmer et al. 1993).
Only 15% of applied imazapyr was translocated to the roots (Tucker et al. 1994), while translocation of glyphosate to roots has been reported as 0.9% (Tucker et al. 1994), 7% (Bowmer et al. 1993) and 1.6 to 3.3% (Bowmer and Eberbach 1993). This poor translocation results in low concentrations of herbicide within the roots, which can be below the phytotoxic threshold (Bowmer et al. 1993). Imazapyr is phytotoxic at lower doses than glyphosate (Tucker et al. 1994), and this combined with the increased translocation of imazapyr to roots and underground storage tissues may be the mechanism responsible for the observed susceptibility of alligator weed to imazapyr relative to glyphosate.
Given that such small amounts of herbicide reach the below ground plant parts, plant size can have a large effect on the concentration that is achieved. Bowmer and Eberbach (1993) showed that the concentration of glyphosate in tissues was smaller in large plants relative to small plants, resulting in glyphosate values below phytotoxic levels for large plants. This data suggests that although large plants provide lots of foliage for optimal herbicide interception, this benefit could be outweighed by the subsequent dilution of the adsorbed glyphosate within the plant tissues.
The results above for alligator weed are unusual, particularly for glyphosate, which in most plants is rapidly translocated from leaves to metabolic sinks, especially meristematic and storage tissues (Bowmer et al. 1993). Another species poorly controlled by glyphosate (Horsetail, Equisetum arvense L.) also has poor translocation of glyphosate to the roots, while susceptible species have high translocation to roots (for discussion see Bowmer and Eberbach 1993).
Another two factors that may reduce the efficacy of glyphosate on alligator weed are exudation by roots and metabolism to non-toxic metabolites. These were investigated by Eberbach and Bowmer (1995), who found that 29% of applied glyphosate was converted to CO2 within 14 days and released through root exudation. Although they found no evidence for breakdown of glyphosate to known metabolites in below ground tissue, they could not rule out direct breakdown into unknown metabolites (and therefore unmeasurable) or very rapid subsequent breakdown of known metabolites to unknown ones.
Translocation of other herbicides in alligator weed has also been investigated. Only ‘minute’ proportions of metsulfuron and bensulfuron were translocated to alligator weed’s below ground parts (Bowmer et al. 1991). Funderburk and Lawrence (1963) found downward translocation of a number of foliar applied herbicides was poor for alligator weed. There was ‘little’ movement of foliage applied simazine, ametryne and prometryne; ‘slight’ downward movement of fenac, diquat and paraquat; and ‘considerable’ downward movement of 2,4-D (acid) and 2,4-D butoxyethyl ester, which accumulated in the nodal portions of the stems. Despite the effective translocation of 2,4-D, it did not result in good control of the plant (55% control at 5 ppm). In contrast, Gangstad, (1978) reported more upward movement of C14 labelled 2,4-D than downward movement with Earle et al. (1951) who reported downward movement of only one to two internodes and upward movement to the top of the plants.
Zurburg et al. (1961) observed that phenoxy herbicides rapidly kill top parts of alligator weed but regrowth usually occurs within 5 to 8 weeks from viable apical buds. They collected stem material from treated alligator weed and found that regrowth occurred from those buds that were below the water line at the time of treatment, indicating that the water protects submersed stems from herbicide and that there was ineffectual translocation down from the aerial portions of the stems. This is somewhat in disagreement with Funderburk and Lawrence (1963), who report ‘considerable’ movement of 2,4-D acid and ester in alligator weed.
Observation of rapid regrowth from axillary buds on stems in herbicide-treated alligator weed (Pate et al. 1965, Langeland 1986a, Dugdale et al. 2010, Clements et al. 2012) indicates a further potential mode of tolerance to herbicides. Pate et al. (1965) found that herbicide damage by dichlobenil and dicamba did not occur in inactive axillary buds, whose vascular connections with the stem were not differentiated, apparently preventing expression of the herbicidal activity. This represents a further example where poor translocation of herbicide within alligator weed allows parts of the plant to survive herbicide application.
Rapid leaf and nodal abscission as a response to herbicide application has been observed by Dugdale et al. (2010) and Langeland (1986a) who speculated that much of the herbicide may be lost, or prevented from further translocation away from the parts of the plant that are directly exposed to the herbicide spray (i.e. under the canopy, the water or the ground), by auto-abscission of the parts of these plant tissues that have the highest herbicide concentrations. Therefore, active abscission of exposed tissues may be an additional factor that provides alligator weed with herbicide tolerance. An alternative explanation may be that, because herbicide activity is quite commonly concentrated at nodes, the herbicide damages vascular tissues here and prevents further translocation and results in stems breaking at the nodes.
In summary, there are a number of factors that probably result in the tolerance of alligator weed to herbicides: 1) poor translocation out of leaves; 2) poor translocation to roots resulting in sub-lethal tissue concentrations; 3) exudation of glyphosate (and possibly other herbicides) by underground tissues (possibly with intermediary metabolites); 4) poor translocation to quiescent buds; and 5) auto-abscission or breakage of nodal tissue with high herbicide concentrations.
All of the above translocation studies have been undertaken on foliar applied herbicides. Interestingly, Solymosy (1968) has shown that absorption and translocation of herbicide through alligator weed roots is very efficient and in some cases may be much better than foliar absorption. Such a method provides a potentially great benefit, by simultaneously circumventing the problem of poor translocation and the problem of poor herbicide coverage on underwater plant parts. Granular dichlobenil application does fit this mode of application and provides excellent control of rooted-emerged alligator weed but not mats of floating alligator weed (Weldon et al. 1968). Additional research on efficacy of herbicides taken up by roots would be valuable.
This section summarises available literature on control programs for alligator weed from around the world. The aim is to give an indication of what has been done and with what degree of success to guide weed managers in making decisions about alligator weed control. By virtue of the authors’ affiliations, more information is provided on Australian and New Zealand control programs than those from other countries. Such an approach does not allow for direct comparisons between programs or herbicides used.
United States of America
Alligator weed was first recognised as a problem at the start of the 20th century in Alabama, where it was recorded completely filling a creek in 1897 (Durden et al. 1975) and in 1901 Major William Rossel, United States Army Corps of Engineers, recognised its potential as a major obstruction to navigation and drainage. At the time, water hyacinth (Eichhornia crassipes (Mart.) Solms) was a superior competitor because it began its growth earlier in the growing season and so shaded out and retarded the alligator weed. However, after the systematic removal of water hyacinth with 2,4-D, alligator weed became the threat that Major Rossel had warned of (Lawrence and Funderburk 1965, Gangstad 1978). Preliminary studies were made on chemical control of water hyacinth and alligator weed in 1950 and until 1975 phenoxy herbicides were widely used to control these two species. However, herbicide application for alligator weed was quite limited because the application rate required was high (and so costly) and control was never satisfactory, usually requiring retreatment after a short period (Gangstad et al. 1975), although frequent applications were considered successful at reducing the spread of alligator weed (Blackburn 1963). For example, single applications of 2,4-D amine salt (9 kg ha-1) gave inadequate control except where it was growing in deep water, even then requiring two applications to bring it under control (Eggler 1953). By 1975 control programs integrating herbicide (2,4-D or 2,4,5-T) with biological control (using flea beetles (Agasicles hygrophila Selman and Vogt), thrips (Amynothrips andersoni O’Neill) and moths (Vogtia malloi Pastrana)) had been established (Durden et al. 1975). Fortunately, good control of floating alligator weed was achieved with the flea beetle, particularly when integrated with herbicide programs (Durden et al. 1975). Attempts to establish the alligator weed flea beetle have been unsuccessful in North Carolina due to cold sensitivity (Langeland 1986a). This means that floating aquatic alligator weed poses a much greater problem in these areas and herbicide control programs are required for the aquatic ecotype.
In the 1960’s repeat applications of picloram and picloram + 2,4-D were used in Louisiana (Lapham 1966). Then, 2,4,5-T (PGBEE) was considered the only consistently effective herbicide for control of alligator weed. Approximately 27,000 kg active ingredient (a.i.) were used in 1978 to treat ~4,000 ha of alligator weed. However, because of concerns over its environmental and health risks, all uses of it were banned by the United States Environmental Protection Agency in 1983 (Gangstad 1984).
Langeland (1986a) reviewed the alligator weed control program in North Carolina. He reports 2,4,5-T (PGBEE) has been effective in controlling alligator weed in many areas but not in North Carolina. Based on an assessment of 3,000 herbicide treatments that were evaluated over a five year period in North and South Carolina; 2,4,5-T was the most consistent of these, but still provided less than 50% control when applied as a spray solution. Granular application of 2,4,5-T gave ‘acceptable’ control when alligator weed was emersed, flowering and rooted. ‘Excellent’ control was achieved with 2,4-D and fenac applied in combination but only when applied 8-10 weeks before frost and then flooded just after treatment. Other herbicides to be evaluated in the program were: 2,4,5-T; 2,4-D; MCPA; [2(2,4-DP); dalapon]; silvex esters and amides; fenac; amitrole and 2,3,6-TBA; TCA; AMS; endothall; simazine; neburon; fenuron; monuron; diuron; monuron TCA; fenuron TCA and erbon. The results were erratic with effective control only occurring when flooding occurred shortly after herbicide application. Control was 90 to 100% for maleic hydrazide, amitrole, dalapon, erbon, silvex, MCPA and 2,4-D acetamide but highly ineffective in areas not subject to flooding (Langeland 1986a). This suggests that the stress of being flooded acts in combination with the stress provided by the herbicide. Stresses likely to be caused by flooding include anoxia, reduced light and therefore photosynthetic capacity, and reduced and altered gas exchange.
Later herbicide evaluations in North Carolina indicated that glyphosate could be used in management programs to control alligator weed growing in lakes and rivers, while imazapyr was most suitable for non-irrigation ditch banks (Langeland 1986a). Preliminary investigations also found that triclopyr was very effective and could be useful for aquatic sites and ditch banks (Langeland 1986a). Langeland (1986a) recommended two applications of glyphosate in the first year of treatment, at recommended rates, in lakes and rivers. For ditch banks, he recommended imazapyr at 0.06 to 0.12 g a.e. (acid equivalent) 100 L-1 be applied on a spray to wet basis. He suggested that with two consecutive applications, alligator weed will be “essentially eliminated” from some ditch banks with re-colonisation occurring from dispersal of vegetative propagules. After this initial effort an annual visit, at most, to spot spray would be required. Triclopyr amine and imazapyr are currently used in control programs to suppress alligator weed (Allen et al. 2007).
In New Zealand, alligator weed control is undertaken for suppression or eradication, depending on location, and most control is carried out with herbicides. Metsulfuron-methyl is used at aquatic and urban sites, glyphosate is used for the aquatic sites (often in conjunction with metsulfuron-methyl) and triclopyr ethoxyethyl ester with picloram at pasture and cropping sites. The alligator weed management programs are reducing both the number and extent of alligator weed sites, although it is still regarded as a problematic plant by waterway managers with limited or ineffective control options (Hofstra and Champion 2010). There are no products that provide effective reduction of alligator weed biomass over a season with one application (Hofstra and Champion 2010) so the control programs require two or three herbicide applications per year to be effective (Champion 2008). A key to the success of this program is that site-specific management plans are developed, with control methods differing depending on whether sites are aquatic or terrestrial, urban or rural, cropping or pasture, risk of spread and potential vectors. Other key drivers of success are the experience of the personnel contributing to the program, sufficient funding, relevant permits to allow appropriate herbicides to be used, and use of legislation to prevent movement of contaminated soil, machinery or garden waste (Champion 2008). Aquatic sites are treated with aerial, boat or land-based hose applications of metsulfuron-methyl or glyphosate. Urban sites are treated by injecting metsulfuron-methyl into the hollow stems to provide control without damaging surrounding vegetation. The propriety mixture of picloram and triclopyr ethoxyethyl ester gives the best results in terrestrial situations, with significant reductions in abundance at all sites treated. Wherever an intensive program of treatment and follow-up has been implemented, the near eradication of alligator weed has been achieved over a two to three year period. Careful manual excavation of remaining tap roots at terrestrial sites is recommended when sites are reduced to the last few remaining plants (Champion 2008).
In Australia, control of alligator weed with herbicide is the most commonly used method, while physical and mechanical removal are also widely adopted (Burgin et al. 2010). Despite widespread and intensive control in the state of New South Wales, there was an increase in the number and extent of infestations between 2001 and 2007 (Burgin et al. 2010). Control of aquatic alligator weed is achieved with the alligator weed flea beetle in large infestations with permanent water in warmer areas of Australia but it fails in temperate areas (i.e. Victoria and southern New South Wales), in small or ephemeral waterways and in terrestrial situations (van Oosterhout 2007). In some situations, where alligator weed is present in a small part of its potential range, control programs may aim to eradicate all alligator weed from a region or jurisdiction. This represents a much more difficult target than maintaining biomass at low levels.
In efforts to improve alligator weed control, multiple herbicide applications per year are used (van Oosterhout 2007) in a resource depletion strategy. A comprehensive management strategy is in place in Australia where infestations are regarded as ‘core’ or ‘non-core’ (van Oosterhout 2007). The former are areas where eradication is not feasible and they are managed to reduce abundance and prevent further spread. Aquatic infestations in the warmer climate ‘core’ areas are suppressed by the flea beetle. For ‘non-core’ areas, suppression leading to eradication or immediate eradication are the goals. Manual excavation is recommended for small patches, while larger patches should be sprayed with a broad spectrum herbicide (to allow easy detection of alligator weed regrowth) and retreated at least three times per year. Complete eradication of these sites has proven difficult so excavation of the remnant plants is suggested once the infestation becomes small. In terrestrial areas, selective herbicides are recommended to encourage competition from monocots.
The eradication approach in Australia acknowledges that translocation of herbicide to the underground portions of the plant is poor and regrowth will occur. The eradication approach is to damage above ground stems by repeated and frequent herbicide application with three applications of metsulfuron-methyl per year, typically for six years. After each herbicide application it is expected that the plant will respond by establishing new shoot growth, fuelled by energy and nutrient reserves in the underground roots and rhizomes. Because the reserves are depleted regrowth may be slow after two years, and from then on only two applications per year may be possible. If each successive application is applied before there is significant downward translocation of carbohydrates and nutrients, then the plant will eventually exhaust its reserves and die (van Oosterhout 2007). However, if the frequency of herbicide application is such that it always or occasionally allows replenishment of the underground parts, eradication will not be achieved. Although we are unable to say in general terms how many applications are required over what period to achieve sufficient depletion to kill an infestation, nor how this might vary in different climatic zones, evidence of nutrient and biomass depletion in roots has been obtained by Schooler et al. (2007, 2008, 2010).
Schooler et al. (2008) also provides evidence for the resource depletion strategy where suppression is the primary target in certain ‘non-core’ areas. Here the regular application of dicotyledon-specific herbicides is recommended. Schooler et al. have shown in a temperate pasture situation that dicotyledon-specific herbicides (metsulfuron-methyl and triclopyr triethylamine) resulted in a greater reduction in alligator weed biomass over the long term (15 months after final treatment) and an increase in monocot biomass compared to the broad spectrum herbicide used (glyphosate). This was attributed to competition between the grasses and the alligator weed and suggests that prevention of spread may be enhanced when using dicotyledon-specific herbicides. Further, in terms of suppressing alligator weed, they found that there was no advantage of four applications over three, nor was there an advantage of using high rates over low, so they recommended applying low concentrations of selective herbicides relatively infrequently (one to three times per year).
Cook and van Oosterhout (2008) provide a technique to suppress alligator weed in pasture based on tipping the competitive advantage from alligator weed to pasture species. It is achieved by reducing the canopy height by slashing or grazing, thus promoting new growth of alligator weed and more importantly opening the pasture canopy to allow greater herbicide coverage on the alligator weed; metsulfuron-methyl is then applied 2-4 weeks later, which kills much of the alligator weed and allows the pasture species to increase in abundance. This process is repeated three times per year and can maintain alligator weed cover <10%. A prerequisite for success is that there must be at least 10% cover of a stoloniferous grass that grows to at least 30 cm tall (i.e. kikuyu (Pennisetum clandestinum Hochst. ex Chiov.), Swazi grass (an improved variety of Digitaria didactyla Willd., formerly D. swazilandensis Stent) and Rhodes grass (Chloris gayana Knuth.), to provide strong competion to the alligator weed.
Based on success in the USA (e.g. Allen et al. 2007), a permit has recently been granted to use imazapyr in New South Wales to control alligator weed in terrestrial situations, but it is too early to determine how successful this has been. Glyphosate and dichlobenil have also been used for eradication and suppression programs in Australia, but these are not recommended for eradication programs. Glyphosate has not been as effective as metsulfuron-methyl, and dichlobenil suppresses alligator weed growth near the soil surface, but when the effect of this residual herbicide ceases, regrowth occurs from material below the strata that the herbicide penetrates to. A mix of glyphosate and metsulfuron has also been used in the past to little effect (van Oosterhout 2007).
In Victoria, initial treatment programs in urban backyards usually used dichlobenil (4.05 kg a.i. ha-1), but metsulfuron-methyl (0.06 g a.i. L-1), glyphosate (3.6 g a.i. L-1), or a mixture of the later two herbicides were also used (Gunasekera and Adair 1999). The treatment program began in 1997/1998 and 65 to 75% of 800 infestations were successfully “eradicated” by a single treatment with dichlobenil (eradication was defined as the weed not being present on a subsequent visit, Gunasekera et al. 2006). By 2008, only five backyard sites out of a total of 805 were thought to contain regrowth following initial treatment in 2006/2007 (Gunasekera and Bonilla 2008). However, a high proportion (>50%) of these sites were found to still have alligator weed in subsequent site visits during 2009/2010 (E. Cox, Department of Primary Industries Victoria, personal communication). Despite being implicated in eradciation of ~50% of Victroian sites, dichlobenil is currently not recommended as a tool for the eradication of alligator weed because it is a residual herbicide, suppressing the subterranean roots for some time, rather than killing it, and resulting in the site being declared eradicated (van Oosterhout 2007). However, its residual nature may be an advantage, because it needs to be applied less frequently, and all of the few efficacy studies in the literature report excellent medium and long term control (Table 1). Therefore, with proper monitoring programs this effective herbicide need not be excluded from use.
Control of alligator weed infestations in urban backyards has been undertaken in the Hawkesbury-Nepean Catchment, New South Wales. In 2004, re-inspections of 66 of the backyards where alligator weed has been controlled in the past revealed 30 infestations that were still present (45%). Metsulfuron-methyl was used to control alligator weed in 80% of these cases, with glyphosate and a combination of MCPA and dicamba also used. All three herbicides were effective in controlling the above ground parts of the plant in the short term but no information is provided regarding long term control (Meyer et al. 2007).
For aquatic sites, aquatic formulation glyphosate is a good candidate for alligator weed control. The reasons for this are that it: is effective at killing other perennial weeds with underground storage organs; is generally translocated within a few hours from the foliage to the roots, has low aquatic toxicity; and rapidly deactivates in the soil (Bowmer et al. 1993). Indeed, it has proven effective at aquatic sites in the USA (Langeland 1986a) and at controlling floating mats of alligator weed in Australia. Free floating alligator weed was controlled effectively using standard rates of glyphosate (3.6 g a.i. L-1) in Botany Wetlands in Sydney and at Barren Box swamp in Griffith (both New South Wales; Sainty et al. 1998). Glyphosate is far less effective on terrestrial or rooted alligator weed, which have an extensive underground root system making them harder to control (Cook and Storrie 2008, Anonymous 1994). Bowmer et al. (1993) used glyphosate at high volume (400 L ha-1 or up to 3.2 kg ha-1) in late spring, late summer or autumn. Excellent short term control was achieved, but regrowth always occurred.
At a high value site in Sydney that consisted of 11 interconnected ponds and dams, spot-treatments of herbicide were applied at regular intervals for ten years. On average, application occurred at monthly intervals during cooler months and bi-monthly during warmer months (September to May). Aquatic infestations have been treated with 3.6 g a.i. L-1 glyphosate while terrestrial infestations have been treated with metsulfuron-methyl (rate not specified). Where possible, water levels were lowered to increase the amount of alligator weed exposed to herbicides. In addition, manual removal of floating mats occurred. Over ten years this has resulted in an estimated 90 to 95% control of the initial alligator weed infestation (from 5,000 m2 to less than 500 m2, Chandrasena and Pinto 2007). The authors suggested that poor control was achieved with glyphosate because of inadequate plant uptake. Often there were few leaves exposed above the water to apply herbicide to, access was difficult in soft mud and poor retention of spray on the leaves was observed. Such sub-lethal doses caused stem fragmentation that increased spread. Apparently, much better control was observed with metsulfuron-methyl. The description of the alligator weed at this site indicates that the remaining alligator weed was rooted into the substrate and therefore it was expected to be much more difficult to control than the floating mats described by Sainty et al. (1998).
In Victoria, the application of glyphosate (3.6 g a.i. L-1) at two-monthly intervals during the growing season over several years was generally needed to eradicate an average sized infestation (10 m2) growing along the margins of watercourses (Gunasekera et al. 2006; for these naturalised infestations, unlike the previously mentioned program from this paper, “eradication” was considered achieved when two repeat visits failed to find alligator weed). However, many sites in Victoria remain intact, though smaller in size. A recent review of the program has shown that in all cases regular applications of glyphosate or metsulfuron-methyl resulted in substantial reductions in abundance and cover but that eradication was not achieved (Dugdale et al. 2008). Like Sainty et al. (1998), Dugdale et al. (2008) found the problem lay in finding and killing the last remaining plants (which may be dormant roots or shooting plants hidden by competing vegetation).
Although glyphosate and metsulfuron together provide good control of alligator weed in aquatic and terrestrial situations, eradication of infestations remains problematic. Julien and Bourne (1988) report that six of 33 sites were eradicated, however the importance of herbicide in achieving this is probably minor. Four of these eradications were achieved by excavation, one by a combination of residual herbicide and excavation, and one by a single application of herbicide (dicamba). The later site was a creek bed site of 50 m2 and scouring of the creek bed was suspected in contributing to its eradication. It should be noted that the information provided in Julien and Bourne (1988) is not detailed enough to know if the herbicides were used appropriately, i.e. applied repeatedly over several years in a resource depletion strategy. Therefore, we do not know if eradcaition failed at the 28 other sites that were treated with herbicide because the herbicides (the actual herbicides used were not listed) were ineffective or because they were not applied properly. As already stated, a major problem with achieving eradication is reliably finding and killing the last plants.
A number of similar case histories were summarised by Sainty et al. (1998). For example, at Woomargama, New South Wales, glyphosate was applied repeatedly for 20 years but shoots continued to sprout (the frequency of application was not provided by Sainty et al. 1998). This site remains under a control program. Other herbicides applied there included 2,4,5-T, flupropanate and tetrapion (Julien and Bourne 1988). At Barren Box swamp near Griffith, many hectares of free-floating alligator weed were controlled using a single application of glyphosate in the mid 1990’s. Rooted alligator weed growing along the 22 km perimeter of the swamp required three herbicide applications per year for three growing seasons. In the first two years, a tank mix of glyphosate and metsulfuron at recommended rates was used to kill Typha spp. that the alligator weed was growing amongst. In the third year, metsulfuron was used alone. Alligator weed has not yet been eradicated from this site.
In response to poor control in Australia, Chandrasena et al. (2004) suggested that site specific management plans need to be developed for each site to achieve better control. This was enacted in New Zealand, and where they exist, excellent control of alligator weed was achieved (Champion 2008).
Although significant resources are directed to alligator weed management in China (US$72 million per annum; Liu and Diamond 2005) there are no publications that describe control programs.
The control of alligator weed with herbicide is challenging. Alligator weed has an extensive underground root system and one of the key impediments to successful control is poor translocation of foliar applied herbicides to below ground portions of the plant. There are a number of factors that probably result in the tolerance of alligator weed to herbicides, these being: 1) poor translocation out of leaves; 2) poor translocation to roots resulting in sub-lethal tissue concentrations; 3) exudation of glyphosate (and possibly other herbicides) by underground tissues (possibly with intermediary metabolites); 4) poor translocation to quiescent buds; and 5) auto-abscission or breakage of nodal tissue with high herbicide concentrations. There is some evidence that soil-applied herbicide provides better control than foiliar-applied herbicide, because the issue of poor translocation is circumvented. However, additional research is required to confirm this.
The efficacy of forty-nine herbicides, or herbicide combinations, have been reported in the literature, but many more have been tested without rigour or sufficient clarity to be included here. Floating aquatic alligator weed can be controlled effectively with glyphosate (Langeland 1986a, Sainty et al. 1998, Chandrasena and Pinto 2007), although management of subsequent fragments may be needed (Dugdale et al. 2010, Clements et al. 2012). However, almost universally, control of rooted alligator weed has been difficult to achieve, with most herbicides applied once or infrequently only scorching the above ground or above water parts of the plant. Regeneration proceeds rapidly, typically from roots and rhizomes that are protected from herbicide application (by the soil or water shielding them and poor translocation through the plant). Substantial biomass reductions are achieved when multiple applications of herbicide are made per year with the best herbicides being imazapyr and metsulfuron-methyl, although satisfactory results have also been reported with glyphosate, and triclopyr TEA. Despite the wealth of information available, it is difficult to determine which herbicide is the best because, the impacts of herbicides on key alligator weed response metrics (above ground biomass, below ground biomass and stem and root viability) are unknown for many herbicides. For example, it is not known if viable stem fragments are produced after application for herbicides other than glyphosate and metsulfuron-methyl, and the effect of single or multiple herbicide applications on below ground biomass and root viability is not well documented for most herbicides. Furthermore, how these differ for aquatic versus terrestrial alligator weed is often not well understood.
Control programs rely on multiple applications. Current management approaches are based on a resource depletion strategy. Repeat herbicide applications are made at short time intervals to kill re-sprouting alligator weed before it has begun replenishing its underground reserves, thus eventually exhausting the plant. Although there is limited information on the reduction of below ground biomass following herbicide application, the studies of Schooler et al. (2008) and Bowmer et al. (1991) provide strong evidence that the resource depletion approach to eradicating terrestrial alligator weed is effective at reducing below ground biomass and viability. Despite this, eradication remains difficult to achieve with herbicides alone and mechanical removal of the last remaining plants is sometimes required.
We thank Melbourne Water and the Department of Primary Industries Victoria for funding this project and Trevor Hunt and Daniel Clements for reviewing earlier drafts of the manuscript. Two anonymous reviewers provided valuable suggestions that improved this paper.
Allen, S.L., Hepp, G.R. and Miller, J.H. (2007). Use of herbicides to control alligator weed and restore native plants in managed marshes. Wetlands 27, 739-48.
Anonymous (1994). Alligator weed demands guerrilla tactics. ECOS 81, 10-4.
Bassett, I.E., Beggs, J.R. and Paynter, Q. (2010). Decomposition dynamics of invasive alligator weed compared with native sedges in a Northland lake. New Zealand Journal of Ecology 34, 324-31.
Blackburn, R.D. (1963). Evaluating herbicides against aquatic weeds. Weeds 11, 21-5.
Blackburn, R.D. and Durden, W. (1974). Control of terrestrial and rooted alligatorweed. Procceedings of the Southern Weed Science Conference 22, 283.
Bourke, C.A. and Rayward, D. (2003). Photosensitisation in dairy cattle grazing alligator weed (Alternanthera philoxeroides) infested pastures. Australian Veterinary Journal 81, 361-2.
Bowmer, K.H. and Eberbach, P.L. (1993). Uptake and translocation of l4C-glyphosate in Alternanthera philoxeroides (Mart.) Griseb. (alligator weed). II. Effect of plant size and photoperiod. Weed Research 33, 59-67.
Bowmer, K.H., Eberbach, P.L. and McCorkelle, G. (1993). Uptake and translocation of l4C-glyphosate in Alternanthera philoxeroides (Mart.) Griseb. (alligator weed). I. Rhizome concentrations required for inhibition. Weed Research 33, 53-7.
Bowmer, K.H., McCorkelle, G. and Eberbach, P.L. (1991). Alligator Weed Control Project 86/85. Final Report 31 January 1991. National Water Research Program. Commonwealth Scientific and Industrial Research Organisation, Division of Water Resources, Griffith, New South Wales.
Boyd, C.E. (1987). Evapotranspiration/evaporation (E/Eo) ratios of aquatic plants. Journal of Aquatic Plant Management 25, 1-3.
Burgin, S., Norris, A. and Karlson, D. (2010). Alternanthera philoxeroides in New South Wales, Australia: Are we closer to control of alligator weed? Weed Technology 24, 121-6.
Champion, P.D. (2002). Alligator weed in Lake Whangape and implications for its containment in the Waikato Region. National Institute of Water and Atmospheric Research, NIWA Client Report EVW02220.
Champion, P.D. (2008). Review of alligator weed (Alternanthera philoxeroides) management in the Waikato region. National Institute of Water and Atmospheric Research, NIWA report HAM2008-001.
Chandrasena, N., Gunasekera, L. and Coventry, R. (2004). Improved strategic planning and management of alligator weed (Alternanthera philoxeroides (Mart.) Griseb.) – myth or reality? Proceedings of the 14th Australian Weeds Conference, eds B.M. Sindel and S.B. Johnson, pp. 214-7. (Weed Society of New South Wales, Wagga Wagga, New South Wales).
Chandrasena, N. and Pinto, L. (2007). Integrated management of alligator weed [Alternanthera philoxeroides (Mart.) Griseb] at Botany wetlands, Sydney – A case study. Proceedings of the 21st Asia-Pacific Weed Science Society Conference, eds B. Marambe, U.R. Sangkkara, W.M. De Costa and A.S.K. Abeysekara, pp. 59-64. (University of Peradeniya and Chemical Industries, Colombo, Sri Lanka).
Cheeseman, T.F. (1906). ‘Manual of the New Zealand flora’. (Government Printer, Wellington).
Clements, D., Dugdale, T.M. and Butler, K.L. (2012). Using plant growth regulators to limit herbicide-induced stem fragmentation of aquatic alligatorweed (Alternanthera philoxeroides). Weed Technology 26, 89-94.
Clements, D., Dugdale, T.M. and Hunt, T.D. (2011). Growth of aquatic alligator weed (Alternanthera philoxeroides) over 5 years in south-east Australia. Aquatic Invasions 6, 77-82.
Cook, T. and Storrie, A. (2008). Tactics for the control and possible eradication of terrestrial alligator weed (Alternanthera philoxeroides). Proceedings of the 16th Australian Weeds Conference, eds R.D. van Klinken, V.A. Osten, F.D. Panetta and J.C. Scanlan, pp. 469-71. (Queensland Weeds Society, Cairns).
Cook, T. and van Oosterhout, E. (2008). Suppresion of alligator weed in pastures: For areas of NSW where alligator weed is declared a Class 3 noxious weed. Primefact 726. New South Wales Department of Primary Industries, Orange.
Dugdale, T.M., Clements, D., Hunt, T.D. and Butler, K.L. (2010). Alligatorweed produces viable stem fragments in response to herbicide treatment. Journal of Aquatic Plant Management 48, 84-91.
Dugdale, T.M., Hunt, T.D., Clements, D. and Gunasekera, L. (2008). Evaluation of aquatic alligator weed (Alternanthera philoxeroides) eradication in Victoria. (Department of Primary Industries, Primary Industries Research, Victoria).
Durden, W.C., Blackburn, R.D. and Gangstad, E.O. (1975). Control program for alligatorweed in the southeastern states. Journal of Aquatic Plant Management 13, 27-30.
Earle, T.T., Riess, K. and Hidalgo, J. (1951). Tracer studies with alligatorweed using 2,4-D-C14. Science 114, 695-6.
Eberbach, P.L. and Bowmer, K.H. (1995). Conversion of l4C-glyphosate to carbon dioxide by alligator weed. Journal of Aquatic Plant Management 33, 27-9.
Eggler, W.A. (1953). The use of 2,4-D in the control of water hyacinth and alligator weed in the Mississippi Delta, with certain ecological implications. Ecology 34, 409-14.
Emerine, S.E., Richardson, R.J., True, S.L., West, A.M. and Roten, R.L. (2010). Greenhouse response of six aquatic invasive weeds to imazamox. Journal of Aquatic Plant Management 48, 105-11.
EPPO, European Plant Protection Organisation. (2012). Alternanthera philoxeroides (Amaranthaceae) Alligatorweed. http://www.eppo.int/INVASIVE_PLANTS/iap_list/Alternanthera_philoxeroides.htm (accessed 15 August 2012).
Funderburk, H.H. and Lawrence, J.M. (1963). Absorption and translocation of radioactive herbicides in submersed and emersed aquatic weeds. Weed Research 3, 304-11.
Gangstad, E.O. (1978). ‘Weed control methods for river basin management’. (CRC Press, Florida).
Gangstad, E.O. (1984). Aquatic use pattern of silvex cancellation. Journal of Aquatic Plant Management 22, 78-80.
Gangstad, E.O., Spencer, N.R. and Foret, J.A. (1975). Towards integrated control of alligator weed. Hyacinth Control Journal 13, 30-3.
Garbari, F. and Pedulla, M.L. (2001). Alternanthera philoxeroides (Mart.) Griseb. (Amaranthaceae), a new species for the exotic flora of Italy. Webbia 56, 139-43.
Geng, Y.-P., Pan, X.-Y., Yuan, C., Zhang, W.-J., Li, B., Chen, J.-K., Lu, B.-R. and Song, Z.-P. (2007). Phenotypic plasticity rather than locally adapted ecotypes allows the invasive alligator weed to colonize a wide range of habitats. Biological Invasions 9, 245-56.
Gunasekera, L. and Adair, R. (1999). The alligator weed battle in Victoria. Proceedings of the 12th Australian Weeds Conference, eds A.C. Bishop, M. Boersma and C.D. Barnes, pp. 547-50 (Tasmanian Weed Society, Hobart).
Gunasekera, L., Ainsworth, N. and Bonilla, J. (2006). Alligator weed eradication in Victoria – a developing success story. Proceedings of the 15th Australian Weeds Conference, eds C. Preston, J.H. Watts and N.D. Crossman, pp. 784-6. (Weed Management Society of South Australia, Adelaide).
Gunasekera, L. and Bonilla, J. (2001). Alligator weed: tasty vegetable in Australian backyards? Journal of Aquatic Plant Management 39, 17-20.
Gunasekera, L. and Bonilla, J. (2008). Alligator weed: from backyards to nature in Victoria. Proceedings of the 16th Australian Weeds Conference, eds R.D. van Klinken, V.A. Osten, F.D. Panetta and J.C. Scanlan, pp. 358-9. (Queensland Weeds Society, Cairns).
Hofstra, D.E. and Champion, P.D. (2010). Herbicide trials for the control of alligatorweed. Journal of Aquatic Plant Management 48, 79-83.
Holm, L., Doll, J., Holm, J., Pancho, J. and Herberger, J. (1997). ‘World Weeds: Natural histories and distribution’. (John Wiley and Sons, New York).
Julien, M.H. and Bourne, A.S. (1988). Alligator weed is spreading in Australia. Plant Protection Quarterly 3, 91-6.
Julien, M.H., Bourne, A.S. and Low, V.H.K. (1992). Growth of the weed Alternanthera philoxeroides (Martius) Grisebach, (alligator weed) in aquatic and terrestrial habitats in Australia. Plant Protection Quarterly 7, 102-8.
Julien, M.H. and Broadbent, J.E. (1980). The biology of Australian weeds 3. Alternanthera philoxeroides (Mart.) Griseb. The Journal of the Australian Institute of Agricultural Science 46, 150-5.
Julien, M.H., Skarratt, B. and Maywald, G.F. (1995). Potential geographical distribution of alligator weed and its biological control by Agasicles hygrophila. Journal of Aquatic Plant Management 33, 55-60.
Kay, S.H. (1992). Response of two alligatorweed biotypes to quinclorac. Journal of Aquatic Plant Management 30, 35-40.
Kay, S.H. (1999). Evaluation of SP1001 (pelargonic acid) in combination with glyphosate on cattail and alligator weed. Journal of Aquatic Plant Management 37, 29-31.
Kay, S.H. and Haller, W.T. (1982). Evidence for the existence of distinct alligatorweed biotypes. Journal of Aquatic Plant Management 20, 37-41.
Langeland, K.A. (1986a). Management program for alligatorweed in North Carolina. UNC-WRRI-86-224. United States Geological Survey Project No. 04. Water Resources Research Institute Project No. 70039.
Langeland, K.A. (1986b). Evaluation of herbicides for the control of rooted alligator weed (Alternanthera philoxeroides). Proceedings of Southern Weed Science Conference 39, 384.
Lapham, V.T. (1964). Notes of the control of alligator weed using dichlobenil. Proceedings of the Southern Weed Science Conference 17, 325-7.
Lapham, V.T. (1966). Control of alligator weed with picloram. Proceedings of the Southern Weed Science Conference 19, 409-13.
Lawrence, J.M. and Funderburk, H.H. (1965). Effectiveness of herbicidal combinations on alligator weed growing in flowing water. Proceedings of the Southern Weed Science Conference 18, 558-566.
Li, J. and Ye, W.-H. (2006). Genetic diversity of alligator weed ecotypes is not the reason for their different responses to biological control. Aquatic Botany 85, 155-8.
Liu, J. and Diamond, J. (2005). China’s environment in a globalizing world. Nature 435, 1179-86.
Meyer, R., Coventry, R., Chandrasena, N. and Ferry, E. (2007). Managing backyard and creek line infestations of alligator weed (Alternanthera philoxeroides (Mart.) Griseb) in the Hawkesbury-Nepean Catchment, New South Wales, Australia. Proceedings 21st Asia-Pacific Weed Science Society Conference, eds B. Marambe, U.R. Sangkkara, W.M. De Costa and A.S.K. Abeysekara, pp. 303-7. (University of Peradeniya and Chemical Industries, Colombo, Sri Lanka).
Pan, X.Y., Zhang, H., Dong, H.Q. and Li, B. (2010). Impact of invasive alligatorweed on species diversity and composition of native insect in wetland vegetation: A field study. Journal of Aquatic Plant Management 48, 130-3.
Pate, D.A., Funderburk, H.H., Lawrence, J.M. and Davis, D.E. (1965). The effect of dichlobenil and dicamba on nodal tissues of alligatorweed. Weeds 13, 208-10.
Sainty, G., McCorkelle, G. and Julien, M. (1998). Control and spread of alligator weed Alternanthera philoxeroides (Mart.) Griseb., in Australia: lessons for other regions. Wetlands Ecology and Management 5, 195-201.
Sandberg, C.L. and Burkhalter, A.P. (1983). Alligator weed control with glyphosate. Proceedings of Southern Weed Science Society 36, 336-9.
Schooler, S.S., Cook, T., Bourne, A., Prichard, G. and Julien, M. (2008). Selective herbicides reduce alligator weed (Alternanthera philoxeroides) biomass by enhancing competition. Weed Science 56, 259-64.
Schooler, S.S., Cook, T., Prichard, G. and Yeates, A.G. (2010). Disturbance-mediated competition: the interacting roles of inundation regime and mechanical and herbicidal control in determining native and invasive plant abundance. Biological Invasions 12, 3289-98.
Schooler, S.S., Yeates, A.G., Wilson J.R.U. and Julien, M.H. (2007). Herbivory, mowing, and herbicides differently affect production and nutrient allocation of Alternanthera philoxeroides. Aquatic Botany 86, 62-8.
Sculthorpe, C.D. (1967). ‘The Biology of Aquatic Vascular Plants’. (Edward Arnold, London).
Shen, J.Y., Shen, M.Q., Wang, X.H. and Lu, Y.T. (2005). Effect of environmental factors on shoot emergence and vegetative growth of alligator weed (Alternanthera philoxeroides). Weed Science 53, 471-8.
Solymosy, S.L. (1968). Comparative root versus top treatment of Alligatorweed. Proceedings of the Southern Weed Science Conference 21, 304-8.
Spencer, J.L. (1968). The effects of herbicides on seven species of aquatic plants in the Mobile Delta. Proceedings of the Southern Weed Science Conference 21, 319-26.
Spencer, N.R. and Coulson, J.R. (1976). The biological control of alligator weed, Alternanthera philoxeroides, in the United States of America. Aquatic Botany 2, 177-90.
Steenis, J.H. and McGilvery, F.B. (1961). Environmental factors effecting the control of alligatorweed. Proceedings of the Southern Weed Science Conference 14, 289-92.
Tegan, M. (2009). Beware of the ‘alligator’. The Sunday Times October 4th. http://sundaytimes.lk/091004/Plus/plus_09.html (accessed 24 July 2012).
Tucker, T.A., Langeland, K.A. and Corbin, F.T. (1994). Absorption and translocation of 14C-imazapyr and 14C-glyphosate in alligatorweed Alternanthera philoxeroides. Weed Technology 8, 32-6.
van Oosterhout, E. (2007). Alligator weed control manual: Eradication and suppression of alligator weed (Alternanthera philoxeroides) in Australia. New South Wales Department of Primary Industries, Orange.
Wang, B.R., Li, W.G. and Wang, J.B. (2005). Genetic diversity of Alternanthera philoxeroides in China. Aquatic Botany 81, 277-83.
Weldon, L.W. and Blackburn, R.D. (1968). The control of floating aquatic weeds with ametryne. Proceedings of the Southern Weed Science Conference 21, 312-8.
Weldon, L.W., DeRigo, H.T. and Blackburn, R.D. (1968). Control of alligatorweed with dichlobenil. Proceedings of the Southern Weed Science Conference 21, 287-91.
Wilson, J.R.U., Yeates, A., Schooler, S. and Julien, M.H. (2007). Rapid response to shoot removal by the invasive wetland plant, alligator weed (Alternanthera philoxeroides). Environmental and Experimental Botany 60, 20-5.
Zeiger, C.F. (1967). Biological control of alligator weed with Agasicles n. sp. in Florida. Hyacinth Control Journal 6, 31-4.
Zurburg, F.W., Nelson, I.S. and Foret, J.A. (1961). An aspect of the herbicidal action of 2,4-D and related compounds in alligatorweed (Alternanthera philoxeroides). Proceedings of the Southern Weed Science Conference 14, 293-4.