Brazilian Water Milfoil

Myriophyllum aquaticum (Vell. Conc.) Verdc.

Synonyms - Enydria aquatica, Myriophyllum brasiliense, Myriophyllum proserpinacoides.
Family: Haloragaceae. 

Myriophyllum is from the Greek myrios meaning many and phyllon meaning leaf and refers to the many leaf segments on this plant.
Aquaticum is Latin for living in the water.
Brazilian refers to the country of origin and Milfoil is Latin meaning a thousand leaves.

Other names:

Parrots Feather refers to the feather like structure of the leaves.
Thread of Life.


A long-tangle-stemmed, mat forming, perennial, fresh water plant with many feather like leaves in rings of 4-6 with larger leaves below the water line and smaller ones above.




First leaves:


Bright green or with a waxy bloom. Leaves are finely divided and feather like both above and below the water surface. Arranged in rings of 4-6 around the stem. Rings very close together near the top but separated near the base. Submerged leaves tend to rot leaving bare underwater stems.
Stipules -
Petiole -
Blade - Parallel sided, egg shaped or oval in outline.
Submerged leaves 35-40 mm long, obtuse tipped, deeply cut into 25-30 thread like segments up to 7 mm long.
Emergent leaves have a waxy bloom, 15-35 mm long with 18-36 linear or awl shaped segments, 4.5-5.5 mm long.


Mats of tangled stems, yellow green, slender, 2000-5000 mm long. Hairless and smooth. Root at the nodes near the base. Stems are erect where they emerge from the water and are trailing underwater. Brittle and break off easily. Hairless.

Flower head:

Single flowers on a short stalk between 2 small bracts in the leaf axil.


Male and female flowers on separate plants usually.
In Australia all flowers are female.
Bracts - 2 small, white, awl shaped bracts, 1.2-1.5 mm long with one or two teeth or 3 lobes near the base.
Female flowers.
Have a short stalk.
Ovary - Prominent, pear shaped, 4 celled. White style and stigma with many fine translucent hairs.
Sepals - 4. White. Translucent. 0.4-1.5 mm long. Triangular.
Petals - None.
Male flowers.
May have a short stalk.
Sepals - Egg to triangular shaped, 0.7-0.8 mm long.
Petals - Yellow, weakly hooded and keeled, 3 mm long.
Stamens - 8
Anthers - Narrowly oblong.


None recorded in Australia.
Cylindric to egg shaped, 1.7 mm long with cylindric mericarps.


Doesn't produce seed in Australia.


Fibrous. Threadlike. Arise from stem nodes. Attach to the stream bed.

Key Characters:

Robust perennial plants with stems 100-5000 mm long. Leaves all whorled and pectinate. Emergent leaves with 18-36 pinnae.


Life cycle:

Perennial. Stem fragments produce roots to anchor the plant, then new stems emerge from buds. Grows mainly in summer and is semi-dormant in winter. Grows until it forms a dense mat on the water surface and waterlogged banks. Rarely grows in water more than 1.5 m deep. Flowers recorded in September, November and May.


Tolerates a wide range of temperatures including freezing. Grows best at high temperatures.
Uptake of nutrients and herbicides appears to be passive rather than active.
Increasing N, P, K, Ca, Mg and S increased growth (Domingos et al., 2005).
Carbohydrate levels are probably lowest in autumn (Madsen, 1997).
A bicarbonate medium was better than the standard Gerloff medium for growing Myriophyllum (Smith, 1993).


Vegetative reproduction only in Australia.

Flowering times:

November to April in Perth.
Spring to autumn in WA.

Seed Biology and Germination:

No seed formed in Australia.
Seeds from M. spicatum loose some viability when dried but will tolerate at least 36 weeks of drying with 53% survival (Standifer and Madsen, 1997).

Vegetative Propagules:

Stem fragments break from the parent plant to form new plants.


Other species of Myriophyllum hybridise and the hybrids can be more invasive than the parents (Bugbee and White, 2005)


Not much allelopathy (Elakovich, 1989).

Population Dynamics and Dispersal:

Local spread is by stem fragments moving with the water flow. Wave action, boats and mechanical harvesting fragment stems.
Long distance spread is by dumping of discarded aquarium plants into water ways.
Prefers slow moving or stagnant water.
It has high growth rates (Rejm?nkov?, 1992).
Growth and development of Myriophyllum spicatum. Studies in a controlled environment have shown two distinct seasonal growth forms. In summer (150C-260C) a fine hair-like root system develops accompanied by rapid top growth; in winter (140C-60C) root growth is coarse and fibrous and the slow top growth is deep red; at the higher winter temperature, adventitious roots develop and the branches then abscise, producing the material for future infestations. Control may be possible mechanically or chemically in autumn and early winter to kill existing plants (Anonymous, 1980)

Origin and History:

South America.
First recorded as naturalised in Centennial Park in Sydney in 1908.


Invasive alien species in Africa (Shaw, 2003), Australia (Arthington and Mitchell, 1986), Britain (Clarke and Newman, 2002), Japan (Muranaka et al., 2005), South Africa (Hill, 2003), USA (Pine, 1983).

Courtesy Australia's Virtual Herbarium.


Banks to water several metres deep.
It is rarely a problem in water more than 1.5 m deep (Cilliers, 1999a).
A similar species M. spicatum (Eurasian Milfoil) is more invasive in eutrophic lakes and water with high P concentrations (Cilliers, 1999a).


Warm temperate and sub-tropical regions.


Water saturated mud and gravels close to slowly moving or static fresh water.

Plant Associations:



Ornamental used in aquaria and ponds.
Used as stock feed overseas.
It is capable of quickly transforming trinitrotoluene (TNT) in contaminated water (Medina et al., 2003) and used for phytoremediation of munition sites (Hitchcock et al., 2003).
A related species M. triphyllum helps break down 1080 quickly (Ogilvie et al., 1995).
May be useful for absorbing cadmium from water (Sajwan and Ornes, 1996).
Extracts of the related Myriophyllum spicatum reduce algal growth (Nakai et al., 1996), reduce mosquito larvae (Graham et al., 1984) and reduces mosquito populations (Schultz et al., 1983).


Forms dense mats that impede water flow. Sections of floating mats break off and block irrigation and hydro-electric equipment. Invades rice fields reducing yields. Mats of weed interfere with recreational use of waterways.
Competes with native water plants and provides ideal breeding grounds for mosquitoes.
Slows stream and drain flows, which may result in flooding.
Weed of freshwater streams, dams and drains.


Not recorded as toxic.




Noxious weed WA and Tas.

Management and Control:

Prevent movement of stem fragments, especially by dumping of aquarium refuse. Control of existing infestations is highly desirable because of the possibility of explosive spread if male forms of the plant establish in Australia. Mechanical removal provides useful temporary control. Covering small dams with black plastic is effective (Toscani, 1983).
Reduce the quantity of nutrients (especially phosphorous) entering the water body.
Chlorsulfuron and 2,4-D provide better control than glyphosate.
The effect of boats and management plans on spread has been modelled (Macpherson et al., 2006) and management strategies (Dearden, 1983).
It is unlikely to establish in streams or drains with more than 13 high-flow disturbances per year (Riis and Biggs, 2003) or in areas shaded by trees (Brookes, 1987).
Herbicide trials
50 kg /ha 2,4-D granules gave good control in US on M. heterophyllum and M. spicatum without exceeding irrigation threshold of 100 ppb (Bugbee and White, 2005).
2,4 D at 1340 and 670 g/ha provided 100% control (Negrisoli et al., 2003).
Spring to autumn applications of 2,4-D butoxy ethyl ester (19%) at 227 kg/ha provided good control whereas 2,4-D Na salt at 114 kg/ha only provided control in shallow areas and poor control in deeper open areas on M. heterophyllum (Bugbee et al., 2003).
6.48 kg 2,4-D amine applied in summer and repeated 2 months later provided better control than diquat, glufosinate or glyphosate (Monteiro and Moriera, 1990).
The effect on algae after treatment depended on the species of alga present (Kobraei and White, 1996).
Severe Eurasian water milfoil injury occurred when exposed to 0.5 mg for 72 h, 1.0 mg for 36 h and 2 mg for 24 h. The threshold levels for control of Eurasian water milfoil was established in the 1.0 mg for 48 h and 2.0 mg for 36 and 48 h exposures (Green and Westerdahl, 1990).
Controlled release formulations provide good extended control (Van et al., 1986).
1.5 kg diquat + 1.5 kg 2,4-D ester lasted about 60 days (Toscani et al., 1983).
The distribution of Myriophyllum spicatum in Washington, USA, and its control with an invert oil emulsion of 2,4-D (Pine, 1983).
2,4-D did not adversely affect the non-target components in the sampled ecosystem (Couch and Nelson, 1982).
20-100 ug/L bensulfuron provided good control of M. spicatum over a 12 week period. The length of exposure was more important the concentration (Getsinger et al., 1994).
Bensulfuron methyl at 100-300 g/ha applied to dry irrigation channels controlled M. variifolium for 4-5 months after water was returned a week after application (McCorkelle et al., 1990).
200 ug/L carfentrazone did not provide adequate control (Glomski et al., 2006).
20 kg chlorfenac/ha seemed a promising chemical treatment but 3 ppm in the water was ineffective (Toscani, 1983).

EC50 = 10-9 M (Turgut et al., 2003)
18.75-37.5 g chlorsulfuron/ha containing 0.5% by vol. of a non- ionic surfactant provided effective control of Myriophyllum aquaticum. An infestation sprayed in Jan. 1983 with 40 g chlorsulfuron/ha in 500 L water plus surfactant through a fine jet spray pistol was controlled 100% with no regeneration >12 months after spraying. The herbicide was also effective against Zantedeschia aethiopica, Salvinia molesta, Eichhornia crassipes and Pistia stratiotes.
Cyanoacrylate herbicides
These have potential for Myriophyllum control (Huppatz and Phillips, 1981).
Myriophyllum spicatum and Hydrocotyle umbellata were controlled with glyphosate at 3 kg/ha and Banvel 720 at 14 L/ha., respectively (Haller and Ramaprabhu, 1981)
Diquat at 204 g a.i./ha gave good initial control but regrowth was occurring by 36 days after treatment (Negrisoli et al., 2003).
3 ppm in the water was ineffective (Toscani, 1983).
Control with 2.64 kg diquat and 1.5 kg diquat + 1.5 kg 2,4-D ester lasted about 60 days (Toscani et al., 1983).
0.9 kg hexazinone + 0.8 kg diuron controlled the weeds for >145 days (Toscani et al., 1983).
Severe M. spicatum injury (>85% biomass reduction) occurred after exposure to 0.5, 1.0, 3.0 and 5.0 mg/litre endothal for 48, 36, 18 and 12 h, respectively (Netherland et al., 1991).
Within 96 hours endothal in sediments and water falls below detectable levels (Reinert et al., 1988).
Localized applications of Aquathol® (endothal granules) used to control Myriophyllum spicatum infesting >550 acres of Pat Mayse Lake had no direct or indirect effects on non-target species or abiotic water quality, even though the weed was temporarily eradicated (Hinman et al., 1984).
Not available in Australia but is giving selective control in US (Valley et al., 2006).
3 kg fluridone/ha lasted about 60 days (Toscani et al., 1983).
1.0-2.4 kg glufosinate ammonium gave good initial control but regrew by 4 months after treatment (Monteiro and Moriera, 1990).
Generally not very effective (Negrisoli et al., 2003).
Myriophyllum spicatum were controlled with glyphosate at 3 kg/ha (Haller and Ramaprabhu, 1981).
0.9 kg hexazinone + 0.8 kg diuron controlled the weeds for >145 days (Toscani et al., 1983).
Hydrogen peroxide
20 ppm gave control of algae but little effect on Milfoil (Fowler and Barrett, 1986).
Generally not very effective (Negrisoli et al., 2003).
EC50 = 10-6 M (Turgut et al., 2003).
0.83 kg paraquat/ha controlled the weeds for >145 days (Toscani et al., 1983).
0.2 ppm gives control in static water with long exposures for Myriophyllum propinquum. Higher rates required in the field. Terbutryn residues in the water decreased with 1st order half-life of about 9-20 days (Bowmer et al., 1985).
Thifensulfuron is less effective than chlorsulfuron or metsulfuron (Turgut et al., 2003)
Tribenuron is less effective than chlorsulfuron an metsulfuron (Turgut et al., 2003)
Triclopyr is more effective than glyphosate endothal, dichlobenil, fluridone or clopyralid in NZ (Hofstra et al., 2006).
About 2 mg/L triclopyr applied in autumn provided 99% control in the first year and took about 7 days to drop to potable levels. Biodiversity doubled, non target biomass increased 5-10 times and adequate control lasted 3 years (Getsinger et al., 1997).
Triclopyr at 0.25 mg/L for 84 h and at 2.5 mg/L for 18 h reduced weed biomass by 98% (Netherland and Getsinger, 1993).
Triclopyr triethylamine salt formulation - excellent control (>85% biomass reduction) was achieved at concentration/exposure time combinations of 0.25 mg for 72 h, 0.5 mg for 48 h, 1.0 mg for 36 h, 1.5 mg for 24 h, and 2.0 and 2.5 mg for 18 h. Treatments of 2.5 mg for 2 h, 1.0 mg for 6 h, and 0.25 and 0.5 mg for 12 h were ineffective for M. spicatum (Netherland and Getsinger, 1992).
Variable success with disturbance based weed control methods (Abernethy et al., 1996).
Hand harvesting and suction harvesting needs to be repeated every 3 years and milfoil invades within a year or two of removal of benthic barriers (Boylen et al., 1996).
Mechanical harvesting reduced species diversity in 3 out of 4 sites (Nichols and Lathrop, 1994).
Harvesting for 4 years didn't reduce P levels enough to reduce growth (Painter, 1988).
Rotary hoeing reduced stem densities by 25-70% in the year following treatment (Gibbons and Gibbons, 1988).
Hand removal gave about 25% decrease in biomass the following year (Nicholson, 1981).
Dredging or covering sediments (with sand, gravel and plastic) failed to provide long-term suppression of M. exalbescens (Engel and Nichols, 1984).


Eradication strategies:

Prevent the sale and dumping of aquarium material. If possible lower water levels then treat with herbicides and repeat as required.
Hazard models for control have been developed (Dearden, 1985).

Herbicide resistance:

Biological Control:

One or two triploid grass carp can be used in Water Lily (Nymphaea) production ponds if they are removed when there is little weed left and they start eating the lilies (Santha et al., 1994).
A number of weevils and fish have some potential for biocontrol.
A leaf-feeding beetle, Lysathia (Chrysomelidae) has been introduced to South Africa for biological control (Cilliers, 1999b).
Introducing water fowl gave initial control but after 2 years recolonization ensued (Dutartre and Dubois, 1986).
Root and stem rot of parrot feather (Myriophyllum brasiliense) caused by Pythium carolinianum (Bernhardt and Duniway, 1984).

Related plants:

Coarse Water Milfoil (M. elatinoides)
Hooded Water Milfoil (M. muelleri)
Red Water Milfoil (M. verrucosum)
(M. variifolium)
There are 16 native species of Myriophyllum in WA.

Plants of similar appearance:

Other Milfoil species have serrated edges on the leaves above water, whereas Brazilian Water Milfoil has feathery leaves both above and below the water surface.


APB Advisory leaflet No. 85. (1982)1

Hussey, B.M.J., Keighery, G.J., Cousens, R.D., Dodd, J. and Lloyd, S.G. (1997). Western Weeds. A guide to the weeds of Western Australia. (Plant Protection Society of Western Australia, Perth, Western Australia). P168. Photo.

Lamp, C. and Collet, F. (1990). A Field Guide to Weeds in Australia. (Inkata Press, Melbourne).

Lazarides, M. and Hince, B. (1993). CSIRO handbook of economic plants of Australia. (CSIRO, Melbourne). #859.1.

Marchant, N.G., Wheeler, J.R., Rye, B.L., Bennett, E.M., Lander, N.S. and Macfarlane, T.D. (1987). Flora of the Perth Region. (Western Australian Herbarium, Department of Agriculture, Western Australia). P363.

Parsons, W.T. and Cuthbertson, E.G. (1992). Noxious weeds of Australia. (Inkata Press, Melbourne). P485-487. Photos.


Collated by HerbiGuide. Phone 08 98444064 or for more information.