Cookies on Plantwise Knowledge Bank

Like most websites we use cookies. This is to ensure that we give you the best experience possible.

Continuing to use means you agree to our use of cookies. If you would like to, you can learn more about the cookies we use.

Plantwise Knowledge Bank
  • Knowledge Bank home
  • Change location
Plantwise Technical Factsheet

purple loosestrife (Lythrum salicaria)


L. salicaria is a perennial herb 30-200 cm tall with a persistent woody rootstock. In North America and exceptionally in the southern limits of its native range, taller plants (up to 350 cm) can be found. Stems are erect and quadrangular in section with evenly spaced nodes in opposite pairs or in whorls of three. Leaves are 3-10 cm long, sessile, lanceolate to ovate and arising from each node (Thompson et al., 1987; Mal et al., 1992). The stem can be without lateral branches but plants usually form branches in the mid to lower part of the stem (Hegi, 1925). The length and number of lateral branches is variable, depending upon environmental conditions, probably soil nutrient status. Leaves are glabrous to pubescent on the stem and branches, or sub-tomentose on the inflorescence. Inflorescences are purple, in a dense terminal spike up to 1m long. In the first year or under poor nutrient conditions, plants usually have one shoot only, which dies at the end of growing season. In older plants, herbaceous stalks with lateral branches, each with terminal spike of flowers, arise from the rootstock to make a wide-topped crown (Thompson et al., 1987). Fruits are oblong-ovoid capsules (3-4 mm long) with two valves. Seeds are very small (200-400 µm in size, 0.5-0.6 mg in mass), thin-walled with two cotyledons and no endosperm (Thompson et al., 1987). The species is heterostylous with three distinct arrangements of pistils and stamens. The flowers are categorized according to stylar morphs as short-, medium- and long-styled (Mal et al., 1992).

Prevention and control


Cultural control

Use of flooding and competition from other plant species has been attempted with very limited success. However, flooding to a depth of up to 50 cm for two years had little effect on the stature and reproductive characteristics of L. salicaria and appeared not be a successful method of control (Malecki and Rawinski, 1985). High water level may additionally present stresses for native plant communities. Also, plant competition was only partly successful and results differed greatly depending on the plant species used. Echinochloa frumentacea (Japanese millet) and Polygonum lapathifolium (nodding smartweed) grew well and out-competed L. salicaria (Rawinski, 1982) although Echinochloa walteri (Walter's millet) and Phalaris arundinacea (reed canarygrass) were unsuccessful (Malecki and Rawinski, 1985; Balogh, 1993). Typha x glauca is another potential species on permanently flooded sites (water level always greater than 40 cm) possibly combined with damage by carp (Rawinski and Malecki, 1984). Growth form, habitat type and phenology of L. salicaria indicate that it will not be susceptible to control with fire (Thompson et al., 1987).

Mechanical control

Mechanical control, such as mowing, ploughing or hand-pulling give only limited success (Malecki and Rawinski, 1985). Small populations may be successfully controlled by hand pulling, but this method should be avoided after flowering so as not to scatter seeds and plants should be bagged at the site to avoid fragments being dropped along the exit route. Burning is the preferred method for disposal of cut or pulled plants. The date of cutting has an important role in reducing the number of shoots but does not result in permanent control.

Chemical Control

Due to the variable regulations around (de-)registration of pesticides, we are for the moment not including any specific chemical control recommendations. For further information, we recommend you visit the following resources:
- PAN pesticide database (
- Your national pesticide guide

Biological control

Since the 1990s, great attention has been paid to the biological control of L. salicaria in Canada and the USA. In 1992/3, three insects were approved for release as L. salicaria control agents after laboratory tests in Europe and North America (Hight et al., 1995), although as many as 59 species of phytophagous insects were collected on L. salicaria plants in the north-eastern USA (Hight, 1990). The released agents were two herbivorous insects, Galerucella calmariensis and Galerucella pusilla (Coleoptera: Chrysomelidae) and a root weevil Hylobius transversovittatus (Coleoptera: Curculionidae) and all three established successfully (Hight et al., 1995; Blossey et al., 2002). Malecki et al. (1993) predicted reduction of L. salicaria abundance to approximately 10% of original levels, however, it is expected that plant refugia will remain at sites with high water levels which prevent the development of the root-mining weevil and limit recruitment of the Galerucella spp, and also that plants growing in the shade will also be less likely to be attacked by these biological control agents. They were released in 1684 different wetlands in the USA invaded by L. salicaria (Blossey et al., 2001). Of these, about 50% are regularly monitored using standardized monitoring protocols to assess the impact of the biocontrol agents on L. salicaria. Released herbivore species were established in over 10 provinces of Canada and 30 states of the USA (Blossey et al., 2001) and at several sites had selectively reduced L. salicaria biomass by 95% (Blossey et al., 2002). However, 10 years after release of the two Galerucella species in the state of New York, USA, Grevstad (2006) found no significant impacts at all on L. salicaria populations. Also, mass emergence of adults of Galerucella spp. was also found to result in localized, short-term attack on native species including Decodon verticillatus, Potentilla anserina and Rosa rugosa at some sites (Blossey et al., 2001).
Integrated control

Henne et al. (2005) found that an integrated vegetation management (IVM) strategy using herbicides and biological control outperformed either technique used in isolation. An integrated approach for effective, long-term management through the integration of physical, cultural, mechanical, biological and chemical control strategies was also recommended by Woo et al. (2002).


Where L. salicaria is common, direct economic losses are caused by reduced livestock stocking rates and reduction of water flow in irrigation systems. Livestock forage value was reduced in meadows where L. salicaria had established monospecific stands and the palatability of cut hay containing L. salicaria is reduced. L. salicaria is a threat to wetland pastures that typically occur along floodplains and on the peripheries of glacial marshes in central USA and is threatening the production of wild rice (Zizania aquatica) in such areas (Thompson et al., 1987).

Summary of invasiveness

L. salicaria, an Old World native, is a highly invasive species of wetlands in North America, beginning to spread rapidly about 140 years after its accidental introduction around 1800. It is a very variable species with an ability to occupy numerous habitats and substrates with the exception of dry places. Its spread and persistence in ecosystems is supported by very high seed production, a vigorous and persistent root system and rapid growth. It is an invasive species and/or noxious weed in almost all states and provinces of Canada and the USA where it is a serious threat to many sensitive wetland ecosystems.

Hosts / species affected

As an exotic invasive species, L. salicaria is not generally a weed of agricultural land as it prefers moist to wet habitats, but it may occur at disturbed edges of crop fields in the vicinity of wetlands. The establishment of L. salicaria adjacent to stands of wild rice (Zizania aquatica) in northern California and Wisconsin, USA, means that it may be or become a pest of this crop (Thompson et al., 1987).

Zoomed image