Impatiens glandulifera Royle

Common Name: Ornamental jewelweed

Synonyms and Other Names:

Impatiens roylei Walp., Himalayan balsam, Indian balsam, purple jewelweed, Policeman’s helmet, custodian helmet, touch-me-not, Washington orchid

Copyright Info

Identification: Impatiens glandulifera is an herbaceous annual that is succulent and glabrous (smooth and hairless) and typically grows to 6.5 ft, but can reach 10 ft (Campbell et al. 2010, Forest Service 2007, Toronto Region Conservation 2012, WSNWCB 2008). The upright, hollow stems are easily broken, have a purplish /reddish tinge, and are separated by nodes (King County 2007). The leaf arrangement can be opposite or whorled, with (usually) three leaves to a node. The simple leaves are approximately six inches long, three inches wide, and oblong or elliptic (egg shaped). The leaf margins are sharply serrate with ≥ 20 teeth along each side. Each leaf has a stout petiole (stem), with small, glandular stalks found at the base of these petioles. Several solitary flowers terminate an elongated axillary stalk. The flowers are irregular, with five petals (two fused), three sepals (two fused), and five stamens with connate (fused) filaments. The overall flower shape resembles an English policeman’s helmet. The fused sepals form a spur less than 6 mm long. The flower color ranges from white to a variety of shades of pink and purple. The mature fruit is a five chambered capsule that when touched will eject many seeds. A single plant can produce and disperse 800–2,500 seeds (Beerling and Perrins 1993). The roots extend four to six inches deep, with adventitious roots found along the lower stem nodes—sometimes buttressing (WSNWCB 2008).

No other plants are likely to be confused with I. glandulifera. The native touch-me-not (Impatiens noli-tangere) is yellow flowered (Hultén 1968).

Size: 1-2 m high. Leaf length: 6-15 cm

Native Range: Native to India and the western Himalaya (WSNWCB 2008)

Map Key
This map only depicts Great Lakes introductions.

Great Lakes Nonindigenous Occurrences: Ornamental jewelweed is naturalized in 31 countries. It is widespread in Europe, North America, and Asia between the latitudes of 30 and 64°N (Beerling and Perrins 1993, Lid and Lid 1994). Within North America, ornamental jewelweed has been recorded in Alaska, California, Connecticut, Idaho, Indiana, Maine, Massachusetts, Michigan, Montana, New York, Oregon, Vermont, Washington, British Columbia, and Ontario (Hitchcock and Cronquist 1973, USDA 2008).

First Great Lakes report: Lake Huron, 1912 (Mills et al. 1993).


Table 1. Great Lakes region nonindigenous occurrences, the earliest and latest observations in each state/province, and the tally and names of HUCs with observations†. Names and dates are hyperlinked to their relevant specimen records. The list of references for all nonindigenous occurrences of Impatiens glandulifera are found here.

State/ProvinceFirst ObservedLast ObservedTotal HUCs with observations†HUCs with observations†
MI1912201810Betsy-Chocolay; Detroit; Lake St. Clair; Lake Superior; Lower Grand; Northwestern Lake Huron; Northwestern Lake Michigan; Southeastern Lake Superior; St. Clair; St. Marys
NY200820083Indian; Oneida; Seneca

Table last updated 6/19/2024

† Populations may not be currently present.

Ecology: Impatiens glandulifera tolerates a wide variety of soils (sediment type, nutrient availability, pH, etc.), but requires high soil moisture. It is partially shade tolerant, and is found in lowland, riparian areas, including moist forests, swamps, stream sides, ditches, and roadside thickets (Campbell et al. 2010, WSNWCB 2008). Impatiens glandulifera is susceptible to frost, both in the early seedling stage and as a mature plant, potentially limiting its non-native distribution. However, within its native range, Himalaya, this species is frost tolerant, and it found at elevations of 1,800 to 4,000 meters (Beerling and Perrins 1993). Willis and Hulme (2002) found that plants were smaller and produced less seeds with increasing elevations.

Impatiens glandulifera germinates early in the spring. Pollinators include several species of bees, moths, and wasps (Beerling and Perrins 1993, Mumford 1988). Roots develop 12 days after germination and photosynthesis begins in the leaves four weeks after germination (WSNWCB 2008). Flowers bloom from June to October or until the first frost (all life stages of this plant are frost intolerant), with delayed flowering in shaded environments. Seed set occurs about 13 weeks after flowering. When seed capsules are mature, they split along the five seams of the fused stamens. Seeds are viable for 18–24 months, can float and germinate under water, and require cold temperatures to break seed dormancy (Beerling and Perrins 1993). Mumford (1998) determined that fully hydrated seeds kept at 20°C can remain dormant for several years. Impatiens glandulifera seedlings often require small open spaces (e.g., as created by localized disturbances) in order to compete with nearby vegetation and successfully mature (Perrins et al. 1993).

Means of Introduction: Deliberate release (USEPA 2008). Impatiens glandulifera is frequently sold or shared as a garden ornamental (King County 2004). Seeds ejected from mature capsules may disperse up to 5 m; they also can travel along waterways and be carried by small mammals (Beerling and Perrins 1993, King County 2004).

Status: Established and rapidly expanding in North America (Prots and Klotz 2004).

Great Lakes Impacts:
Summary of species impacts derived from literature review. Click on an icon to find out more...



Current research on the environmental impact of Impatiens glandulifera in the Great Lakes is inadequate to support proper assessment.
Impatiens glandulifera was historically transported to many countries for horticulture and is now considered to be one of the most invasive plants of the world (Weber 2000 in Vervoort et al. 2011). This species aggressively spreads through moist woodlands, wetlands, stream edges, and beaches (King County 2007, Ontario Invasive Plant Council 2009, Perrins et al. 1993). Such rapid spread in non-native environments is due to prolific seed production and the ability to explosively disperse seeds up to 5 m under favorable conditions (Beerling and Perrins 1993, Forest Service 2007). In Britain, its average rate of range expansion is 2–5 km per year, while its historic peak rate was 38 km per year (Beerling and Perrins 1993, Perrins et al. 1993). Perrins et al. (1993) named I. glandulifera the most invasive species of its genus in Britain. It has also been listed as one of the top five most aggressive invasive plants in Sweden (Larsson and Martinsson 1998 in Helmisaari 2010). However, studies of six riparian communities in the Czech Republic indicated that I. glandulifera has a negligible effect on existing communities and therefore is not a threat to plant diversity there (Hejda and Pyšek 2006).

As access to nutrients increases, I. glandulifera individuals will allocate extra resources and energy for seed production (Willis and Hulme 2004). It is able to out-compete beneficial, and often native, plants for physical space, light, and nutrients (Tanner 2011). If it displaces perennial vegetation along water bodies, soil erosion is likely to occur (IPANE 2004). The adventitious roots of ornamental jewelweed can obstruct waterways and wetlands, which can alter hydrology of the ecosystem. This altered hydrology can lead to increased erosion or flooding (Forest Service 2007, King County 2007). A larger sediment load in the river could in turn reduce available habitat and smother benthic communities (Tanner 2011).
High nitrate concentrations and light levels—conditions that are typically found at disturbed woodland sites—may result in the rapid development of I. glandulifera seedlings (Andrews et al. 2009). Early spring emergence, plus the ability to capitalize on disturbance opportunities, can lead to dense monospecific stands (Perrins et al 1993). At maturity, ornamental jewelweed is taller and has relatively larger leaves than most grasses and forbs. This shading effect creates bare patches nearby, facilitating germination and emergence of additional I. glandulifera seedlings (Centre for Aquatic Management 2004, King County 2007).

The stems of I. glandulifera have high holocellulose content (insoluble carbohydrates) that does not fully decompose over the winter. As a result, litter still present in the spring can suppress other plant seedlings (Beerling and Perrins 1993).

Impatiens glandulifera has large, bright flowers that allow easy access to its high-sugar content nectar (Vervoort et al. 2011). These attributes attract many insects, and as visitation increases, so does the probability of pollen deposition and seed production (Titze 2000). Preferential visitation of ornamental jewelweed by pollinators could ultimately lead to a reduction in fitness of neighboring species (Chittka and Schürkens 2001 in Tanner 2011). In a study conducted by Vervoort et al. (2011) in Belgium, I. glandulifera was visited by potential pollinators up to 250 times—substantially more than other Impatiens species studied (< 10 visits). It is believed that this higher visitation was due to the quality and quantity of the nectar and resulted in a higher seed production (Vervoort et al. 2011). However, in a study conducted in Germany, there was no significant evidence to suggest that I. glandulifera out-competes native plants for pollinators during periods of simultaneous blooming (Bartomeus et al. 2010).
Due to its genetic variation, I. glandulifera has the ability to adapt to local environments within a few generations and has a strong probability of expanding northward into previously unoccupied niches (Kollmann and Bañuelos 2004). Ornamental jewelweed also seems to react positively to increases in carbon dioxide and temperature. With increasing global mean temperature, I. glandulifera could expand its range northward by several degrees latitude (Beerling 1993).

Monospecific stands also alter the aboveground insect community. As native plants become displaced, specialized herbivorous insects leave the area and subsequently lead to a shift in the predatory insect community (e.g., spiders). There is the potential for further trophic shifts, as well (Tanner 2011). However, alterations to predator-prey cycles or trophic shifts have not been recorded in the Great Lakes.

There is little or no evidence to support that I. glandulifera has a significant socio-economic impact in the Great Lakes.

There is little or no evidence to support that I. glanulifera has a significant beneficial effect in the Great Lakes.
In the Czech Republic, I. glandulifera is an important source of nectar and pollen because its bloom period lasts longer than many of the native plant species. Starý and Taklcú (1998) have concluded that the present of I. glandulifera has contributed to the conservation of several bumblebee species.

When I. glandulifera and Lythrum salicaria (purple loosestrife, another major invasive of the Great Lakes) share a pollinator community, I. glandulifera will out-compete L. salicaria for pollinators, ultimately decreasing pollen deposition and seed production in the latter (Thijs et al. 2012).

Management: Regulations (pertaining to the Great Lakes region)

Impatiens glandulifera is ranked as having “low environmental invasiveness” by the New York State Office of Invasive Species and is therefore unregulated in that state. Regionally, the Great Lakes Indian Fish & Wildlife Commission (GLIFWC) classify this species as capable of causing moderate to severe ecological impacts and/or having limited effective control options available (Falck and Garske 2003).

Note: Check federal, state/provincial, and local regulations for the most up-to-date information.

Regardless of method, all control methods should be adaptive and involving monitoring of the site for several years to ensure plants do not germinate from the seed bank (King County 2010).

It should also be noted that in experiments conducted by Wadsworth et al. (2000), I. glandulifera was so prolific that even scenarios with 99% control efficiency were as ineffective as scenarios with no management action.

To date, no specific biological control methods are available for I. glandulifera (Sheppard et al. 2006). However, allowing cattle or sheep access to areas infested with I. glandulifera will control the population and the spread of the species either by direct grazing or by trampling of young seedlings (Centre for Aquatic Management 2004).

The species Aphis fabae, Impatientinum balsamines, and Deilephila elpenor are known to feed on ornamental jewelweed, but their capacity to act as biological control agents is still unknown (Beerling and Perrins 1993). Although an initial experiment by Tanner (2011) indicated that Deilephila elpenor exhibited lower biomass and survivorship when raised on I. glandulifera.

In its native range, I. glandulifera has been known to harbor Puccinia komarovii (a rust pathogen) which is currently undergoing research as a control agent (Tanner 2011).

A good control method for small infestations is the removal of ornamental jewelweed by pulling or digging. Efforts should be concentrated prior to seed-set in the spring, while the soil is still moist (King County 2010).

Hartmann et al. (1995) found that mowing, mulching or soil cultivation were successful in controlling I. glandulifera populations in Germany. Mowing infestations of I. glandulifera also causes less soil erosion than hand-pulling or digging (King County 2010). Ornamental jewelweed should be cut close to the ground, preferable below the lowest node to prevent regrowth (Centre for Aquatic Management 2004). If the vegetative parts are to be left on-site to decompose, plant material should be allowed to dry out completely or the stems should be crushed (by walking or jumping on them) to prevent regrowth. Flower heads and seed capsules should not be left on site (King County 2010).

Herbicides are more effective on populations of I. glandulifera that are large or inaccessible to other equipment (King County 2007). Application of any herbicide needs to be carefully timed. Seedlings need to be large enough that they will be covered by the herbicide, but before flowers are produced (King County 2010). Glyphosate is most effective if applied to growing leaves. Care should be taken not to get the herbicide on desirable plants because it is non-selective and will damage any foliage it comes in contact with (King County 2007). Herbicides containing triclopyr (Renovate3), 2,4-D, or metsulfuron are more selective and will not harm most grass species (King County 2007). These herbicides may be preferable if I. glandulifera infestations are in mixed communities and/or near water bodies (Centre for Aquatic Management 2004).

When using herbicides, do not cut or mow treated plants until they have died completely, which may take two weeks (King County 2007).

Note: Check state/provincial and local regulations for the most up-to-date information regarding permits for control methods. Follow all label instructions.

Remarks: Given the number of common names for Impatiens glandulifera and the practice of referring to both native and non-indigenous Impatiens spp. as “touch-me-not” or “jewelweed”, only sources that included its scientific name were included in this factsheet.

Associated species in western Washington sites include: blackberries (Rubus spp), Japanese knotweed (Polygonum cuspidatum), elderberry (Sambucus racemosa), Spirea douglassi ssp douglassi, reed canarygrass (Phalaris arundinacea), willows (Salix spp.), and ivy (Hedera helix).

Destroying riparian stands of Himalayan Balsam can open up the habitat for more aggressive invasive plants such as Japanese knotweed and aid in seed dispersal by dropped seeds sticking to shoes (Hejda & Pyšek 2006).

References (click for full reference list)

Other Resources:
Author: Cao, L., L. Berent, and A. Fusaro

Contributing Agencies:

Revision Date: 9/23/2012

Citation for this information:
Cao, L., L. Berent, and A. Fusaro, 2024, Impatiens glandulifera Royle: U.S. Geological Survey, Nonindigenous Aquatic Species Database, Gainesville, FL, and NOAA Great Lakes Aquatic Nonindigenous Species Information System, Ann Arbor, MI,, Revision Date: 9/23/2012, Access Date: 6/19/2024

This information is preliminary or provisional and is subject to revision. It is being provided to meet the need for timely best science. The information has not received final approval by the U.S. Geological Survey (USGS) and is provided on the condition that neither the USGS nor the U.S. Government shall be held liable for any damages resulting from the authorized or unauthorized use of the information.