Clidemia hirta (Koster's Curse)

Scientific name

Clidemia hirta (L.) D. Don


Clidemia crenata DC.; Clidemia elegans (Aubl.) D. Don; Melastoma elegans Aubl.; Melastoma hirtumL.

Common names

Koster's curse, clidemia, soap bush




This species is native to much of tropical America.

Naturalised distribution (global)

Locations within which Clidemia hirta is naturalised include tropical Africa, Madagascar and some oceanic islands with warm climates.

Introduced, naturalised or invasive in East Africa

Clidemia hirta is invasive in parts of Tanzania (Tropical Biology Association 2010).  The editors are not aware of records of the presence of this species in Uganda, though this does not necessarily mean that it is absent from the country.


This species prefers humid tropical climates and may invade both disturbed and undisturbed habitats. It is a potential weed of wetter pastures, open grasslands, plantations, roadsides, wetter open woodlands, riparian zones (banks of watercourses), forest margins and rainforests.


Clidemia hirta is a densely branching long-lived (perennial) shrub normally growing 0.5-3 m tall, but sometimes reaching 5 m in height, depending on habitat. In more shaded habitats it grows much taller than it does in exposed areas, where it typically grows less than 1 m tall.

The younger stems are rounded and are covered in large, stiff, brown or reddish-coloured hairs (they are strigose). The oppositely arranged simple leaves (5-18 cm long and 3-8 cm wide) are borne on stalks (petioles) 5-30 mm long. They are oval (elliptic) or egg-shaped in outline with broad end at base (ovate), with pointed tips (acute to shortly acuminate apices), and almost entire (sub-entire) to finely toothed (crenulate or denticulate) margins. Their upper surfaces are sparsely covered in hairs, similar to those found on the stems (they are sparsely strigose), while their lower surfaces and margins are more densely hairy. The leaves also have a somewhat wrinkled (rugose) appearance and five distinct veins that run in an almost parallel fashion from the leaf bases to their tips.

The flowers are arranged in small clusters in the leaf forks or at the tips of the branches (in axillary or terminal clusters). Each flower is borne on a very short stalk (pedicel) 0.5-1 mm long and has five white, or occasionally pale pinkish, petals (6-11 mm long and 4-5 mm wide). The base of the flower is swollen into a cup-shaped structure (a hypanthium) about 3-3.5 mm long, which is moderately to sparsely covered with a mixture of bristly and sticky (glandular) hairs. The flowers also have five sepals, but these are very small and inconspicuous (about 0.5 mm long), and five distinctive stamens that have a claw-like appearance. Plants flower and fruit prolifically throughout the year, except during dry periods.

The small, rounded (globular), fruit (4-9 mm across) are berries and are either dark blue, purplish or blackish in colour. Each of these berries contains over 100 light brown coloured seeds (0.5-0.75 mm long). These fruit are also covered in stiff spreading hairs, especially when they are young.

Reproduction and dispersal

This plant reproduces by seed, which are principally dispersed by fruit-eating birds. Other animals moving through thickets of this species may carry seeds away with them (e.g. feral pigs) and the fruit are also dispersed by floodwaters. Long distance dispersal may also be brought about by human activities.

Similar species

Clidemia hirta is quite easily to recognise. Young plants, with their stiff hairs and crinkled leaves, may vaguely resemble a stinging nettle (Urtica spp. and in Kenya this is Urtica massaica Mildbr.), but mature plants can be distinguished by their shrubby habit and white flowers.

  • C. hirta can be distinguished by the reddish tinge in the young branches which characteristic is not found in U. massaica.
  • The hairs in C. hirta do not cause the burning sensation as those of U. massaica.
  • The stems of U. massaica are 4-angled, unlike those of C. hirta.

Economic and other uses

Clidemia hirta has been widely introduced as an ornamental plant.

Environmental and other impacts

Clidemia hirta is regarded as an environmental weed in many parts of the world. It has been nominated as among 100 of the "World's Worst" invaders by the IUCN Invasive Species Specialist Group and it has been listed as a noxious weed in Queensland and the Northern Territories (Australia) and Hawaii.

This species has already caused significant environmental damage in other parts of the world. It is a highly invasive shrub in the montane rainforests and cloud forests. The impact of this weed on native species and ecosystems is devastating and the rate at which it is spreading is alarming. It competes with native plants in gaps in undisturbed forests and has the potential to alter forest regeneration. The spread of C. hirta in islands such as Hawaii has been linked to soil disturbances, particularly those caused by wild pigs.

In Hawaii it is replacing the endemic species that formerly dominated the forests and threatens their extinction. Elsewhere, it is regarded as one of the most problematic invasive species in the Comoros Archipelago, on La Réunion, in the Seychelles and on Mauritius.


The precise management measures adopted for any plant invasion will depend upon factors such as the terrain, the cost and availability of labour, the severity of the infestation and the presence of other invasive species. Some components of an integrated management approach are introduced below.

The best form of invasive species management is prevention. If prevention is no longer possible, it is best to treat the weed infestations when they are small to prevent them from establishing (early detection and rapid response). Controlling the weed before it seeds will reduce future problems. Control is generally best applied to the least infested areas before dense infestations are tackled. Consistent follow-up work is required for sustainable management.

Once established Clidemia hirta is extremely difficult to control. Because of it large seed bank and the ability of detached leaves to root in forested areas. Traditional cultivation methods can help to prevent the establishment of C. hirta on arable land.

Over the past four decades extensive searches of biological control agents have been made to control C. hirta in Hawaiian forests.


Not listed as a noxious weed by the state or governments in Kenya, Tanzania and Uganda.


Binggeli, P. (1997). Clidemia hirta (L.) D. Don. (Melastomataceae). Woody Plant Ecology - Invasive Woody Plants. Pierre Binggeli.

CABI Invasive Species Compendium online data sheet. Clidemia hirta (Koster's curse). CABI Publishing 2011. Accessed March 2011.

Germplasm Resources Information Network (GRIN). National Germplasm Resources Laboratory, National Genetic Resources Program, Agricultural Research Service (ARS), United States Department of Agriculture (USDA), Beltsville, Maryland, USA. Accessed March 2011.

Global Compendium of Weeds. Hawaiian Ecosystems at Risk Project. Accessed March 2011.

Global Invasive Species Database online data sheet. Clidemia hirta (shrub). Invasive Species Specialist Group. Accessed March 2011.

Maundu, P. M., Ngugi, G. W. and Kabuye, C.H.S, .(1999). Traditional Food Plants of Kenya. National Museums of Kenya, Nairobi.

Pacific Island Ecosystems at Risk (PIER). Clidemia hirta (L.) D. Don, Melastomataceae: plant threats to Pacific ecosystems. Institute of Pacific Islands Forestry, Hawaii, USA. Accessed January 2011.

Tropical Biology Association (2010). Usambara Invasive Plants - Amani Nature Reserve -


Agnes Lusweti, National Museums of Kenya; Emily Wabuyele, National Museums of Kenya, Paul Ssegawa, Makerere University; John Mauremootoo, BioNET-INTERNATIONAL Secretariat - UK.


This fact sheet is adapted from The Environmental Weeds of Australia by Sheldon Navie and Steve Adkins, Centre for Biological Information Technology, University of Queensland. We recognise the support from the National Museums of Kenya, Tropical Pesticides Research Institute (TPRI) - Tanzania and Makerere University, Uganda. This activity was undertaken as part of the BioNET-EAFRINET UVIMA Project (Taxonomy for Development in East Africa).


BioNET-EAFRINET Regional Coordinator: