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

hydrilla (Hydrilla verticillata)

Host plants / species affected
Oryza sativa (rice)

H. verticillata is a submerged, monoecious or dioecious perennial. Its stems are branched, about 1 mm thick and up to 3 m long; the internodes are 3 to 50 mm long. The sessile leaves are formed in whorls at the nodes; there are 3-8, sometimes up to 12 leaves in a whorl. The leaves are 7-40 mm long, linear to lanceolate, with a conspicuous midrib. They have sharply toothed margins and spines on the vein on the lower side of the leaves; a few teeth may also be formed on this vein. These leaf characteristics are commonly used to distinguish H. verticillata from similar submerged plants in the Hydrocharitaceae, like Egeria and Elodea spp.

The inflorescences are unisexual, arising from spathes situated in the leaf axils, each flower has three sepals and three petals. All six perianth parts are clear or translucent green (the sepals usually slightly reddish).The male spathe is about 1.5 mm long, solitary in the leaf axils, somewhat spiny. The female spathe is about 5 mm long, solitary in the leaf axils. There are three petals, three stamens and three styles. The ovary is cylindrical to narrowly conical and is enclosed in the base of a hypanthium; the style is as long as the hypanthium and there are three stigmas. For further information, see Cook et al. (1974) and Aston (1977).

The fruit is cylindrical, about 7 mm long and 1.5 mm wide. It contains 2-7 oblong-elliptic seeds. For further information, see Cook and Lüönd (1982); Swarbrick et al. (1981); and Yeo et al. (1984).

Prevention and control

Hydrilla can be controlled by physical, chemical and biological methods, or by a combination of these methods (integrated pest management).

Physical Control

The removal of plants either manually, using hand tools, or mechanically, using machines, is relatively expensive. Several machines, developed for aquatic weed control, can be used to remove H. verticillata plants from irrigation canals and drains; these include mowing buckets attached to a tractor or hydraulic excavator.

Different harvesters can be used to remove H. verticillata from lakes; these harvesters collect the plant material and dump it on the shore. One problem in the use of these harvesters is that cuttings of hydrilla, which are not removed from the water, help to spread the weed. Using mechanical control of hydrilla on large lakes without the use of herbicides or other control methods has not been feasible because of high cost, short-term effects and logistical constraints (Hetrick and Langeland, 2013).

In the USA, lake drawdowns are occasionally used to expose the plant and dry it out (USDA, 2011).

Chemical Control

The application of herbicides in or near waterbodies may have serious consequences for the environment and can endanger the health of local people if the water is used for drinking, swimming or washing. As a consequence, there are strict regulations for the use of chemicals to control aquatic weeds in many countries. The risk is greatest when the herbicide is introduced directly into the water, which is necessary for the control of submerged weeds.

There are many chemicals which are effective against H. verticillata but only a few of these compounds are reasonably safe for the environment and human health. One additional problem is that the use of herbicides can lead to the build up of masses of decaying plant material and, as a consequence, a sudden decrease in the oxygen content of the water.

Until recently, the main herbicide options for H. verticillata were endothal, diquat and copper. The costs of applying these contact herbicides twice a year in Florida amounted to approximately $400 per ha (Haller, 1995).

A systemic herbicide, fluridone, has been developed which can provide effective control of hydrilla; this control usually lasts for about one year. The costs in the 1990s amounted to approximately $200 per ha per year (Haller, 1995). Hydrilla biotypes with varying levels of fluridone resistance, however, have been documented in Florida (Puri et al., 2007). Benoit and Les (2013) reported an alarming increase in fluridone resistance in Florida, and the first case of resistance outside this state, from Lake Seminole in Georgia. Schmid et al. (2010) stated that the herbicide resistance in Florida had resulted in the inability to economically control large infestations of this weed. Even eight years after the last applications of fluridone in the Kissimmee Chain of Lakes in Florida, most hydrilla remained resistant (Netherland and Jones, 2015). This herbicide resistance has led to renewed interest in biological control methods (Cuda et al., 2008).

The herbicide, bensulfuron-methyl, has been tested against H. verticillata (Van and Vandiver, 1994). After one month, effective concentrations caused severe damage but regrowth occurred rapidly where herbicide exposure was limited to periods of less than 14 days. Langeland (1996) reported that while this herbicide showed promise, an Experimental Use Permit was not renewed in 1992 and efforts to register the compound were discontinued.

Since 2007, four new herbicides have been registered for hydrilla control in Florida. They include the active ingredients bispyribac-sodium, flumioxazin, penoxsulam, and imazamox (Hetrick and Langeland, 2013).

Biological Control

Biological control is theoretically the best method of controlling aquatic weed problems because the effect is lasting and relatively inexpensive. The most promising biological control agent for H. verticillata is the phytophagous fish, grass carp or white amur (Ctenopharyngodon idella). It feeds on many aquatic plants, in particular, submerged species. It is presumed that the grass carp will not breed outside its native habitat in China and Siberia as it requires special conditions for spawning. However, artificial reproduction is possible in fishery stations. In the USA, 180-320 kg/ha of grass carp is required for effective control of H. verticillata, whereas 80-70 kg/ha is sufficient in India (Pieterse, 1981).

A sterile, triploid grass carp has been bred because there is some controversy in the USA over the possibility of grass carp reproducing naturally in sufficient quantities to interfere with fisheries and waterfowl populations. This hybrid grass carp is the result of a cross between a female grass carp and a male bighead carp (Hypophthalmichthys nobilis). Control of H. verticillata using grass carp and hybrid grass carp is highly economical. In the USA, this control amounts to approximately $12 per ha per year (Haller, 1995). However, the fish needs to be retained within a certain area using fencing to prevent movement from the target site. This also allows the fish to be removed if the population needs to be managed. Fouts et al. (2017) found stakeholder support for methods of hydrilla control, including the use of sterile grass carp (Ctenopharyngodon idella) in public reservoirs in the southern USA where a neurotoxid cyanobacterial species growing on hydrilla was linked to a fatal wildlife disease.

Four insects of the genera Bagous and Hydrellia have been introduced into the USA by entomologists. These insects have eliminated large infestations of H. verticillata. Bagous affinis from India was released in Florida, USA, in 1987 (O'Brien and Pajni, 1989). This species was recently shown to have potential as a biological control agent of H. verticillata in California as it survived the winter season (Godfrey et al., 1994). Research on these insects and other potential candidates for biological control continued, and early introductions were reviewed by Center (1992).

The development of herbicide resistance in fluridone populations in Florida led to renewed interest in biological control methods (Cuda et al., 2008; Schmid et al., 2010). The study by Schmid et al. (2010) found that the leaf-mining fly Hydrellia pakistanae survived at similar levels on fluridone resistant and susceptible genotypes of hydrilla, whereas a stem-mining midge Cricotopus lebetis had lower emergence when reared on herbicide resistant genotypes. Center et al. (2013) report that Bagous hydrillae released in the southern USA had persisted and dispersed widely in the southeastern USA, but found no evidence that it has had a suppressive effect on hydrilla.

Bownes (2014; 2015) reports on the possible use of Hydrellia spp. for biological control of H. verticillata in South Africa. H. pakistanae was rejected as a biocontrol agent in favour of a Hydrellia species collected in Singapore. Bownes and Deeming (2016) suggest that Hydrellia purcelli from Singapore is a candidate biocontrol agent for this weed in South Africa.

Integrated Pest Management

An integrated approach to the control of H. verticillata had been followed in Fish Lake in Florida, USA, an area with 22 lakes in the headwaters of the Kissimmee River. This approach involved applying the herbicide, fluridone, and subsequently releasing hybrid grass carp. By 2000, however, sustained use of fluridone resulted in dominance of hydrilla strains resistant to this herbicide throughout these lakes. Last large-scale applications of fluridone were conducted in 2004, but in 2012 most of the hydrilla remained resistant (Netherland and Jones, 2015).

Summary of invasiveness

H. verticillata is a submerged fast-growing aquatic herb. It has a highly effective survival strategy that makes it one of the most troublesome aquatic weeds of water bodies in the world. It has the potential to alter fishery populations, cause shifts in zooplankton communities and affect water chemistry. It forms dense masses, outcompeting native plants and interfering with many uses of waterways. It is readily dispersed by movement of plant fragments and can produce up to 6,000 tubers per m2. Tubers can remain viable for several days out of water or for over 4 years in undisturbed sediment. They are not impacted by most management activities, and a small percentage can sprout throughout the year making the species very difficult to manage or eradicate. It can be spread by water flow, waterfowl and recreational activities and is sold as an aquarium plant. Currently, this species is considered as one of the most aggressive invasive species in aquatic habitats. In the USA it has been listed as a Federal Noxious Weed since 1976, and is regarded as one of the worst invasive aquatic weed problems in Florida and much of the country. Its import is prohibited in Western Australia and Tasmania, and it is on the EPPO alert list.

Zoomed image