Introduction

Aquatic macrophytes provide food, shelter, and substrate for a variety of organisms in the aquatic system. Macrophytes also alter sediment, water quality, physical condition, and population dynamics of the system by plant growth, metabolism, and decay. Overabundance of aquatic plants often occurs in most lakes during summer in the United States. The excessive growth of aquatic plants may interfere with fishing, boating, and swimming activities and be aesthetically displeasing. Besides, respiration by large plant masses in the littoral zone during hours of darkness may significantly reduce oxygen concentrations of the lake (Moore, 1994). In addition, nutrient uptake by aquatic macrophytes may significantly reduce sediment nutrient availability. In the United States, several species of aquatic macrophytes are considered as nuisance weeds such as Myriophyllum spicatum and Hydrilla verticillata. Aquatic macrophytes control methods have been studied intensively for years in the United States. Recently, in the United States and Canada, there is a growing interest in controling Myriophyllum spicatum due to overgrowth of this plant population. Myriophyllum spicatum (Eurasian watermilfoil), a nuisance aquatic macrophyte, was introduced into the United States from Eurasia in the late 1800's (Reed, 1977). The plants are often found in lakes, rivers, drainage and irrigation canals, ponds, and streams (Jones, 1993). Severe infestations of watermilfoil can restrict boat traffic, interfere with fisheries, block water flow, and displace native species. According to Carpenter (1980), "a typical invasion of milfoil is characterized by a pattern of explosive growth, persisting for 5 to 10 years, followed by declining abundance." This type of plant growth is often exhibited when exotic species invade an area without their natural competitors. The absence of natural enemies in the new habitat enables the invading species to compete with and often dominate existing native species (Painter and McCabe, 1988). Carpenter (1980) assessed the following credible hypotheses that the decline is caused by: toxin accumulation, herbicides and harvesting, climate, nutrients, epiphytes, competition with other macrophytes, and parasites or pathogens. He concluded that no single factor could account for the decline of biomass; i. e., a multi-factor synergistic mechanism was involved. In this paper, aquatic macrophytes control methods are discussed with special emphasis on Eurasian watermilfoil control.

Macrophyte Community Dynamics

Aquatic macrophytes alter the physical environment by intercepting water movements and sunlight. Dense vegetation provides relatively quiet area near shore and avoids turbulence from breaking waves, longshore currents, and runoffs. High energy shores, exposed to erosion from winds and waves, become calm depositional plains after plant growth (Engel, 1990). Plants at the water surface blunt the wind, reducing its potential to stir the bottom. In terms of casting shade and blocking water movement, plant beds reduce the amount of heat transfer to the bottom. The foliage of acts as a barrier to separate warm water at the surface and cool water at the bottom. Also, density differences between cold and warm water masses help maintain this separation. This temperature difference can reach 10ûC on hot summer days (Dale and Gillespie, 1977). The temperature gradients create warm and cool microclimates for colonizing organisms (Engel 1985). Animals unable to withstand warm water can approach the shore under macrophytes; others can be repelled by heat radiated from surface foliage. Cool bottom water can also reduce photosynthesis in lower leaves and delay hatching of fish eggs, whereas warm surface water can speed development of invertebrates. Microclimates which is caused by macrophytes growth thus restructure the ecosystem (Engel, 1990). Aquatic macrophytes provide food, shelter, and substrate for aquatic animals. Bacteria, algae, protozoan, rotifers, and invertebrates inhabit underwater leaves and stems (Engel, 1990). Rotifers and small crustaceans filter bacteria, algae, and detritus from water flowing by the plants. Snails, certain leeches, and many insect larvae scrape algae and detritus on the foliate or beneath it. They in turn are consumed by other aquatic insects, fish, water birds, and some furbearers. Besides, waterfowls need aquatic macrophytes as food. A diverse macrophyte community offers the best feeding for a variety of waterfowls. They consume plant tissues such as seeds, tubers, foliage, and plant-dwelling invertebrates. Sediment supplies nitrogen (N), phosphorus (P), and micronutrients to aquatic macrophytes. Evidence from field studies suggests that rooted submersed macrophytes, e. g., Hydrilla verticillata and Myriophyllum spicatum, are capable of markedly depleting N and P in sediment. Researchers report that various species of freshwater aquatic macrophytes not only extract nutrients from the sediment but also release them into the surrounding water (Carignan and Kalff, 1980). As a result, high productivity and biomass turnover of rapidly growing macrophytes can result in high rates of sediment nutrient loss (Barko, 1993). Thus, nutrient uptake by aquatic macrophytes may significantly reduce sediment nutrient availability. Aquatic macrophytes growth also alters the sedimentation rate of the water. Sedimentation rates are generally much greater in macrophyte beds than in the open water systems due to the filtering capacity of macrophytes (Barko, 1993). As a result, aquatic macrophytes not only reduce the turbulence in the water but also stabilize the sediment. As suspended sediment is removed from the water column, light penetration increases, so more light becomes available for photosynthesis by aquatic macrophytes. The value of pH increases in macrophyte beds during periods of photosynthesis. Elevated values of pH near the sediment surface tend to increase rates of phosphorus release from sediment, which can stimulate algal production. Algal production on macrophyte leaf surfaces becomes an excellent source of food for grazing invertebrate communities. The growth of invertebrate population can result in an increase in fish population. Besides, the structural complexity of macrophyte beds can influence fish communities of aquatic systems directly. These beds provide refugees for smaller fish, and are thus important in predator-prey relationships. To sum up, aquatic macrophyte community can potentially influence bed geometry as well as biomass distribution in aquatic systems (Barko, 1993).

Hand Harvesting

Hand harvesting; manual uprooting, pulling, and removing plants by workers; has been used in many locations for aquatic weed control. It is often applied in conjunction with bottom barrier applications as a part of intensive localized control efforts. Hand removal methods are usually preferred in particularly sensitive environments such as fish spawning areas (Cook, 1993). Sutherland (1990) indicated that hand harvesting was preferred in the area which has scattered populations of low plant density. Hand harvesting of nuisance aquatic plants is practiced widely in the U.K. and underdeveloped countries. It is usually limited to small, shallow water courses such as ditches and canals, and not practical for lakes and reservoirs. Murphy (1988) found that consistent hand harvesting was more expensive than mechanical harvesting or herbicidal alternatives in the U.K. Hand harvesting was tested in a luxuriant watermilfoil-dominated area that is located at the south of Chautauqua Institution (Nicholson, 1981). Plants were removed by manual uprooting. The results of this experiment showed that shoots and total biomass of both watermilfoil and all macrophytes combined were lower in most instances in treatment plots than in control plots. Nicholson concluded that uprooting can effectively reduce the amount of Eurasian watermilfoil growth on a long term basis in small areas. However, uprooting has generally been considered too labor intensive to be widely applicable (Daniel, 1974; Dunst, 1974). The ecological impacts of uprooting may be less serious than those of most other aquatic weed control techniques. Selective manual removal would have limited effects on animals. Invertebrates may remain undisturbed by this technique and would likely benefit from the removal of watermilfoil.

Drawdown

Water level drawdown is a well-established multipurpose technique to control certain aquatic plants, to manage fish populations, to repair structures such as dams or docks, and to carry out other improvement procedures such as dredging or installation of sediment covers (Cook, 1980). According to Cook (1993), water level drawdown is to kill the aquatic macrophyte by exposing the plants, especially root systems, to dry, freezing, or dry, hot conditions. Water level drawdown has been used to control various species of nuisance plants. Eurasian watermilfoil is one of the aquatic weeds which appears to be successfully controlled by water level drawdown. Moreover, Stanley (1976) found in laboratory experiments that exposing dewatered watermilfoil to subfreezing temperature (-1ûC) for 96 hours severely restricted the regrowth of the species. Beard (1973), Davis et al.. (1964), Goldsby et al. (1978), and Smith (1971) all confirm the success of controlling watermilfoil with winter water level drawdown. Besides weed population control, water level drawdown can also provide riparians an opportunity to repair docks and dams, clean and repair shorelines, and deepen areas such as swimming beaches. However, there are some negative environmental impacts of water level drawdown application. Such potential negative impacts on water quality include increase in total suspended solids and turbidity, decrease in Secchi disk transparency, increase in chlorophyll a, and increase in total nitrogen and phosphorus concentrations (Geiger, 1993). Besides, impoundment may lose their attractiveness to waterfowls if drawdown eliminates desirable food plants. Also, fishkills may occur if the volume of water remained does not hold sufficient oxygen through winter or summer stratification. In addition, algal blooms that often accompany reflooding after drawdown can be considered as another negative environmental impact of this method.

Dredging

Dredging means removing sediments and deepening a lake to a depth below the light compensation point to eliminate plant growth. Nowadays, modern dredging equipment is reasonably efficient at moving large volumes of sediment. Therefore, all deepening projects that used dredging were considered successful at the time of their completion (Pierce, 1970). However, shallow water dredging to remove nutrient sources for aquatic plant growth has little lasting impact on the abundance of plants. It can also have impacts on the other species present in the system and on successional trends (Nichols and Shaw, 1993). Dunst (1974) reports that milfoil (Myriophyllum exalbescens and Myriophyllum spicatum) invaded into Lilly Lake, Wis., the first year after dredging, and was common to abundant in water depths to 4 meters by the second year after dredging.

Mechanical Harvesting

Mechanical harvesting is normally viewed as a cosmetic measure to improve the appearance of the water surface or to remove plants interfering with human uses of water bodies such as boating, swimming, and fishing (Minnesota Department of Natural Resources, 1994). Mechanical harvesting has developed gradually over the past 30 years to be a common aquatic weed control technique in the Northeast, Upper Midwest, and West Coastal regions of the United States. The method is applied during the growing season of aquatic weeds, when submersed macrophytes have grown to or topped out the water surface. Multiple mechanical harvesting procedures have been designed to perform cutting, collection and processing, transportation, storage, and shore disposal functions. The simplest variation of mechanical harvesting is to cut and permit vegetation to float and to be moved to a suitable collection area by natural processes such as water currents (Koegel et al., 1978). The efficacy of harvesting is related to the biology of individual plant species. Information about potential biomass, regrowth rates, and methods of reproduction of the plant species is needed to determine the longevity of a harvest treatment. Moreover, the target weeds need to be soft enough to be cut so that it will float to the top of the water, where they can be removed with conventional harvesting equipment. Kimbel and Carpenter (1979) state that multiple harvests reduce the amount of regrowth and recovery of Myriophyllum spicatum more effectively than a single harvest during a growing season. Most authors also agree (Peverly et al., 1974; Wile, 1978; Newroth, 1980) that more than one harvest is necessary to control wartermilfoil regrowth throughout the growing season. At least two harvests at monthly intervals are recommended to sufficiently control watermilfoil. More harvests would probably be needed in regions with longer watermilfoil growing seasons. Nichols and Cottam (1972) found that intensive harvesting (three times per summer) for two years significantly reduced the biomass of watermilfoil in the third year. Peterson et al. (1974) also concluded that intensive harvesting on Lake Sallie, Minn. in one year reduced aquatic weed growth in the second year. However, there is also an observation which questions the effects of multiple harvesting. A research in British Columbia (Anonymous, 1981) indicates that harvesting removes the shading plant canopy and allows light penetration down to basal shoots, so harvesting may actually stimulate the growth of watermilfoil. Like other weed control methods, mechanical harvesting alters plant community composition. It removes competitors and opens the lake bed to new growth. Besides, macroscopic algae, spreading from spores and filaments, can become dominant plant species (Nichol, 1973). In addition, mechanical harvesting removes fish and invertebrates trapped in weeds (Wile, 1978), thus many non-target species can be eliminated (Mikol, 1985; Engel, 1990). Mechanical harvesting can remove as few as 2-3% (Mikol, 1985) or up to 30% (Haller et al., 1980) of fish populations in the lake. Avoiding fish spawning and nursery areas, slowing harvester speed, raising the front cutting bar of the harvester, and leaving escape routes for fish can reduce fish losses due to mechanical harvesting. Besides fish, many snails and aquatic insect larvae are also removed with the vegetation during the harvesting (Engel, 1990). Amphipods, oligochaetes, and larvae of midges, caddisflies, and damselflies are frequently eliminated from the lake. Moreover, a mechanical harvester produces noise which may frighten water birds. For instance, common loons sometimes abandon their nesting and brooding sites when they are bothered by the noise (Zimmer, 1979; Titus and VanDruff, 1981). The major positive environmental effects of mechanical harvesting are: (1) organic material removed by mechanical harvesting is no longer available to deplete oxygen supplies upon decay; (2) nutrients are not available for recycling upon decay of the plant; (3) foreign material of a chemical or biological nature is not being introduced into the aquatic ecosystem; (4) unlike herbicide treatment, local regulations do not ban access to the water for 10 to 14 days after treatment; (5) by choice of timing and depth of harvest, it may be possible to selectively remove specific plant species; (6) multiple use of the water body may continue during mechanical harvesting, (7) Mechanical harvesting may be less expensive than other types of physical control (Cook, 1993; Nichols and Shaw, 1993). The negative impacts include: (1) a temporary increase in turbidity; (2) loss of animal habitat; (3) potential of plant spread; (4) an increase in weed growth by removing the canopy (Cook, 1993; Nichols and Shaw, 1993).

Sediment Covers and Surface Shading

Various attempts have been practiced to control rooted aquatic plants by covering sediment with materials such as flyash, sand, gravel, clay, and plastic or rubber sheeting. These applications have failed because root systems remained and produced shoots after the applications. Shoots eventually penetrate earthen covers and regrow. Besides, many aquatic plants reestablish an infestation through the growth of fragments carried to the treated site from other lake areas (Cook , 1993). In addition, sediment collects rapidly on top of the covering, and plants will invade into the treated site as soon as the sediment can support plant growth. However, if coverings are cleaned regularly or in heavily used and intensively managed areas such as beaches, it is possible to avoid plant regrowth, so covering can be considered as an alternative plant control method. Floating black plastic sheeting on the water surface as a shade was first reported by Mayhew and Runkel (1962). Surface shading technique was also used in a Wisconsin farm pond by Nichlos (1974) and in New York by Peverly et al. (1974), and is one of the standard practices recommended as an aquatic weed control method in Missouri (Whitely, 1964). Nichols (1974) found that the black plastic shade sufficiently controlled M. spicatum population: the weeds turned brown and then dead after four weeks. After the shade was removed from the experiment pond, there was little or no regrowth of the plant in the rest of the summer. However, Peverly et al. (1974) found that although an application of a floating shade in mid-June had reduced the plant biomass (about 90% of which was watermilfoil) by 90 percent by mid-August (from 518 g/m2 to 83 g/m2), the biomass had increased to 92 g/m2 by mid-September. The study by Peverly et al. shows that there is a possibility of regrowth after the application of surface shading. More recently, a commercial product, Aquascreen (Menadri-Southern Corp.), has been designed for surface shading. It is a vinyl-coated fiberglass screen material with a mesh size of 64 apertures/cm2. Because Aquascreen is more dense than water, it is placed on the bottom or draped over existing weed beds (Nichols and Shaw, 1993). Mayer (1978), Goldsby (1980), Perkins et al. (1980), and Nichols and Shaw (1993) all reported that Aquascreen was effective at controlling weeds. Mayer found that 95 percent of all the plant material, including Myriophyllum spicatum, Potamogeton crispus, and Elodea canadensis, was destroyed 3 weeks after an Auqascreen application. However, there are negative effects of surface shading application. Floating shades make the water underneath them unusable. It is conceivable that water underneath a large shade could become stagnant and depleted of oxygen. To date, the benthic fauna and water chemistry underneath the screen have not been studied in detail (Nichols and Shaw, 1993). Further investigations on ecosystem dynamics and water chemistry under the shading will be necessary.

Dyes

"Aquashade" (Aquashade), "Mariner Blue Pond Dye," (3-M Corp.), and "Sierra Blue" (Aquat. Sys.) are commercially developed dyes for aquatic weed control. The products should be applied only when the depth of the water is over 70 cm. Besides, dyes should be applied early in the growing season, march or early April (Nichols and Shaw, 1993). The first application of dyes was in 1947 by Eicher. He used the commercial aniline dye, nigrosine, to control Potamogeton. crispus in hatchery ponds and in Granite basin Lake, Ariz. Initially, the dye was applied at the rate of 11 kg/ha in May. This application successfully controlled P. crispus in water depths over 2 meters. Another application by Levardson in 1953 was not able to control elodea using nigrosine dye which was applied at the rate of 17 kg/ha in a lake in Michigan. This application was not made until July, and most of the weed growth was in less than 150 cm of water. The failure of this experiment may have been due to the timing of application. Another example was practiced by Nichols in 1974 to control M. spicatum using Aquashade, an commercial dye, in a Wisconsin farm pond. Aquashade was applied early in the spring at the manufacturer's recommended level. However, an extremely wet spring caused a high turnover of water in the experimental pond which probably caused the treatment failure. Besides the dilution of dye in a flowing system, other factors can also remove dye from the water and cause failures. Levardson (1963) indicated that "nigrosine dye could be removed from the water by plants, by turbidity, and by a chemical reaction between salts in the water and the dye." Moreover, if the dye does not degrade rapidly via microbial or photo-degradation, it may bind to clay-sized particles and be lost from the system in turbid waters (Nichols and Shaw, 1993). Environmental changes may occur after the application of dyes. Levardson (1963) reported oxygen depletion problems in the lake where he treated with a commercial dye, nigrosine. Surber and Everhard (1950) indicated a slight increase in water temperature in their experimental ponds. Moreover, Buglewicz and Hergenrader (1977) stated that although the toxicity of aniline dyes to other organisms is unknown, they are extremely toxic to humans. The water is not usable for a period of time after the application. Besides, the aesthetics of colored water should be factored into consideration when people use dyes.

Chemical Control

A variety of chemicals including 2,4-D, Diquat, Endothal, Simazine, Fenac, Dichlobenil, Acrolein, Fluridone, and copper compounds are available for aquatic weed control. Combined use of herbicides and metal ions has advantages such as increased efficiency of weed control and lower dosages of chemicals required. For example, combined applications of Diquat with CuSO4 and Endothal with CuSO4 have been shown to be effective at lower concentrations than those required by single-applications of Diquat and Endothal (Sutton et al., 1970; Yeo and Dechoretz, 1977). Besides, the use of invert and bivert emulsions of herbicides has been evidenced to effectively control weeds with 80 percent less amount of herbicides than the use of ordinary application methods (Bitting, 1974; Baker et al., 1975). The invert or bivert adheres to the plants, putting the herbicide in direct contact with the plant and minimizing the concentration of the herbicide in the water. Another method of herbicide application that can reduce the herbicide concentration in the environment and provide season-long control is the use of slow-release formulation of herbicides (Harris, 1973; Steward, 1981). In addition, the physical conditions of the water body can also alter the efficiency of herbicide uses. Herbicides are usually most effective at water temperatures about 15 to 18ûC, in water with low turbidity, and on young plants. Water hardness and high calcium concentrations have been shown to increase the efficiency of herbicides (Parker, 1960; Stanley, 1975). To sum up, a good timing, appropriate technique, and suitable physical conditions to apply herbicide can reduce the amount of herbicide use. Chemical applications for submersed plant control often result in the rapid dispersion and dilution of herbicide residues from the treatment area (Fox et al., 1991; Getsinger, Fox, and Haller, 1992). Rapid residue dissipation is generally considered to be desirable from an environmental standpoint. However, residues that disperse too quickly (via gravity flow, tides, thermal and wind induced circulation patterns, etc.) can result in a failure of plant control because of insufficient exposure to the herbicide. Due to the unique properties of each compound, for example, rate of application, mode of action, environmental half-life, and species selectivity, the concentration/exposure time (CET) requirements can vary greatly among herbicides (Netherland, Getsinger, and Turner, 1993). Among the herbicides mentioned above, the herbicide 2,4-dichlorophenoixyacetic acid (2,4-D) is widely used to prevent and control the excessive growth of aquatic weeds. The application of 2,4-D has been proved to be the most suitable Eurasian watermilfoil control method in some locations, and the use of this herbicide is expected to continue for site-specific aquatic weed control by local agencies (Adams, 1982). Another herbicide which has been studied to control Eurasian watermilfoil is fluridone. In 1992, the State of Minnesota Department of Natural Resources began to evaluate the potential of selectively controlling Eurasian watermilfoil with fluridone, the active ingredient in Sonar herbicide. Sonar was tested on two Minnesota lakes in 1994. Applications of Sonar showed declines in both the frequency of submersed and floating-leave aquatic plants and the number of plant species present in the lakes (Minnesota Department of Natural Resources, 1994). Moreover, Netherland, Getsinger, and Turner (1993) reported that the key to a successful fluridone treatment is to maintain herbicidally active concentrations for periods exceeding 60 days. It was concluded that applications of Sonar at low rates could effectively control Eurasian watermilfoil and have minimal effects on native plant species. In addition, Garlon 3A, triclopyr, was tested under an Experimental Use Permit from the EPA. The initial results indicated that these applications produced excellent control of Eurasian watermilfoil and little damage to other native submersed aquatic plants (Minnesota Department of Natural Resources, 1994). Hence, Garlon may become a favorable option in herbicide control of Eurasian watermilfoil in the future. However, herbicides are toxic not only to plants but also to animals. Environmental impacts of herbicide uses have been observed and intensely studied. Using chemicals to control excessive aquatic weeds either directly affects the aquatic environment through herbicide toxicity or causes secondary effects resulting from loss of weeds (Brooker and Edwards, 1975). There is a possibility of minimizing herbicides' direct toxic effects on aquatic life by carefully selecting herbicides and applying them properly (Nichlos and Shaw, 1993). Another example to potentially minimize the harm of herbicides use is an escape route for fish. The secondary effects of herbicide uses due to weed destruction may affect aquatic life more drastically than the herbicide itself. The disturbance of the oxygen-carbon dioxide balance in the aquatic system is one of the secondary effects. The decline in photosynthesis and the increase in metabolism of dying vegetation may result in a deficiency of oxygen and an excessive amount of carbon dioxide in the aquatic system (Brooker and Edwards, 1973). This phenomenon may result in massive fish and invertebrate deaths where a fast-acting herbicide is used over a large portion of the lake. Another effects is that the decomposition processes of plant tissues release nutrients into the water. Not only the released nutrients but also carbon dioxide resulting from the decrease in photosynthesis and improved light penetration resulting from the decline in aquatic weed populations can be quickly used for growth by nonsusceptible species. Besides, planktonic algal blooms was observed following chemical control of aquatic weeds (Brooker and Edwards, 1973). This phenomenon may be also due to the released nutrients. In addition, effects of herbicide treatments on aquatic fauna are most dramatic on the invertebrates (Brooker and Edwards, 1974). The destruction of aquatic weeds may also destroy the habitat of invertebrates. This may cause a decline in the invertebrate population. However, some studies have shown an increase in benthic invertebrate population. Benthic invertebrates may increase in abundance as a result of detritus associated with death and decay of plants.

Insect

The first utilization of insect biocontrol agents to manage a noxious plant in the United States was in 1902. A beetle, Aerenicopsis champoini, was released in Hawaii to control Lantana (Weber, 1956). To date, biocontrol projects have been initiated on over 100 species of weeds in more than 70 countries (Julien, 1987). In the United States, the U.S. Department of Agriculture (USDA), the USACE, the Florida Department of Natural Resources, and the University of Florida are the leaders in research programs to identify biological control methods (Cook, 1993). Sanders (1981) and A. J. Leslie (1983) have described the steps which must be taken in order to release an insect from quarantine to the environment as a potential biological control agent. They are (1) seek for candidate species as potential biocontrol agents, usually in the native habitat of the plant, (2) while in the native country, expurgate candidate insects for host specificity and for possibility of attack on crop species, (3) transport candidate insects under quarantine to laboratories for studies on host specificity and biocontrol potential, (4) petition a technical advisory group (TAG) for permission to release the species from quarantine, and (5) release to field sites to establish a population. As Eurasian watermilfoil became a severe problem in the United States, entomologists searched its native range for suitable insect predators as biological control agents. Spencer and Lekic (1974) reported 25 insect species that feed on Eurasian watermilfoil. However, many of them are polyphagous and could not be imported into the United States. Some of the promising species identified were Parapoynx stratiota, P. allionealis, Acentria nivea, Litodactylus leucogaster., Bagous feniculatus, B. vicmus, Triaenodes tarda, Cricotopus sp., Euhrychiopsis lecontei, and Litodactylus griseomicans (Baloch et al., 1972; Spencer and Lekic, 1974; Anonymous, 1981). Some of the insect species have been studied, and significant predations have been observed. Those are described in the following paragraphs.

Triaenodes tarda

Triaenodes , a genus of Caddisflies, is generally regarded as herbivorous and uses aquatic plants for both food and material for case construction (Wiggins, 1977). One species of Triaenodes, T. tarda constructs an elegant tapered case from living Myriophyllum leaflets (Kangasniemi, 1993). The loss of leaflets from any part of the plant has effects on the growth of the shoot. T. tarda was first observed grazing on M. spicatum in Magic Lake in 1979 (Kangasniemi, 1993). An extremely abundant population of T. tarda in Magic Lake virtually eliminated the standing crop of M. spicatum population in 1979. Only bare stems and roots remained throughout the entire weed beds. In addition, close observation of T. tarda 's feeding behavior in aquaria indicates that most leaf material is cut and dropped into the water. Only a relatively small portion of leaf material was used as food or for case building.

Cricotopus sp.

Chironomid larvae were first observed grazing on M. spicatum apical meristems in 1980. The larvae were later identified as one species of the genus Cricotopus. It is more widespread and abundant than the other insect biological control agents of Eurasian watermilfoil. Cricotopus sp. has demonstrated the greatest impact on M. spicatum and exerts grazing pressure throughout the growing season (Kangasniemi, 1993). For example, in Vernon Arm of Okanagan Lake, an average of 80 percent of plant apices were being grazed by Cricotopus sp. between April and September, the growing season of M. spicatum. Cricotopus sp. larvae usually cause the most damage to the stem, apical bud, and surrounding leaves. The net effect of persistent grazing of the apical bud region is to retard or stop stem elongation and flower formation. This may result in a significant decline in the standing crop of M. spicatum population.

Weevil

Lekic and Mihajiovic (1971) reported several species of weevils associated with M. spicatum including Bagous longitarsus, Eubrychiopsis velatus, and Litodactylus leucogaster. Hatch (1971) also documented Eubrychiopsis lecontei as being associated with Myriophyllum and Potamogeton in British Columbia and Washington. According to Kangasniemi (1993), "adult weevils appear to cause some damage to the plant stems in the form of small holes or superficial epidermal damage and may benefit from the O2 available in the extensive air-filled lacunal system of M. spicatum stems." Moreover, there has been a growing interest in using the weevil Eubrychiopsis lecontei as a biological control agent for M. spicatum in North America. It was discovered that Eubrychiopsis lecontei was associated with declining watermilfoil populations (Creed and Sheldon, 1991). Eubrychiopsis lecontei adults and larvae were found "to suppress watermilfoil growth (adults and larvae), remove significant amounts of leaf tissue (adults), weaken the stem and remove vascular tissue (larvae), and reduce watermilfoil buoyancy (adults and larvae) in laboratory experiments" (Creed and Sheldon, 1991).

Fish

Numbers of fish have been suggested as biological control agents of aquatic weeds. These include the common carp (Cyprinus carpio) and the Israeli strain of the common carp, roach (Rutilus rutilus), rudd (Scardinus erythopthalmus), various species of tilapia including Tilapia zillii and Tilapia mossambica, silver dollar fish (Metynnis roosevelti and Mylossoma argenteum), white amur (Ctenopharyngodon idella), and a variety of hybrids of the white amur. The most promising fish appears to be the triploid hybrid, between the white amur and the bighead carp (Hypopthalmichys nobillis). All of these species are not native to the United States (Cook, 1993). Among those fish, the most extensively studied herbivorous fish is the white amur or Chinese grass carp. At least two States, Arkansas and Iowa, use white amuras as a standard management tool (Nichols and Shaw, 1993). Under good conditions, a white amur eats the same amount of weeds as its body weight every day (Stott, 1972). However, the environmental concerns about using white amur center on four areas: "(1) the possibility to become a pest; (2) the potential for recycling nutrients or causing other water quality problems; (3) the selectivity of feeding habits; and (4) the ability to destroy aquatic habitat (Nichols and Shaw, 1993). Grass carp provide long periods of plant control (8 to 10 years or more for diploids, 3 to 4 years for triploids). However, the disadvantage of using fish as a biological control agent is that fish are difficult to capture and remove from the lake once they are stocked. As a result, when grass carp are adopted as a plant management technique, lake users are committed to a long term change in their lake (Cook, 1993).

Plant Pathogens

The most well-known plant pathogen as biological control agent of Eurasian watermilfoil was isolated by researchers at the University of Massachusetts in 1979 (Theriot, Cofrancesco, and Shearer, 1993). The fungal pathogen was identified as Mycoleptodiscus terrestris (Mt). A joint testing program of the pathogen on Eurasian watermilfoil was initiated between the University of Massachusetts and WES (U.S. Army Engineer Waterways Experiment station). The pathogen appeared generally specific for watermilfoil, and its impact on watermilfoil in laboratory and greenhouse tests was enormous. The fungal pathogen significantly reduced 80 percent of watermilfoil biomass in laboratory and small-scale greenhouse studies (Theriot, Cofrancesco, and shearer, 1993). This pathogen was then applied on the field. An Experimental Use Permit was received in December 1991 from the U.S. Environmental Protection Agency which allowed application of the mycoherbicide, Mycoleptodiscus terrestris, to bodies of water larger than 1 acre (Shearer, 1993). In this experiment, four replicate test plots were established on a 62-acre watermilfoil-infested pond at Tennessee Valley Authority's Guntersville Reservoir Aquatic Ecosystem Facility in Guntersville, AL, in the summer of 1992. A formulation of the pathogen was applied to the test plots in early July after the watermilfoil had grown to the water surface and topped out. One month after the application, there were no visible differences between the treated plots and the untreated control plots. The plants were then collected, and it was confirmed that there were no significant differences in above ground biomass between the control and treated plots (Shearer, 1993). In addition, laboratory analysis revealed that the fungus did not become well established in watermilfoil stem tissues. Hence, the failure of the mycroherbicide to control watermilfoil in the field test could be attributed to problems with the fungus and/or the formulation. The fungus was known as a weak pathogen. The viability and virulence of the fungus could well account for poor field performance and result in the failure of the experiment (Shearer, 1993).

Conclusion

Most problem plants in the United States, particularly in aquatic and wetland environments, are exotic species. The plants usually have been introduced into favorable environments without their natural enemies. These exotic plants have the ability to increase rapidly, outcompeting the native vegetation for habitat and resources. Eurasian watermilfoil is one example of such plants and has become one of the most serious water issues in the North America and Canada. Many aquatic macrophytes control methods have been applied to control the watermilfoil population, and evidences of significant influences of the control methods on the macrophyte population have been found in some of the studies reviewed in the above. At the same time, other studies in this review present counter-evidences of ineffectiveness of the methods in controlling the macrophytes. This amalgam of seemingly contradicting results indicates that there is not, and likely there will never be, the single best solution to macrophyte control for all aquatic systems: each method has its advantages and disadvantages; each aquatic system has suitable and unsuitable methods to control macrophyte overgrowth. Therefore, it is extremely important to carefully choose suitable control methods to each aquatic ecosystem. Sound understanding of the ecosystem of a treatment site is indispensable, thus a detail physical, chemical, and biological pretreatment investigation is strongly recommended.

References

Adams, M. S., and R. T. Prentki. 1982. Biology, metabolism and functions of littoral submersed weedbeds in Lake Wingra, Wisconsin, USA: a summary and review. Arch. Hydrobiol. Suppl. 62:333-409.

Anonymous. 1981. A summary of biological research on Eurasian milfoil in British Columbia.Vol. XI: Inf. Bull. Aquat. Plant Manage. Progr. Br. Colum. Ministr. Environ.

Baker, G., et al. 1975. Improved application techniques for aquatic herbicides. Hyacinth control J. 13:21-24.

Baloch, G. M., A. G. Khan, and M. A. Ghani. 1972. Phonology, biology and host specificity of some stenophagous insects attacking Myriophyllum spp. in Pakistan. Hyacinth Control J.10:13-16.

Barko, J. W. 1993. Ecology of submersed aquatic macrophytes (overview 1992): advances, management implications, and related needs.  Proceedings, 27th Annual Meeting, APCRP. 197-200.

Beard, T. D. 1973. Overwinter drawdown. Impact on the aquatic vegetation in Murphy Flowage, Wis. Tech. Bull. No. 61. Dep. Nat. Resour. Madison, Wis.

Bitting, L. E. Sr. 1974. Field use of invert emulsion. Hyacinth Control J. 12:34-35.

Brooker, M. P., and R. W. Edwards. 1973. Effects of the herbicide paraquat on the ecology of a reservoir. I. Botanical and chemical aspects. Freshwat. Biol. 3:157-175.

Brooker, M. P., and R. W. Edwards. 1974. Effects of the herbicide paraquat on the ecology of a reservoir. III. Fauna and general discussion. Freshwat. Biol. 4:311-333.

Brooker, M. P., and R. W. Edwards. 1975. Aquatic herbicides and control of water weeds. WeedRes. 9:1-15.

Buglewicz, E. G., and G. L. Hergenrader. 1977. The impact of artificial reduction of light on a eutrophic farm pond. Trans. Neb. Acad. Sci. 4:23-33.

Carignan, R. and K. Kalff. 1980. Phosphorus sources for aquatic weeds: water or sediments? Science. 207:987-989.

Carpenter, S. R. 1980. The decline of Myriophyllum spicatum  in a eutrophic Wisconsin lake, U.S.A. lake. Can. J. Bot. 58:527-535.

Cook, G. D. 1980. Lake level drawdown as a macrophyte control technique, Water Res. Bull.16:317-322.

Cook, G. D. 1993. Restoration and management of lakes and reservoirs. 2nd. ed Lewis Publishers, Boca Roton. 548 p.

Creed, R. P., Jr., and S. P. Sheldon. 1991. The potential for biological control of Eurasian watermilfoil (Myriophyllum spicatum): Results of the Research Programs initiated in 1990. Prepared for Region 1, U.S. EPA, Boston, MA. 119 p.

Dale, H. M. and T. J. Gillespie. 1977. The influence of submersed aquatic plants on temperature gradients in shallow water bodies. Can. J. Bot. 55:2216-2225.

Daniel, T., S. A. Nichols, and J. W. Clark. 1974. Controlling waterweeds. Univ. Wisconsin- Extension, Madison. 6 p.

Davis, J. T., J. S. Hughes, and H. E. Schafer. 1964. water level fluctuation to control aquatic vegetation. Proc. 17th St. Weed Conf. 32-38.

Dunst, R. C., S. M. Born, P. D. Uttomark, S. A. Smith, S. A. Nichols, J. O. Peterson, R. D. Knauer, S. L. Serns, D. R. Winter, and T. L. Wirth. 1974. Survey of lake rehabilitation techniques and experiences. Dep. Nat. Resources Tech. Bull. No. 75, Madison, Wisc. 179 P.

Eicher, G. 1947. Aniline dye in aquatic weed control. J. Wildlife manage. 11:193-197.

Engel, S. 1990. Ecosystem responses to growth and control of submerged macrophytes: a literature review. Technical Bulletin No. 170. Department of Natural Resources Madison, Wisconsin.

Fox, A. M., Haller, W. T., Getsinger, K. D., and Green, W. R. 1991. Characterization of water exchage in hydrilla infested tidal canals of the Crystal River, Florida. Miscellaneous Paper A-91-2, U.S. Army Engineer Waterways Experiment Station, Vicksburg, MS.

Geiger, N. S. 1993. Winter drawdown for the control of Eurasian water milfoil in an Oregon oxbow lake (Blue Lake, Multnomah County). Proceedings, 27th Annual Meeting,  APCRP. 193-197.

Getsinger, K. D., A. M. Fox, and W. T. Haller. 1992. Controlling submersed plants with herbicides in flowing water systems. Proceedings, 26th Annual Meeting, Aquatic Plant Control Research Program. Miscellaneuos Paper A-92-2, U.S. Army Engineer Waterways Experiment Station, Vicksburg, MS. 103-105.

Goldsby, T. L., A. L. Bates, and R. A. Stanley. 1978. Effect of waterfluctuation and herbicide on  Eurasian water milfoil in Mellon Hill Reservoir. J. Aquat. Plant Manage. 16:34-38.

Goldsby, T. L. 1980. TVA testing of Aquascreen barrier for aquatic weed control. Tenn. Valley   Author. Mimeo Rep.

Haller, W. T., J. V. Shireman, and F. F. DuRant. 1980. Fish harvest resulting from mechanical  control of hydrilla, Trans. Am. Fisheries Soc. 109:517-520.

Harris, F. W. 1973. Controlled-release herbicide polymer formulations for aquatic weed control.  Hyacinth Control J. 11:61-63.

Hatch, N. H. 1971. Beetles of the Pacific Northwest. Part 5. Univ. of Washington Press, Seattle WA. 662 p.

Jones, H. L. 1993. The allelopathic ability of three species of aquatic plants to inhibit the growth of Myriophyllum spicatum . Proceedings, 27th Annual Meeting, APCRP. 124-127.

Julien, M. H. 1987. Biological control of weeds: A world catalogue of agents and their target weeds. 2nd ed., CAB International Institute of Biological Control, Wallingford, Oxon,   UK.

Kangasniemi, B. J. 1993. Observations on herbivorous insects that feed on Myriophyllum spicatum in British Columbia. Proceedings, 27th Annual Meeting, APCRP. 214-218.

Kimble, J. C., and S. R. Carpenter. 1979. The dynamics of Myriophyllum spicatum  biomass following harvest. In J. E. Breck, R. T. Prentki, and O. L. Loucks, eds. Aquatic Plants, lake Management and Ecosystem Consequences of Lake Harvesting. Inst. Environ. Stud.    Univ. Wisconsin, Madison.

Koegel, R. G., D. F. Livermore, and H. D. Bruhn. 1978. Improvement and evaluation of techniques for the mechanical removal and utilization of excess aquatic vegetation. Tech.    Rep. 78-02. Water Resources Center, University of Wisconsin, Madison.

Lekic, M., and Lj. Mihajlovic. 1971. Entomofauna of Myriophyllum spicatum L. (Halorrhagidacease), on aquatic weed on Yugoslav territory. J. Sci. Agric. Res. 23:59074.

Leslie, A. J., Jr., L. E. Nall, and J. M. Van Dyke. 1983. Effects of vegetation control by grass carp on selected water-quality variables in four Florida lakes. Trans. Am. Fish. Soc.112:777-787.

Levardson, N. O. 1953. Experiments with nigrosine dye in aquatic plant control. Prog. Fish Cult.15:109-113.

Mayer, R. J. 1978. Aquatic weed management by benthic semi-barriers. J. Aquat. Plant Manage.16:31-33.

Mayhew, J. K., and S. T. Runkel. 1962. The control of nuisance aquatic vegetation with black plastic sheeting. In Proc. Iowa Acad. Sci. 69:303-307.

Mikol, G. F. 1985. Effects of harvesting on aquatic vegetation and juvenile fish populations at Saratoga Lake, New York. J. Aquat. Plant Manage. 23:59-63.

Minnesota Department of Natural Resources. 1994. Ecologically harmful exotic aquatic plant and wild animal species in Minnesota. Annual Report 1994.

Moore, B. D. 1994. Water quality, fishery, and biologic characteristics in a shallow, eutrophic lake with dense macrophyte populations. Lake and Reserv. Manage. 8 (2): 175-188.

Murphy, K. J. 1988. Aquatic weed problems and their management: a review. II. Physical control measures. Crop Protection. 7:283-302.

Netherland, M. D., K. D. Getsinger, and E. G. Turner. 1993. Herbicide concentration/exposure time requirements for Eurasian watermilfoil and Hydrilla. Proceedings, 27th Annual Meeting, APCRP. 146-155.

Newroth, P. R. 1980. Case study of aquatic plant management for lake restoration and preservation in British Columbia, Canada. Pages 146-152 in Proc. Int. Symp. Restoration of Lakes and Inland Waters. EPA-440/5-81-010. U.S. Environ. Prot. Agency, Washington, D.C.

Nichols, D. S., and D. R. Keeney. 1973. Nitrogen and phosphorus release from decaying water milfoil. Hydrobiologia. 42:509-525.

Nichols, S. A., and G. Cottam. 1972. Harvesting as a control for aquatic plants. Water Res. Bull. 8:1205-1210.

Nichols, S. A. 1974. Mechanical and Habitat Manipulation for aquatic plant management. A review of techniques. Tech. Bull. No. 77. Wisconsin Department of Natural Resources, Madison.

Nichols, S. A. and B. H.. Shaw. 1993. Review of management tactics for integrated aquatic weed management of Eurasian water milfoil (Myriophyllum spicatum ), curlyleaf pondweed (Potamogeton crispus) and elodea (Elodea canadensis). Proceedings, 27th Annual Meeting, APCRP. 181-192.

Nicholson, S. A. 1981. Effects of uprooting on Eurasian watermilfoil. J. Aquat. Plant Manage.  19:57-59.

Painter, D. S. and McCabe, K. J. 1988. Investigation into the disappearance of Eurasian  watermilfoil from the Kawartha Lakes, Canada. J. Aquat. Plant Manage. 26:3-12.

Parker, C. 1960. Influence of water hardness on paraquat. Nature 212:1465-1466.

Perkins, M. A., L. Boston, and E. F. Curren. 1980. The use of fiberglass screens for control of  Eurasian water milfoil. J. Aquat. Plant Manage. 18:13-19.

Peverly, J., G. Miller, W. H. Brown, and R. L. Johnson. 1974. Aquatic weed management in the Finger Lakes. Tech. Rep. 90. Cornell Univ. Water Resour. Mar. Sci. Center, Ithaca, N.Y.

Peterson, S. A., W. L. Smith, and K. W. Malueg. 1974. Full-scale harvest of aquatic plants: nutrient removal from a eutrophic lake. J. Water Pollut. Control Fed. 46: 697-707.

Pierce, N. D. 1970. Inland lake dredging evaluation. Tech. Bull. 46. Wisconsin Dept. Nat.. Res., Madison.

Reed, C. F. 1977. History and distribution of Eurasian water milfoil in the United States and Canada. Phytologia. 36:417-436.

Sanders, D. R. 1981. Biological control technology development. an overview. in Proceedings  15th Annual Meeting Aquatic plant Control. Res. Program. Misc. Pap. A-81-3. U.S.     Army Corps Engineers, Vicksburg, MS. 139-143.

Shearer, J. F. 1993. Biocontrol of hydrilla and milfoil using plant pathogens. Proceedings, 27th     Annual Meeting, APCRP. 79-81.

Smith, G. E. 1971. Resume of studies and control of Eurasian watermilfoil (Myriophyllum spicatum  L.)  in the Tennessee Valley from 1960-1969, Hyacinth Control J. 9:23-25.

Spencer, N. R., and M. Lekic. 1974. Prospects for biological control of Eurasian watermilfoil.  Weed Sci. 22:401-404.

Stanley, R. A. 1975. Interaction of calcium and 2,4-D on Eurasian watermilfoil. Weed Sci. 23:182-184.

Stanley, R. A. 1976. Response of Eurasian water milfoil to subfreezing temperature. J. Aquat. Plant Manage. 14:36-39.

Steward, K. L. 1981. Improving technology for chemical control of aquatic weeds. Tech. Rep. A-81-2. U.S. Army Corps Eng. Waterways Exp. Sta. Vicksburg, Miss.

Surber, E. W., and M. H. Everhart. 1950. Biological effects of nigrosine used for control of weeds in hatchery ponds. Prog. Fish Cult. 12:135-140.

Sutherland, J. W. 1990. 1990 Interim report on the Lake George nuisanca aquatic plant management program. New York state Department of Environmental Conservation,  Albany, NY.

Sutton, D. L., L. W. Weldon, and R. D. Blackburn. 1970. Effect of diquat on uptake of copper in aquatic plants. Weed Sci. 18:703-707.

Theriot, E. A., A. F. Cofrancesco, Jr., and J. F. Shearer. 1993. Pathogen biocontrol research for aquatic plant management. Proceedings, 27th Annual Meeting, APCRP. 82-84.

Titus,J. R., and L. W. Van Druff. 1981. Response of the common loon to recreational pressure inthe Boundary Waters Canoe Area, northeastern Minnesota. Wildl. Monogr. 79. 59 p.

Weber, D. W. 1956. Recent introductions for biological control in Hawaii XVI. Proceedings of the Hawaiian Entomological society. 16:162-164.

Whitely, J. R. 1964. Control of aquatic weeds in farm fish ponds. Proc. N. Center. Weed Control  Conf. 20:31-32. Abstr.

Wiggins, G. B. 1977. Larvae of the North American Caddisfly Genera (Trichoptera). University of Toronto Press, Toronto.

Wile, I. 1978. Environmental effects of mechanical harvesting. J. Aquat. Plant mange. 16:14-20.

Yeo, R. R., and N. Dechoretz. 1977. Acute toxicity of a herbicidal combination of diquat, plus  copper ion to eggs, alevins and fry of rainbow trout and to aquatic invertebrates. J. Aquat. Plant Manage. 14:57-60.

Zimmer, G. E. 1979. The status and distribution of the common loon in Wisconsin. Univ. Wis-Stevens Point. M.S. Thesis. 63 p.

Comments or questions, email