In: Van Driesche, R., et al., 2002, Biological Control of Invasive Plants in the Eastern United States, USDA Forest Service Publication FHTET-2002-04, 413 p.
Pest Status of Weed
Although M. quinquenervia is a major pest in south Florida, it is considered threatened in its native Australia. Therefore, conservation groups in Australia advocate its protection. Melaleuca quinquenervia habitat in Australia comprises low-lying, high-rainfall areas, primarily in coastal regions (Resource Assessment Commission, 1992). Most of the remaining and remnant M. quinquenervia woodlands in Australia are located on private land, where clearing for commercial development continues.
Nature of Damage
Economic damage. Melaleuca quinquenervia flowers several times a year producing large amounts of pollen, allegedly a mild respiratory allergen (Morton, 1962; Lockey et al., 1981; Stanaland et al., 1986), from which as much as 20% of the population may suffer allergic reactions (Diamond et al., 1991). In addition, M. quinquenervia trees burn with extremely hot crown fires due to high foliar concentrations of essential oils. These fires are difficult to extinguish, often threatening buildings near M. quinquenervia-infested areas and causing local municipalities to incur additional fire fighting costs (Diamond et al., 1991).
Melaleuca quinquenervia infested areas become less attractive and monocultures become impenetrable to tourists, hikers, boaters, and other recreational users. Such impacts result in decreased revenues for parks and harm the economies of surrounding communities that rely on tourism associated with wilderness areas (Diamond et al., 1991).
Ecological damage. Prolific seed production, tolerance of brackish water, flooding, and fire enable M. quinquenervia to aggressively invade various wetland habitats and diminish the value of these habitats for native plant communities and associated wildlife (Meskimen, 1962; Crowder, 1974; Myers, 1975; Hofstetter, 1991). Melaleuca quinquenervia may accelerate loss of groundwater due to increased evapotranspiration (Alexander et al., 1977), although this view has been challenged (Allen et al., 1997). Trees produce allelopathic chemicals (Di Stefano and Fisher, 1983), which may enhance their ability to displace native flora. Melaleuca quinquenervia invasion has resulted in significant (60 to 80%) losses of biodiversity in freshwater herbaceous marsh communities in south Florida (Austin, 1978).
Extent of losses. The extent of the M. quinquenervia infestation in southern Florida (the area south of Lake Okeechobee) has been estimated at 0.20 to 0.61 million ha of the total 3.04 million ha (7 to 20% of the total) in the region (Bodle et al., 1994). It has been suggested that many of the remaining natural areas within this region will be overtaken by uncontrolled growth of M. quinquenervia within 30 years (Bodle et al., 1994).
The National Park Service, the U.S. Fish and Wildlife Service, the U.S. Army Corps of Engineers, and the South Florida Water Management District repeatedly conduct costly and labor-intensive operations to control M. quinquenervia. Mechanical removal of a moderately thick stand (about 988 trees/ha) cost $2,080/ha (McGehee, 1984), whereas ground herbicide treatment using tree injection techniques cost $1,330/ha (Laroche et al., 1992). Aerial treatments are less expensive but less effective and cause more damage to non-target plant species. Additionally, seed release is accelerated among trees stressed by herbicides and new infestations are created by dispersal of these seeds (Laroche and Ferriter, 1992). This regeneration of treated stands necessitates repeated herbicidal applications, which compromises environmental preservation. According to a recent estimate, the South Florida Water Management district alone spent more than $13 million from 1991 to 1998 for M. quinquenervia control in water conservation areas, Lake Okeechobee, and Loxahatchee Wildlife Refuge (Laroche, 1999). Millions of dollars also have been spent by other agencies such as Loxahatchee National Wildlife Refuge, Big Cypress National Preserve, Everglades National Park, Lee County, Miami-Dade County, and Palm Beach County.
Melaleuca quinquenervia trees benefit Florida’s beekeeping, chip, mulch, lumber, and pulp industries. However, M. quinquenervia honey is less valuable than that from fruit tree sources. Furthermore, access to M. quinquenervia infestations limits commercial utilization. Failure to control expanding M. quinquenervia infestations would result in an estimated loss of $168.6 million/year in revenue from reduced ecotourism of the Florida Everglades (Schmitz and Hofstetter, 1999).
Native range. Melaleuca quinquenervia is the southern-most member of the M. leucadendra complex. It is distributed within a 40-km-wide zone along Australia’s northeastern coast from Sydney in New South Wales to the tip of Cape York peninsula in northern Queensland; in New Guinea; and in New Caledonia (Fig. 2) within the range of 11 to 34°S latitude (Boland et al., 1987). Its altitudinal range is from sea level to about 100 m, sometimes up to 165 m (Boland et al., 1987). It occurs in coastal wetlands that are at least seasonally inundated, such as freshwater swamps, stream banks, and in brackish water behind mangrove swamps. The center of diversity for the M. leucadendra complex is on the Cape York Peninsula in northern Queensland (Turner et al., 1998). Barlow (1988) noted the genus Melaleuca to be of northern Australian (tropical) origin, with
Florida. The main infestations of M. quinquenervia exist along both coasts of southern Florida with scattered infestations in between (Fig. 3). The northernmost records (ca. 30°N latitude) are in Gainesville (Turner et al., 1998).
Background Information On The Pest Plant
The taxonomic position of M. quinquenervia according to Cronquist (1988) is as follows: class Magnoliopsida (Dicot), subclass Rosidae, order Myrtales, family Myrtaceae, subfamily Myrtoidea, tribe Leptospermae, genus Melaleuca, species quinquenervia (Cav.) Blake. Synonyms include M. leucadendra L., M. viridiflora var. angustifolia (L.f.) Byrnes, M. viridiflora var. rubiflora Brong. and Gris., and Metrosideros quinquenervia Cav.
Linnaeus coined the genus name Melaleuca (Greek melas = black and leucos = white), probably in reference to the fire-charred white bark (Holliday, 1989). The genus contains at least 219 and perhaps more than 250 species (Craven and Lepschi, 1999) and is the third largest angiosperm genus in Australia (Barlow, 1986). Melaleuca quinquenervia is a member of the M. leucadendra (L.) L. group, 15 species that are mostly large shrubs or trees and occur primarily in northern and northeastern Australia (Boland et al., 1987; Craven, 1999).
Holliday (1989) and Bodle et al. (1994) describe M. quinquenervia as being erect trees up to 25 m tall with multi-layered, thick white or grayish papery bark that insulates the trunk and branches. Leaves are lanceolate-elliptical to oblanceolate with five prominent longitudinal veins; up to 3 cm broad and 10 cm long; and flat, stiff, and leathery when mature. In general, woody biomass constitutes a major portion (83 to 96%) of dry weight across habitats (non-flooded, seasonally flooded, and permanently flooded), with the remaining portion (4 to 17%) being comprised of foliage and seed capsules (Rayachhetry et al., 2001). In Florida, some M. quinquenervia trees become reproductive within a year of germination, and flowering events occur several times a year (Meskimen, 1962). Inflorescences are indeterminate, 2 to 5 cm long, and arranged in bottlebrush-like spikes (Holliday, 1989). Flowers are white or cream colored, with tripartite ovaries surrounded by five sepals. Five petals surround 30 to 50 stamens, and a pistil. Capsular fruits are persistent, arranged in the series of clusters, and may remain attached to the trunks, branches, or twigs for several years (Meskimen, 1962). In Florida, a flower spike can produce 30 to 70 sessile capsules (Meskimen, 1962). In excess of seven linearly occurring capsule clusters (each separated by series of leaves) have been recorded from M. quinquenervia branches (Rayachhetry et al., 1998). Seed capsule biomass (dry weight) on trees in permanently flooded habitats is two-fold greater compared to seasonally flooded habitats (3 to 4% vs. 2% of total biomass). These serotinous capsules release seeds when their vascular connections are disrupted by increased bark thickness or stresses such as fire, frost, mechanical damage, herbicide treatments, or self-pruning of branches (Woodall, 1982; Hofstetter, 1991). The canopy of a mature tree (38 cm diameter at breast height and 12 m height) may hold up to 1.4 kg of seeds (about 56 million seeds) (Rayachhetry, unpub. data). While massive, synchronous seed release occurs in response to various stresses, some capsules open successively in a non-synchronous manner, resulting in a light but constant seed rain (Woodall, 1982; Hofstetter, 1991). In Florida, capsules contain 200 to 350 seeds each (Meskimen, 1962) and each seed weighs ca. 25 g (Rayachhetry, unpub. data). Only about 15% of the canopy-held seeds in Florida are filled (embryonic). Overall, about 9% of seeds are viable and 7% can germinate, suggesting that ca. 2% remain dormant (Rayachhetry et al., 1998). Enforced dormancy under field conditions is suggested by the fact that a small proportion of buried seeds remain germinable for more than two years (Van, unpub. data). Forest fires reduce competition, prepare ash-enriched forest floors, and promote establishment and rapid growth of seedlings, provided the soil remains wet and the canopy is open. Seed germination occurs in both shade and sun, as well as under submerged conditions (Meskimen, 1962; Lockhart, 1995). However, prolonged submergence (six to 12 months) and fire can kill smaller seedlings (Myers, 1975; Woodall, 1981).
Because of the massive seed release from mother trees, extremely dense (more than 250,000/ha of 3- to 4 m-tall trees) M. quinquenervia stands are common (Hofstetter, 1991; Van et al., 2000). Standing biomass of 129 to 263 metric ton/ha has been reported for M. quinquenervia in the United States and Australia (Van et al., 2000). Melaleuca quinquenervia is fire adapted (Stocker and Mott, 1981; Ewel, 1986). It has layers of thick, spongy bark; dormant epicormic buds on trunks that regenerate new shoots; and is capable of sprouting from roots (Turner et al., 1998).
Analysis of Related Native Plants in the Eastern United States
The Myrtaceae is a large, diverse plant family with approximately 100 genera and 3,000 species worldwide (Stebbins, 1974). It is almost entirely tropical in distribution. The group achieves maximum diversity in Australia, where several hundred species are known, but it is also quite diverse in the New World tropics. The family was formerly divided into two groups (the Myrtoideae and Leptospermoidae) based upon characteristics of the fruit. The Myrtoideae, which are centered in tropical America, produce berries whereas the Leptospermoideae, which are centered in Australia, produce serotinous capsules. Eight indigenous species of Myrtoideae occur in the continental United States and Florida (Tomlinson, 1980), including the genera Eugenia, Calyptranthes, Psidium, and Myrcianthes, commonly referred to as “stoppers.” Some species such as red stopper (Eugenia rhombea Krug and Urban) and long-stalked stopper (Psidium longipes [Berg] McVaugh) are rare and considered endangered. All native Florida species are threatened by loss of habitat to development. At least 30 non-native species of Eugenia (sensu latu), as well as species in other Myrtoideae genera, are cultivated in Florida for their edible fruits and for ornamental uses (Menninger, 1958). Besides M. quinquenervia, about 56 additional Melaleuca species have been imported to Florida, of which at least 16 and 14 species were common in California and Florida, respectively, during the first decade of the 20th century (Dray, pers. comm.). Current status of these additional Melaleuca species in both states is not known. No native species of Leptospermoideae occur in North America. Besides Melaleuca, the only representatives of Leptospermoideae present in Florida are a few species of Australian native Callistemon (bottlebrush, some of which have been recently transferred to the genus Melaleuca). These have been widely planted as landscape ornamentals.
History of Biological Control Efforts in the Eastern United States
Area of Origin of Weed
Australia is clearly the center of origin for the genus Melaleuca, but a few tropical species within this genus extend into New Guinea, New Caledonia, Malaysia, and Burma (Holliday, 1989; Craven, 1999). The M. leucadendra group consists of broad-leaved Melaleuca species, including M. quinquenervia and 14 closely related species (Craven, 1999). This group is widespread along the eastern coast of Australia, from Sydney to Cape York. It also occurs in New Caledonia and the southern parts of Papua New Guinea and Irian Jaya (Blake, 1968). In Australia, M. quinquenervia is more common in the southern part of its range, mainly growing along streams and in swamps (Holliday, 1989), or in seasonally inundated, low-lying areas. Five separate sources (Nice, France; Ventimiglia, Italy; Tamatave, Madagascar; Sydney, Australia; and Burringbar, Australia), mostly botanical gardens or plantations, have been identified for the M. quinquenervia seeds that were introduced into Florida (Dray, pers. comm.).
Areas Surveyed for Natural Enemies
Preliminary surveys to locate biological control agents for M. quinquenervia were conducted in New Caledonia and southeastern Queensland in 1977 (Habeck, 1981). The United States Department of Agriculture, Australian Biological Control Laboratory (USDA ABCL), started a long-term exploration program in 1986. Surveys have been conducted from south of Sydney in New South Wales, along the eastern seaboard of Australia, to Cape Flattery in northern Queensland. Searches for biological control agents on other broad-leaved Melaleuca spp., closely related to M. quinquenervia, also were conducted near Darwin in the Northern Territory, and in southern Thailand. During November 1999, several species in the M. leucadendra complex were surveyed for pathogens in southern and northern Queensland and northeastern New South Wales in Australia. A number of microorganisms have been found associated with M. quinquenervia in Florida and in Australia.
Natural Enemies Found
More than 450 plant-feeding insect species have been collected from M. quinquenervia in Australia, and an additional 100 species have been collected from closely related Melaleuca spp. (Balciunas et al., 1994a, 1995). Of the major herbivores (Table 1), seven species have been intensively studied, but only five have been introduced into domestic quarantine facilities. Only the Melaleuca snout beetle (leaf weevil), Oxyops vitiosa Pascoe, and the Melaleuca psyllid, Boreioglycaspis Melaleucae (Moore), have been released. The bud gall fly, Fergusonina n. sp. Malloch, is currently undergoing host range testing. The Melaleuca defoliating sawfly, Lophyrotoma zonalis (Rohwer), is being tested for vertebrate toxicity. The mirid bug Eucerocoris suspectus Distant and the tip wilting bug, Pompanatius typicus Distant, though very damaging (Burrows and Balciunas, 1999) were found to be insufficiently host specific for introduction. Other insects, including a leaf-galling cecidomyiid (Lophodiplosis indentata Gagné), several flower-feeding tortricids (Holocola sp., Thalassinana species group), and the tube-dwelling pyralid moth Poliopaschia lithochlora (Lower), are currently undergoing preliminary host range testing in Australia.
Table 1. Insects under investigation for Biological Control of Melaleuca quinquenervia
Previously, a few fungal species had been reported from M. quinquenervia and its allies in Florida, Australia, and some other parts of the world (Alfieri et al., 1994; Rayachhetry et al., 1996ab, 1997). Four additional fungal species (Fusarium sp., Pestalotiopsis sp., Phyllosticta sp., Guignardia sp.) have recently been found to be associated with M. quinquenervia and its close relatives in Australia (Rayachhetry, unpub. data).
Host Range Tests and Results
Three herbivorous insect species (O. vitiosa, L. zonalis, and B. Melaleucae) have been subjected to intensive host specificity tests. These host range studies have shown O. vitiosa, L. zonalis, and B. Melaleucae to be specific to M. quinquenervia. Small amounts of feeding and development through only one generation in the laboratory were found on a few test plant species, mostly Callistemon spp. (Balciunas et al., 1994b; Buckingham, 2001; Center et al., 2000; Wineriter and Buckingham, 1999). However, extensive field studies in Australia determined that Callistemon spp. are not host plants of O. vitiosa, L. zonalis, or B. Melaleucae (Balciunas et al., 1994b; Burrows and Balciunas, 1997; Purcell et al., 1997). Currently, federal authorities are preparing biological and environmental assessments for the field release of B. Melaleucae. Australian host range studies have demonstrated that the bud-gall fly, Fergusonina sp., also is highly specific to M. quinquenervia and it is undergoing further host specificity testing at the quarantine facilities in Florida.
Botryosphaeria ribis (a canker fungus) and Puccinia psidii (a rust fungus) were evaluated in south Florida as potential M. quinquenervia biological control agents. Botryosphaeria ribis appeared to be a plurivorous pathogen (Smith, 1934) that attacked stressed plants (Punithalingham and Holiday, 1973; Rayachhetry et al., 1996c,d ), while P. psidii attacked vigorously growing M. quinquenervia branch tips (Rayachhetry et al., 1997) and had a host range restricted to the family, Myrtaceae (Rayachhetry et al., 2001b).
Of the five insects imported into Florida quarantine, only O. vitiosa and B. Melaleucae have been released (Center et al., 2000). Adults and/or larvae of O. vitiosa were released during spring 1997 at both permanently and seasonally flooded habitats. By winter 2000, more than 47,000 adults and 7,000 larvae had been released at more than 97 locations in south Florida. Oxyops vitiosa established at all but the permanently flooded sites. Even small releases of 60 adults successfully produced viable populations when site conditions were favorable (Center et al., 2000). However, populations have dispersed slowly. Ease of establishment and slow dispersal suggested an optimal introduction strategy of numerous small releases at carefully selected but widely dispersed sites. Currently, the distribution of O. vitiosa is limited compared to the vast area occupied by M. quinquenervia. Therefore, concerted establishment and redistribution efforts are ongoing to ensure the widespread colonization of M. quinquenervia in south Florida. Boreioglycaspis Melaleucae was first released during Spring 2002 but it is not yet certain whether populations have established.
Biology and Ecology of Natural Enemies
Oxyops vitiosa (Coleoptera: Curculionidae)
Oxyops vitiosa larvae prefer to feed on relatively new foliage (Fig. 4a), while adults feed on both young (Fig. 4b) and old (Fig. 4c) foliage. The resultant damage stunts growth of saplings and reduces foliage production in older trees. Larvae are most damaging, feeding on one side of a leaf through to the cuticle on the opposite side, which produces a window-like feeding scar (Fig. 4a). This damage may persist for months, ultimately resulting in leaf drop (Fig. 5). Adult feeding on young and mature leaves is characterized by holes (Fig. 4b) and narrow scars along the leaf surfaces (Fig. 4c), respectively. Oviposition occurs mainly during daylight hours from September to March in Florida (Center et al., 2000). Eggs are laid singly, or in small clusters, on the surface of young leaves, usually near their apex, or on stems of young shoots. A hardened black-to-tan coating of frass and glandular materials covers individual eggs. In Florida, larvae are absent or uncommon from April to August unless damage-induced regrowth is present (Center et al., 2000). Pupation occurs in the soil, usually beneath the host plant. Egg-to-adult development requires about 50 days. Females survive up to 10 months and can produce more than 1,000 eggs. Adults can be collected year round.
Boreioglycaspis Melaleucae (Homoptera: Psyllidae)
The Melaleuca psyllid, B. Melaleucae, severely damages M. quinquenervia, especially in the absence of its predators and parasites. Nymphs are parasitized by Psyllaephagus sp. (Hymenoptera: Encyrtidae) and preyed on by coccinellids (Coleoptera) and lygaeids (Hemiptera) in Australia. This psyllid was collected in northern and southeastern Queensland and northern New South Wales during field surveys in Australia. Collection records also exist for Western Australia and the Northern Territory. Psyllids, both adults and nymphs, reportedly feed on phloem sap through the stomata (Clark, 1962; Woodburn and Lewis, 1973); however, nymphs cause the most damage by inducing defoliation and sooty mold growth on excreted honeydew. Populations of B. Melaleucae grow rapidly, causing moderate leaf curling, discoloration, defoliation, and plant mortality.
Adults of B. Melaleucae mate throughout the day and the male grasps the female with large abdominal claspers (parameres) before mating. Females oviposit on leaves or stems of host plants and lay an average of 78 eggs. Each egg is attached to the leaf by a pedicel that is inserted into the plant tissue to absorb water (White, 1968). Most eggs hatch within 18 days. Nymphs of B. Melaleucae congregate on leaves and secrete white, flocculent threads, which can completely cover the nymphs. These secretions facilitate easy detection at field sites. Like all psyllids, B. Melaleucae has five instars (Hodkinson, 1974) and development from egg to adult takes 28 to 40 days. Purcell et al., (1997) present a complete biology of B. Melaleucae.
Lophyrotoma zonalis (Hymenoptera: Pergidae)
The defoliating melaleuca sawfly, L. zonalis, was the most damaging insect observed on Melaleuca spp. in Australia. It was collected from Mackay in central Queensland to the Daintree River in north Queensland, and near Darwin in the Northern Territory. Records also indicate its presence in New Guinea (Smith, 1980). Larvae are voracious leaf feeders and dense populations cause complete defoliation. Defoliation stresses trees and reduces flowering during subsequent years (Burrows and Balciunas, 1997). Adults do not feed on the plant tissue. They are frequently observed swarming around the bases of trees. Larvae burrow into the papery bark of M. quinquenervia to pupate, unlike many other pergid sawflies that pupate in soil. It therefore should be an excellent agent for use in wetter areas, where other agents are less effective. Females are parthenogenic, producing all males when unmated, while mated females produce both males and females.
Burrows and Balciunas (1997) provide a detailed description of the life history of L. zonalis. The life cycle from egg to adult takes approximately 12 weeks. Females insert eggs into the tissue along the edges of leaves using their saw-like ovipositors. The subsequent egg batches form a line along the leaf margin, and harden and turn brown with age. Females oviposit up to 140 eggs in their lifetime, which are heavily parasitized in Australia. The neonate larvae feed gregariously, forming a feeding front across the leaf; later instars become solitary feeders. Three unidentified fly species (Diptera: Tachinidae) and one wasp species (Hymenoptera: Ichneumonidae) parasitize larvae. The final instar, or prepupa, does not feed and burrows into the bark of the trunk and lower branches to excavate a chamber in which it enters the pupal stage. In Australia, L. zonalis is mainly found during the summer months and a resting, possibly diapausing, prepupal stage occurs during winter. Larval outbreaks also occasionally occur during cooler months.
The toxic peptides lophyrotomin and pergidin, which have been reported in three other sawflies from around the world (Oelrichs et al., 1999), have recently been detected in L. zonalis larvae (Oelrichs, pers. comm.). Consumption of large quantities of larvae of a related sawfly from Eucalyptus sp. causes cattle mortality in Australia, although L. zonalis has never been implicated in livestock or wildlife poisonings (Oelrichs et al., 1999). Therefore, the decision to release L. zonalis in Florida awaits assessment of the risk of this insect to wildlife and livestock.
Fergusonina sp. (Diptera: Fergusoninidae)
The M. quinquenervia bud-gall fly, Fergusonina sp., forms galls in vegetative and reproductive buds of M. quinquenervia in a unique, mutualistic association with nematodes of the genus Fergusobia Currie (Nematoda: Tylenchida: Sphaerulariidae). Preliminary data indicate that the nematode initiates gall production (Giblin-Davis et al., 2001). Fergusonina sp. have been reared from most broad-leaved Melaleuca spp. in Australia, although the flies on each plant species appear to be unique (Taylor, pers comm.). Galls on M. quinquenervia vary greatly in size and color, depending on growth stage and type of buds being attacked, and on developmental stage of the gall. They have the potential to impede branch and foliage growth, and retard flower formation resulting in reduced seed set. These galls also may act as nutrient sinks, reducing plant vigor (Goolsby et al., 2000). However, the gall production is seasonal, with highest densities occurring during periods of maximum leaf bud production, usually during winter and spring (Goolsby et al., 2000). The flies are heavily parasitized by several species of parasitic Hymenoptera in Australia.
Botryosphaeria ribis (Pleosporales: Botryosphaeraceae)
Grossenbacher and Duggar (1911) first described B. ribis from currants (Ribes sp.) in New York. Taxonomy, biology, and ecology of this fungus are discussed in Punithalingam and Holliday (1973), Morgan-Jones and White (1987), and Rayachhetry et al. (1996a). It belongs to a group of fungi that produce conidiospores (asexual spores) in stromatic pycnidia and/or ascospores in ascomata on the surface of stems, leaves, and fruits. The mode of entry into stem tissues is assumed to be through wounds, frost-induced cracks, sun-scorched bark, lenticels, or branch stubs. Stems of healthy plants callus rapidly, and the fungus may remain latent under the callus tissues, causing perennial cankers when trees are stressed. Stems and branches of stressed trees are girdled quickly due to the plants’ inability to callus and compartmentalize the fungus. Infected plants may die back, show vascular wilt, or crown thinning. Affected vascular tissues usually appear brown to black in color (Rayachhetry et al., 1996d).
Puccinia psidii (Uredinales: Pucciniaceae)
Puccinia psidii, commonly known as guava rust, has been reported on 11 genera and 13 species in the family Myrtaceae in Central America, Caribbean Islands, and South America (Laundon and Waterston, 1965; Marlatt and Kimbrough, 1979). In 1996, P. psidii was found to attack healthy new growth of M. quinquenervia (Rayachhetry et al., 1997). Figueiredo et al., (1984) studied the life cycle of P. psidii on Syzygium jambos (L.) Alston, and reported three spore stages (uredospore, teliospore, and basidiospore) in its life cycle. Only uredinial pustules have been observed on M. quinquenervia in Florida, but other stages also may exist. No alternate host has been discovered and it is assumed to be autoecious (Figueiredo et al., 1984). Guava rust attacks both foliage and succulent stems of vigorously growing M. quinquenervia saplings. Rust disease on M. quinquenervia is usually severe during winter and spring. Severe infections cause foliage distortion, defoliation, localized swellings on twigs, and tip diebacks (Rayachhetry et al., 2001b).
Evaluation of Project Outcomes
Establishment and Spread of Agents
Oxyops vitiosa is now established at many locations in south Florida where larvae or adults were released; however, rate of spread is limited (Center et al., 2000). Slowly expanding O. vitiosa populations now exist in Dade, Broward, Lee, Collier, Palm Beach, Martin, Monroe, Sarasota, and Glades Counties. Habitats with short hydroperiods, dry winter conditions, and abundant young foliage favor growth and development of O. vitiosa. Oxyops vitiosa populations did not establish in permanently aquatic sites because of the soil requirement for pupation (Center et al., 2000). Dispersal occurs more rapidly at sites where the trees are scattered savannah-like in open areas. Other factors such as geographical location, hydroperiod, wind direction, life stage released, or date of release do not affect the rate of overall dispersal (Pratt, unpub. data). Also, adults seem to move from unsuitable trees (tall, dense stands with a paucity of young foliage) onto trees that provide acceptable foliage (smaller, bushier, open-grown trees with an abundance of young foliage) (Center et al., 2000).
Suppression of Target Weed
Oxyops vitiosa adults feed on both old and new foliage as well as on emerging vegetative and reproductive buds (Fig. 4). Early instars feed only on young succulent foliage, while late instars are less discriminating (Fig. 4a). Adults feed on both young and mature leaves (Figs. 4b, c). Severe adult or larval feeding results in tip dieback and defoliation (Fig. 5). Repeated damage of growing tips removes apical dominance and induces lateral growth from axillary buds. Subsequent new growth acts as a nutrient sink and sustains continual adult and larval weevil populations. Foliar damage, and the subsequent diversion of photosynthetic resources to the development of new foliage, appears to limit reproductive performance of M. quinquenervia. In preliminary studies, flowering of severely damaged M. quinquenervia trees was reduced more than 90%, (Pratt, unpub. data).
Repeated defoliation weakens the trees’ defense mechanisms, predisposing them to attack by other insects and pathogens. As a result, existing populations decline as their regenerative capabilities become reduced. The diverse community of insects that damage M. quinquenervia in Australia probably suppresses the regenerative potential of native Melaleuca forests. For example, the number of seed-capsules per unit of infructescense length is three and eight capsules/cm in Australia and Florida, respectively. Similarly, the viability (9.1 vs. 3.3%) and germinability (in 14 days, 7.8 vs. 2.8%) of M. quinquenervia seeds are significantly higher in the United States than in Australia. The reduction in seed production, and thus the invasibility of M. quinquenervia, is the primary objective of the biological control program. While removal of existing stands may be best accomplished by other means (herbicides and mechanical removal), a reduction in canopy seed production through biological control should enhance the efficacy of the overall management program (Laroche, 1999).
Recovery of Native Plant Communities
The diversity and abundance of native plant species in areas invaded by M. quinquenervia should begin to recover as M. quinquenervia canopies open due to crown thinning and/or tree mortality resulting from feeding by biological control agents. Long-term monitoring programs have been initiated by establishing permanent plots in M. quinquenervia-infested sites to document such events in dry, seasonally inundated, and aquatic habitats.
The containment and/or elimination of M. quinquenervia monocultures should produce economic benefits by sustaining the tourist industry, permitting the recovery of native flora and fauna, decreasing the risk to human health, and reducing the fire hazard to urban areas near highly flammable M. quinquenervia stands.
Recommendations for Future Work
Currently, the M. quinquenervia biological control program is focused on procuring additional biological control agents. Additional quarantine space is needed to improve and accelerate host testing of additional agents. Construction of a new facility designed for this purpose began at Fort Lauderdale during December 2001. The primary focus of the Fort Lauderdale Invasive Plant Research Laboratory has been the release of new agents as they become available, and the evaluation of those agents that establish. To combat the M. quinquenervia invasion and successfully reduce its impact, state and federal agencies will need to (1) continue foreign exploration for new biological control agents, with special emphasis on those that will complement the effects of existing agents; (2) continue to evaluate host specificity and efficacy of promising agents; (3) import selected agents into quarantine for further evaluation; (4) accelerate release programs through development of efficient testing facilities and reduction of avoidable delays; (5) develop a thorough understanding of the biology and ecology of the host as well as the candidate biological control agents, both in Florida and Australia, to enhance agent selection and subsequent establishment; (6) acquire necessary permits for field release of the bud-gall fly into M. quinquenervia populations in south Florida; (7) continue to monitor field populations of established agents and redistribute them to new locations as needed; and (8) monitor the impact of released agents at individual plant, community, and landscape scales.
Because M. quinquenervia is a large perennial tree, the effect of biological control agents likely will be slow and cumulative over an extended period of time. In addition to O. vitiosa, and B. Melaleucae which have already been released, other insects are either waiting for field-release permission or undergoing evaluation in Australia or in U.S. quarantine. Therefore, evaluation of the performance of released agents in the field and their relationship with predators and pathogens in Florida should continue with special emphasis on (1) measuring changes in the reproductive potential of existing trees and monitoring for signs of population decline and habitat recovery; (2) assessing the impact of predators, parasitoids, and pathogens on the released biological control agent populations; (3) monitoring other plant species to validate host specificity research and determine whether non-target effects occur; and (4) developing and integrating selected fungal agents into the suite of herbivorous biological control agents.
Alfieri, S. A., K. R. Langdon, J. W. Kimbrough, N. E. El-Gholl, and C. Wehlburg. 1994. Diseases and
Allen, Jr., L. H., T. R Sinclair, and J. M. Bennett. 1997. Evapotranspiration of vegetation of Florida:
Alexander, T. R., R. H. Hofstetter, and F. Persons. 1977. Comparison of transpiration of cajeput
Austin, D. F. 1978. Exotic plants and their effects in southeastern Florida. Environmental Conservation.
Balciunas, J. K., D. W. Burrows, and M. F. Purcell. 1994a. Insects to control Melaleuca I: status of
Balciunas, J. K., D. W. Burrows, and M. F. Purcell. 1994b. Field and laboratory host ranges of the
Balciunas, J. K., D. W. Burrows, and M. F. Purcell. 1995. Insects to control Melaleuca II: Prospects for
Barlow, B. A. 1986. Contribution to a revision of Melaleuca (Myrtaceae): 1-3. Brunonia 9: 163-177.
Barlow, B. A. 1988. Patterns of differentiation in tropical species of Melaleuca L. (Myrtaceae).
Blake, B. A. 1968. A revision of Melaleuca leucadendron and its allies (Myrtaceae). Contributions from
Bodle, M. J., A. P. Ferriter, and D. D. Thayer. 1994. The biology, distribution, and ecological
Boland, D. J., M. I. H. Brooker, G. M. Chippendale, N. Hall, B. P. M. Hyland, R. D. Johnston, D. A.
Buckingham, G. R. Quarantine host range studies with Lophyrotoma zonalis, an Australian sawfly of
Burrows, D. W. and J. K. Balciunas. 1997. Biology, distribution and host-range of the sawfly,
Burrows, D. W. and J. K. Balciunas. 1999. Host-range and distribution of Eucerocoris suspectus
Center, T. D., T. K. Van, M. B. Rayachhetry, G. R. Buckingham. F. A. Dray, S. Wineriter, M. F.
Clark, L. R. 1962. The general biology of Cardiaspina albitextura (Psyllidae) and its abundance in
Craven, L. A. 1999. Behind the names: the botany of tea tree, cajuput and niaouli, pp. 11-28. In
Craven, L. A. and B. J. Lepschi. 1999. Enumeration of the species and intraspecific taxa of Melaleuca
Cronquist, A. 1988. The Evolution and Classification of Flowering Plants. New York Botanical Garden,
Crowder, J. P. 1974. Exotic pest plants of South Florida. A study appendix to the South Florida
Di Stefano, J. F. and R. F. Fisher. 1983. Invasion potential of Melaleuca quinquenervia in southern
Diamond, C., D. Davis, and D. C. Schmitz. 1991. Economic impact statement: The addition of
Ewel, J. J. 1986. Invasibility: lessons from south Florida, pp. 214-230. In Mooney, H. A. and J. A. Drake.
Figueiredo, M. B, L. N. Coutinho, and J. F. Hennen. 1984. Studies on the determination of the life cycle
Giblin-Davis, R. M., J. Makinson, B. J. Center, K .A. Davies, M. Purcell, G. S. Taylor, S. J. Scheffer, J.
Gifford, J. C. 1937. The cajeput tree in Florida. The American Eagle 32(29): 1.
Goolsby, J. A., J. Makinson, and M. F. Purcell. 2000. Seasonal phenology of the gall-making fly,
Grossenbacher, J. G. and B. M. Duggar. 1911. A contribution to the life history, parasitism, and biology
Habeck, D. H. 1981. Potential for biological control of Melaleuca, pp. 125-128. In Geiger, R.K. (ed.).
Hodkinson, I. D. 1974. The biology of Psylloidea (Homoptera): a review. Bulletin of Entomological
Hofstetter, R. L. 1991. The current status of Melaleuca quinquenervia in southern Florida, pp. 159-176.
Holliday, I. 1989. A Field Guide to Melaleucas. Hamlyn, Port Melbourne, Australia.
Laroche, F. B. 1994. Melaleuca Management Plan for Florida. South Florida Water Management
Laroche, F. B. 1999. Melaleuca management efforts, South Florida Water Management District, pp.
Laroche, F. B., and A. P. Ferriter. 1992. The rate of expansion of Melaleuca in south Florida. Journal of
Laroche, F. B., D. D. Thayer, and M. J. Bodle. 1992. Melaleuca response to various herbicides and
Laundon, G. F. and J. M. Waterson. 1965. C. M. I. Description of Pathogenic Fungi and Bacteria. No.
Lockey, R. F., J. J. Stablein, and L. R. F. Binford. 1981. Melaleuca tree and respiratory disease.
Lockhart, C. S. 1995. The effect of water level variation on the growth of melaleuca seedlings from the
Marlatt, R. B., and J. W. Kimbrough. 1979. Puccinia psidii on Pimenta dioica in south Florida. Plant
Menninger, E. A. 1958. Florida’s 9 native species of Eugenia. Florida State Horticultural Society,
McGehee, J. T. 1984. Melaleuca research and control strategy for Lake Okeechobee. In Takekawa, J.,
Meskimen, G. F. 1962. A silvical study of the Melaleuca tree in south Florida. M.S thesis, University of
Morgan-Jones, G., and J. F. White, Jr. 1987. Notes on Coelomycetes. II. Concerning the Fusicoccum
Morton, J. F. 1962. Ornamental plants with toxic and/or irritant properties. II. Proceedings of the Florida
Myers, R. L. 1975. The relationship of site conditions to the invading capability of Melaleuca
Oelrichs, P. B., J. K. Macleod, A. A. Seawright, M. R. Moore, F. Dutra, F. Riet-Correa, M. C. Mendez,
Punithalingam, E. and P. Holliday. 1973. Botryosphaeria ribis. Description of Pathogenic Fungi and
Purcell, M. F., J. K. Balciunas, and P. Jones. 1997. Biology and host-range of Boreioglycaspis
Rayachhetry, M. B., G. M. Blakeslee, R. S. Webb, and J. W. Kimbrough. 1996a. Characteristics of the
Rayachhetry, M. B., G. M. Blakeslee, and R. Charudattan. 1996b. Susceptibility of Melaleuca
Rayachhetry, M. B., G. M. Blakeslee, and T. D. Center. 1996c. Predisposition of Melaleuca (Melaleuca
Rayachhetry, M. B., G. M. Blakeslee, and T. Miller. 1996d. Histopathology of Botryosphaeria ribis in
Rayachhetry, M. B., M. L. Elliot, and T. K. Van. 1997. Natural epiphytotic of a rust fungus (Puccinia
Rayachhetry, M. B., T. K. Van, and T. D. Center. 1998. Regeneration potential of the canopy-held
Rayachhetry, M. B., T. K. Van, T. D. Center, and F. B. Laroche. 2001a. Aboveground biomass
Rayachhetry, M. B., T. K. Van, T. D. Center, and M. L. Elliott. 2001b. Host range of Puccinia psidii, a
Resource Assessment Commission. 1992. Coastal Zone Inquiry Draft Report: Summary and Interim
Schmitz, D. C. and R. H. Hofstter. 1999. Environmental, economic and human impacts, pp. 17-21. In
Smith, D. R. 1980. Pergidae (Hymenoptera) from New Guinea and Australia in the Bishop Museum.
Smith, O. C. 1934. Inoculations showing the wide host range of Botryosphaeria ribis. Journal of
Stanaland, B. E., R. N. Gennaro, S. D. Klotz, M. J. Sweeney, and R. S. White. 1986. Isolation and
Stebbins, G. L. 1974. Flowering Plants. Evolution above the Species Level. Balknap Press, Cambridge,
Stocker, G. C., and J. J. Mott. 1981. Fire in the tropical forests and woodlands of northern Australia, pp.
Tomlinson, P. B. 1980. The Biology of Trees Native to Tropical Florida. Harvard University Press,
Turner, C. E., T. D. Center, D. W. Burrows, and G. R. Buckingham. 1998. Ecology and management of
Van, T. K., M. B. Rayachhetry, and T. D. Center. 2000. Estimating aboveground biomass of Melaleuca
White, T. C. R. 1968. Uptake of water by eggs of Cardiaspina densitexta (Homoptera: Psyllidae) from
Wineriter, S. A. and G. R. Buckingham. 1999. Biological Control of Melaleuca—Insect Quarantine
Woodburn, T. L. and E. E. Lewis. 1973. A comparative histological study of the effects of feeding by
Woodall, S. L. 1981. Site requirements for Melaleuca seedling establishment, pp. 9-15. In Geiger, R. K.
Woodall, S. L. 1982. Seed dispersal in Melaleuca quinquenervia. Florida Science 45: 81-93.
|Bargeron, C.T., D.J. Moorhead, G.K. Douce, R.C. Reardon & A.E. Miller |
(Tech. Coordinators). 2003. Invasive Plants of the Eastern U.S.:
Identification and Control. USDA Forest Service - Forest Health
Technology Enterprise Team. Morgantown, WV USA. FHTET-2003-08.