Specialized Terms
haplodiploidy A genetic system in which unfertilized eggs develop into males, while fertilized eggs develop into females.
propagule The initial group of individuals that found an introduced population.
An introduced species is one that did not achieve its current taxonomic status in some current location. Such species are often called nonindigenous or adventive there. Among non-indigenous species, those carried to the location by humans, either deliberately or inadvertently, are called “introduced species,” whereas those that arrived on their own are termed “immigrant.” However, all nonindigenous species are often colloquially termed “introduced.” “Exotic” and “alien” are sometimes used for all nonindigenous species.
Introduced insects are numerous and generate major ecological and economic impacts by a variety of means. They span most insect orders. Their means of introduction are varied, but most arrive with human help. Only a minority become problematic, and the reasons why a newly arrived species survives or fails to establish a population, and, if established, has or does not have a major effect, are often mysterious. Some problematic introduced species can be eradicated, and several management procedures can adequately control others.
NUMBERS
Introduced species often comprise a substantial fraction of a regional entomofauna. For example, Great Britain has ca. 21,000 native insects and ca. 200 established, introduced ones (comprising ca. 1% of the total). Israel has ca. 15,000 native species and 220 introduced ones (ca. 1.5% of the total). For the contiguous United States, ca. 84,000 native species have been identified, as have 1862 introduced species (ca. 2% of the total). Florida has relatively more introduced species: 11,509 native and 993 introduced (8%). For oceanic islands, the introduced proportion can be much greater. Thus, the Hawaiian Islands have ca. 5400 native insects and 2600 introduced ones (32% introduced), whereas the mid-Atlantic island Tristan da Cunha has 84 native insects and 32 introduced ones (28% introduced). The greater proportions of introduced species on islands probably result more from the mathematics of smaller numbers of native species and relatively larger numbers of attempted introductions than from some inherent invasibility of island communities or stronger “biotic resistance” of continental ones.
SYSTEMATICS
Established introduced insects are not a systematically random group. An order may be over- or underrepresented, reflecting primarily why it was introduced and/or how it arrived. This fact is exemplified in Fig. 1. which depicts the fractions, by order, of insects of
FIGURE 1 Fractional representation of different insect orders of the world (cross-hatched; data from R. H. Arnett, 1983, “Status of the taxonomy of the insects of America north of Mexico: A preliminary report prepared for the subcommittee for the insect fauna of North America project”; privately printed) and established introduced insects of the United States.
the world compared to introduced insects of the contiguous United States. Far more species of homopterans, hemipterans, and thysa-nopterans are introduced than would be expected based on their numbers in the world, because they are associated with introduced agricultural and ornamental plants. Hymenopterans are overrep-resented, perhaps partly because they are often used in biological control and partly because their haplodiploidy allows a single female to found a population, as she can produce both male and female offspring. Haplodiploidy may also lower inbreeding depression in founding populations. Haplodiploidy is also found among homopter-ans and thysanopterans, both overrepresented groups. At the family level, 26 families are overrepresented, and other patterns become clear. In the contiguous United States, coccinellid beetles and anthocorid bugs are both overrepresented; species in both groups were introduced as biological control predators. Tephritid flies were introduced deliberately to control knapweeds and inadvertently on cultivated plants. Ten overrepresented families are Hymenoptera; seven of these were introduced for biological control. Two thrips families are associated with cultivated plants, as are the adelgids (pine and spruce aphids). Oestrid flies are overrepresented, and these are internal parasites of mammals, including livestock; they probably arrived with their hosts. Dermestid and anobiid beetles are overrepresented; these stored product pests probably arrived with their food sources.
WHY, WHEN, AND HOW THEY ARE
INTRODUCED
Some introduced insects arrive in new locations on their own and are true immigrants. Any simple range extension could bring a new introduced species in this sense, but we generally restrict introduced status to species arriving at a distant location by a discontinuous dispersal,rather than gradually diffusing from a neighboring site. Thus, the monarch butterfly (Danaus plexippus) in Australia might not be considered introduced. It arrived around 1870 and established a population, having spread through the Pacific during the 19th century largely unaided by humans. The monarch migrates long distances in its native North America, is an occasional straggler in Europe, and was recorded successively in Hawaii, the Caroline Islands, Tonga, and New Zealand before reaching Australia.
Some insects are introduced deliberately by humans. A few arrive as pets or pet food; recent pet price lists include mantids, walkingsticks, spider wasps, velvet ants, and dung and blister beetles. In Florida in 1989, giant Madagascan hissing cockroaches ( Gromphadorhina portentosa) became popular as pets; at least some were released to the wild, where they survive. The European honeybee (Apis mellifera) has been introduced worldwide for both the production of honey and crop pollination. The Asian silkworm ( Bombyx mori) has been widely introduced along with its host plant, mulberry, for silk production. Even introductions that fail to establish a commercial industry can nonetheless establish a population. The Asian ailanthus moth (Samia walkeri) was brought to the United States in an attempt to found a silk industry on the ailanthus tree. Although the industry foundered, the moth remains. The gypsy moth ( Lymantria dispar) was brought to North America to establish a silk industry; its predictable escape established one of North America’s major pests.
Most deliberate insect introductions are for biological control. Although weeds and insect pests of agriculture are the usual targets, there are others. Thus, the Brazilian ant, Paratrechina fulva, was introduced to Colombia to control poisonous snakes, and over 45 dung beetle species were introduced to Australia to break down droppings of introduced livestock (ca. 1/3 established populations). Insects introduced to attack insects are either predators (mainly coccinellid beetles, but including other beetles, hemipterans, and neuropterans) or parasitoids (mostly hymenopterans, but including some dipterans). Insects introduced to control plant pests include primarily flies, beetles, and moths, although others such as bugs and thrips have been employed. The number of insect species introduced for biological control purposes is substantial. For example, of ca. 2600 introduced insect species established in the Hawaiian Islands, ca. 400 were introduced for biological control. Further, about twice as many species introduced there for this purpose perished.
Far more insects are introduced inadvertently than deliberately by humans. Pathways are myriad. Soil ballast was an early predominant mode of entry to North America—many of the first introduced insects were soil beetles from southwestern England. This pathway is less common now, but insects are still carried in rootballs around cultivated plants and soil on heavy equipment. Phytophages, particularly homopterans, dominated introductions to North America in the 19th century with the advent of fast steamships and a proliferation of imported nursery stock, and imported plants are still a major means of introducing insects worldwide. The phylloxera (Daktulosphaira vitifo-liae) that devastated French vineyards in the 19th century arrived on saplings or cuttings of American vines. Insects can also be carried in water. The yellow fever mosquito (Aedes aegypti) probably arrived in colonial America in drinking water casks, whereas the Asian tiger mosquito (Aedes albopictus) reached North America in the 1980s in scrap tires from Japan. This is probably the route taken by Ochlerotatus japonicus, which arrived in the United States in 1998 and vectors West Nile virus. These latter two mosquitoes have recently been detected in used tires in New Zealand. Wooden packing material brought the Asian longhorned beetle (Anoplophora glabripennis) and probably the emerald ash borer (Agrilus planipennis) to North America, whereas the European elm bark beetle (Scolytus multistria-tus) that vectors Dutch elm disease arrived on unpeeled veneer logs of European elm. The growth of international tourism can enhance the rate of insect introduction; in 1992, an Australian tourist returned from South America with a wound containing maggots of the New World screw-worm fly (Cochliomyia hominivorax).
ESTABLISHMENT AND SPREAD
The great majority of introduced insects do not survive, although only for biological control introductions are there substantial data on failed introductions. For parasitoid species introduced to control insect pests, only about 30% establish populations, whereas for all insects introduced for plant control, the comparable figure is about 60%. Because biocontrol candidates are chosen and often tested for survival in the target environment, one might expect failure rates for inadvertently introduced introductions to be even higher. For most taxa, invasion biologists believe that 5-20% of introduced species establish populations, although many of these may remain for years or in perpetuity near the point of introduction, and a large fraction are restricted to anthropogenous habitats such as human habitations or agricultural fields.
An arriving propagule must be large enough to survive the initial threat of demographic stochasticity—that is, random elimination of so many individuals during the first few generations that the population fails. Data from the biological control literature show that probability of establishment increases with propagule size and number of attempts, but many very small propagules have established large, widespread populations. For example, a single fertilized female of the cochineal insect, Dactylopius opuntiae, from Ceylon initiated a large, ongoing population on Mauritius. The European solitary bee, Lasioglossum leu-cozonium, probably colonized North America as a lone singly-mated female; it is now widespread in Canada and parts of the northeastern United States. In Puerto Rico, two females of a unisexual race of the encyrtid wasp, Hambletonia pseudococcina. were used to rear 7000 individuals, which were released, established, and quickly spread. For a parthenogenetic species, at least the difficulty of finding a mate is obviated, but demographic stochasticity has other components.
Even assuming an adequate propagule size, the environment, both physical and biotic, must be suitable for a species to survive and spread. Predators, parasites, competitors, and pathogens can eliminate an introduced species or restrict its ambit. For example, the Asian aphelinid wasp parasitoid Aphytis fisheri. introduced to California to control California red scale (Aonidiella aurantii), failed to establish because of competition from previously introduced A. melinus and A. lingnanensis. Conversely, the absence of natural enemies from its native range is often posited as the reason for the success of some invader, as for the cynipid Andricus quercuscali-cis into Great Britain. By contrast, the presence of some other species, such as a food plant or a symbiont, might be necessary for an invader to survive. The monarch butterfly would not have survived in Australia but for the prior introduction of its host milkweeds. The physical environment is probably even a more frequent reason an introduced species perishes. For instance, a temperate climate does not augur well for a newly arrived tropical insect.
IMPACTS
Quantifying the impact of a new species is an unsolved challenge, but it is safe to say that most introduced species do not generate major impacts. However, some are enormously damaging, whereas others are highly beneficial. The variety of impacts is staggering.
Many introduced insects prey on natives. This activity can be useful, as in biocontrol introductions such as that of the Australian vedalia beetle (Rodolia cardinalis) to attack the cottony cushion scale (Icerya purchasi). Predation can also be extremely damaging; on Christmas Island, the introduced crazy ant, Anoplolepis gracilipes, has locally devastated populations of the dominant red crab Gecarcoidea natalis. Because the crab controls seedling recruitment and litter breakdown, the entire community is affected. Parasites can also be beneficial or harmful. The wasp, Aphytis melinus, has effectively controlled California red scale in parts of California. Alternatively, sheep blowfly (Lucilia cuprina), introduced to Australia from Africa, caused massive losses. Herbivory can similarly be beneficial or detrimental. The South American flea beetle, Agasicles hygrophila, effectively controls alligatorweed in Florida. However, phytophagous insect crop pests impose enormous costs. The alfalfa weevil (Hypera postica) caused $500 million in losses in the United States in 1990 alone.
Resource competition is subtler than predation, parasitism, and herbivory, but many introduced insects outcompete natives. The European seven-spotted ladybeetle (Coccinella septempunctata), introduced to the United States for control of the Russian wheat aphid (Diuraphis noxia), has locally outcompeted several native ladybeetles. European honeybees outcompete the native bee, Osmia pumila, for pollen in the New York State. Introduced insects can even outcompete vertebrates. The introduced wasps Vespula ger-manica and V vulgaris in New Zealand outcompete an endemic parrot for honeydew produced by a scale insect (Ultracoelostoma assimile) and have locally lowered parrot populations.
Introduced insects can vector or be reservoirs of diseases of humans, domestic animals, cultivated plants, and wild animals and plants. The spread of yellow fever and dengue as the vector mosquito Aedes aegypti dispersed throughout the tropics, and of malaria to Brazil with the introduction of its vector Anopheles gambiae, are notable examples of human diseases vectored by introduced insects. Cat fleas (Ctenocephalides felis) and dog fleas (C. canis), introduced to Australia with their hosts, are intermediate hosts for the dog tapeworm (Dipylidium caninum). The mosquito, Culex quinquefascia-tus, was accidentally introduced to the Hawaiian Islands in 1826. Subsequently it vectored avian malaria, introduced with resistant Eurasian songbirds, to susceptible native birds and helps to exclude them from low elevations. The mosquito, Ochlerotatus japonicus, transmits West Nile virus to both birds and humans in the northeastern United States. Animal disease vectors may be useful biocontrol introductions. For example, the rabbit flea (Spilopsyllus cuniculi), the main vector of myxomatosis in Europe, has been introduced (so far unsuccessfully) in Australia to attempt to boost disease transmission. In the lab, it also transmits calicivirus.
Among plants, Dutch elm disease dispersed to and through North America with the European elm bark beetle (Scolytus multistriatus). Beech bark disease spread throughout northeastern North America after the causal fungus was introduced ca. 1890 from Europe with its vector, the beech scale (Cryptococcus fagisuga). The California wine industry is plagued by the recent introduction of the glassy-winged sharpshooter (Homalodisca coagulata) from the southeastern United States. The sharpshooter spreads an incurable bacterial disease of grape vines, a malady long present but rarely a problem until this vector arrived.
Introduced species often exacerbate one another’s impacts, a process termed “invasional meltdown.” Sometimes this interaction occurs when coevolved mutualists invade a region separately.
In South Florida, over 60 species of ornamental figs were not invasive because their obligatory pollinating wasps were absent. Since the 1970s, three such wasps (Parapristina spp.) have arrived, and three formerly innocuous fig species have begun spreading in natural areas. However, invasional meltdown need not involve coevolved species. In California citrus orchards, the Argentine ant ( Linepithema humile) tends and protects the California red scale, thereby exacerbating its impact. Similarly, in Hawaii, the African big-headed ant (Pheidole megacephala) protects the tropical American gray pineapple mealybug (Dysmicoccus neobrevipes) from coccinel-lids introduced for biological control.
Relative to the numbers of species introduced, insects rarely cause enormous ecological (as opposed to economic) damage. Introduced species whose impacts ripple through entire communities usually do so by changing the habitat dramatically, and it is mostly plants that do this (by becoming structural dominants or modifying fire regimes) or pathogens that attack dominant plants. Occasionally, mammals can generate an enormous ecosystem impact by trampling or grazing. A recent list of the world’s 100 worst introduced species included 15 insects, but at most one would qualify as having a huge ecosystem-wide impact: the crazy (or long-legged) ant, because it removes the keystone red crab species on Christmas Island. Of the 15, five are ants. In addition to the crazy ant, the Argentine ant, the big-headed ant, the little fire ant (Wasmannia auropunctata), and the red imported fire ant (Solenopsis invicta) all affect other ants greatly, and sometimes other insects, but to date none has had the dramatic impact of certain plants and mammals. Several species among the 15 transmit human diseases (Aedes albopic-tus and Anopheles quadrimaculatus), and others are agricultural pests, such as the sweet potato whitefly (Bemisia tabaci) and several of the ants. The Formosan termite (Coptotermes formosanus shiraki) has caused enormous damage to housing in New Orleans. With respect to ecosystem-wide damage to natural areas, however, the only members of the list that might qualify, aside from the crazy ant, are the gypsy moth in North America, by virtue of its devastating impact on dominant trees, and the Argentine ant, because it has greatly lowered densities of native seed-carrying ants in the fynbos of South Africa. Other insects may have ecosystemic impacts by removing dominant plants. The beetle transmitting Dutch elm disease has already been noted. The Asian balsam woolly adelgid (Adelges piceae) has eliminated the dominant Fraser fir throughout the high southern Appalachians, and the hemlock woolly adelgid (Adelges tsugae) has killed most hemlocks in much of eastern North America and is spreading to the south and west.
Impacts can occur after a substantial lag period during which an introduced species can appear innocuous. For example, the beetle Chrysolina quadrigemina, introduced to Australia in 1939 to control St John’s wort, seemed to die out but resurfaced and spread in 1942. Such lags are mysterious; they are often attributed to favorable changes in the environment or to evolution of the invader, but evidence for these phenomena is generally lacking. Some introduced insects achieve great numbers and appear to have a major impact, but the population suddenly crashes, again often for reasons poorly understood. After rapid increase and spread, most populations of the European browntail moth (Euproctis chrysorrhoea) crashed in New England and eastern Canada, probably because of parasitism by a tachinid fly (Compsilura concinnata) introduced to control the gypsy moth.
EVOLUTION
The conditions under which a species is introduced to a new region [isolation from parent population, small propagule size (usually), and different physical and biotic environment] should conduce to rapid
evolution. There has been little study of this phenomenon, but some striking examples have emerged. Drosophila subobscura, introduced to the Americas from the Old World ca. 1980, has spread widely and, by 2000, evolved a cline of increasing total wing length with latitude in North America phenotypically similar to that in its native range. The ichneumonid Bathyplectes curculionis, introduced to the western United States for biological control of introduced alfalfa weevils (Hypera spp.), evolved in less than 10 years to become less susceptible to the encapsulation reaction of its host.
A phenomenon widely reported among introduced vertebrates and plants, particularly in North America and Eurasia, is hybridization with native species, sometimes to the point of a sort of genetic extinction of the latter. Although hybridization is known to have played an important role in insect evolution, hybridization between native and recently introduced insects is rarely if ever reported. This fact may reflect biological differences or simply less genetic study of insects. There are instances of introduced populations hybridizing with one another, most notably the Italian and African strains of the honeybee. The red imported fire ant hybridizes extensively with the previously introduced black imported fire ant (Solenopsis richteri) in Tennessee.
ERADICATION AND MANAGEMENT
Eradication is one possible response to an introduced species, particularly if it has not dispersed widely. Many introduced insects have been eradicated, some from substantial areas. Perhaps most impressive is the chemical eradication of the African malaria mosquito Anopheles gambiae from 31,000km2 of northeastern Brazil in 1939-1940. The medfly (Ceratitis capitata) was eradicated over 18 months from a 20-county region of Florida by a strict quarantine, destruction of produce and plants, trapping, and insecticide sprays. More recent attempts to eradicate the medfly in both Florida and California may not have been so successful. Although victory has been declared repeatedly, reappearances are frequent and may constitute either new invasions or simply recovery by uneradicated remnant populations.
The development of the sterile male technique in the United States against the New World screw-worm fly gave tremendous impetus to the eradication approach. Release of massive numbers of sterile males so reduced the probability of fruitful mating by females that this species disappeared from the island of Curagao totally in 1954-1955, and this method greatly aided eradication of this fly from the southeastern United States in 1958-1959. The melon fly (Bactrocera cucurbitae) was eliminated from Rota Island by this method. The male annihilation method, in which males are attracted and destroyed, has also succeeded in eradicating introduced fruitfly populations from islands, including the Oriental fruit fly (Dacus dorsalis) from Rota and Guam and the melon fly from Nauru. Male annihilation followed by release of sterile males eradicated the melon fly from the entire Ryukyu Archipelago. The white-spotted tussock moth (Orgyia thyellina) was eradicated from greater Auckland, New Zealand by pheromone lures plus spraying of Bacillus thuringiensis, while the Australian painted apple moth (Teia anartoides) was eradicated from the same area by use of virgin females in lures, insecticides, B. thuringiensis, and sterile males.
There have also been disastrous failures of expensive eradication campaigns, such as the $200 million attempt to eradicate the red imported fire ant chemically from the southeastern United States, which imposed greater mortality on native insects than on the invader. Quick detection, rapid response, sufficient resources to finish the project, and adequate regulatory power to enforce cooperation have proven to be important for successful eradication.
If eradication fails or is not attempted, chemical and biological control are the two methods most commonly attempted to manage introduced insects. There are successes and failures for both the methods. The nontarget and human health impacts of early-generation insecticides such as DDT are legendary. Although more recent chemicals minimize or eliminate this problem, chemical control frequently is problematic for two main, related reasons. First, insects evolve resistance to chemicals; second, expense can be far too great, especially when amounts used must increase because of resistance. For large natural areas, expense of continued chemical applications can be particularly prohibitive.
Biological control is attractive because the expense of development and deployment is lower, and because, although a host may evolve resistance, the biological control agent itself can evolve countermeas-ures (as witness Bathyplectes curculionis, discussed earlier). There are many striking successes, such as the rescue by the predatory South American coccinellid beetle Hyperaspis pantherina of St Helena’s endemic national tree species from extinction by the South American scale insect Orthezia insignis. However, the success rate of biological control is rather low. For instance, for parasitoids introduced for insect control, only 10% have been effective. Further, biocontrol agents can affect nontargets, as has the seven-spotted ladybeetle discussed earlier, and these nontarget impacts can be generated by established biocon-trol agents that are not even effective against their targets (about three times as many biocontrol parasitoids establish populations as actually control the target pest). The tachinid fly Compsilura concinnata has failed to control gypsy moths in North America, but it is believed to be responsible for the decline of several large native moths.
