Studies in Avian Biology No. 22:291-304, 2001.

NEWLY EMERGENT AND FUTURE THREATS OF ALIEN TO PACIFIC AND ECOSYSTEMS

LLOYD L. LOOPE, FRANCIS G. HOWARTH, FREDERICK KRAUS, AND THANE K. PRATT

Ahstracr. Although the devastatingeffects of established alien species to Pacific birds and ecosystems are generally well recognized by the avian conservation community, we raise the under appreciated issue of effects of incipient and future invasions. Although special attention to Pacific species “on the brink” is to a certain extent appropriate and necessary, a comparable focus on stopping new invasions appears desperately needed. All indications suggest that introductions will escalate with the trend toward ever increasing commerce and unrestricted trade unless stronger preventative measures are implemented very soon. The threat to Pacific island avifaunas from the brown tree snake (Bniga irregularis) is well-known, but as many as several hundred of the world’s snake species, some of which are repeatedly smuggled illegally as pets, might have similar impacts on native birds if trans- ported to Pacific islands. We touch upon a sampling of obviously severe potential future threats, with the hope of raising awareness and resolve to fix the current woefully inadequate system for prevention of and rapid response to new invasions. Key Words: biological invasions; invasion and biological diversity; invasive amphibians; invasive invertebrates; invasive plants; invasive reptiles; invasive vertebrates: newly emergent alien species; quarantine; snake invasions of islands.

The biotas of oceanic islands in general, and the Located near the middle of the Pacific Ocean, Hawaiian Islands and other Pacific islands in Hawai‘i is increasingly important as an inter- particular, are highly susceptible to damage national transportation hub. Honolulu Interna- caused by alien plants, , and microor- tional Airport is the seventeenth busiest airport ganisms transported by humans. The high sus- in the world in terms of total passenger traffic; ceptibility is related to the evolutionary history military air traffic is also substantial. The state of island organisms that generally evolved with is a social melting pot, with much movement of reduced exposure to certain physical (e.g., fire) cultural trappings such as ethnic fruits and veg- and biotic (e.g., ungulates, snakes, ants) forces etables as well as the ever increasing repertoire (Loope and Mueller-Dombois 1989). Although of the international horticulture and pet trades. habitat destruction by humans was a very im- Tourism is the primary industry, and visitors ar- portant factor in the decimation of Hawaiian rive from all over the world. Agriculture is also landbirds in the past, the greatest current threats an important industry, which routinely moves are from alien species. The most important living material into and out of Hawai‘i. All these threats include avian diseases transported by activities result in the frequent arrival of new mosquitoes; predation by rats, cats, dogs, and alien species (Holt 1996). Furthermore, the in- mongooses; competition for food and other re- creasing globalization of the world economy and sources by alien species, especially arthropods the increasing scope of free trade agreements and birds; and habitat degradation by feral un- promise to expedite the flow of species (Jenkins gulates, especially pigs, which also facilitate 1996). spread of alien plants (Cuddihy and Stone 1990, Our focus on this topic was heightened by re- Jacobi and Atkinson 1995). Alien species also cent experience on a technical panel convened prevent the recovery of native ecosystems after by the U.S. Federal Aviation Administration, the disturbances, thus seriously exacerbating the ef- Hawaii Department of Transportation, and the fects of habitat destruction. We concur that the U.S. Fish and Wildlife Service to examine likely current focus of conservation agencies on miti- impacts on endangered species of an expanded gating these threats, with special attention to airport at Kahului, Maui. We were asked to pre- “species on the brink,” is appropriate and nec- dict what new species might arrive, particularly essary. However, we aim in this paper to call on new direct flights from Asia and eastern Unit- attention to the intuitively obvious but seldom ed States, and how their arrival might challenge mentioned reality that although existing inva- the currently minimal quarantine system. sions pose formidable threats, the situation The analysis below makes no attempt to be promises to get much worse as additional inva- comprehensive, but only to highlight a range of sive species are introduced and established. We taxonomic groups and pathways posing obvious ask the conservation community and public threats, with emphasis on potential vertebrate in- agencies to recognize and address the problem vaders, especially snakes. Our aim is to present of continued alien species introduction. a range of examples in sufficient detail to illus-

291 292 STUDIES IN AVTAN BIOLOGY NO. 22 trate the scope of the problem and to highlight half the native lizard fauna, and had left the the urgency of finding solutions. Although most three surviving forest bird species and remaining examples are for Hawaii,‘ since that is the island fruit bat highly endangered (Savidge 1987a; group with which we are most familiar, conclu- Wiles 1987a, 1987b; Rodda and Fritts 1992, sions largely apply to other Pacific islands. The Rodda et al. 1998). Huge reductions have also analysis includes suspected modes of entry, po- been observed in the populations of introduced tential threats, and examples of high-risk species birds, mammals, and lizards on Guam (Savidge not yet known to be established. 1987a, Rodda et al. 1998). The loss of the avi- fauna has had unknown affects on the native for- VERTEBRATES est ecosystem, but loss of pollinators and fruit Nearly all alien vertebrate species in Hawaii‘ dispersers are likely to have important repercus- are pests in some situations, although some also sions over several decades (Savidge 1987b). For have economic benefits (Stone 1985). The mam- example, there has been a dramatic bloom in mals are the best known and provide the best spider populations coincident with the loss of examples of the dilemma created by alien spe- the insectivorous avifauna (Rodda et al. 1998). cies introductions; this is especially clear with It is sometimes claimed or implied that the the ungulates. Ungulates have been the most de- brown tree snake is somehow unique in its abil- structive group for native ecosystems but are ity to wreak ecological devastation on island among the most important groups economically communities and that other snake species would (Cuddihy and Stone 1990). Other herbivorous not present similar problems (e.g., McKeown and frugivorous vertebrates are often important 1996: 144- 145). But this argument derives from threats to native species and are pests of agri- ignorance of snake ecology and the fact that the culture as well. The insectivorous and carnivo- brown tree snake invasion of Guam is the only rous species are potentially extremely detrimen- snake invasion to be well studied to date. In fact, tal to native birds and other animals. several snake species have invaded other islands Vertebrates as a group are particularly de- (or, in the case of peninsular Florida, areas eco- structive because of their relatively large size logically similar to islands), and damage to na- (which gives them more food and water reserves tive biotas has been documented or inferred in and consequently wider environmental toler- some instances. Additional snake invasions in- ances than smaller animals). They are also often clude the wolf snake Lycodon aulicus on Re- generalist feeders, more mobile, and thus more union and Mauritius in the 1800s (Cheke 1987), effective competitors than most invertebrates. on Christmas Island in the 1980s (Fritts 1993), On the other hand, an observational bias makes and perhaps throughout the Philippines and it easier for humans to recognize impacts from western Indonesia in the past few centuries vertebrates, especially for the larger species. A (Leviton 1965); Elaphe guttata (corn snake) on few terrestrial vertebrates (especially the smaller Grand Cayman Island (Schwartz and Henderson ground and den-inhabiting species) can disperse 1991); E. taeniura (striped racer), Protobothrops as stowaways in cargo and aircraft, and aquatic elegans, and cobras (of an unspecified species) species may arrive in ballast. However, by far on Okinawa (Rodda et al. 1997, Ota 1998); Na- the most important avenue of dispersal of ver- trix maura (viperine watersnake) on Mallorca tebrates into Hawaii‘ has been purposeful intro- (Corbett 1989); Boa constrictor in Florida and ductions for economic, recreational, or cultural Cozumel (Dalrymple 1994, Butterfield et al. purposes, often by persons or groups unfamiliar 1997; T Fritts, pers. comm.); and possibly Ac- with the potential negative consequences of such rochordus in southern Florida (I? Moler, pers. introductions. comm.). Especially successful has been the spread of the parthenogenic (lacking a require- SNAKES ment for fertilization) blind snake Ramphotyph- Hawaii‘ and virtually all other oceanic Pacific lops braminus throughout the tropics over the islands lack native terrestrial snakes (Loveridge past century, primarily as a stowaway in potting 1945, Allison 1996). Consequently, the native soil associated with horticultural shipments. The birds lack adaptive behaviors to deal with these invasion of Lycodon on Reunion is thought to predators. The apparently inadvertent introduc- have resulted in the near extinction of a native tion after World War II of the brown tree snake lizard (Cheke 1987). An endemic frog, Alytes (Boiga irregularis) into Guam well illustrates muletensis, is endangered on Mallorca, appar- the effects alien snakes may have on native is- ently because of the introduction there of Natrix land ecosystems. Within 40 years of introduc- maura (Corbett 1989). The recent introduction tion, the brown tree snake had attained peak of B. constrictor to Cozumel is expected to pose densities of lOO/ha, had exterminated nine of a serious threat to the survival of nesting sea- Guams’ 12 native forest birds and approximately birds and several endemic birds and mammals EMERGENT AND FUTURE THREATS OF ALIEN SPECIES--Loope et al. 293

(T. Fritts, pet-s. comm.). But most snake inva- Lycodon and Ramphotyphlops have moved sions remain poorly studied, and their ecological around the Indo-Pacific region. The second is by consequences remain largely undocumented. deliberate introduction as pets followed by es- There is every reason, however, to be con- cape or intentional release. Most of the free- cerned with snake invasions more generally; the roaming snakes captured in Hawaii‘ each year brown tree snake may be only the vanguard of are clearly in the latter category (based on ex- a potentially great ecological problem. There is amination of Hawaii Department of Agriculture nothing especially remarkable about the ecology records), as are the foreign snakes established or of the brown tree snake. Its clutch size of 4-12 commonly seen in Florida (Dalrymple 1994; P. (mean = 8) eggs (Shine 1991, Rodda et al. Moler, pers. comm.). 1998) is unexceptional and lower than that of The number of snake species that would many snakes (Fitch 1985, Seigel and Ford 1987, prove adept at hitchhiking is unknown but prob- Shine and Seigel 1996). The brown tree snake ably fairly small. Secretive and nocturnal spe- apparently produces at most a single clutch per cies having high densities and with facultative year in its native range (Shine 1991) but may (Schuett et al. 1997) or obligate (McDowell produce two per year in Guam (Rodda et al. 1974, Nussbaum 1980) parthenogenesis are like- 1998); hence, its intrinsic rate of increase is ly to make the most successful hitchhikers. probably fairly low. It is not adapted to extremes While the group of snakes meeting these speci- of either temperature or humidity, judging from fications is relatively small, it has nevertheless its natural geographic, elevational, and ecologi- furnished the most accomplished invasive snake cal range (McCoy 1980, Cogger 1992, OShea’ agents of ecological destruction so far. 1996, Rodda et al. 1998). The most noteworthy In Hawaii,‘ the past three decades have seen features of the ecology of the species are its a dramatic increase in the rate of pet reptile in- catholic diet of vertebrates (Savidge 1988, troduction, release, and establishment. Given the Greene 1989, Campbell 1996); its arboreal pro- burgeoning number of species bred and avail- clivities, which allow it greater access to forest able within the mainland pet trade, Hawaii‘ and birds than most snakes would have; and its noc- other Pacific islands remain highly vulnerable to turnal habits. But these features are by no means further introductions. Many snake species intro- unique to brown tree snakes: many snakes are duced for the purpose of furnishing pets may genera1 vertebrate predators, many are arboreal, well prove just as great a threat to native avi- and many are nocturnal, especially in the tropics faunas as has the brown tree snake, judging from and subtropics. Many arboreal snakes specialize their ecological attributes. Among the common- on birds or feed on them opportunistically and ly kept species, boas, pythons, rat snakes (Ela- could be expected to devastate Pacific avifaunas phe), bullsnakes (Pituophis), and most pit vipers if they were to become established. Lastly, any (Crotalinae) specialize on endothermic prey, and snakes to become established on oceanic islands many of the rat snakes and pit vipers have an would be in environments largely free of pred- ontogenetic switch from ectothermic to endo- ators and disease organisms, as is the brown tree thermic prey. King snakes (Lumpropeltis) are snake in Guam (the widely touted, terrestrial and vertebrate generalists. Many boas, pythons, pit diurnal mongoose would have no affect on noc- vipers, and rat snakes are arboreal and feed pri- turnal or arboreal snakes, nor on pit vipers, marily, or to a large extent, on avian prey. All whose strike is faster than the mongoose). these taxa have clutch sizes of the same mag- Hence, introduced snakes on most oceanic is- nitude as brown tree snakes or, in the case of lands could be expected to lack significant pred- the commonly kept Boa, Eunectes (anaconda), ators or other sources of premature mortality. A and Python species, are much larger (30%>lOO; reasonable estimate is that several hundred of Fitch 1985, Stafford 1986, Seigel and Ford the worlds’ approximately 3,000 snake species 1987). Several of these species can potentially could prove damaging to island avifaunas pre- produce two or more clutches per year when viously unexposed to snakes, although the major food is freely available (Tryon and Murphy effects of many of these would be primarily on 1982, Tryon 1984), as it is in Hawaii,‘ where ground-dwelling birds. Several potentially in- the environment is artificially enriched with an vasive snake species are dangerously venomous abundance of alien rodents, lizards, and birds. and could be expected to have negative conse- Furthermore, some species are suspected to be quences for humans too. facultatively parthenogenic (Schuett et al. 1997), Snakes are likely to be introduced to islands an attribute whose significance for colonizing in two ways. The first is by hitchhiking in cargo oceanic islands should be obvious. Most of these or on vessels used for transportation. This is species are nocturnal. The only ecological pa- how the brown tree snake is thought to have rameter for which some of these common pet arrived on Guam (and other islands) and how species cannot match brown tree snakes is ele- 294 STUDIES IN AVIAN BIOLOGY NO. 22 vational range. In its native New Guinea, brown early 198Os, and is now found on most or all of tree snakes can live at elevations from sea level the main islands. In an effort to curb the spread to 1,400 m (OShea’ 1996). Most commonly kept of the species, Hawaii Department of Agricul- pythons and boas probably cannot live at such ture prohibited the keeping of Jacksons’ cha- high altitudes, although many Elaphe, Lampro- meleons in the state until 1994, when the regu- peltis, Pituophis, and pit vipers would have no lation was rescinded because of its ineffective- trouble doing so, judging from their native lati- ness. However, during this same time, sale and tudinal and elevational ranges. The significance export were allowed, providing an economic in- of these considerations for Hawaii,‘ and perhaps centive for people to move the lizards around other islands, is that most of the snakes captured surreptitiously to begin new populations that and identified in Hawaii‘ are in the genera Boa, could serve as a source of saleable animals. Python, Elaphe, and Pituophis. That these Consequently, the spread of the species to other snakes have not elicited the same level of con- islands and to new localities within islands was cern in Hawaii‘ that brown tree snakes have is rapid, despite its illegality. remarkable and probably attributable to the gen- Jacksons’ chameleon is native to cloud forest eral ignorance about snakes and their biology (l,SOO-2,400 m) in Kenya, where temperatures that prevails at any location in which they are range from 2.5” C during the day to 10” C at naturally absent. night (McKeown 1996). It forms dense popula- tions at lower elevations (400-l) 100 m) in Ha- OTHER REPTILES waii’ and can be expected to invade forested up- A host of other alien reptile species could also land habitats, perhaps as high as the upper tree be expected to have negative consequences for line. There was an unconfirmed sighting of an native Pacific avifaunas, though they perhaps individual at Hosmer Grove (elev. 1,830 m) of may not be as damaging as snakes. A handful Haleakala National Park, Maui, in June 1994. In of large aquatic turtles are noteworthy for their 1996, Haleakala Chief Ranger K. Ardoin found predation upon waterbird chicks (Ernst et al. one crossing the road at 1,800 m elevation in 1994). Included in this group are several soft- ranchland just below the park boundary. shelled turtles (Trionychidae), of which two spe- In addition to being a voracious and efficient cies have been introduced to Hawaii‘ (Mc- predator of arthropods, Jacksons’ chameleon at- Keown 1996), and other turtles such as the snap- tains sufficiently large size that there is concern ping turtle (Chelydra serpentina) that have been about its potential ability to take native forest established elsewhere (McCoid 1995). Several bird nestlings as prey items, although this con- monitor lizards (Varanus) grow to large sizes, cern has yet to be scientifically investigated. feed to some extent on birds, and are adept at Other chameleons attain a larger size, are known climbing trees (Daniel 1983, Green and King to eat nestling birds (Schmidt and Inger 1957; 1993). These could be expected to have negative C. Raxworthy, pers. comm.), and are available consequences for at least some native birds. Re- in the pet trade. ports of wild monitor lizards occur occasionally Another concern is that any introduced lizard in Hawaii‘ and frequently in Florida (Dalrymple species could serve as an additional food source 1994). for many species of introduced snakes, thereby Of potentially significant impact is the intro- serving to keep introduced snake populations at duction of arboreal insectivorous lizards because an artificially high level and thus maintaining a these species can often reach high population high predation pressure on native birds. This is densities and may seriously impact the food re- one means by which brown tree snakes have sources of native insectivorous birds. Especially maintained phenomenally high population den- problematic in this regard is Jacksons’ chame- sities on Guam, even after the extirpation of leon (Chamaeleo jacksoni), which provides an most native birds (Campbell 1996, Rodda et al. illustrative case history of illegal alien vertebrate 1997). The high densities of alien geckos and establishment in Hawaii’ and ineffectual man- skinks in Hawaii‘ suggest a similar scenario agement response to the threat. The Jacksons’ could obtain there should snakes become estab- chameleon became popular in the international lished. pet trade in the 1970s. Some reached Oahu‘ le- gally in 1972 under a pet store import permit. AMPHIBIANS These were illegally released in the importers’ Frogs represent another under appreciated po- Kaneohe‘ backyard (McKeown 1996), subse- tential threat to native Pacific avifaunas. Bull- quently became free-ranging, and served as the frogs (Rana catesbeiana), already introduced to source for a rapidly expanding distribution and Hawaii,‘ attain a large size and will consume trade. The species spread throughout Oahu‘ dur- anything they can cram into their mouths, in- ing the 1970s and 198Os, reached Maui by the cluding all classes of vertebrates (Bury and EMERGENT AND FUTURE THREATS OF ALIEN SPECIES--Loope et al. 295

Whelan 1984). In the western United States, comprehend fully the potential ecological im- where this species has also been introduced, it pacts of a species before it is introduced. How- has been observed to eat adult passerines, ever, an invasion is unlikely to benefit most na- snakes, frogs, fish, and bats, and is partly re- tive species because all species engage in a web sponsible for the endangered status of one snake of interactions with a large host of other species. and several frogs (Rosen and Schwalbe 1995). The nonlinear nature of many of these interac- It is reasonable to expect bullfrogs to exert some tions makes complete prediction of a species ’ predation pressure on waterbird chicks where effects inherently difficult, but breaking estab- they co-occur. lished webs or creating energy sinks (e.g., am- A more insidious threat may be posed by a phibians discussed above) will inevitably be det- variety of arboreal tropical frogs, loosely termed rimental to some native species. Consequently, “treefrogs” but representing a diverse array of blanket claims that particular aliens are taxonomically unrelated species. It is reported “kamaaina”‘ (native-born) species or “harm- that the Cuban hylid Osteopilus septentrionulis less” or “helpful” (e.g., throughout McKeown is established on Oahu‘ (McKeown 1996). In 1996) can clearly be seen to be unscientific 1997, two species of leptodactylids were report- statements deriving from a different agenda than ed for the first time in the Hawaiian Islands: the impartial description of reality. Dissemina- Eleutherodactylus coqui and E. planirostris tion of such complacent ignorance of and un- (Kraus et al. 1999). The first is arboreal and has concern for the native biota is perhaps the great- a loud, piercing call; the latter is terrestrial with est long-term threat to the Pacific avifauna. a quieter chirp. Both species originated in the Caribbean and are associated in Hawaii,‘ as else- BIRDS where (Conant and Collins 1991, Kaiser 1992, Alien birds threaten native birds directly Dalrymple 1994), with greenhouses and nurs- through competition and transmission of dis- eries. Both species are currently spreading from eases and parasites, and indirectly through aid- nurseries to surrounding areas and are also being ing habitat conversion and ecosystem alteration transported and established by landscaping of re- by dispersal of seeds of alien plants (Stone and sorts and residential areas with plants from in- Loope 1987). Over 150 species of alien birds fected nurseries. These species are easily spread have been introduced to the Hawaiian Islands in plants and associated soil because eggs are hidden in these areas and directly develop into (Hawaii Audubon Society 1989) but only 54 small froglets, bypassing a tadpole stage and, have successfully established breeding popula- tions (Pyle 1997). hence, any need for standing water. E. coqui oc- curs to elevations of 1,200 m in its native ranges In contrast to Florida (James 1997), opportu- (Schwartz and Henderson 1991) has already es- nities for alien bird stowaways and “natural” tablished at higher elevations in Hawaii,‘ and, colonizations of Pacific islands are highly lim- hence, has potential to invade upland rain forest ited. In Hawaii,‘ all successfully established in Hawai i.‘ alien bird species were deliberately brought to Both species of established Hawaiian Eleu- the islands for some purpose. Therefore, limiting therodactylus and many other species of hylid, introductions has mainly relied on regulating leptodactylid, and rhacophorid “treefrogs” form trade in live birds. If this trend continues, future high-standing biomass and can be expected to introductions of alien birds are likely to be de- exert a significant impact on native insect faunas rived from four main groups: waterfowl, galli- and, indirectly, on the insectivorous birds depen- forms (chickenlike birds), psittacids (parrotlike dent upon them (Kraus et al. 1999). Because birds), and passerines (perching birds), especial- they will have few or no predators in the Pacific, ly finches. Birds are also increasingly smuggled such species may serve as energy sinks, produc- as eggs, which means that almost any species ing large quantities of biomass that do not get might be introduced in the future. transferred to higher trophic levels and, conse- Escaped Mallards (Anus platyrhynchos) quently, may exert ecosystem-level changes as threaten the endangered Hawaiian Duck or Ko- well (Dalrymple 1994). loa (A. wyvilliana) ecologically and genetically Again it needs to be emphasized that unsup- through hybridization (USFWS 1985, Rhymer ported claims that only a single species of tree- this volume). Apart from Mallards and Cattle frog “has the capacity to do great harm to island Egrets (Buhulcus ibis), all populations of alien ecosystems” (McKeown 1996:20) and that other waterfowl have been ephemeral (Berger 1981). alien species would be beneficial if introduced New wild populations derived from collections to Hawaii‘ are simply statements of faith com- belonging to resorts or private individuals are bined with a studious disregard for general eco- likely, but their long-term establishment is less logical principles. It is usually impossible to likely, because Hawaii‘ has relatively few wet- 296 STUDIES IN AVIAN BIOLOGY NO. 22 lands and the largest of these are managed for birds, an Oahu‘ population of Red-crowned Am- native wildlife by government agencies. azons (Amazona viridigenalis) grew to more All wild populations of galliforms in Hawaii‘ than 30 individuals by the late 1980s (T. K. (12 species) have been authorized releases. Such Pratt, pers. obs.). While there are no recent es- releases still take place, but now for the purpose timates for this parrot, a flock of 40 was seen in of providing birds to shoot rather than to estab- 1998 (E. VanderWerf, pers. comm.), suggesting lish new populations. This unmonitored practice a slow rate of increase. Rapidly growing colo- appears to involve mainly the Ring-necked nies of Mitred Conure (Aratinga mitrata and, Pheasant (Phusiunus colchicus) and Wild Tur- possibly, related species) appeared on Oahu,‘ key (Meleagris gallopavo), but the practice has Maui, and Hawaii‘ in the 1990s (T K. Pratt and the potential for escalating. The small trade in L. L. Loope, pers. obs.). The Maui, and perhaps “ornamental” galliforms (tropical pheasants. the Hawaii,‘ population of conures stem from etc.) could be a latent source of new introduc- deliberate releases. The population in the Huelo tions. area of Maui is expanding, and now numbers at Over the history of releases in Hawaii,‘ there least 80 (E Duvall, pers. comm.). This species has occurred an important shift in the taxa re- may well become numerous, being adapted in leased. Besides establishing the Mallard and its original range to open and disturbed habitats. many galliforms, early acclimatization projects We expect escapes and releases of parrots to in- and escapes successfully introduced the Cattle crease as the number of parrots bred in captivity Egret, Chestnut-bellied (Pterocles exceeds the demand for these long-lived birds as exustus), Barn Owl (Tyto &a), Guam Swiftlet pets. (Aerodrumus bartschi), 13 species of insectivo- Parrot invasions would seem to pose a threat rous and frugivorous passerines, and 19 species of additional diseases to Pacific birds. Certainly, of finches (Pyle 1997). The egret, owl, and swif- agricultural and ecological problems caused by tlet were government releases for biological con- parrots are well-known. Viewed by farmers as trol. little more than winged rodents, parrots damage Since 1970, all new introductions have been seed and fruit crops. In Hawaii,‘ parrots have unauthorized and have included parrots and depredated crops of corn, mangos, and lichee; finches. This shift has resulted from a tightening permits have been issued for their control (T. K. of restrictions imposed on importers by the state Pratt, pers. obs.). A very serious concern is the of Hawaii‘ and by changes in federal laws re- potential ecological role of parrots as seed pred- garding importation and quarantine of birds into ators and herbivores of native trees. Sulfur-crest- the country, particularly by the Wild Bird Con- ed Cockatoos (Cacatua plerita) introduced to servation Act of 1992. Importers prefer seed-eat- Palau depredate native palms. Cockatoos “fed ing birds that can survive the long wait through heavily on the heart of two species of endemic quarantine rather than the more delicate insec- palms, and large stands of these trees have been tivores and frugivores with their difficult diets; destroyed” (Engbring 1992:32). The native Ha- however, other passerines continue to appear oc- waiian flora includes many trees and shrubs with casionally in Honolulu pet stores. large seeds potentially vulnerable to predation Hundreds of species of parrots and other psit- by parrots. The role of parrots in the spread of tacids are available through the pet trade, and alien plants is unclear. Research is needed to de- their availability is increasing as new popula- termine which seeds are digested and which are tions are established in captivity. Fortunately, viable after passage. Seeds that pass undigested members of the most potentially damaging will be transported long distances by parrots. group, the lories and lorikeets, are prohibited Nineteen species of finches (Fringillidae and from legal importation into Hawaii,‘ yet in the Passeridae) now swarm the gardens and grass- past small colonies have been illegally held by lands of Hawaii‘ (Pyle 1997). Yet only the private breeders. Many species of these aggres- House Finch (Carpodacus mexicanus), Northern sive nectar feeders would thrive in high-eleva- Cardinal (Cardinalis cardinalis), Yellow-faced tion rain forests and compete for food with en- Grassquit (Tiaris olivacea), Nutmeg Mannikin dangered Hawaiian honeycreepers. (Lonchuru punctulata), and Common Waxbill Unfortunately, there is little or no account- (Estrilda astrild) reside in edges or gaps in na- ability for releases of parrots. Three species have tive forests (Scott et al. 1986; T K. Pratt, pers. established breeding populations, and others are obs.). Apart from the House Finch, which serves trying. The worlds’ most successfully invasive as a reservoir of avian poxvirus (Warner 1968, parrot, the Rose-ring Parakeet (Psittacula kra- Docherty and Long 1986), introduced finches meri) now inhabits Kauai,‘ Oahu,‘ Maui, and play an undetermined role in transmission of Hawaii,‘ though no population estimates exist diseases. (Hawaii Audubon Society 1993). From three Another important component of alien bird

298 STUDIES IN AVIAN BIOLOGY NO. 22 the most destructive of alien mammals there, se- known from Hawaii‘ (references in Miller and verely damaging native trees and epiphytes and Eldredge 1996), but many additional harmful dispersing alien plants (Strahm 1996). We note species could be introduced with alien hosts with alarm that three species of monkeys have brought in through the pet trade (Nilsson 1981). become established in Florida (Layne 1997), in- The major groups of arthropods affecting na- cluding the squirrel monkey (Sairniri sciureus). tive birds are the parasitic and blood-feeding A fenced colony of squirrel monkeys maintained species (including several mites, fleas, and flies) by Panaewa‘ Zoo in Hilo could potentially be and the insectivorous species (especially wasps) freed by the next hurricane to strike the island that compete for avian food (Howarth 1985a; of Hawaii.‘ Ferrets and other small wild carni- G.J. Brenner, pers. comm.). The parasitic and vores, which are commonly kept as exotic pets blood-feeding species affect birds not only by (although illegally in Hawaii;‘ Tomich 1986), causing disease and worrying their hosts, but also represent serious threats to native birds if also by serving as vectors for avian diseases any become established. (van Riper 1991). Mosquitoes, especially Culex quinquefasciatus, are considered among the INVERTEBRATES most severe current threats to Hawaiian land- Nearly 4,000 species of alien invertebrates birds because they are the vector for malaria, have been recorded from Hawaii‘ (Miller and bird pox, and other diseases among wild bird Eldredge 1996), but many species remain undis- populations (van Riper and van Riper 1985, Jar- covered, especially the smaller, cryptic ones. vi et al. this volume, Shehata et al. this volume). More than three-quarters of the established alien Only five blood-sucking mosquitoes are estab- invertebrates in Hawaii‘ are arthropods, which lished in Hawaii,‘ but several hundred more are represented by over 2,500 insect species and could potentially invade if given the chance. over 500 other arthropods (Nishida 1994). Be- Many of these are associated with leaf axils and tween 15 and 20 alien species of arthropods are could be imported with bromeliads, which are added to the list each year (Beardsley 1979), and currently popular in horticulture. The Central one or more become pestiferous. Most alien in- American mosquito, Wyeomyia mitchellii, is be- vertebrates arrived inadvertently through com- lieved to have arrived in this way in the 1980s. merce, or associated with their purposely intro- Unlike many introduced vertebrates, most in- duced hosts. Over one-fifth of the insects and a vertebrates are narrowly specialized to exploit few other invertebrates were purposefully intro- particular environments; thus, to succeed, invad- duced for biological pest control. ing invertebrates generally must find a new en- For most of the recorded alien species of in- vironment that closely matches their require- vertebrates little is known of their biology and ments. Hawaii,‘ with its benign, perpetual even less of their impacts on native species. spring-like climate and great range of elevation, What is known indicates that some species can temperature, and moisture regimes, could host a affect native ecosystems in profound ways (Ho- large percentage of the worlds’ tropical, subtrop- warth 1985a, Howarth and Ramsay 1991). Alien ical, and warm temperate invertebrates since invertebrates have invaded virtually all ecosys- they could find a suitable environment if given tems from the seacoast to the summits of the the opportunity. With increasing travel and highest mountains, and probably few native spe- world commerce, the pool of potential invaders cies escape at least some feeding damage. Some is immense, and a thorough analysis of their po- change ecosystem processes; for example, earth- tential threats daunting. Thus, we describe ex- worms change nutrient cycling in soils thus fa- amples from just two arthropod groups to illus- voring invasion by alien species (Vitousek and trate the scope of the problem in hopes of divin- Walker 1989). Over two-thirds of the 750 native ing long-range solutions. land snails are extinct or endangered, and alien predators (particularly the purposefully intro- ANTS (HYMENOPTERA:FORMICIDAE) duced predatory snail Euglandina rosea) are be- Ants are notorious invaders and recognized as lieved to be the major culprits in their decline a cause of native species extinctions, both in Ha- (Cowie et al. 1995). Euglandina has not yet waii‘ and elsewhere (Cole et al. 1992, Gillespie reached its full potential range, and as it ex- and Reimer 1993, Hiilldobler and Wilson 1994, pands, it threatens additional populations. Reimer 1994, Wilson 1996). Ironically, many of Four phyla contain species that can potentially the same invasive ant species are also regarded invade and directly affect the survival of land- as beneficial for their role as biocontrol agents birds in Hawaii.‘ Three of these (Platyhelmin- (Way and Khoo 1992), and some species have thes, Acanthocephala, and Nematoda) include been purposely introduced to new areas for bio- parasitic worms capable of causing disease in control (Greenslade 1965, Way and Khoo 1992, birds. A few alien bird-infecting species are Zenner de Polania and Wilches 1992). For ex- EMERGENT AND FUTURE THREATS OF ALIEN SPECIES--Loope et ul. 299 ample, in the 1980s a group of businessmen in- near the bird colonies. Adult ants also feed on troduced the ant, Paratrechina ,fulva, into Co- sweets such as nectar and honeydew. Thus, they lombia, South America, in an effort to control could disrupt reproduction and survival of native snakes at lumber mills. Subsequently, Zenner de plants and favor invasions of certain alien plants Polania and Wilches (1992) reported that species and honeydew-producing insects. Many plant richness decreased over 90% in areas invaded and extinctions would be expected to oc- by the ant. Native ant species were especially cur in invaded habitats. affected, but other arthropods and some verte- A colony of S. invicta was intercepted in Ho- brates also declined or completely disappeared nolulu in a package from Texas in 1991 from invaded areas. (CGAPS 1996). As the species expands its range There are no native ants known in Hawaii.‘ in North America, it will have greater opportu- About 40 species of alien ants are established nity to be transported to Hawaii.‘ S. invictu re- (Nishida 1994), and of those the 16 species with produces in two ways: individual fertile queens large, aggressive colonies are the most trouble- establishing new colonies, and polygyne colo- some (Howarth 1985a, Reimer 1994). There are nies dividing and part of the colony walking to numerous other ant species that could invade a new nest site (Shoemaker and Ross 1996). Po- new habitats or attack different prey if they be- lygyne colonies pose a greater invasive threat came established in Hawaii.‘ Two examples are and are more likely to establish if transported, described: the fire ants, which are currently se- but they are also far less likely to disperse long rious invaders of southern North America, and distances, although they might be transported to the weaver ants, which are dominant forest can- Hawaii‘ in a containerized shipment or in soil opy predators in the Old World tropics and sub- on earth-moving or construction equipment. Fer- tropics. tile females, on the other hand, could become stowaways in planes, cargo, and containers. Fire ants Two species of fire ants were inadvertently Weaver ants introduced from South America into southeast- The Asian arboreal weaver ant (Oecophylla ern United States: Solenopsis richteri, which ar- smaragdina [Fab.]) is widely distributed from rived about 1918, and S. invicta, which arrived Asia to Australia, where it occupies a wide range about 1940, and both have become problem in- of forest habitats from savanna and monsoon vasive species (U.S. Congress 1993, Callcott dry forests to more mesic habitats and rain for- and Collins 1996). S. invictu, especially, has ests (Hblldobler and Wilson 1994). A closely re- been implicated in the extirpation of native spe- lated species lives in Africa. Weaver ants use cies in areas where it has invaded. Solenopsis their larvae as spindles to weave nests in the nests in the ground, usually in open habitats and canopy, and their ability to select an optimal en- open woodlands. If these warm temperate spe- vironment within the canopy for their nests cies established in Hawaii,‘ they probably would gives the group a wide tolerance for different invade at least low- and mid-elevation dry for- forest types. Given the Asian weaver ants’ ests and open country. Their upper elevation known distribution and preferred environments, limit is unknown, but their subterranean nests it would be able to invade all forested habitats are protected from most frosts. in Hawaii‘ except perhaps the wettest and Two species of fire ants already occur in Ha- coldest rain forests. waii‘ and are widespread on all the main islands. The ant is a voracious arboreal predator, The native North American fire ant, S. geminatu, which can exclude all sensitive animals from its prefers to nest in loose soil and sandy areas, and nest tree as well as closely neighboring trees. in Hawaii‘ it remains confined to sandy coastal Colonies can contain 500,000 or more workers habitats and in dry leeward areas up to 300 m and can control a territory of a dozen or more altitude, mostly in disturbed sites (Huddleston large trees (Hiilldobler and Wilson 1994). They and Fluker 1968, Reimer 1994). S. pupuana pre- control the entire tree surface from the ground fers wetter habitats and forests, nesting under up and kill virtually all animals found within rocks or wood on the ground in wet to mesic their territory (Holldobler and Wilson 1994). forests between 300 and 1,100 m. Its large po- Native forest birds would be naive to such a lygyne (multiple-queen) colonies may contain competitor and probably would be unable to nest over 1,000 workers (Reimer 1994). or forage near an active ant nest. Both native Fire ants are voracious predators of small an- invertebrates and several native forest bird spe- imals, feeding the protein to their larvae. Few cies, as well as the endangered tree-roosting na- native invertebrates would escape their depre- tive bat, would be severely affected, and the ex- dations. Naive ground-nesting birds would be tinction of many currently listed species as well especially vulnerable, if the ants can survive as many currently nonendangered species would 300 STUDIES IN AVIAN BIOLOGY NO. 22 be expected if this species established in Ha- and Davies 1971) and attracted to lights at night waii.‘ (Howarth 1985b); thus they could become stow- The Asian weaver ant is often considered ben- aways on aircraft departing infested areas at eficial by farmers, who have lionized the ants night. Leaf axil breeding species could be intro- and introduced them to their orchards for pest duced in bromeliads and other plant material. control for centuries (Way and Khoo 1992). The Culicoides are very small; most adults are less species has been introduced to south Pacific is- than 2 mm long. Unless the species bit humans lands for biocontrol of palm pests (Greenslade (which many do) or otherwise became conspic- 1965). However, its effects on either the intend- uous, their impact on endangered birds would ed target or potential nontargets have not been go unnoticed until too late. To illustrate the po- recorded. It could be introduced into Hawaii‘ il- tential impacts of these alien species in Hawaii,‘ legally by well-intentioned gardeners returning the potential threats posed by two species will from Asia. Less likely is the possibility that fer- be described: C. arakawae and C. obsoletus. tile queens could arrive as stowaways in aircraft or in shipments of cut flowers or other plant ma- Culicoides arakawae terial. C. arakawae is widespread in Asia from Ja- The weaver ants’ exceptionally complex be- pan south to the Indonesian islands and west to havior makes them popular research animals. India (Arnaud 1956, Wirth and Hubert 1989). The related African species is established in en- The species does well in both tropical and tem- tomological laboratories in the continental Unit- perate climates, but whether its range results ed States (Holldobler and Wilson 1994) and from different strains is unknown. Arnaud could be moved to Hawaii.‘ Hiilldobler and Wil- (1956) reported it to be the most abundant and son (1994) describe a method to transport small widely distributed Culicoides in Japan. It breeds colonies within hand luggage on aircraft. in mud and soil at water margins, especially where polluted, such as animal wallows, ditches, BITING MIDGES (DIPTERA: CERATOPOGONIDAE: flumes, streams, and pools (Kitaoka and Morii CULICOIDES) 1963, Howarth 1985b). Near Tokyo (35”-36” N), Biting midges in the Culicoides are im- the species has two to three generations per year portant veterinary and public health pests in with a minimum life cycle of 30 days (Kitaoka most areas of the world (Linley and Davies and Morii 1963). The species probably can 1971). There are over 1,000 valid species, and breed continuously in the tropics; adults were many more still to be discovered and described collected in most months of the year in Laos (Borkent and Wirth 1997). Over 175 species are (Howarth 1985b). known from Japan and Southeast Asia (Arnaud The adults readily attack birds and sometimes 1956, Wirth and Hubert 1989), and about 135 mammals (Arnaud 1956), and the species is con- from North America (Wirth 1965). The biology sidered to be the most important vector of the of most species remain unknown. The larvae are bird protozoan parasite Leucocytozoon caulleryi, scavengers or predators on tiny invertebrates in a serious disease of poultry in east Asia (Kitaoka semi-aquatic and aquatic habitats; larval sub- 1978), and fowl poxvirus (Fukuda et al. 1979). strates include damp rotting plant material, ani- Fowl pox is already recognized as a severe dis- mal dung, mud, and soil in tree holes, leaf axils ease among Hawaiian endangered birds (van compost heaps, rotting vegetation, margins of Riper and van Riper 1985); thus the establish- water bodies, and a variety of aquatic habitats ment of an efficient new vector would pose a (Jamnback 1965, Howarth 1985b). Each species significant new risk. Adult C. arakawae are prefers particular larval habitats, and in concert readily attracted to lights (Arnaud 1956) and are most potential larval substrates are exploited. easily transported on the wind; they are, there- Adult females of many species are specialized fore, potential stowaways on aircraft departing to suck vertebrate blood: some generalists, some from infested areas at night. attacking birds, others small or large mammals, Female C. arakawae disembarking in Hawaii‘ reptiles, amphibians, or even larger arthropods would find abundant ideal breeding habitats in (Jamnback 1965, Wirth and Hubert 1989). They the immediate area surrounding most island air- are important transmitters of diseases, including ports. For example at Kahului, Maui, Kanaha blood protozoans (especially the primitive bird Pond and the irrigation ditches and pools in and malarias), filarial worms, viruses, and other par- near neighboring cane fields would be ideal. asites among birds (Kettle 1965, Wirth and Hub- From these lowland habitats the species could ert 1989). In addition, they also would increase easily disperse on the wind to rain forest habitats the spread of mechanically transmitted diseases on both east and west Maui. The endangered of birds (e.g., avian pox). Adult females of most waterfowl at Kanaha Pond and other wetlands species are readily dispersed by wind (Linley could be severely impacted both from exsangui- EMERGENT AND FUTURE THREATS OF ALIEN SPECIES--Loope et al. 301 nation and from exposure to new diseases. PLANTS Breeding habitats may be more limited in the Invasions by alien plants can alter the popu- upland rain forests, except for pig wallows and lation dynamics and community structure of na- some natural pool margins; however, the species tive species and change the large-scale function- might adapt over time to breed in the constantly ing of native ecosystems (Vitousek 1992). The moist soil in the wet forests of Hawaii.‘ If it did prevention of recruitment of native plant species become abundant, it could cause the declines of by invasive alien plant species is often the mech- several native forest birds. anism of long-term conversion of ecosystem structure and function (Macdonald et al. 1989). Culicoides obsoletus Alien plant invasion in Hawaii‘ frequently alters C. obsoletus is one of the most widespread ecosystems, jeopardizing and eventually elimi- species of biting midges, occurring in North Af- nating habitat for most native birds (e.g., Scott rica, Eurasia, and North America (Jamnback et al. 1986, Cuddihy and Stone 1990, Stone et 1965). It is recorded from both South Korea and al. 1992). For example, invasion of the vine ba- Japan, where it is widespread on Honshu and nana poka (Passi$ora mollissima) reaches ele- Hokkaido (Arnaud 1956). In North America C. vations as high as 1,500 m and smothers koa obsoletus is found from southern Canada to and ohi‘ a‘ forest, killing mature trees and pre- North Carolina and Tennessee in the east and venting recruitment, and degrading habitat for from British Columbia and Alberta to northern native birds (Warshauer et al. 1983, Jacobi and California in the west (Jamnback 1965). It is a Scott 1985). Shrubs and trees such as clidemia serious pest of humans and animals on Hokkai- (Clidemia hirta), strawberry guava (Psidium do (Arnaud 1956) and in North America (Jamn- cattleianum), kahili ginger (Hedychium gardner- back 1965). There are two generations a year ianum), firetree (Myrica faya), Australian tree (Kitaoka and Morii 1963). fern (Cyathea cooperi), and miconia (Miconia The wide range of larval breeding habitats in- calvescens) can potentially reach similarly high elevations, alter ecosystems, and degrade bird dicates that the species could become invasive habitat. At Kanaha and Kealia ponds on Maui, in Hawaii.‘ Suitable breeding habitats include dense thickets of fleabane (Pluchea indica) con- stream and pond margins and irrigation ditches vert extensive areas of habitat for Hawaiian in the lowlands, as well as moist forest floor in Stilts (Aeo;‘ Himantopus mexicanus knudseni) rain forests. If overwintering larvae diapause, and Hawaiian Coots (Alae‘ keoke‘ o;‘ Fulica they would not be successful in lowland habi- alai) to nonhabitat. In Tahiti, 40-50 species of tats, except as continual re-invaders from upland the 107 plant species endemic to the island are sites, but this species would probably survive believed to be on the verge of extinction pri- very well in cool upland forests where the major marily because of invasion of miconia (Meyer populations of endangered forest birds survive. and Florence 1997). Effects on bird habitat in Emerging females do not require a blood meal Tahiti remain unanalyzed. to develop their first clutch of eggs, making es- Alien plant invasions of Hawaii’ and Pacific tablishment of colonizers more likely but per- islands already pose an acute problem in pres- haps decreasing their role in disease transmis- ervation of ecosystems and bird habitat. Much sion. In suitable habitats, they can become in- effort is expended in Hawaii‘ and elsewhere on credibly abundant, severely worrying their hosts. weed control. Managers of natural areas and Like C. arakawae, adult C. obsoletus are readily agencies are struggling to address immediate attracted to lights and are potential stowaways problems through manual, chemical, and biolog- on aircraft. Additionally, immatures of this and ical control of invasive alien plants. However, other problematic species could be inadvertently most weed control programs get underway only imported on sphagnum or other moist materials after an alien species is an obvious problem. used to pack shipments of living organisms and Managers and agencies normally have their re- cut flowers. sources directed at dealing with the major weed In summary, the prospect is grim for future problems that are already highly conspicuous. invertebrate introductions unless we can learn There is a concurrent urgent need for dealing how to prevent them and are given the political with incipient and future plant invasions which support (including adequate funding for quar- is only beginning to be addressed. Whereas ap- antine) to apply what we have learned. The ex- proximately 100 plant species are currently rec- amples above give only the merest glimpse of ognized as serious invaders of native ecosystems the thousands or tens of thousands of potentially in Hawaii‘ (Smith 1985, Stone et al. 1992), over damaging species with potential to reach Pacific 8,000 plant species had been introduced to Ha- islands. waii‘ by the late 1980s (Yee and Gagne 1992) 302 STUDIES IN AVIAN BIOLOGY NO. 22

Time to full range occupation (%)

FIGURE 1. Stvlized reuresentationof the sureadof an invasiveplant speciesover time (Hobbs and Humphries 1995). and at least 861 had been recognized as exhib- experiments in invasive potential becomes enor- iting reproduction in the wild (Wagner et al. mous. One proponent of enriching Hawaii‘ s’ flo- 1990a,b). An ongoing up-to-date analysis of ra with more introductions recently wrote (Be- plant introductions in Hawaii‘ places the number zona 1996), “After visiting Ecuador, I realize we at over 13,000 (G. Staples, Bishop Museum, have barely tapped the potential for new plant pers. comm.), or roughly 3-4% of the worlds’ materials, including bamboo in Hawaii.”‘ A re- known vascular plant species. A substantial cent effort in Hawaii‘ at developing defoliator- number of the worlds’ most invasive plant spe- resistant, nitrogen-fixing trees which can aggres- cies are already present in Hawaii‘ but not yet sively invade degraded lands of the tropics has widely perceived to exhibit alarming invasive- hybridized 22 species in the genus Leucaena on ness. Examples include Arundo donax (giant Oahu‘ (Brewbaker and Sorensson 1994), creat- reed), Cinchona pubescens (quinine), Cryptos- ing a source for a new wave of invasion in the tegia grand@ora (rubber vine), Hiptage bengh- Pacific by that genus. alensis, Ligustrum spp. (privet), Lonicera japon- A large amount of literature on alien plant bi- ica (Japanese honeysuckle), Pittosporum undu- ology, impacts, and management exists in Ha- latum, and Thunbergia grandiflora. Some of waii‘ (e.g., Smith 1985, Stone et al. 1992) and these were not included in the 861 species re- worldwide (e.g., Cronk and Fuller 1995, Hobbs garded as naturalized by Wagner et al. and Humphries 1995). Experience in Hawaii‘ (1990a,b), and others were included as “spar- and elsewhere suggests that plant species which ingly naturalized.” Most of them are probably have proved invasive when introduced to one in a so-called “lag phase” (see below). part of the world are highly likely to be invasive Furthermore, whereas there is currently gov- when introduced to similar habitats elsewhere ernment scrutiny of proposed legal introductions (Cronk and Fuller 1995, Loope and Stone 1996, of animal species in Hawaii‘ and many other Reichard and Hamilton 1997). However, there is Pacific islands, there is still almost no govern- often a “lag phase,” in which a newly intro- ment-sponsored effort to prevent the potentially duced potentially invasive species is slow in invasive plant species which have not yet spreading and therefore easily controllable (Figs. reached the shores from being introduced. The 1, 2). phasing out of sugar cane and pineapple in Ha- Recognizing (1) the desirability of early de- waii‘ is contributing to a quest for agricultural tection and local eradication of such species (as diversification and experimentation. And with advocated by Hobbs and Humphries 1995, West- increasing travel combined with botanical curi- brooks and Eplee 1996, Loope and Stone 1996), osity and industry, the number of possible future and (2) the increasing danger of arrival of ad- EMERGENT AND FUTURE THREATS OF ALIEN SPECIES--Loope et al. 303

Calry ringcapacity

Quarantine Control Effective control Invader priority priority abundance unlikely without stage stage massive resource inputs I

L Time

FIGURE 2. Phases of weed invasion and priorities for action at each phase. Ease of treatment of an invasion problem declines from left to right (Hobbs and Humphries 1995). ditional potentially invasive species because of thickets of miconia had by the 1980s replaced accelerating international trade (Jenkins 1996) the native forest over most of the island, with prompt action to deal with newly emergent and dramatic reduction of biological diversity. After future plant threats is obviously urgently needed. the late E R. Fosberg saw this species in Tahiti Two examples of the consequences of being in 1971, he reported that “it is the one plant that “slow on the draw” follow. could really destroy the native Hawaiian forest.” Firetree, a small tree from the Azores, Ma- Yet because of its attractive purple and green deira, and the Canary Islands, provides a rep- foliage, it had already been brought to Hawaii‘ resentative example of the potential for rapid al- as an ornamental in the 1960s and nobody did teration of natural areas. One of the worst in- anything about it until it got well established. vaders in Hawaii‘ Volcanoes National Park, fi- After its detection on Maui by conservation retree often forms dense stands that shade out agencies in 1990, an alarm was raised; miconia native competitors. It fixes nitrogen in root nod- seemed to be an especially severe threat to the ules and alters early successional ecosystems high-elevation rain forest habitat of many forest through nutrient enrichment (Vitousek and birds. Now miconia has become something of a Walker 1989). Brought to Hawaii‘ in the 1920s household word in Hawaii‘ and an aggressive for reforestation, firetree was an incipient invad- campaign against it is being conducted, especial- er in Hawaii‘ Volcanoes National Park in the ly on the islands of Maui and Hawaii‘ (Conant 196Os, at which time an intense debate arose et al. 1997, Medeiros et al. 1997), at costs that over whether aggressive control or allowing nat- will soon approach $1 ,OOO,OOO.The government ural succession to take its course was the proper of French Polynesia is also aggressively in- response (D. Reeser, National Park Service, volved in preventing miconia from taking over pers. comm.). In an eight-year period between forests in islands neighboring Tahiti (e.g., Meyer 1978 and 1986, firetree expanded its range twen- and Malet 1997). Miconia needs to be stopped, tyfold within the park (Whiteaker and Gardner and we need to watch out for future miconias. 1992). It currently occupies 14,800 ha within Unless a proactive approach is taken by gov- Hawaii‘ Volcanoes National Park in spite of ernments to prevent continued or even acceler- concerted control efforts (Satchel1 1997). ated introduction of invasive weeds, we are very The most dramatic current example of an in- likely to have many more examples like firetree cipient invasive plant threat in the Pacific in- and miconia in the future. volves the invasive tree Miconia calvescens (Melastomataceae), native to neotropical forests PROSPECTS FOR IMPROVEMENT IN at 300-1,800 m elevation, and now known to be SLOWING INVASIONS an unusually aggressive invader of moist island We have attempted to describe the nature of habitats. Introduced to Tahiti in 1937, dense the threats that we believe loom ominously be- 304 STUDIES IN AVIAN BIOLOGY NO. 22 low the tip of the iceberg represented by the cur- can potentially prevent many future problems. rently recognized threats to avian “species on As of late 1997, agencies and individuals on the the brink” and Pacific ecosystems in general. island of Maui, which have been working to- Hawaii‘ is the biological invasions capital of the gether at a grassroots level for six years to deal United States (and consequently the endangered with the weed tree miconia invasion, envision species capital of the United States) and in many evolution toward an interagency working group ways is at the forefront in confronting the prob- with subcommittees dealing with major catego- lem, if not yet in effectively dealing with it. Yet ries of invaders. The group sees itself as a grass- biological invasions constitute a national and roots component of CGAPS. An island-wide global problem (Vitousek et al. 1997), and Ha- plan would establish categories (exclusion, erad- waii‘ could well be regarded as a laboratory for ication, containment, large-scale management), addressing alien species issues. and set priorities and responsibilities for pest Hawaii‘ is a microcosm-a small world in it- management. The greatest challenge appears to self where boundaries are clear, allowing oppor- involve obtaining funding and personnel to do tunity as well as challenge in dealing with alien the control work in an era of shrinking govern- species problems. Lessons learned in Hawaii‘ ment. Is success possible? All agree that public are highly relevant to other Pacific islands and education is a crucial ingredient of the anti-alien to continental situations. The state of Hawaii,‘ species strategy, to gain broad political support. dominated by urban politics (with 75% of the Direct public involvement in selected eradica- states’ population on Oahu,‘ which has < 10% of tion efforts is an important tool. Achieving and the states’ land area), is clearly overwhelmed publicizing success stories is an effective strat- with the problem. Much more attention to the egy. Given much more resources than are cur- problem from the federal government as well as rently on the political horizon, Mauis’ successes from the state government is clearly warranted and failures could guide efforts statewide. and desperately needed. Concurrent research is needed to (1) examine The Honolulu-based interagency Coordinat- and explain the lag phase phenomenon for both ing Group on Alien Pest Species (CGAPS) is an plants and animals; (2) detect and predict what alliance of biodiversity, agriculture, health, and specific incipient invader populations need at- business interests that has been working since tention statewide; (3) determine the specific 1995 to seriously address the alien pest crises in pathways by which these recent invaders arrived Hawaii‘ (Holt 1996). A major public relations and are being spread in the state; (4) develop campaign was launched in late 1996 to increase techniques for eradicating various groups of in- public awareness of alien species problems vaders once detected; and (5) develop the bio- (CGAPS 1996). The intentions of CGAPS are logical basis for needed legal tools to ameliorate extremely good, but their effectiveness remains current problems and prevent future problems. to be demonstrated, largely because of an inad- ACKNOWLEDGMENTS equate political (and thus bureaucratic) response to the challenge. LLL thanks D. Reeser for inspiration, practical A better-funded, better-staffed, better-equipped, progress, and insight for how to realistically confront alien species issues in Hawai‘i. FGH thanks N. L. Ev- and better-legislated quarantine system for Ha- enhuis, S. E. Miller, and R. H. Cowie of Bishop Mu- waii‘ is desperately needed (yet the Hawaii‘ leg- seum for reviewing earlier drafts of the invertebrate islature and agencies involved are not pushing section and for locating critical references. FK thanks for it). Additionally, early detection and treat- E. Campbell, R. Crombie, G. Rodda, I? Rosen, and A. ment of invaders before explosive spread occurs Wolf for assistance in locating relevant references.