EXOTIC PLANT INVASIONS IN TROPICAL FORESTS: PATTERNS … › psw › publications › denslow › psw... · Chapter 24 EXOTIC PLANT INVASIONS IN TROPICAL FORESTS: PATTERNS AND HYPOTHESES

Chapter 24

EXOTIC PLANT INVASIONS INTROPICAL FORESTS: PATTERNSAND HYPOTHESES

Julie S. Denslow and Saara J . DeWalt

OVERVIEW

In I lie tropics, exotic plants have been widely introduced for industrial timber. for land reclamation and lorage crops. andas orriamentals. In spite of the apparent opportunity for naturalization and spread. invasive exotic plants are scarce inmany continental tropical Forests. We examine several conditions under which exotic species do pose substantial threatsto tropical ecosystems or to their maniigcment. These include island ecosystems. open-canopied thrests,fragincnted ordisturbed ecosystems and forests managed for timber or crops. We explore four h ypotheses to account for the scarcity

of exotic species in manY tropical forests: (1) tropical forests are resistant to invasions b y exotic species because theyare rich in species and functional groups: (2) native rainborest species competitively exclude exotic species: (3) high

pest loads and high pest diversit y in the tropics deter establishment and spread of exotic species: and (4) low propaguleavailability contributes to the rarity of exotic species in many tropical forests. While current research suggests thathigh species diversity per se is not likely to be sin impediment to exotic species. Functional group diversity, highcompetitive exclusion rates, and high pest loads all may confer a certain biotic resistance to the establishment andspread of exotic species in tropical forests. Similarly, high junctional diversity and high productivity may increasethe resilience of tropical forests to the kinds of ecosystem changes etkcled by invasive species in other ecosystems.However, we are unable to fully evaluate these hypotheses and their interactions in the absence oh a better assessmentor the actual exposure of tropical forests to exotic propigules and resultsIts from seed addition expermnients to test therelative importance of biotic resistance and dispersal limitation in limiting the spread of exotic species into tropical

Forests.

INTRODUCTIONTropical forests face myriad threats from humanactivities. including land conversion and habitatfragnscntation. altered fire cycles, and defauna-lion (Sala et iii. 2000). With some exceptions.however, few continental tropical forests appearto he affected strongly by invasive exotic plants(e.g.. Rarnakrishnati 1991. Whitmore 1991,RejmItnek 1996. Pine 2002). Rejmsmnek (1996)found only 42 exotic plant species known to invadetropical rainforests: of those, about half wereknown to invade forests onl y on islands and eight

were reported only from treefall gaps. Similarly.exotic species constitute small percentages of' thefloras ol two tropical held stations. La Selva Bio-logical Station in Costa Rica (7.6%) and BurroColorado Island in Panailia (2 1i0, where exoticspecies are confined to pastures, clearings, or otherhighly disturbed sites (Foster and Hubbell 1990.Hammel 1991)). A global survey of threats tohiodiversity suggests that biotic exchange is sec-ondary to other factors such as land-use changefor tropical forests as it is for most forest ecosys-tems (Sala et id. 2000). The apparently low impactOf exotic species on tropical forest ecosystems

1

410Julie S. Denslow and Saara J DeWalt

could reflect biotic resistance (Mack 1996) toexotic invaders and/or historically low exposure topropagules from exotic species (Fine 2002). Bothbiotic resistance (in the form of impact from com-petitors, predators, and pathogens) and propaguleavailability (via reproductive output, vegetativespread, and dispersal) are important componentsof plant community composition (e.g., Turnbullet al. 2000) and have figured in rates of bioticchange throughout evolutionary time. Of inter-est here is their role in the spread and impactsof exotic invasive species in tropical forest ecosys-tems. In this chapter we examine patterns of exoticplant invasions in tropical and subtropical forestsand explore four hypotheses proposed to accountfor these patterns.

Our perception of the vulnerability of an ecosys-tem to invasive species has at least two com-ponents (D'Antonio and Dudley 1995): (1) theease with which exotic species are able to estab-lish and spread, and (2) the tendency for exoticspecies to alter ecosystem and community pro-cesses. Here we will use the term ecosystemresistance" to describe the degree to which com-petition, predation, and disease limit the abilityof exotic species to establish reproducing popula-tions. A community with low ecosystem resistancewill be more highly invasible than a communitywith high resistance. "Ecosystem resilience" willbe used to describe the tendency for ecosystemprocesses to remain unchanged following exoticinvasion. Thus ecosystem processes such as dis-turbance frequency or resource supply rates willremain relatively unchanged following establish-ment of an exotic species in a resilient community.Propagule pressure - a key component of theinvasion cycle (D'Antonio and Dudley 1995) -is a function of sizes of source populations,seed production, and propagule dispersal, all ofwhich reflect the ecology and introduction his-tory of the invader rather than attributes of theecosystem.

Figure 24.1 illustrates some of the processesthat affect the establishment and impacts of apotential exotic invader. The impact of an exoticplant species on an ecosystem will be a func-tion both of its abundance (population size anddensity) and of its capacity, relative to estab-lished species, to alter ecosystem structure and

processes. Propagule pressure, resource availabil-ity, and pressure from natural enemies all influ-ence the probability that an exotic species willestablish a reproducing population and the rateof growth of that population. Habitat fragmen-tation increases exposure of forests to propagulepressure from exotic species in nearby disturbed ormanaged ecosystems. Available resources, such aslight and space, also are increased by disturbanceand habitat fragmentation and decreased whenpre-empted by native species via competition.Rates of competitive exclusion are thought to behighest where primary productivity and growthrates are high (e.g., Rosenzweig and Abramsky1993). Similarly, high species and functionalgroup richness is thought to reduce resourcesavailable to newly establishing exotic species.The complexity of the invasion process andscarcity of appropriate data preclude evaluationof the relative importance of the many factorsaffecting the impact of exotic species on tropi-cal forests. Many of the processes illustrated inFigure 24.1 are interdependent and most stud-ies focus on situations in which invasive speciespresent substantial threats to the biotic integrityof ecosystems.

Moreover, there is a strong historic compo-nent to current distributions of invasive exoticplants. For example, Wu et al. (2004) suggestthat the low number of naturalized exotic plantspecies per log (area) in Taiwan versus Japanreflects Taiwan's shorter history of introductions.The apparent vulnerability of Hawai'i's foreststo invasive species reflects, in part, a historyof large-scale introductions. Between 1910 and1960. some 1026 taxa, all exotic except for78 native species, were out-planted into forestreserves statewide (Woodcock 2007). This enter-prise, carried out to restore Hawai'i's watersheds.also provided opportunity for the establishmentand spread of invasive species into native forests atan unprecedented scale, and certainly affects ourperceptions today of the vulnerability of Hawai'i'sforests to exotic species.

Our objective here is to consider the circum-stances under which invasive exotic plants havehad strong ecological impacts on tropical ecosys-tems and to use these examples to provide insightinto the attributes of some tropical rainforests that

Exotic Plant Invasions in Tropical Forests411

Speciesdiversity

Disturbance Functional

and habitat diversity

fragmentation

Compet P rod U ctivi

Resource availability

Exotic propagule supplyIHerbivores and ]

(,/1 tospathogens

Exotic populationgrowth

Relative percapita exotic

impact

Impact of an invasive species

Figure 24.1 Conceptual diagram of factors affecting the impact of exotic plant invasions on a tropical forestecosystem. Positive effects are shown with solid lines and negative effects with dashed lines. The diagram depicts avariety of interacting processes, some of which reflect attributes of the invaded community and contribute to itsresistance to the establishment of new species. Others are attributes of the invading species that affect its ability toreach suitable establishment sites and to alter local ecosystem processes. HI through H4 refer to hypothesesaddressing these processes discussed in the text.

might account for the scarcity of exotic species inthem. Our examples are drawn from the availableliterature, which necessarily addresses ecosystemsthat may be vulnerable to the establishment ofexotics by virtue of location (islands, peninsulas)or exposure to frequent or historic disturbances.Cited examples are listed in Table 24.1 for ease ofreference. We have found few examples of investi-gations in large tracts of intact continental forests,

so we will draw insights from examples on theirfringes. For the same reason, we have defined trop-ical forests broadly to include forests in the tropicsand subtropics under a wide range of climatic andedaphic conditions. We discuss several hypothesesthat might account for scarcity of exotic plants inmany tropical forest ecosystems, review the avail-able information, and offer suggestions for futureresearch.

U)

Q

a)

Q

0)

Cx;

a)U)a)

-ca-EI

0xi).0

Q)CC)Cl)

-0.90)CC)C0C:

a).?I(I)

S:30

•CC)C:xi)012cx)CC)0) S

.9CC)0C:CC)-J

a)Cx)I-

Ci)0

(a:30S

•CC)0CC)CC)(a

a)C000

CL

a)

U)a)?

20)

=0

(0.9U)13Ca)

Co.2.0 0

0

I

-D Ca)CCx) 0) R CCi) 0) -o0) 0)xi)xx)0) ©o c

Cti

0 00a)Q...- -

c'c'a)°a)Coxi co0 aa)-a)('JU) .04- 0

--c -z-ocCx)C (1) NU) U) U)(0

CC)CC) ) CC) a)a)0) -z-(0

ON c Ci) >S- - a) 0)! Ecxioc=Ca)a)Eci)Oco0

CC)Ci) EE Q _-CI) Ci)-Ci) C

-0) .2 0) 0) 0- co.0 co CC) .-D a)o-= EU) (00 0I(I) 0 LU

C?a)a)0 a)-D.2_Cl)U)U)U)ui.a)Q) C

, E E U) EE0) E a)a--C .9 .92 0> .0CU)CC) 0)(l)CCI)

0)0) -

U)C)Ea a)ci)Ci)Ca >'00Cl)a)0(l) U)888°0U)a)0oC U)LI)a)a)0a)a)-.2a).9-CU) -2-°a)tDa) D0.9-Da)-0-DCa) CL

a) = = 0) CC 13 C CL C -oC C.0 C D0) -C:=.9.9)0U)==a)

0C- coCx)a)U)C0)a)Ca)9ci)Ca)CaCaCC0)o

U)

U)

a)

CCC

0.U)o

C

U)U)

2>C

-

00

C)

= U)

a)U)- 0.U)clC)U)- 0.0 xLU

13Cx)C() a)_U) U)U)

a)a).9a)'-a)=_0 c2

.5 .5.9U)--OU) C a0 CL-

0 00a aa0 00

I

a)a)CC)co -0

LU

Ci).2 .c xi. C .9.0 a) - -D •9 -CO U)t U) 0< C0.90020 LE a) Cfl a) 0co U)

< - > Qc1 < -:<

.0 .0a) 2

.0(DCDI- 0)I-0)

.9.9C:CC)013

Q)a)a)O 0.CC)0Cl)(C)U)

c d

.0C -.9-a)a)CL

Z52 -o o>E Ea)EE

a)0U)0000aao a000g ao0

.9 0)UJCa CWa)0

a) >o,..2.Ct

U()/)9U)C0)913 13 1313 -,) 13 C CC()CU)CCCa0Ci)a)( -a)-a)m6 i50 aD 0.900 i) E.CC) 0

a)0=U)

a)a)2F- F- (I)

ca

S(a

U) 00.9

N

5.9=.0.-o0

o 5.0.0(5(3(.:

0)-Cl)

- a) >_.2 C

CC0 co

oE.00a)I

a)a)

CC)

xi)0a)=

.0:30.

:3ci)

.90a)0.9

(I)CO

(I)U)2

CD

.9a)0

E0ECUEE0C.)0

E0a)

00o00

E0010Q)ZiCZ0E0

1/)0

a)C.)a)(00-00Cc

I)),0

.0

0

.0a)a)U)0.0-0C.)U)

a)C.)0Ea)0-0C.)2

PIP-

Inw0)

.0

E(I)

0-In

- 0c

o-ooO)s a).-0

0a)0)cj0E(NC:Mo -.

o -. E(0 0 0) 0 0)I .-.z a) 0)(N

(C) F'-C) .0 .0 < . tao In In00Mooc:-

UUI fl

a) I> mC

2io

.9

2.9-2

0CL0 c6F= 70E )a) g )a)'-U)a)Q) co 0)10a)"-OI)a)

a)U) U)21/)2 U)U)°2 ca

'aU)t0a)10'9 0_

co I0 •0 -U)0D0O0000°a) o 0'Qo__-oa)0)0a)00_0a)=0a)= 00)a)0 O.0

0 ,.0t) a) C- a)C)0 0t) o C)a)00=0

o U) :3 Q= 0 (/) -oa)= U) 0:3 U) D:3O:3 0 O 0000

0)--- a) -

a) U)U) .9c:E

'5a)a) -Da)-a)a)E.9.2E .2

-0- -- -01/) 0U)0) - 1/)U) 1/)1)) 1/) (0 0cc'5 •5•5'55'5E-oEEEEEa-----.--_3--a::a)

cts00)00)0001/)01/) 000)00

52'5-2 'E • 5-5-2 5-2'5-.5--S0000 0 000 000°E 0-F--H HHHF-HH

0II)Uo- 0

CO(I

j5a)0(0Ca)0a) e-0:3Oq)='-00Ca)a).2a)'-.-o0-o0LL

(I)0•

2E2o-o0CMo0EE

--0 CO0 0

a 5-2 5-o00 0,:i-

-=a)0(0=

tCl)

-0-•(0a)Ca)

;-<(0

a)a)

.0 Q)a)

2(I)I->

Eoto-ca)a)

o oa)a)a)(3.5a)(0 E

U)C0a)0a)_5•0

a)C--0-

M a) 0 9_ 00) aj 0 U)DoQ

q)0a) 00=.0 a)a) I-C

0U) 0'a)

-oc a)- t7

E.9-sO0a) 0 C--a)U) -a)0a) -a)z9j2C.9 .9Eo2-U)

CO(/) .9Q2D= -Ca)U))O__=U)a)._O)0. Ca' U) .9 0 :3 0 .2U) .0 a)- 0 0

Cf)LL ocr<.9<

a)a).0 U)

a)a)a)(I)

.0>(I)I (I)ItCD

Ca0.a)

Ea)C.)

0

a)1/) IL

000C')II)Ua)0.co

E0

'3)07Ca)

'3)a)a)a)a)C0Ca)a),0)00I-)a)C1/i

a)CO

'-a)

ClO- C

.9a) a)

AM

414Julie S. Denslow and Saara J DeWalt -fl,I

INVASIBLE TROPICALECOSYSTEMS

While many tropical forests appear to be sub-stantially weed-free, invasjves can have strongimpacts on mainland forest ecosystems wherecanopy structure is naturally open, where rain-forests are fragmented or disturbed, or whereforests are exploited for crops or timber, andon island ecosystems, where both disturbed andintact forest ecosystems are vulnerable.

Islands

Tropical islands are often seen as invasive-specieshotspots because of both the abundance of exoticspecies and their impacts in those ecosystems(D'Antonio and Dudley 1995, Sax ci al. 2002,Denslow 2003, Wu ci al. 2004). The effectsof these species are not confined to highly dis-turbed areas. For example, the flora of Hawaiicontains similar numbers of native (989) andnaturalized exotic (1044) species (Wagner ci al.1999), among which are many that invadeand alter native forests (Smith 1985). Theseinclude subcanopy trees, such as Psidiiim cat-tielanuni Sabine (Myrtaceae), Morella Jinja (Ait.)Wilbur (Myricaceac), and Schinns terebin(hiJb/iusRaddi (Anacardiaceae); large herbs like Hedy-chinni çjardnericmum Ker Gawl (Zingiberaceae);shrubs such as Clidemia hirta (L) D. Don andMc!asioma candidum D. Don (Melastornataceae);and vines such as Passiflora tarminiana Cop-pens & Barney (Passifloraceac). Examples offorest invaders abound from other islands aswell. I'ittospornm undnhitum Vent. (Pit losporaceae)invades montane rainforests of Jamaica (Belling-ham et al. 21)05); Cinchona pnbesccns VahI (Rubi-aceae) is a major forest conservation concernin the Galapagos highlands (MacDonald et al.1988): Rubus aceifolins Poiret (Rosaceae) invadesthe forests of Christmas Island and Reunion(Baret et al. 2003): Cinnamo,,,i,,i zci/!anicumBlume (Lauraceae) invades forest in the Seychelles(Fleischmann 1997): and the neolropical under-story shrub Piper aduiwwn L. (Piperaceae) isspreading in lowland forest in Papua New Guinea(Rogers and 1-lartemink 2000). Sax ci al. (2002)

note that, on average, islands have about twice asmany exotic plant species as comparable mainlandhabitats.

However, not all tropical islands are character-ized by high densities of exotic species. In theirsummary of 20 island floras, Wit ci al. (2004)show that tropical islands do not have notablymore naturalized exotic species per unit log (area)than islands elsewhere nor do oceanic islands havea higher species density than continental islands(see also Rejrnánek 1996, Sax ci al. 2002). Thesepatterns suggest that factors other than isola-tion or latitude likely affect invasibility of islandecosystems.

Open-canopied forests

Invasive exotic species strongly affect some main-land tropical forests as well, especially thosewith naturally open canopies, even when rela-tively undisturbed. Melaleuca qninqnencrvia (Cay.S.T. Blake (Myrtaceae( invasion is altering thestructure of the Florida Everglades where itinvades scrub cypress habitats in the ecotonebetween upland pine forests and cypress swamps(Ewel 1986). Annona qiara L. (Annonaceae),a native of Florida and Central America, cre-ates dense thickets where it invades Queensland(Australia) Melaleuca swamp forests (Humphriesci al. 1991). lalcataria inolnccwia (Miq. ) Barnebyand J.W. Crimes (Leguminosae), a large nitrogen-fixing tree, invades undisturbed but open-canopied Met rosideros forests on recent lava flowsin Hawaii (Hughes and Denslow 2005). Theseexamples also emphasize that high-stress habitats,such as frequently flooded or shallow soils, arealso vulnerable to invasions if exposed to exoticspecies with appropriate physiological tolerances.

Disturbed forests

Chronic disturbances open forest canopies andprovide opportunities for the spread of aggres-sive exotics. Such disturbances long have beenrecognized to predispose plant communities toexotic species establishment, in part because ofthe increase in resource availability they cause(Rcjmhnek 1989, Kit avama and Mueller-Dombois


1995. Mueller-Dombois 1995. Davis ci al. 2000.Mack i't al. 2002). On the continental island ofSingapore, numbers of exotic species are positivelycorrelated with canopy openness, but intact rain-forest appeared to be resistant (Teo et al. 2003).In Florida (Horvitz et aL 1995) and Jamaica(Bellingham et al. 2005). canopy opening blow-ing hurricanes facilitated the growth of exoticspecies already present in the seed and seedlingpooi. The forests of the Western Ghats in India sup-port large populations of native ungulates )Bagchici al. 2004) and indigenous human populationspracticing swidden agriculture.'['he understoricsOF these forests are dominated by dense standsof Lantana camaro I,. Verbenaceae ) . Mikan inoncron( ha FIB. K. (Asteraceae). and C/i ro!no!oenoodorota (L.) R.M. King and H. Robinson (Aster-aceae), all of neotropical origins (Chandrashekaraand Ramakrish nan 1994, Chandrasckaran andSwamy 2000).

Browsing and rooting by exotic ungulates fre-quently is associated with invasions of exotic plantspecies. In Hawai'i. pigs contribute to tree andshrub death, churn the soil, and disperse seedsOf exotic species, facilitating the spread of Psithinn

cauleuinien into moist forests (Aplet et (!. 1991).Fensham et al. (1994) described high densitiesof Lantaiia camaro in dry rainforest in northernAustralia following impacts of pig digging andground lire. In Pasoh Forest Reserve in PeninsularMalaysia, the native pig Sus scram, strongly modi-lies the forest understory (Ickes et al. 2001) whichmay facilitate the spread of the neotropical under-story shrub C'!idemia 10cm. Thus the disturbancesassociated with ungulate foraging are associatedwith the spread of exotic plant species both whereungulates are recent introductions and whereungulates are a historic component of the forestecosystem.

Fragmented forests

Where rainforests are highly fragmented. they aresubject to edge encroachment from grass fires,penetration of wind and light into the forest inte-rior, high rates of canopy daniage, and seed rainfrom adjacent clearings, all of' which facilitatethe establishment of disturbance-adapted species

)Laurance 199 7, DiStefaflo et al. 1998). Naturallyfragmented riparian forests in Seychelles areheavily invaded ( Fleisehniann 1997). In NorthQueensland. Australia. forest fragments aredegraded further by a suite of exotic vines whichsmother canopy trees and underslory (Humphrieset (l!. 1991). a common pattern iii fragmentedtropical forests (Laurance 1997). Native vinesand lianas have similar impacts on fragmentedforests in Brazil (Tabáncz ci al. 199 7). In contrast,extensive intact Queensland rainforest appearsresistant to invasions by exotic species. evenwhen disturbed by occasional severe windstorms(Humphries and Stanton 1992). One conse-quence of forest fragmentation has been thealteration of successional trajectories by exoticspecies and the establishment oil agri-cultural land of new forest types sometimes domi-nated initially by exotic species )Lugo and Helmer2004).

Managed ecosystems

When coupled with exotic seed sources, distur-bance and canopy opening due to logging andswidden agriculture also facilitate the spreadof invasive species. Cliromoku'na odorata invadestropical dry forest. in Thailand after extractionof Sliorea siwiiensis Miq. DipterocarpaCeae) forlimber (Chazoul 2004). In South and CentralAmerica, invasive African grasses, such asPanicoin i naxim ooi Jacq. ( Poaceac ) and ii rocli!oaniotica (Forssk.) T.Q. Nguyen (Poaceae), impedeforest regeneration following logging or swiddenagriculture and carr y lire into the edges of intactforest (Uhl and Kauffman 1990. l)'Antonio andVitousek 1992). The high productivity, aggres-sive spread, and nutritious loliage of these exoticgrasses have been important drivers of the conver-sion of Central and South American tropical forestto cattle pasture (I)arsons 1972).

The considerable cost of controlling weeds inmanaged ecosystems in the tropics also is indica-tive of their potential to affect the course ofsecondary succession. barest restoration projects,and long-term forest management. For exam-ple, competition from exotic grasses and relatedincreased lire frequencies are major impediments

416Julie S. Denslow and Saara J. DeWalt

to the restoration of tropical dry and mesic for-est ecosystems (l)'Antonio and Vitousek 1 992, Cabin ci al. 2000). Along with nutrient depletion,weed encroachment is a principal cause of fieldabandonment in swidden agriculture (Nye andGreenland 1960). Early fallow vegetation in trop-ical rainforest environments often is dominatedby native pioneer species, but Ramakrishnan andVitousek (1989) note that reducing the timebetween clearings in northeast India increasedthe dominance of invasive exotics such as Lantanaand Chromolaeiio and other pantropical weeds. Intropical tree plantations, competition from bothexotic weeds and native pioneer species is animportant determinant of the success or failureof tree establishment (Wadsworth 1997). Tropi-cal forest ecosystems may he especially vulnerableto the spread of exotic plants from swidden orlogging operations. These activities not only pro-vide disturbances that facilitate the establishmentof exotic species, but are also the vehicle for thefl() reinent of novel species and varieties into landsadjacent to forest reserves (Denslow 2002).

This brief review suggests several generaliza-tions: (] ) exotic species are not perceived asa threat to most continental tropical forests;2) nonetheless, invasive alien species do affect

some tropical forest communities severel notablythose on islands, those with an open canopy struc-ture, and those frequently disturbed or highlyfragmented; and (1) invasive alien species presentsubstantial problems in managed ecosystems intropical environments where they alter succes-sional trajectories, impede restoration, and maybecome propagule sources driving invasion intointact ecosystems.

IMPACTS OF EXOTIC PLANTS ONTROPICAL FORESTS

These examples stand in strong contrast to intactclose-canopied hrests where exotic plants are rare.even in treef'all gaps. We discuss several hypothe-ses to account for the apparently high resistance of'intact continental tropical forests to the establish-ment of invasive exotic plants. These hypothesesaddress different processes in exotic invasions asindicated on Figure 24.1.

Hypothesis 1. Tropical forests areresistant to invasions by exotic speciesbecause they are rich in species andfunctional groups

The idea that species-rich communities are lessinvasible than species-poor communities datesfrom the writings of Elton (1958), who suggestedthat more resources were likely to he pre-emptedand more niches filled in species-rich than inspecies-poor communities. It has been offered asone of the central organizing tenets of invasionecology (see reviews by Levine and D'Antonio1999. Mack ii al. 2002) and is an often-usedexample of the effects of diversity on ecosystemprocesses (Hooper et al. 2005). The relationshipbetween native and exotic species diversity isnegative when plot sizes are small (e.g.. Fridleyet al. 2004) and experimental manipulations ofcommunity structure have shown that species-rich communities resist establishment of newspecies more effectively than do less rich corn-niunities (Levine and D'Antonio 1999, Levine2000. Tilman et al. 2001, Kennedy et al. 2002).In contrast, studies of grasslands (Stohlgreri ci al.1999). riparian ecosystems (Levine 2000), islands(Lonsdale 1999, Sax ci al. 20)12), and conti-nental ecosystems (Stark et al. 2006) show thatat regional scales both native and exotic speciesrichness are similarly correlated wilh environ-mental gradients - that is, native and exoticspecies richness are positively correlated and bothincrease along gradients of increasing resourcesupply. In the absence of direct evidence, how-ever, these patterns are not sufficient to documentcompetitive exclusion or resistance of diversecommunities to exotic invaders. Two recent stud-ies have shown that these patterns of negativeand positive correlations do not differ from thatpredicted by a neutral model of no species inter-actions and that the relationship between exoticand native species richness depends on the areaand/or number of individuals sampled (Fridleyet al. 2004. Herben et al. 2004). At small plotsizes, native and exotic richness are negativelycorrelated because the number of individualsand species sampled is necessarily limited. Atlarge plot sizes, the number of' individuals andspecies sampled in a plot is more variable and

or


native and exotic species richness are positivelycorrelated.

Thus there appears to be little support forthe hypothesis that species richness makes corn-munities more resistant to the establishmentof exotic species. Diverse communities can heinvaded where propagule pressure is high (levineand l)'Antonio 1999). At Semliki Forest Reserve.I Iganda. high-diversity forests were no more resis-tant to exotic invasion than were the low-diversityplantations (Rejniânek 1996). The effectivenesswith which the native communit y pre-empts avail-able resources may be more important thandiversity per se in impeding the establishment ofexotic species (Davis ci a). 1998, 2000. Shea andChesson 2002. Denslow 200 i and see Hypothesis2). Although key functional groups are more likelyto be present in species-rich than in species-poorassemblages (Hooper ci a). 200), diverse ecosys-tems may be highly invasible when key functionalgroups are missing, as has been proposed fortropical islands (Kitavamna 1996. Kitayama andItow 1999, Lonsdale 1999, Fine 2002, Denslow2003).

I li p h diversity of functional groups may buflercontinental tropical forests against the kindsof ecosystem and structural alteration causedby exotic species iii other ecosystems. Wheninvasive exotic species represent novel func-tional groups they are likely to alter communitystructure, disturbance regimes, or soil processeswith ecosystem-wide consequences (Vitousek1986). Tropical forests may be resilient to suchecosystem-altering consequences of exotic inva-sions when exotics do not add new functionalgroups to the plant community or have highper capita impacts relative to native species.Ecosystem processes such as nutrient and mois-ture supply rates are not easily altered in rainforestenvironments where moist ore supply. nitrogenturnover rates, and net primary productivity arenaturally high (Vitousek and Sanford 1987). Theprincipal impact of exotic species on rainforestcommunities thus is likely to he through competi-tion for space and resources rather than throughalteration of ecosystem processes or disturbanceregimes. Exotic vines and lianas ma y he an excep-tion to these generalizations. While this is awell-represented functional group in mainland

tropical forests, a heavy infestation of banns -exotic or native - can kill or damage canopytrees, prevent sapling growth. and contribute tothe gradual degradation of rainforest structure(Humphries ci a). 1991. Tabánez ci a). 1997, Horvitz et al. 1998).

Hypothesis 2. Native rainforestspecies competitively exclude exoticspecies

The ability of the native plant community toexclude potential invaders also will be a function,in part, of growth and dispersal rates of nativespecies and of their ability to rapidly pre-emptresources. Two functional groups in particularmay play important roles in reducing seedlingestablishment, thus contributing to invasion resis-tance in tropical forests (e.g.. Rejniänck 1996,1999, Fine 2002): (1 (fast-growing pioneers ofseveral growth forms that effectivel y occupy spaceand pre-empt resources in high-light environ-ments, and (2) broad-leaved understory speciesthat are able to persist in low-light environments.

I ugh-light-demanding pioneers, including fast-growing trees. shrubs, large herbs, and lianas,are important components of forest regenera-tion processes because they quickly establish inlarge trecfall openings or other disturbed areas,such as landslips or riparian corridors. 'l'hesesites also provide establishment opportunities forexotic species in forest ecosystems (e.g.. Rejmiinek1996, Knapp and Canham 2000. Webb ci al.2000. McDowell and Turner 2002). In I lawai'iseveral exotic Eubus species compete with thenative Rnbmis lmowam'n.sis A. Gray for gap estab-lishment sites (Gerrish ci al. 1992) and, in theBudongo Forest, Uganda, spread of the exoticsCassia spectuhilis DC. (Fahaceae ) and Bronssonctiapajmjrifi'ra (I.) L'l-lér. ex Vent. (Moraceae) is facili-tated by gap openings (Shell ci al. 2000).

High rates of competitive exclusion in pro-ductive wet tropical forests have been sug-gested as a niechanismn for the hump-shapedpattern of diversity across productivity gradi-ents (e.g.. Rosenzweig and Abranisk y 1993) andhigh competitive exclusion rates likely reduceestablishment success of exotic species as well as


Ilahive Species. For cXaniiJIc. IICSV hir'chill ()f)cJi-iugs are colonized by a combination of large-caved herbs, vines, palms. and last-growing.light-demanding trees, which rapidly reduce lightlevels near the ground (Walker et al. 1996,Dcnslow eta!, 1998. Schnitzer et(i!. 2000). Generasuch as Ceeropia, 'I'rema, Balsa, and Maearanqaexhibit some of the highest growth rates observedamong woody plants (Wadsworth 1997). withlife-history characteristics similar to many inva-sive exotics. Where they are abundant, vinesand lianas suppress seedling establishment andsapling growth in gaps (Putz 1991. Schnitzerci al. 2000). Where such pioneer species arerare, forests may be particularly vulnerable tothe establishment of exotics. E-lorvitz etal. (1995)comment on the scarcity of pioneer species inthe hardwood hammock flora of south Floridaand speculate that exotic species (especially vines)iii these hammocks usurped that role follow-ing the passage of llurricane Andrew in 1992.Oil such as the Galapagos and Hawaii,species with pioneer growth characteristics aresparse. possibly contributing to the invasibilityof these ecosystems (Kitayama and [tow 1999,l)cnslow 2003). These species are able to converthigh resource availability into rapid growth andhigh rates of production. Where invasive speciesincrease productivity, such as through nitrogenaddition to nitrogen-poor soils, competitive exclu-sion rates are expected to increase as well. Forexample, where the nitrogen-fixing tree 1'aleatariaoio/nccana invades Hawaiian Metrosideros for-est on nitrogen-poor soils, productivity increasesbut the native Met rosideros declines (Hughes andl)cnslow 2005).

At the other end of the shade-tolerance spec-trum, understory palms, shrubs, ferns, andherbs produce dense shade at ground levels(Montgomery 2004). Where these species arecommon, seedling establishment is suppressedand seedling densities are low (Denslow ci al.1991, Farris-Lopez et al. 2004, Harms et(i!. 2004.Wang and Augspurger 2004). The inhibitoryeffect is strong on native species and likel y wouldaffect exotic species as well. The combined effectof fast-growing pioneers and shadc-tolerani herbsand shrubs is to reduce site occupancy by seedlingsand increase the importance of recruitmentlimitation in rainforest d ynani ics. While density

lily be more critical than the number of species,redundancy within hinctional groups is likely toincrease their distribution and impact.

Hypothesis 3 High pest loads and highpest diversity in the tropics deterestablishment and spread of exoticspecies

lii tropical forests, high diversit y and abundanceof natural enemies (herbivores and pathogens)occasionally may lead to high impacts of nativePests on exotic species (e.g., Nair 2001. Novotnyet a]. 2003). Certainly, rates of leaf dam-age by herbivores and pathogens tend to hehigh, although variable among species, in trop-ical ]crests (Coley and Aide 1991. Coley andBarone 1996), Thus, native generalist herbivores,pathogens. and viruses may provide a biotic bar-rier to invasion (Mack 1996, Parker et (i]. 2006)if they are able to exploit exotic plants. Highdiversities of both pests and host plants. whichreach their peak in the wet tropics for manytaxa, may increase the probability that an exoticplant is suppressed by native herbivores, as shownby Prieur-Richard et al. (2002) in a Mediter-ranean plant community. In addition, generalistPest species may play important regulatory roles inpopulation dynamics of tropical plants. For exam-ple, Augspurgcr (1984) describes the importanceof damping-off fungi as a source of seedling mor-talit y in the tropical forest understory. ()omycetes,a common group of clamping-off fungi, can per-sist in the soil in the absence of hosts andexhibit low host specificity (Augspurger 1984,Hood etal. 2004). Among insect herbivores, mostspecies are not monophagous but feed on multi-ple species within a genus or famil y (Coley andBarone 1996). Some common foliage-feeding her-bivores. such as leaf-cutter (Attinc) ants (Fowleret al. 1989, Farji-Brener 2001, Wirth et (i!. 2003)and orthoptera (Novotny ci al. 2004), have broaddiets. Leaf-cutier ants in particular are seriouspredators of a number of exotic crops, includ-ing citrus, eucalyptus, coffee, and cacao: Chei'ett(1989) suggests that the susceptibility of so manycrop species to this pest is clue in part to theirlack of defenses. A meta-anal ysis by Parker ci al.(2006) showed that exotic invaders often are


repelled by native herbivores. We suggest thatsuch generalist natural enemies may contributeto the apparent invasion resistance of tropicalforests.

This biotic barriers hypothesis runs counter toone of the principal explanations of invasiveness -the enemy release hypothesis (ERH).Thts hypothe-sis proposes that the accidental or even intentionalintroduction of plants awa y from their nativerange is accomplished most often without con-comitant introduction of the specialist herbivores,pathogens, and viruses that attack and limit theirpopulations in their native range. II' native speciesare limited by their own suite of natural enemieswhere exotics are introduced, then exotics mayproliferate because of their relatively lower pestloads (Elton 1958, Maron and Vila 2001, Keaneand Crawley 2002). Indeed, there is evidence that.where introduced, some exotic tropical specieshave lower pest loads in their invasive than intheir native range and lower pest loads than localnative species in their invasive range. For exam-ple. invasive species oil the main islandof the tropical Seychelles, suffered less leaf arealoss to herbivores than native woody species (Dietzetal. 2004). Native pioneer species were especiallysusceptible to herbivores (C. Kueffer pers.cornin.).In addition, a biogeographical comparison of theimpact of natural enemies on the neotropicalshrub Clideuna !iirta in its native and introducedranges found that plants were heavily attackedby insect herbivores and fungal pathogens in thenative range, particularly in forest undersiory. butthat they were relatively pest free in the introducedrange (DeWall et al. 2004). The consequences ofpest-load reduction to C. Idita appear to includenot only proliferation in the introduced range, butalso invasion of forest understor y, where it doesnot occur in its native range (DeWalt ci al. 2004).The relative success of plantations of exotic speciessuch as rubber (lievea brasilicnsis )Willd.) Much.-Arg. !Euphorbiaceiie!) and Eucalyptus (Myrtaceae)in the tropics is also attributable in part to theirescape from heavy pest loads, particularly fromspecialists in their native ranges (Rosenthal et (1!.1979. (1adgil and Bain 1999). Thus, some tropi-cal plant populations may he regulated by naturalenemies to the extent that release from theseenemies leads to their proliferation in introducedranges (DeWalt 21)05).

Does a biotic barrier in the form of high pestloads contribute to the relative dearth of invasiveexotic species in tropical forests? Does escape fromnatural predators give exotic species an advantagewhere they are introduced? Both of these h ypothe-ses are compelling and supported by examples.Further evaluation awaits more information onthe population-level effects of natural enemies onexotic species.

Hypothesis 4. Low propagule availabilitycontributes to the rarity of exoticspecies in many tropical forests

The distribution and abundance of man y forestplant species, in temperate as well as in tropicalforests, are limited by failure to recruit seedlings tosites otherwise suitable to their establishment andgrowth (Clark ci al. 1998, 1 999a, Turnbull ('i a).2000. Beckage and Clark 200). Svenning andWright 2005). Factors contributing to recruit-ment limitation include those affecting the sizeof the seed crop (fecundit y and the density anddistribution of adult trees), close and distant dis-persal (Clark ci of. 1999h), and post-dispersalFactors such as pests and pathogens which affectgermination and seedling establishment (Clarkci al. 1998, Nathan and Muller-Landau 2000.Turnbull et al. 2000. Zimmerman ci al. 2000).Dispersal and recruitment limitation increasinglyare seen as major determinants of the relativeabundances of species in forest ecos ystems (Clarket a). 1999h, Harms ci al. 20€ 0. Nathan andMuller-Landau 2000, Hubbell 2001. Denslowet al, 2006).

Propagule supply also plays a major role inexotic species invasions (Von I toile and Simberloff2005). For example. some of the strongest predic-tors of invasiveness are those that alicet propag-ule distribution and abundance, including theduration, frequency, and area at exotic speciesintroductions (Richardson 1999. Lockwood et al.2005). Forests most likely to be free of exoticspecies are those with low exposure to propag-ules from urban or agricultural areas (Aragon andMorales 2004. Sullivan ci al. 2005). Disturbedforests may be free of exotic species if propag-ule input is low. In Hawaii Volcanoes NationalPark. montane rainforests heavily damaged by pig


browsing are little affected by exotic species wherehuman traffic is low ('L Tuntson personal commu-nication). Forest fragmentation, road cons( ruc-tion, and other sources of human disturbance arelikely to expose adjacent forests to seed rain fromexotic species. Fine (2002) has suggested that thescarcity of exotic species in tropical forests mayreflect a more recent history of disturbance andfragmentation in tropical than in temperate for-est. Although forest clearing and fragmentationis a more recent phenomenon in tropical than intemperate forests, human populations have livedin and exploited tropical forests for centuries. Fur-ther, natural disturbance regimes in wet tropicalforests are high (Denslow 1987). It is unlikely thatdisturbance alone accounts for the distribution ofexotic species.

Low exotic species abundances may reflect ahistoric lack of seed availability from species withappropriate physiological characteristics, such asshade tolerance, for establishment in tropical rain-forest. Exotic plants, especially woody species, havebeen widely introduced in the tropics, often atgrand scales. Extensive forestry, agricultural, andland-reclamation projects juxtapose large popu-lations of exotic species with native forest. Manyof these planted exotic species have life-historyattributes similar to those of native pioneers andgap colonizers (Wadsworth 1997). For example,fast-growing species in the genera l'inns. Tectona.Eucahjptns, and Ginelioo commonly have beenplanted for industrial timber (Wadsworth 1997),while many species and genotypes of Leuraeno.Albizia. Acacia, and Calbandra have been intro-duced as utility species to rehabilitate degradedsoils and provide fodder and firewood (Hughesand Styles 1989, Hughes 1994). Tropical forestsmay not be immune to the spread of suchhigh-light-demanding exotic species, which mayappear in natural forest clearings where they canimpede regeneration of native species )Rejinánck1996).

A more likely source of shade-tolerant speciesis ornamental plants which have been widelyintroduced into tropical habitats and are thesource of many invasive species (Meyer andLavergne 2004, Wu ci (i!. 2004). Daehlei- (per-sonal communication) estimates that 39% of theapproximately 100 naturalized exotic species that

pose the greatest threat to Hawaii's native ecosys-tems were introduced as ornamentals. Note-worthy examples of shade-tolerant ornamentalsthat have become invasive in tropical forestsinclude Micon ia cal vesrm'us DC, (Melastonlataceue),a neotropica I tree invasive in native forests ofFrench Polynesia (Meyer 1996) and F-lawai 'i(Conant ci al. 1997); /lrdisia elliptica Thunh.(M yrsinaceae), which has become invasive inhammocks of south Florida )Horvitz ci al. 1998):and numerous vines and lianas with shade-tolerant juvenile stages. such as Merremia tnherosaL.) Rendle )Convolvulaceae) and Jasininion

dichotomnoi Vab I (Oleaceae) (I Iorvitz ci al. 1998),also invasive in south Florida.

In the absence of experimental additions ofseeds (Turnbull et al. 2000). it is dtl'licult to eval-uate the role of propagule availability in limitingexotic species in tropical forests. Tropical forests,like their temperate counterparts, are not likely tobe strongly affected by exotic plant invasions if theforests are buttered from exposure to propagulesources (e.g., Pyck etal. 2002).

CONCLUSIONS

The scarcity of exotic plants recorded from intactcontinental tropical forests suggests that tropi-cal forests may be resistant to invasions of exoticplants. I ligh species and functional group diver-sity, high competitive exclusion rates, and highpest loads all ma y confer a certain biotic resistanceto the establishment and spread of exotic species intropical forests. Similarly, high diversity and highproductivity may increase the resilience of tropicalforests to the kinds of ecosystem changes effectedby invasive species in other ecosystems. However.we are unable to evaluate these hypotheses fullybecause we cannot evaluate exposure levels oftropical forests to propagules of exotic species.'there are exceptions to the general pattern ofsparse exotic species in tropical forests. Preventionand control of invasive species o il in frag-mented or disturbed ecosystems, and in managedecosystems are all major conservation and man-agement concerns in tropical forest environments.Lessons from islands and exploited ecosystemssuggest that control of invasive exotic species

pExotic Plant Invasions in Tropical Forests421

Will assume increasing importance in the con-servation of forest preserves Lis habitats becomefragmented and exposed to increasing varieties otexotic species.

Our review highlights several areas in whichfurther research would be productive. For exam-ple, what is the role of natural enemies (herbiv-orous pests and pathogens) in regulating nativeand exotic plant populations? We know littleabout the extent to which top-down factors con-trol tropical plant populations, macli less abouttheir role in plant invasions. Basic research onthe role of pests and pathogens in regulatingspecies abundances may provide insight intothe mechanisms of invasion as well as eluci-dating factors structuring plant communities.These issues also are related directly to the devel-opment of sale and effective biological controlagents for wildiand weeds (l)enslow and Johnson201)6).

What are the roles of seemingly minor speciessuch as unclerstory shrubs and pioneer trees inregulating resource availability in tropical forestecosystems? Our review suggests that competitionfrom native pioneers may be critical in invasionresistance, yet these species are relatively sparse inintact rainforests. Recent research documents theimportance of undcrstory vegetation in reducinglight levels near the ground (Montgomery 2004).Thus shrubs, ferns, and understory palms couldplay important roles in limiting exotic speciesestablishment.

Experimental additions of seeds. propagu Ics, orseedlings in tropical forest environments wouldprovide important insights into processes lim-iting the establishment and spread of species(e.g.. Turnbull ci (!. 2000) antI how those pro-cesses vary within forests and across landscapes.regions, and species.

While farests in general, and tropical rainforestsin particular, often appear more resistant to theestablishment of exotic species than many otherecosystem types (Rejminek 1989. Fine 2002).global change is likely to increase their vul-nerability. Changing climate, altered disturbanceregimes. and increased forest fragmentation andexploitation (Sala et (i!. 2000) may open for-est canopies, and increased global movement ofspecies, biotypes. pests, and diseases will provide

exposure to new species capable of taking advan-tage of local environmental opportunities.

ACKNOWLEDGMENTS

We are pleased to acknowledge our debt to thefaculty and students of the Department of Botanyof the University of Hawai'i at Manoa duringi.S.O's tenure as CE Wilder Visiting Professor andto the Department of Ecology and EvolutionaryBiology at Rice University during S.J,l).'s appoint-ment as a Huxley Research Fellow. We are alsograteful for comments from C.C. Kueffer, two ad hocreviewers and from W. Carson which were helpfulin improving the manuscript.

REFERENCES

Aplet . C.H .. Anderson, SI . . and Stone. CE 1991 ) Asso-ciation between feral pig disto rba nec and the coin-pos ion of some alien plant assemblages in Eiawa iiVolcanoes National 'ark. Veqeta(io 9 5. 55-02.

Aragon. R. and Morales, J.M. (2004) Species compo-sition and invasion in NW Argentinian secondaryforests: effects of land-use history, environmentmid landscape. Journal of Veqelation Science 14,195-204.

Augspurger. C. 1984) Seedling survival of tropical treespecies: interactions of dispersal distances, light gaps.and pathogens. Ecohyjy 65. 1 705-I 712.

Bagchi, S., Coyal. SE.. and Sankar, K. (2004) Herbivoredensit y and biomass in a semi-arid tropical dry dccid-uous lorest of western India. Journal of Tropical Ecology)0,475-478.

tlaret. S.. Nicolini, F., I.e Bourgeois. 'I.. and Strashcrg. U.(201)1) Developmental patterns of the invasive bram-ble (Ri /sis alceijiAiiis Poiret. Rosaceae( in Rcunionisland: an architectural and morph met nc analysis.A ,inals vi Bolwuj 9 1. 39-48.

Iteckage. B. antI Clark, J.S. (2001) Seedling survival andgrowth of three loresi tree species: the role of spatialheterogeneity. L'cologi 84. 1849-] 561

ltelhingham. Ej.. Tanner. E.VJ.. and Heale y. J.R. (2005)H urrica lie disturbance accelerates invasion by thealien tree Pitlosporitol iindulaiwu in Jamaican mon-tane rain f rcsts.Joii 'mi of Vt'qei (((toil Science 1 6.675-684.

Cabin. RI., Weller. S.C., l.orencc. D.H. eta). (2000) Effectsof long-term ungulate exclusion and recent alienspecies control on the preservation and restoration


of a Hawaiian Tropical Dry Forest. (o,oi'r ya(i(n, BEd-ogy 14.439-453.

Chandrasekaran. S. and Swarny, P.S. (2000) Changes inherbaceous vegetation following disturbance due tobiotic interference in natural and man-made ecosys-tems in Western Chats. Tropical Ecohatij 36, 213-220.

Chandrashekara, U.M. and Ramakrishnanan. P.S.(1994) Successional patterns and gap phase dynam-ics of a humid tropical forest of the Western Chatsof Kerala. India: ground vegetation, biomass. pro-ductivity and nutrient cycling. Forest Eco!oqy andManagement 70. 2 5-40.

Chereit, J.M. (1989) Leaf-cutting ants. In H. 1.ieth andM.J.A. Weiger (eds). Ecosystems of the World. Elsevier.New York-, pp. 473-489.

Clark, J.S., Bcckage, B., Cumill. Pet al. (I 999a) Interpret-ing recruitment limitation in forests. American Journalof Botamj86, 1-16.

Clark. J.S., Macklin. E.. and Wood, L. (1998) Stagesand spatial scales of recruitment limitation in south-ern Appalachian forests. Ecological Monographs 68,213-235.

Clark. IS., Silman. M., Kern, R., Macklin. E.. andHilleRisLambers, J. (1999h( Seed dispersal near andfar: patterns across temperate and tropical forests.Ecology 80, 1475-1494.

Coley. 21). and Aide. 'EM. (1991) Comparison of her-bivory and plant defenses in temperate and tropicalbroad-leaved forests. In P.W Price. 'EM. l,ewinsohn,C.W. Fernandes, and W.W. Benson (eds). ('loot-Animal Inleractions: Cvohmtionar:j Ecology in iropicaland Temperate Regions. John Wiley & Sons. New York.pp. 25-49.

Coley. P.D. and Barone, J.A. (1996) Herbivnry and plantdefenses in tropical forests. Annual Review of Ecologyand Systeoumtics 27. 305-335.

Conant. P., Medeiros. A.C.. and Loope, L.L. (1997)A multiagency containment program for miconia(Mironia calvescens), an invasive tree in Hawaiian rainforests. In JO. Luken and J.W Thieret (eds). Assessmentand Management of Plant invasions. Springer-Verlag.New York, pp. 249-2 54.

D'Antonio, C.M. and Dudley. T.L. 1995) Biological inva-sions as agents of change on islands versus mainlands.In P.M. Vilousek, L.L. Loope. and II. Adsersen I cds(,Islands: Biological l)iversitij and Ecosrjstenm Function.Springer-Verlag, Stanford, pp. 103-12 1.

D'Antonio. C.M. and Vitousek. P.M. (1992) Biologi-cal invasions by exotic grasses, the grass/fire cycle.and global change. Annual Review of P.cologij andSystematics 23. 65-87.

Davis, MA.. Wrage. K.J., and Reich. PB. (1998)Competition between tree seedlings and herbaceous

vegetation: support tor a theory of resource supplyand demand. Journal of Ecoloqij 86. 6S2-661.

Davis, MA., Crime. I.E. and Thompson. K. (2000)Fluctuating resources in plant communities: a gen-eral theory of invasibilily. Journal of Ecology 88,i28-5 34.

Denslow. J.S. (1987) Tropical rainforest gaps andtree species diversit y. Annual Review of Ecology andSystematics 18. 431-451.

Denslow. J.S. (2002) Invasive alien woody species inPacific island forests. Vnasijlva 53, 62-63.

Denslow, J.S. (2003) Weeds in paradise: thoughts on theinvasihility of tropical islands. Annals of the MissouriBotanical Garden 90. 119-127.

Denslow, J.S., Ellison, A.. and Sanford. R.F. Jr. (1998)Treefall gal) size effects on above- and below-groundprocesses in a tropical wet forest. Journal of' Ecology86,597-609.

l)enslow, J.S. and Johnson, M.T. (2006) Biologi-cal control of tropical weeds: research opportuni-ties in plant-herbivore interactions. Biotropica 38,159-142.

l)enslow, J.S.. Newell, E.. and Ellison, A.M. (199 1 ) Theet'fect of understory palms and cyclanths on thegrowth and survival of inga seedlings. Biotropica 23,225-234.

Denslow. J.S.. tiowo[o. A.l.. and Hughes, RE (2006)Limitations to seedling establishment in a mesicHawaiian forest. Qecoloqia 148, 11 8-128.

DeWalt. S.). (2005) Effects of natural enemies ontropical plant invasions. In I).ER.P. l3urslern. M.Pinard, and S. 13. Hartley (eds). Biotic Interactions inthe 'I'ropics. Cambridge University Press, Cambridge.pp. 459-483.

DeWalt, S.J.. Denslow. J.S., and Ickes. K. (2004) Natural-enemy release facilitates habitat expansion of theinvasive tropical shrub Clide,nia hiria. Ecology 85,471-483.

Diets. H.. Wirth, 1.11.. and Buchmann, 11. (2004)Variation in herbivore damage to invasive and nativewoody plant species in open forest vegetation on tvlimhé.Seychelles. Biological invasions 6. 513-521.

Di Stefano, JR. Fournier, TA., Carranza, J., Mann. W.and Mora. A. (1998) Invasive potential of Syzygiurnjamnbos (Myrlaceae( in forest fragments: the case ofCiudad Colon, Costa Rica. Rem'isla di' Bioloqia Tropical46.5617-573.

Elton. C . S. (1958) The Ecology of Invasions of Animals andPlants. Methuen, London.

Ewel, J . J . (1986) Invasibility: lessons from South Florida.In ILA. Mooney and J.A. Drake (eds). Ecologjj ofBiological Invasions of North America and Dawn, 'i.Springer-Verlag, New York. pp. 214-230.


Ewel. J . J.. O'Dowd. I).J.. Bergelson, I. cI a). (1999)Deliberate introductions of species: research needs.BioSch'nce 49. 6 19 -h W.

Farji-Brener, AC 2001) Why are leaf-cutting antsniorecommon iii earl y secondary forests than in old-growthtropical forests? An evaluation of the palatable foragehypothesis. Dikes 92. 169- 177.

Farris-Lopez, K., Denslow, J ..S.. Moser. I)., andl'assmorc, H. 2004) Influence of a coniniofl palm.Oenocarpus nra pora. on seedling establishment in atropical moist forest in Panama. Jmirrial of TropicalEcology 20, 429-439.

Feosham, RI.. Fairfax. R.I.. and Cannel), R.I. 1994)The invasion of Lanlana caniara I.. in Fort y Mile ScrubNational Park, North Queensland. Australian Journalof Ecology 19, 297-31)5.

Fine, P.VA. (21)02) The hivasibilitv of tropical forests byexotic plants. Journal of Tropical Ecology 18, 687-705.

Fleisehmann, K. 11997) Invasion of alien woody plantson the islands of Mahe and Silhouette. Seychelles.Journal of Vegetation Science 8. 5-1 2.

Foster. R.B. and Hubbell, SR (1990) The floristic com-position of the Barro Colorado Island forest. InA.t I. Gentry (ed.). tour Neotropical Rain Gorests. YaleUniversity Press, New Haven, pp. 85-98.

Fowler, HG., Pagani, Ml., Da Silva, D.A.. Forti, IC.. doSilva. V.P.. and Vasconcelos. H.L. (1989) A pest is a pestis pest?'fhedilemnia of Neutropical leaf cuttin g ants:keystone taxa of natural ecos ystems. Ei,i')ri,nrru'iutalManageiru'nt 13. 671-6

Fridley. ID., Brown. R.L.. and Bruno. J.E (2(11)4) Nullmodels of exotic invasion and scale-dependent pat-terns of native and exotic species richness. Ecology 85.3215-3222.

Gadgil. Pt). and ham, J . (1999) Vulnerability of plantedforests to biotic and ahiotfc disturbances. New forests17.227-238.

Gerrish, C.. Stemmcrmann. I,., and Gardner, D.E. (1992)The distribution of Rulius species in the State OfHawaii. Cooperative National Park Resources StudiesUnit, Honolulu. Hawaii. No. 85.

C hazou I. J . (20114) Alien abduction: disruption of nativeplant-pollinator interactions by invasive species.Biol ropica 36, 1 56-1 64.

Green, PT.. O'Dowd. D.].. and Lakes, P.S. (2004) Resis-tance of island rainforest to invasion by alien plants:influence of microhabitat and herbivory on seedlingperformance. Biological Invasions 6. 1 -9.

Hammel. B. (1991)) The distribution of diversityamong families, genera and habitat types in theLa Sd va Flora. In A.l I. Gentry (ed.). Four Neotrop-ira) Rain Forests. Yale University Press. New Haven.

pp. 75-84.

I harms. KU.. Powers, 1 .5. . and Montgomery. R.A. (2004)Variation in small sapling density. understor y cover,and resource availability in four Neotropical forests.13101 rop)ca 36. 4(1-51.

Harms, KU., Wright. S . J . , Calderon, 0., Hernánidev.. A..and Iterre. E.A. (2000) Pervasive density-dependentrecruitment enhances seedling diversity in a tropicalforest. Nature 404.493-495.

hlerben,'I'.. Mandák. B., Bimovd. K., and Munzbergova. Z.(2004) Invasibilit y and species richness of a commu-nity: a neutral model and a survey of published data.Geology 85. 3223-32 13.

Hood. L.A.. Swaine, MI).. and Mason. PA. (2004) The influence of spatial patterns of damping-off disease and arbuscular mycorrhr,.al coloniza-tion on tree seedling establishment in Ghanaiantropical forest and soil. Journal of Ecology 92.

816823.Hooper. DU., Chapin III, US., Ewcl, J . J . etal. (2005) Effects

of hiodiversity on ecosystem functioning: a consensusof current knowledge and needs for future research.l3cological Monographs 75. 3-35.

Horvitz, CC., McMann. S.. and Freedman. A. (1995)Exotics and hurricane damage in three hardwoodhammocks in Dade Count y Parks, Florida. Journal ofCoastal Research 21. 145 1 58.

Horvitz. CC.. Pascarella.J.B.. McMann. S.. Freedman. A..and Hofstettcr. R.h 1. (1998) Functional roles ofinvasive non-indigenous plants in hurricane-affectedsubtropical h ardwood forests. Ecological Applications8.947-974.

Hubbell. S.P. (200II) The Unified Neutral Theory of Broth-i'rsftrj and Biogeography. Princeton University Press,Princeton. NJ.

Huenneke, I U. and Vitousek. PM. (1990) Seedlingand clonal recruitment of the invasive tree Psid-

iran cat tleianu,n: implications for management, ofnative Hawaiian forests. Biological Conservation 53.

199-213.Hughes. C.E. (1994) Risks of species introductions in

tropical forestry. Co,nnionwealt it l'oresl ry Review 73.243 252.

Hughes. R.F. and Denslow, J.S. (2005) Invasion by anN -fixing tree, l-idcataria inoluccana. alters function andstructure of wet lowland forests of Hawaii. EcologicalAmrli(af ions 15, 1615-1628.

Hughes, C.E. and Styles. B.T. (1989) The benefits andrisks of woody legume introductions. In C.H. Stirtonand 1.1.. Za rruch i ieds), Advances in legume Biologic.Monographs in Systematic Botany. Missouri BotanicalGarden 29. pp. 5(15-531.

Humphries. S.F.. Groves, Rh.. and Mitchell. D.S. ( 199 1)Part I. Plant invasions of Australian ecosystems: a


status review and management directions. Koa'ari 2,1-134.

Humphries, S.F. and Stanton. J.P. 11 992) Weed Assess-went in the tAh1 Tropics World I leritage Area of NorthQueensland. Report to the Wet Tropics ManagementAgency Report.

Ickes. K.. l)eWalt. SI.. and Appariah. S. (2001) Efkctsof native pigs (Sus scrofa( oil understorey veg-etation in a Malaysian lowland rain forest. Journal ofTropical Ecology 1 7. 19 1-206.

Keane. K.M. and Crawley, M.J. (2002) Exotic plant inva-sions and the enemy release hypothesis. Trends inEcology and Evolution 17. 164-170.

Kennedy, TA.. Naeem. S.. howe, KM., Knops, J.M.H.,Tilman. I)., and Reich. PB. (2002) Biodiversityas a harrier to ecological invasion. Nature 417,636-638.

Kitayaina, K. (1996) Patterns of species diversity onan oceanic versus a continental island mountain: ahypothesis on species diversification. Journal VegetationScience 7. 879-888.

Kitayama, K. and Itow. S. (1999) Above-ground biomassand soil nutrient pools of a Scalesia peduncalatamontane forest on Santa Cruz, Galcpagos. EcologicalResearch 14. 405-408.

Kitayaina. K. and Muellcr-I)omhois. 1). (1995) Biologicalinvasion on an oceanic island mountain - do alienplant species have wider ecological ranges than nativespecies? Journal of Vegetation Science 6, 667-674.

Knapp, L.B. and Canham. C.D. (20001 Invasion of anold-growth ti/rest in New York by Ailanthus aliissima:sapling growth and recruitment in canopy gaps.Journal of the Torrez Botanical .Societij 12 7, 307-315.

Laurance. W.F. (1997) Hyper-disturbed parks: edgeeffects and the ecology of isolated rain forestreserves in tropical Australia. In WE Laurance andR.O. Bierregaard, Jr. (eds). Tropical I"orest Remnants:Ecology. Management, and Conservation of fragmentedCommunities. University of Chicago Press, Chicago,pp. 71-83.

Levine. J.M. (2000) Species diversity and biological inva-sions: relating local process to community pattern.Science 288. 852-854.

Levine. J.M. and ])'Antonio, C.M. (1999) Elton revisited:a review of evidence linking diversit y and invasibilityOikos 87, 15-26.

Lockwood. J.l... Cassey, P. and Blackburn. T.M. (2005)The role of propagute pre ssure in explaining speciesinvasions. Trends in Ecology and Evolution 20.223-228.

l,onsdale. W.M. 11999) Global patterns of plant inva-sions and the concept of invasibilit y. Ecology 8(1.1522-1536.

Lugo, A.E. and Helmer. F. (2004) Emerging fl/rests ona abandoned land: Puerto Rico's new forests. ForestEcology and Management 190. 145-161.

MacDonald. LAW, Ortiz, L., I,awesson, J.E. , andNowak, J.B. (1 988)fhe invasion of highlands in GaUl-pagos by the red quinine tree. Cinchona succirithra.Eriviromnen(al Conserval ion I S. 2 15-220.

Mack. R.N. (1996) Biotic barriers to plant natural-ization. In VC. Moran and J.l I. Hoffmann (eds),IX International Sijmnposncoi on Biological ComUisI.Universit y of Cape 'town, Stellenbosch, South Africa,pp. 39-46.

Mack. RN., Simherloti, D., Lonsdale. WM., tvans, Fl.,and Clout. M. (2002 ( Biotic invasions: causes, epi-demiology, global consequences and control. Fcoloq-hal Applications 10, 689- 710.

Maron. 1.1.. and Vita, M. (2001) When do herbivoresaffect plant invasion? Evidence lor the ri/it ural ene-mies and biotic resistance hypotheses. Oikos 95.361-373.

McDowell. C.L. and Turner, l).P. (20112) Reproductiveeffort in invasive and nun-invasiveRubus. ()ecoloqia133,102-111.

Meyer. J.-Y. (1996) Status of Miconia calc'escens (Melas-tomatac'eae), a dominant invasive tree in the SocietyIslands (French t'otvnesia(. Pacific' Science 50, 66-76.

Meyer, J.-Y. and Lavergne, C. (2004) llc'aute.s fatales:Acanthaceae species as invasive alien plants on tropi-cal Indo-Pacific Islands. Diversity and Distributions It).333-347.

Montgomery. R.A. (2004) Effects of understor y foliageon patterns of light attenuation near the forest floor.Biotropica 36. 3 3-40.

Mueller-liombois, D. (1995) Biological diversity anddisturbance regimes in island ecosystems. In P.M.Vitousek, L.L. Loope, and H. Adserscn (eds(, Islands:Biological Diversity and l:'coszjstern Punc -lion. Springer.Berlin, pp. 163-175.

Nair. K.S.S. (2001) Pest Outbreaks fit tropical Forest Plan-tations: Is there a Greater Risk for Exotic free Species?('enter for International Forestry Research. Bogor,Indonesia.

Nathan. R. and Muller-landau, tIC. (2000) Spatialpatterns of seed dispersal, their determinants andconsequences for recruitment, trends in Ecology andEvolution 15.278-285.

Novotny, V. Basset, 'F. Miller, S.E. c'l al. (2004) localspecies richness of leaf-chewing insects feeding onwoody plants from one hectare of a lowland rainforest.Biological Conservation 18, 227-2 17.

Novotny. V.. Miller. SE., Cizck. I.. et al. (2003)Colonising aliens: caterpillars (lepidoptera) feedingon Piper aduncum and P. umbellaluni in rainforests


of Papua New Guinea. Geological Pot oniology 28.

704-716.Nye, PH. and Greenland. D.J. (1960) The Soil Under

SIn)) i,iq Cultivation. Technical Communication No. 51Consinonwealtli Bureau of Soils. I larpenden.

Parker. J.i).. Burkepile. D.E., and Day, ML. (2006)Opposing effects of native and exotic herbivores onplan! invasions. Scienn' 11, 1459-1461.

Parsons. J.J . (1972) Spread of African pasture grassesin the American tropics. Journal of Range Management

25. 12-17.Prieur-Richard. AK.. I,avorel....I,inharl. Y.B. and Dos

Santos. A. (2002) Plant diversit y, herhivory andresistance of a plant community to invasions, inMediterranean annual communities. Oecologia 130.96-104.

Puts, RE. (1991) Silvicultural effects of lianas. InFJ'. Puts and H.A. Mooney, (cds), The Biology

of Vines. Cambridge University Press. Cambridge.pp. 49 3-50 1.

Pyek, P.. Jarolek. V. and Kuera, T. (2002) Patternsof invasion in temperate nature reserves. Biological('onservatian 104. 13-24.

Ramakrishnan. P.S. (1991 ) Ecoligg of Biological liii'asionin the Tropics. international Scientific Publications.New Delhi.

Ramakrishnan. US. and Vitousek, P.M. (1989)Ecosystem-level processes and consequences of bio-logical invasions. in j.A. Drake, H.A. Mooney,P Di Castr( ci al. )eds). Ilioleqical Invasions: /I GlobalPerspective. Wiley. New York. pp. 281-128.

Rcjms'inek. M. (1989) lnvasibilitv of plant communities.In ].A. Drake. H.A. Mooney. K IN Castri et al. (cds) Bio-logical Invasions: A Global l'erspec)ive. Wiley. New York.pp. 369-388.

Rejmuinek. M. (1996) Species richness and resistance toinvasions. In C. Orians. R. Dirzo, and J.11. Cushman( eds). Biodiversily and Ecosisteni Processes ill

Forests. Springer Verlag. New York. pp. 15 3-172.Rejmânek. M. (1999) Invasive plant species and inva-

sible ecosystems. In O.1 Sandlund. P. 1. Sehei. andA. \'iken ) eds) . Invasive Species and Bio(livi'rsity Man-aqe,nenl. Kluwer Academic Publishers, Boston. MA.pp. 79-102.

Richardson. 1).M. (1999) Commercial forestr y and agro-forestry as sources of invasive alien trees and shrubs.In O.T. Sandlund, P. J. Schei, and A. \'iken )eds). lni'asivi'Species tit id Biodiversity Management.. Kluwer AcademicPublishers, Boston, MA. pp.2 37-257.

Rogers, H.M. and Hartemink. A.E. (2000) Soil seed hankand growth rates of an invasive species. Piper idun-ciiiii, in the lowlands of Papua New Guinea. Journal ofTropical Ecology 16. 243 251.

Rosenthal, G.A. and Janzen, D.H. (1979) herbivores.

Their Interaction with Secondary Plant Metabolites.Academic Press. New York.

Rosenzweig. M.L. and Ahramsky, Z. (1993) How arediversity and productivity related? Iii R.E. Ricklefsand I). Schluter (eds), Species Diversity in GeologicalConi,nu,nties. University of Chicago Press. Chicago,pp. 52-65.

Sala. OF.. Chapin III. KS., Armesto. J . J . ci al. (2000)Global hiodiversity scenarios for the year 2100. Science287,1770-11-74.

Sax. l).K, Gaines. S.D..osil Brown. ill. (201)2) Speciesinvasions exceed extinctions on islands worldwide:a comparative study of plants and birds. AmericanNaturalist 160, 766-783.

Schnitzer. S.A.. Dalling. JAY, and Carson, WP. (2000)the impact of lianas on tree regeneration in tropicalforest canopy gaps: evidence for an alternative path-way of gap-phase regeneration. Journal of Geology 88.655-666.

Shea. K. and Chesson, P. (2002) Community ecologytheory as a framework for biological invasions. Trendsin hicology and Evolution 1 7. 170-176.

Shell, I). (1994) Naturalized and invasive speciesin the evergreen forests of the East Isambara,Mountains, Tanzania. African Journal of Ecology 32,66-71.

Sheil, t).. Jennings, S.. and Savill. P. 12000) Long-termpermanent plot observations of vegetation dynamicsin Budon go. a II gaodan rain forest. Jou rnal of Tropical

Ecology 16, 765-800.Smith. C.W (1985) Impact ol alien plants on

Hawaii's native biota. In C.P. Stone and J.M. Scott)eds). Hawaii's Terrestrial Ecosyslenis: Preservationand Management. University of Hawaii CooperativeNational Park Resources Studies unit, Honolulu, Fit,PP 180-250.

Space. J.C. (21)1)5) Pacific Island Eeosijslemmis a) Risk. v5.0. tJSI)A Forest Service. Institute of Pacihic IslandsForestry. (www.hesir.org/pier).

Stark, S.C.. Bunker. D.E., and Carson, W.P. (2006) Anull model of exotic plant diversity tested with exoticand native species-area relationships. Ecology letters9.136-141.

Stohlgren. 'I'.J.. Binkley. I)., Chong, G.W. vi al. (1999)Exotic plant species invade hot spots of native plantdiversity Ecological Monographs 69. 2 5-46.

Sullivan, J.J . , l'immins. SM.. and Williams, RA. (2005)Movement of exotic plants into coastal native forestsfrom gardens in northern New Zealand New Zealand

Journal of Ecology 29, 1-11).Svenning. J.C. and Wright. S.J. (211115) Seed limitation in

a Panamanian forest. Journal of Ecology 9 3, 85 3- 862.


Tabáney., i\.A.J., Viana. YM.. and Dias, A.D.S. (1997)ConseqUências da fragmentaç yo e do efeito de bordasobre a estrutura, diversidade e sustentahilidade deurn [ragmento de (loresta dc Planalto de Piracicaba,SF'. Revista Brasileira Ic Biologia 57, 47-60.

Teo. D.H.I... Tan, H.T.W., Corlett. RI, Wong. CM.. andLum, S.K.Y. (2003) Continental rain forest fragmentsin Singapore resist invasion b y exotic plants. Journal ofBiogeoqmphij 30, 305-310.

Tilman, 11, Reich. PB., Knops, J . . Wedin. 1)., Mielke, T.,and Lehman, C. (2001) Diversit y and productivityin a long-term grassland experiment. Science 294,843-845.

Turnbull, L.A.. Crawley. M.J.. and Rees, M. (2000) Areplant populations seed-limited? A review of seed-sowing experiments. Qikos 88. 225-238.

lJhl, C. and Kauffman, J.B. (1990) Deforestation, lire sus-ceptibility. and potential tree responses to lire in theeastern Amazon. Ecoloqmj 71, 437-449.

Vitousek, P.M. (3986) Biological invasions and ecosys-tem properties: can species make a difference? InH.A. Mooney and J.A. Drake (eds), Ecology of Bii,hiy-meal !mmnis ions of North American and Hawai....SpringerVerlag, New York, pp. 163-176.

Vmtousek, P.M. and Sanford, R.L. (1987) Nutrient cyclingin moist tropical forest. Annual Review of Ecology and.Sqsteoiatics 17. 137-168.

Vitoumiek, P.M., Walker, I..R., Whitcaker, L.D., Mueller-Dombois. I)., and Matson, PA. (1987) Biologicalinvasion by MijricaJcmmja alters ecosystem developmentin Hawaii. Science 238, 802-804.

Von Holle. B. and Simherloff. Ii (2005) Ecologicalresistance to biological invasion overwhelmed bypropagule pressure. Ecology 86, 3232-3218.

Wadsworth, F.H. (1997) Forest Production /hr TropicalAmerica. Agriculture Handbook 710. USDA ForestService. Washington, DC.

Wagner, W.L., Herbs!, DR.. and Sohmer, S.H. (1999)Manual of the Howerioy l'lants of Hawaii. 2ndEdmtion.Ilnivcrsity of Hawaii Press. Honolulu, HI.

Walker, L.R.. Zimmerman, J.K.. Lodge. D.J.. andGuznisui-cirijales, S. (1996) An altitudinal com-parison of growth and species composition inhurricane-damaged Forests in Puerto Rico. Journal ofLcoloqij 84. 8 77-890.

Wang, Y.-H. and Augspurger, C.K. (2004) Dwarf palmsand cyclanths strongly reduce Ncotropical seedlingrecruitment. Oiko,c 107, 619-633.

Webb, SI.. Dwyer, M., Kaunzinger. C.K., and Wyckoff,P11. (20110) The myth of the resilient forest: Casestudy of the invasive Norwa y maple (Acer plot anoides).Rhodora 102, 332-354.

Wester, L.L. and Wood, 11.0. (1977) Koster's curseClidenna hmr(a), a weed pest in Hawaiian forests.

Environmental Conservation 4, 35-4 1Whitmore. T.C. (1991) Invasive woody plants in perhu-

mid tropical climates. In P.S. Ramakrishnan (ed.), Ecol-ogy of Biological Invasions in the Tropics. InternationalScientilic Publications, New Delhi, pp. 35-4().

Wirth, R., lleyschlag. W, Rye!, Il., Hers. I-I., and 1-loll-dohier, B. (200 3) Herhim'orq of Leal-Cutting Ants: A CaseStudy on A tta columbica in the 'Tropical Rain Forest ofPanama. Springer-Verlag, New York.

Woodcock, D. 2007) To restore the watersheds:early twentieth-century tree planting in Hawaii.Annals of the Association of American Geographers 31.624—hIS.

Wu, 5.-H.. Hsieh, C-F.. Chaw, S.-M.. and Rejmanek, M.(2004) Plant invasions in 'Taiwan: insights l'rom theflora of casual and naturalized alien species. Dim'm'rsitiand I)isributions 10, 349-362.

Zimmerman, LK., Pmiscarc'lla, .I3.. and Aide, 'LM. (2(1)10)Barriers to forest regeneration in an abandoned pas-ture in Puerto Rico. Restoration Ecology 8 , 35))— 360.

Documents

EXOTIC PLANT INVASIONS IN TROPICAL FORESTS: PATTERNS … › psw › publications › denslow › psw... · Chapter 24 EXOTIC PLANT INVASIONS IN TROPICAL FORESTS: PATTERNS AND HYPOTHESES