Plant and Soil 256: 29–39, 2003.© 2003 Kluwer Academic Publishers. Printed in the Netherlands.

29

Allelopathy and exotic plant invasion

Jose L. Hierro1,2,3 & Ragan M. Callaway1

1Division of Biological Sciences, The University of Montana, Missoula, MT 59812, USA. 2Facultad de CienciasExactas y Naturales, Universidad Nacional de La Pampa, Uruguay 151, (6300) Santa Rosa, La Pampa, Argentina.3Corresponding author∗

Received 11 November 2002. Accepted in revised form 13 May 2003

Key words: allelopathy, competition, exotic plants, invasion, natural enemies’ hypothesis, plant interactions

Abstract

The primary hypothesis for the astonishing success of many exotics as community invaders relative to their im-portance in their native communities is that they have escaped the natural enemies that control their populationgrowth – the ‘natural enemies hypothesis’. However, the frequent failure of introduced biocontrols, weak consumereffects on the growth and reproduction of some invaders, and the lack of consistent strong top-down regulation inmany natural ecological systems indicate that other mechanisms must be involved in the success of some exoticplants. One mechanism may be the release by the invader of chemical compounds that have harmful effects onthe members of the recipient plant community (i.e., allelopathy). Here, we provide an abbreviated compilationof evidence for allelopathy in general, present a detailed case study for Centaurea diffusa, an invasive Eurasianforb in western North America, and review general evidence for allelopathic effects of invasive plants in nativecommunities. The primary rationale for considering allelopathy as a mechanism for the success of invaders is basedon two premises. First, invaders often establish virtual monocultures where diverse communities once flourished,a phenomenon unusual in natural communities. Second, allelopathy may be more important in recipient than inorigin communities because the former are more likely to be naïve to the chemicals possessed by newly arrivedspecies. Indeed, results from experiments on C. diffusa suggest that this invader produces chemicals that long-termand familiar Eurasian neighbors have adapted to, but that C. diffusa’s new North American neighbors have not. Alarge number of early studies demonstrated strong potential allelopathic effects of exotic invasive plants; however,most of this work rests on controversial methodology. Nevertheless, during the last 15 years, methodologicalapproaches have improved. Allelopathic effects have been tested on native species, allelochemicals have been testedin varying resource conditions, models have been used to estimate comparisons of resource and allelopathic effects,and experimental techniques have been used to ameliorate chemical effects. We do not recommend allelopathyas a ‘unifying theory’ for plant interactions, nor do we espouse the view that allelopathy is the dominant waythat plants interact, but we argue that non-resource mechanisms should be returned to the discussion table as apotential mechanism for explaining the remarkable success of some invasive species. Ecologists should considerthe possibility that resource and non-resource mechanisms may work simultaneously, but vary in their relativeimportance depending on the ecological context in which they are studied. One such context might be exotic plantinvasion.

Introduction

Many of the world’s most common and ecologicallydevastating exotic invaders are not as successful at

∗ FAX No: 406-243-4184. E-mail: jhierro@selway.umt.edu

home. Why many exotic plant species competitivelyexclude and eliminate their neighbors in invaded ‘re-cipient’ communities but coexist in relative peace withneighbors in species-diverse systems in their nativehabitat remains one of the most important mysteriesin ecology. Hypotheses include the existence of empty

30

niches in recipient communities, rapid genetic changesin invader populations in response to new selectionpressures in the novel environment, and special ad-aptation to human disturbance by invaders (Mack etal., 2000; Sakai et al., 2001). But by far the primarytheory for the unusual success of invasive plants is thatthey have escaped the natural enemies that hold themin check, freeing them to utilize their full competitivepotential – the ‘natural enemies hypothesis’ (Darwin,1859; Elton, 1958; Gillet, 1962; Williams, 1954).These hypotheses are not mutually exclusive, and eachmay best explain the success of particular invasivespecies. The natural enemies hypothesis for invasiveplants has been informally tested by the release ofhundreds of biological controls (herbivorous insects)around the world. The large majority of biocontrolshave been ineffective (Maron and Vilà, 2001), leavingquestions about mechanisms for the general success ofmany exotic invasive plants open for inquiry.

Allelopathy has also been suggested as a mech-anism for the impressive success of invasive plants,in part because invaders often establish virtual mono-cultures where diverse communities once flourished.Here, we use the term allelopathy as the negativeeffect of one plant on another one through the re-lease of chemical compounds into the environment(sensu Muller, 1969) and consider allelopathy as justone form of non-resource interactions among plants(Callaway, 2002). To many ecologists, the establish-ment of monospecific stands suggests the need forunusually potent mechanisms. However, to explain thestriking biogeographical pattern of many successfulinvaders, these mechanisms must be much more potentin recipient communities than native communities.Many studies suggest that allelopathy may contributeto the ability of particular exotic species to becomedominants in invaded plant communities (e.g., Abdul-Wahab and Rice, 1967; El-Ghareeb, 1991; Fletcherand Renney, 1963; Kanchan and Jayachandra, 1979;Osvald, 1948; Vaughn and Berhow, 1999; Ridenourand Callaway, 2001). Furthermore, recent experimentshave demonstrated that the invader Centaurea diffusamay exude allelochemicals that are relatively ineffect-ive against long-time neighbors in its natural com-munities, but to which plants in invaded communitieslack co-evolved tolerance (Callaway and Aschehoug,2000).

The suspicion that chemicals ‘inimical to plantgrowth’ may be a mechanism by which plants may in-terfere with their neighbors has been around for a longtime. In 1832, De Candolle speculated that plants, spe-

cifically noxious weeds, may exude chemicals fromtheir roots that are detrimental to the growth of otherplants. At the beginning of the last century, Livingstonet al. (1905) and Schreiner and Reed (1907, 1908) con-ducted a series of experiments using ‘carbon black’ tomanipulate toxic substances produced by the roots ofTriticum vulgare (wheat) and Avena sativa (oats). In-creasing recognition of the widespread importance ofthese non-resource-driven interactions rests on studiesthat encompass various non-resource interactions in-cluding inter- and intraspecific root communications(Mahall and Callaway, 1991, 1992; Gersani et al.,2001), wound-stimulated production of compoundsthat induce defenses in unwounded conspecific and in-terespecific neighbors (Arimura et al., 2000; Baldwinand Schultz, 1983; Karban et al., 2000), root-root sig-nals between parasites and hosts (Steffens et al., 1983),and oxidation of gases in smoke or acids from burnedplants that cue germination of other species (Keeleyand Fotheringham, 1997). There is a great deal of evid-ence that plant–plant interactions can go well beyonddirect competition for shared resources such as light,water, nutrients, and space (Agrawal, 2000; Callaway,2002; Schenk et al., 1999).

Allelopathy experienced its heyday in the 1960sand 1970s, a time during which allelopathic explan-ations for community processes and patterns werecommon (Muller, 1966, 1969; Rice, 1974; Whit-taker and Feeny, 1971). However, the demand formechanistic experiments that eliminated alternativeexplanations was not met and the broad acceptanceof allelopathy was replaced with skepticism for its ex-istence (see Williamson, 1990 for a lucid discussionof mechanistic catch-22’s and variable standards re-garding allelopathy and resource competition). Today,despite a much larger body of evidence for allelopathyas an important plant interaction, this skepticism re-mains (for a critical evaluation of contrasting viewson the ecological significance of allelopathy also seeWardle et al., 1998) and many texts present almost‘straw man’ evidence for allelopathy in plant com-munities. If these texts are barometers, allelopathydoes not appear to be important at all (Gurevitch et al.,2002; Pugnaire and Valladares, 1999, but see Barbouret al., 1999). We do not recommend a return to allelo-pathy as a ‘unifying theory’ for plant interactions, nordo we espouse the view that allelopathy is the domin-ant way that plants interact; however, we argue herethat non-resource mechanisms should be returned tothe discussion table as a mechanism for explaining theremarkable success of some invasive species. In this

31

paper we provide an abbreviated compilation of thecurrent evidence for allelopathy in general, present thecase study for C. diffusa in detail, and review generalevidence for allelopathic effects of invasive plants innative communities.

Allelopathy as a mechanism

Does it matter if allelopathy or resource competition(or some combination of the two) drives particularexotic invasions? From a perspective of what inva-sions can teach us about theory and plant communities– certainly. From the perspective of how invasionsmight be controlled – probably. There are severaltheoretical implications to the scenario of invaderssucceeding by employing allelochemicals (also seeCallaway and Aschehoug, 2000). First, disruption ofnative communities by chemicals exuded by an exoticwould suggest that natural plant communities mightbe more tightly knit entities than generally thoughtand that invasion disrupts inherent, co-evolved inter-actions among long-associated native species. Invasivesuccess due to an ability to out-compete locals hasno new bearing on the nature of plant communities.Second, unusually strong allelopathic effects of aninvader against species in recipient communities sug-gest that plant–plant interactions may be somewhatspecies-specific. This may not be the case if invaderssimply out-compete locals for resources. Finally, ifallelopathy is more important in exotic invasion thanin natural communities the possibility is raised thatinteractions among plant species may drive naturalselection in communities, in turn implying that nat-ural biological communities may evolve in some wayas functionally organized units (see Goodnight, 1990;Wilson, 1997). From an applied perspective, if exot-ics succeed because of an unusual capacity for re-source competition then control is a matter of keepingthem from acquiring resources. Herbicide, biocon-trols, grazing, or any other method of reducing themass of invaders should help. If allelopathy is in-volved, the effect of reducing the mass of invasive spe-cies may be much less predictable. In fact, induction ofherbivore defenses may actually increase allelopathiceffects as indicated for Brassica rapa (Siemens et al.,2002).

Effects of Centaurea diffusa on new versus oldneighbors

Callaway and Aschehoug (2000) compared the com-petitive effects of C. diffusa, an invasive Eurasian forb,on three bunchgrass species that co-exist with C. dif-fusa in Eurasia to the effects of C. diffusa on threebunchgrass species from North America. Each of thethree species from North America was paired with acongener (or a near-congener) from Eurasia of a sim-ilar morphology and size. Seeds of C. diffusa, Festucaovina, Koleria laerssenii, and Agropyron cristatumwere collected within an area of several hectares inthe southern foothills of the Caucasus Mountains inthe Republic of Georgia. Seeds of F. idahoensis, K.cristata, and Pseudoroegneria spicata were collectedfrom grasslands in the northern Rocky Mountains inMontana, USA. Until recently Pseudoroegneria wasincluded in the genus Agropyron. Each of the grassspecies was common at their respective sites. In theCaucasus, the cover of C. diffusa was less than 1%,whereas at the Montana site the cover of C. maculosa(which is closely related to C. diffusa) was 10–90%.Each of the seven species was planted alone and in allpairwise grass-Centaurea combinations and all com-binations were grown in sand and in sand mixed withactivated carbon expected to ameliorate any poten-tial allelopathic effects because of its high affinity foradsorbing to organic compounds.

Centaurea diffusa had much stronger negative ef-fects on North American species than it had onEurasian species. Correspondingly, none of the NorthAmerican grass species (nor all species analyzed col-lectively) had a significant competitive effect on thebiomass of C. diffusa, but the Eurasian species K.laerssenii, and all Eurasian species analyzed collect-ively, significantly reduced C. diffusa biomass. Cen-taurea diffusa had no effect on the amount of 32Pacquired by Eurasian grass species, but significantlyreduced 32P uptake of all North American species.Correspondingly, North American grasses had nocompetitive effects on 32P uptake of C. diffusa, but allEurasian species demonstrated strong negative effectson the amount of 32P acquired by C. diffusa.

With the important implications for the role ofallelopathy in exotic plant invasion, activated car-bon had strikingly different effects on the interactionsbetween C. diffusa and the grass species from the dif-ferent biogeographical regions. When growing with C.diffusa the overall effect of carbon on North Amer-ican species was positive. In contrast, the biomass

32

of all Eurasian grass species growing with C. diffusawas reduced dramatically in the presence of activ-ated carbon. Correspondingly, activated carbon putC. diffusa at a disadvantage against North Americangrasses (Centaurea biomass decreased) but an advant-age when with Eurasian grasses (Centaurea biomassincreased). 32P uptake by Eurasian grasses growingwith C. diffusa decreased in the presence of activatedcarbon. The effects of activated carbon on 32P uptakeby grasses corresponded with the effects of activatedcarbon on 32P uptake by C. diffusa. Activated car-bon enhanced uptake by C. diffusa in the presence ofEurasian grasses but reduced uptake in the presence ofNorth American grasses. Ameliorating effects of ac-tivated carbon in general are evidence for allelopathy(Mahall and Callaway, 1991, 1992; Schreiner andReed, 1907), and the strong effects of the place of ori-gin on the competitive ability of grass species againstC. diffusa, and the contrasting effects of activated car-bon, suggest that C. diffusa produces chemicals thatlong-term and familiar Eurasian neighbors have ad-apted to, but that C. diffusa’s new North Americanneighbors have not. The effects of activated carbon onthe interactions among Georgian grasses and C. dif-fusa suggests that allelopathy may also play some rolein the organization of native communities, but moreimportantly Callaway and Aschehoug’s results suggestthat allelopathy may play a role in successful invasion.

Allelopathy and exotic invasive plants

To our knowledge, no other studies have comparedthe relative allelopathic effects of an invader on spe-cies from ‘origin’ versus ‘recipient’ communities;however chemical allelopathy has long been suspec-ted as a mechanism by which invasive plant specieseliminate natives (Fletcher and Renney, 1963; Kel-sey and Locken, 1987; Stachon and Zimdahl, 1980;Yamamoto, 1995). The observation that non-nativeplants often form dense, virtually monospecific stands,whereas natural monospecific stands of any speciesare rare, has been one of the rationales for proposingmechanisms other than resource competition (Abdul-Wahab and Rice, 1967; Alsaadawi et al., 1990; Con-way et al., 2002; Maddox et al., 1985; Muller, 1969;Nath, 1988; Steenhagen and Zimdahl, 1979; Tinninand Muller, 1971; Vaughn and Berhow, 1999).

The literature linking allelopathy to exotic inva-sion includes some of the best known plant invaders inthe world, including Eltrygia repens (ex. Agropyron,

Korhammer and Haslinger, 1994; Osvald, 1948; Wel-bank, 1960; Weston et al., 1987), Bromus tectorum(Rice, 1964), several Centaurea species (Fletcherand Renney, 1963; Muir and Majak, 1983; Riden-our and Callaway, 2001; Stevens, 1986), Cirsiumarvense (Stachon and Zimdahl, 1980), Cyperus ro-tundus (Agarwal et al., 2002; Komai and Tang, 1989;Komai et al., 1991; Quayyum et al., 2000; Tang etal., 1995), Euphorbia esula (Letourneau and Hegge-ness, 1957; Selleck, 1972; Steenhagen and Zimdhal,1979), Parthenium hysterophorus (Kanchan and Jay-achandra, 1979, 1980; Pandey, 1994), Setaria faberii(Bell and Koeppe, 1972), and Sorghum halepense(Abdul-Wahab and Rice, 1967; Elmore, 1985). Al-though the number of studies suggesting allelopathiceffects of exotic plants is impressive, research on al-lelopathy and exotic invasion is less convincing thanthe argument for allelopathy in general. This maybe due to gaps in research approaches rather thanthe biological effects of allelopathy. Indeed, there isreason to hypothesize that allelopathy may be muchmore important as a mechanism in recipient than inorigin communities. A strong argument against allel-opathy as an important mechanism in natural plantcommunities is that plants appear to evolve toleranceto chemicals rapidly (Williamson, 1990). For allelo-pathy to be effective, the argument goes, plants mustevolve new weapons more rapidly than their neighborsevolve defenses. This is thought to rarely happen, as itappears that even Monsanto cannot keep the defensein front. In invasions, however, members of recipientcommunities may be much more likely to be naïve tothe chemicals possessed by newly arrived species.

The argument for allelopathy in exotic invasionhas been enervated primarily by two issues. First,questionable methodological approaches, particularlythe use of petri dish bioassays, have been over-emphasized as assessments for allelopathy, and war-rant the skepticism many express for allelopathy ingeneral (Harper, 1977; Keeley, 1988; Stowe, 1979).Second, target species used to evaluate the potentialallelopathic effects of invaders have been predomin-antly crop species and other exotic weeds rather thanthe native species that are actually excluded. However,during the last 15 years, apparently in response to sur-ging concern about exotic invasion in natural systems,the allelopathic effects of a number of exotic invasiveplants has been tested on natives.

Zoysia-dominated seminatural grasslands of Ja-pan have been heavily invaded by a pasture plantintroduced from Europe, Anthoxanthum odoratum,

33

prompting tests for an allelopathic mechanism poten-tially involved in the invasion process (Yamamoto,1995). Coumarin solutions, the main allelopathic com-pound identified in this invader, inhibited the devel-opment of Zoysia japonica seedlings growing in petridishes even at very low concentrations, and aqueousextracts of leaves of A. odoratum retarded the rootgrowth of the native. Low concentrations of coumarinincreased seedling growth of the exotic invader, butat high concentrations its growth was also inhibited.Although far superior to extract tests on lettuce andcucumber seeds, and suggesting potential allelopathiceffects of A. odoratum on Z. japonica, such petridishes bioassays could greatly overestimate the effectsof allelochemicals relative to realistic field conditions(Keeley, 1988; Stowe, 1979; Wardle et al., 1998).

Using a similar approach, El-Ghareeb (1991)tested allelopathic impacts of the invasive plant Tribu-lus terrestris on surrounding annual vegetation in anabandoned field of Kuwait. Shoot leachates of T. ter-restris retarded the germination and radicle elongationof the majority of the target species, which includedboth native and exotic annuals, in petri dish bioassays.In addition to this laboratory experiment, several abi-otic parameters such as soil water content, texture, andnutrient concentration were recorded in two adjacentsites having a high (Tribulus site) and low (control)cover of T. terrestris. Both soil moisture and con-centrations of nitrogen, phosphorus, and potassiumwere significantly higher in the Tribulus site. Abund-ant evidence supports the idea that higher resourceavailability increases the susceptibility to invasion ofplant communities (Burke and Grime, 1996; Davis etal., 2000; Maron and Connor, 1996). Hence, althoughthe conclusions of this work are that T. terrestris “isable to exert effective allelopathic interference on cer-tain annuals in the surrounding flora and thus gain anadvantage in dominating some sites in the abandonedfields in the desert of Kuwait”, community dominanceof this invasive plant may also be explained, at least inpart, by a resource-based mechanism.

Allelopathic effects of the alien Kochia scopariaon the native grass Bouteloua gracilis were also eval-uated in petri-dish bioassays (Karachi and Pieper,1987). Water extracts from ground tissues (another ex-perimental flaw) of the invader had no effects on ger-mination, but seedling growth of the native grass wassignificantly suppressed. Interestingly ‘regrowth’ ofKochia leaves was much more inhibitory than tissuescollected before damage. This suggests a link betweendefense responses and allelopathy (see Siemens et al.,

2002). It is unknown whether these harmful effectsalso occur in the field.

Goslee et al. (2001) used a simulation model toassess the importance of allelopathy in the invasionof North American shortgrass steppe by the non-native forb Acroptilon repens (Figure 1). In additionto allelopathy, other components of the model in-cluded competition for soil water, soil texture, andweather conditions. Simulation results showed thatA. repens dominated grassland communities only ifnative species were sensitive to allelopathy from theinvader. Importantly, model outputs closely resembledthe composition of plant communities recorded in thefield when simulations incorporated allelopathic inter-actions, while results from the model differed greatlyfrom field observations when simulations were con-ducted without allelopathy. Other studies of extractsand soil concentrations of the suspected polyacetyleneinhibitor indicated the presence of potentially allelo-pathic chemicals in A. repens (Grant, 2000; Kelsey andBedunah, 1989; Stevens, 1986). Considering our casestudy of C. diffusa, it is interesting that extract tests onnatives were more inhibitory than on the non-nativeBromus inermis (see Goslee et al., 2001, p. 34). Mod-els are a far cry from empirical evidence, but Gosleeet al.’s (2001) work certainly should provide impetusfor field investigations and a more open interpretationof other empirical results.

The genus Acroptilon is closely related to Cen-taurea, and the case in the literature for allelopathiceffects of Centaurea species is strong. In additionto the experiments described above for C. diffusa,Ridenour and Callaway (2001) used activated car-bon to manipulate the effects of root exudates of C.maculosa on root elongation rates and growth of thenative bunchgrass Festuca idahoensis. In root observa-tion chambers, Festuca root elongation rates decreasedto ≈50% of the control, beginning four days beforecontacting C. maculosa roots in silica sand. How-ever, when activated carbon was added to the sandthe effects of C. maculosa roots on F. idahoensisroot elongation were reduced. In other experiments,F. idahoensis plants were 50% smaller when grownwith C. maculosa than with conspecifics in pure silicasand. However, F. idahoensis grown with C. maculosain mixtures of sand and activated carbon were 85%larger than F. idahoensis grown with C. maculosa insilica sand without carbon. These results suggest thatallelopathy accounts for a substantial proportion ofthe total interference of C. maculosa on F. idahoen-sis, shifting the balance of competition in favor of

34

Figure 1. ECOTONE model showing linkages between the soil water and allelopathy modules and the plant recruitment, growth, and mortalitycomponents. Reprinted from Goslee et al. (2001) with permission from Ecological Modelling.

C. maculosa. However, C. maculosa outperformed F.idahoensis even in the presence of activated carbon,suggesting the importance of the combined roles ofresource competition and allelopathy.

Substantial research has been conducted on thespecific chemicals in Centaurea root exudates re-sponsible for its inhibitory effects. Fletcher and Ren-ney (1963) assayed extracts obtained from the leaves,stems, seeds, and roots of C. maculosa, and C. dif-fusa on the germination of barley and lettuce seedsand determined the presence of potential allelopathicsubstances. Muir and Majak (1983) found that seedgermination and seedling growth of Lolium perenne(ryegrass) were inhibited by extracts from C. mac-ulosa leaves. They discovered that the inhibitoryeffects were associated with those fractions of ex-tracts containing sesquiterpene lactones, compoundsexperimentally demonstrated to inhibit IAA-inducedelongation growth of plants (Spring and Hager, 1982).Kelsey and Locken (1987) used column chromato-graphy along with a bioassay of C. maculosa shootextracts with lettuce seeds to isolate cnicin. Pure cni-cin was then assayed at varying concentrations againstthe seeds of lettuce, Centaurea, and several native spe-cies. Cnicin was found to significantly retard seedlinggrowth for all species tested.

Bioassays of suspected volatile phytotoxins or al-lelopathic leachates (extracts) do not, however, elim-inate alternative mechanistic explanations, as they donot demonstrate that these toxic substances either ac-cumulate at sufficient concentrations or persist longenough under normal growing conditions to inhibit thegrowth or development of other plants. Root mediatedallelopathy demonstrated under controlled conditionsmay not occur in complex, natural field systems. Inthe field, root-mediated allelopathy would depend onmany factors such as plant densities, root distributions,root densities, soil characteristics, microclimate, andmicrobial activity. Furthermore, the mobility of com-pounds in natural soil might be less due to buffering orimmobilization.

A major step towards understanding the mechan-isms of the allelopathic effects of C. maculosa infield conditions was achieved by Bais et al. (2002).They isolated two forms, or enantiomers, of catechinexcreted from C. maculosa roots. The more com-mon enantiomer, (+)-catechin, exhibited anti-bacterialfunctions, whereas the less common enantiomer, (−)-catechin, has strong allelopathic effects on a varietyof agricultural species and other exotic weeds. Baiset al. (2002) collected exudates from the roots of C.maculosa that had been exposed to fungal cell wallpreparations and roots that had not, and found that the

35

allelochemic activity of the former isolates were muchstronger, inhibiting the germination of all target plantspecies tested. When catechin was isolated from theseexudates its negative effects were even stronger. Cen-taurea maculosa was resistant to its own exudates andpurified catechin. (±)-catechin had negative effects onother species at concentrations of ≈100 µg mL−1, and(±)-catechin was exuded from C. maculosa roots atdoses as high as 83.2 (non-fungal elicited) and 185.0µg mL−1 (fungal elicited). Importantly, they found(±)-catechin in soil extracts in fields containing C.maculosa in concentrations far higher than the min-imum dose required for inhibition. Depending on thedistance from C. maculosa taproots, (±)-catechin con-centrations ranged from 291.6±17.8 to 389.8±28.6µg cm−2. They concluded that the presence of a nat-urally exuded and stable chemical found in soil withphytotoxic abilities was solid evidence for allelopathy.We agree, but look forward to tests of (±)-catechinagainst the native species C. maculosa eliminates fromnatural communities.

Although negative results of allelopathic tests areundoubtedly under-reported, we found three cases inwhich experiments designed to test for allelopathyfailed to support suspected allelopathic effects. Leafwater extracts of Bunias orientalis, an aggressive in-vader of native plant communities in Central Europe(Steinlein et al., 1996), exhibited considerable inhibi-tion of seed germination and radicle growth of severalnative species, commonly found in Bunias-invadedstands, sown on filter paper in petri dishes (Dietz et al.,1996). However, when allelopathy of this invader wasassessed in a number of additional experiments, weakor no effects were detected. For example, the growthof native seedlings was only slightly or not suppressedat all by the invader’s root exudates in a system con-necting native and invader plants growing in pots filledwith sand and through which a nutrient solution wasforced to circulate (recirculating system, Dietz et al.,1996; Stevens and Tang, 1985). In another assay,leaf litter leachates applied to flats filled with eithersand or soil showed an important substrate effect onthe allelopathic influence of B. orientalis on natives.While leachates of the exotic weakly reduced seed-ling growth as compared to leachates of a combinationof plant species when sand was used as a substrate,the growth of native seedlings showed no response or,against expectations, was stimulated in flats filled withsoil. Adsorption of allelopathic compounds by soilparticles (Krogmeier and Bremmer, 1989; Wardle etal., 1998) and/or microbial breakdown of allelochem-

icals (Dalton, 1999; Huang et al., 1999; Kaminsky,1981) may account for the outcome of this experiment.In a third and final experiment, the allelopathic impactof B. orientalis was evaluated on seedlings of nativespecies growing in soil cores collected from a densestand of B. orientalis and an adjacent site without theinvader. Unexpectedly, seedling growth was improvedin soils from the B. orientalis stand as compared to thesite without B. orientalis. Presumably, higher nutrientinputs resulting from the mineralization of abundantplant material, characteristic of dense B. orientalisstands, were responsible for the enhanced seedlinggrowth. Dietz et al. (1996) concluded that factors otherthan allelopathy might be operating in the rapid estab-lishment and persistence of dense stands of this alienspecies. This study clearly highlights the importanceof evaluating whether results from laboratory bioas-says are corroborated by those obtained in experimentsconducted under more natural conditions.

In a second study, Keay et al. (2000) hypothesizedthat the rapid transition from grasslands to woodlandsdominated by Sapium sebiferum in southeastern NorthAmerica may be mediated in part by the chemical in-hibition of native grasses by this exotic tree. However,S. sebiferum aqueous extracts applied to seeds of thenative grass Schizachyrium scoparium sown in cupsfilled with potting soil did not reduce or slow ger-mination, and growth of grass seedlings was increasedwhen watered with the invader’s extracts as comparedto those watered with water.

Results from Keay et al. (2000) were corroboratedby a latter study testing potential allelopathic effectsof S. sebiferum on the native trees Salix nigra andTaxodium distichum (Conway et al., 2002). Aqueousextracts of fresh leaves and litter of S. sebiferum andsoil from a monotypic woodland of this species wereapplied to seeds of the native trees, which were sownin petri dishes. Aqueous extracts did not inhibit ger-mination or seedling length and mass when comparedto seeds receiving distilled water (control). Further-more, germination rates and seedling root length,shoot length, and mass of T. distichum watered withaqueous extracts were the same or greater than con-trol treatments. In addition to seeds from native trees,seeds of S. sebiferum were also used as test material.Similar to results obtain for T. distichum, germinationrates and all seedling measurements of S. sebiferumwere higher for seeds receiving the aqueous extractsthan for the controls. When S. sebiferum leaf, litter,and soil aqueous extracts were subjected to chromato-graphic analysis to examine for the potential presence

36

of allelochemicals, the only chemical detected wasinositol. It is known that inositol has little or no al-lelopathic effects (Rice, 1995), which gives support tooutcomes from germination experiments. Based on theconsistent positive effects of extracts of S. sebiferumon its own germination and seedling growth, Con-way et al. (2002) proposed that invasion success ofthis exotic tree is due to self facilitation rather thansuppression of surrounding vegetation.

Resource availability, allelopathy, and theinvasibility of plant communities

Resource availability has been suggested as being amajor determinant of the amount of secondary meta-bolites in plant tissues (Coley et al., 1985). Plantsgrowing in resource-limited environments or understress often exhibit higher tissue concentrations ofsecondary compounds when compared to those grow-ing under less stressful conditions (Aerts and Chapin,2000; Gershenzon, 1984; Grime, 1977; McKey et al.,1978). Secondary metabolites deter herbivory (Ehr-lich and Raven, 1964; Freeland and Janzen, 1974;Robinson, 1974), and one of the leading mechanismsproposed to explain increased amounts of secondarycompounds in communities with low resource avail-ability is that natural selection favors well defendedindividuals in environments where replacing costs ofnutrients lost to herbivores are too high (Aerts andChapin, 2000; Coley et al., 1985; Herms and Mattson,1992). Allelochemicals are also products of secondarymetabolism (Rice, 1974; Robinson, 1974; Whittakerand Feeny, 1971) and the quantity of these com-pounds also increases under resource-poor conditions(del Moral, 1972; Tang et al., 1995). Accordingly, ithas been proposed that the chemical suppression ofplant’s competitors is also enhanced in environmentswith low resource availability (Tang et al., 1995). Un-like the case for plant defenses against herbivores,evolutionary mechanisms and adaptive advantages forsuch an enhancement of allelopathy in stressful envir-onments remain to be theoretically developed. Non-etheless, the possible intensification of allelopathiceffects in environments with resource deficiencies mayhave important implications for the susceptibility ofplant communities to invasion (i.e., community invas-ibility, Lonsdale, 1999). If allelopathy is an importantmechanism in the plant invasion process and resourcelimitation increases allelopathic effects, increasing inconsequence the invasive potential of a plant spe-

cies, then the invasibility of plant communities shouldincrease as resource availability decreases.

A large number of both descriptive and experi-mental studies, on the other hand, support a diamet-rically different concept. That is, the invasibility ofa plant community increases as resource availabilityincreases – the theory of invasibility and fluctuatingresource availability articulated by Davis et al. (2000).According to this theory the success of an invader de-pends on having access to resources unused by theresident vegetation; hence, any factor either reducingresource uptake by local plants or increasing resourcesupply will promote the invasibility of a plant com-munity. Thus, while studies on allelopathy suggestthat by having greater allelopathic effects, the invas-ive potential of exotic plants increases in resource-limited communities, making these communities morevulnerable to invasion, studies focusing on resourcecompetition argue that plant communities are moresusceptible to invasion when the availability of re-sources increases. Further investigation is required toelucidate these contrasting arguments.

Conclusions

Non-resource mechanisms, such as allelopathy, shouldnot be considered as simple alternative hypotheses toresource competition as the general way plants inter-act. Instead, ecologists should consider the possibil-ity that resource and non-resource mechanisms worksimultaneously, but vary in their relative importancedepending on particular species and the ecologicalcontext in which they are studied. One such contextis exotic plant invasion. The reason why some invas-ive plants are so successful in new environments maybe that they bring novel mechanisms of interactionto the recipient community. Allelopathy is expectedto be an important mechanism in the plant invasionprocess because the lack of co-evolved tolerance ofresident vegetation to new chemicals produced bythe invader could allow these newly arrived speciesto dominate natural plant communities. Additionalwork comparing the relative allelopathic effects ofinvaders on species from ‘origin’ versus ‘recipient’communities should demonstrate the generality of thisphenomenon, making a substantial contribution to theunderstanding of plant invasions. There is a great dealof evidence for strong allelopathic effects of exotic in-vaders on native species, but very little that escapesmethodological quandaries described by Williamson

37

(1990). We suggest that future research efforts shouldevaluate the presumed allelopathy of exotic plants onthe plant species most commonly encountered by theinvader in nature. Furthermore, links between the ef-fects of chemicals found in plant exudates in the laband exudates in natural concentrations in soils, suchas demonstrated by Harsh Bais and Jorge Vivanco atColorado State University provide a model for futureresearch. The hypothesis that allelopathy increasesthe invasive potential of exotic plants in environmentswith low resource availability awaits evaluation, butmay contribute a great deal to general models of in-vasive susceptibility in natural systems. Finally, thepossibility that unique allelopathic effects of someexotic species on naïve, ‘inexperienced’ communit-ies (see Callaway and Aschehoug, 2000) contribute toinvasive success suggests that evolutionary responsesmay occur on the part of invaded communities. In-vestigation of shifts in the tolerance of native speciesto the allelopathic effects of invaders may yield newinsights into the mechanisms that drive the dramatictransformation of the ecology of some species whenthey are introduced to new regions of the globe.

References

Abdul-Wahab A S and Rice E L 1967 Plant inhibition by Johnsongrass and its possible significance in old-field succession. Bull.Torrey Bot. Club 94, 486–497.

Aerts R and Chapin III F S 2000 The mineral nutrition of wild plantsrevisited: a re-evaluation of processes and patterns. Adv. Ecol.Res. 30, 1–67.

Agarwal A R, Gahlot A, Verma R and Rao P B 2002 Effects ofweed extracts on seedling growth of same varieties of wheat. J.Environ. Biol. 23, 19–23.

Agrawal A A 2000 Communication between plants: this time it’sreal. Trends Ecol. Evol. 15, 446.

Alsaadawi I S, Sakeri F A K and Al-Dulaimy S M 1990 Allelopathicinhibition of Cynodon dactylon (L.) Pers. and other plant speciesby Euphorbia prostrata L. J. Chem. Ecol. 16, 2747–2754.

Arimura G-I, Ozawa R, Shimoda T, Nishioka T, Boland W andTakabayaski J 2000 Herbivory-induced volatiles elicit defencegenes in lima bean leaves. Nature 406, 512–515.

Baldwin I T and Schultz J C 1983 Rapid changes in tree leaf chem-istry induced by damage: evidence for communication betweenplants. Science 221, 277–279.

Bais H P, Walker T S, Stermitz F R, Hufbauer R A and VivancoJ M 2002 Enantiomeric-dependent phytotoxic and antimicrobialactivity of (±)-catechin. A rhizosecreted racemic mixture fromspotted knapweed. Plant Physiol. 128, 1173–1179.

Barbour M G, Burk J H, Pitts W D, Gilliam F S and Schwartz MW 1999 Terrestrial Plant Ecology. Benjamin/Cummings, MenloPark, CA.

Bell D T and Koeppe D E 1972 Noncompetitive effects of giantfoxtail on the growth of corn. Agron. J. 64, 321–325.

Burke M J W and Grime J P 1996 An experimental study of plantcommunity invasibility. Ecology 77, 776–790.

Callaway R M 2002 The detection of neighbors by plants. TrendsEcol. Evol. 17, 104–105.

Callaway R M and Aschehoug E T 2000 Invasive plants versus theirnew and old neighbors: a mechanism for exotic invasion. Science290, 521–523.

Coley P D, Bryant J P and Chapin III F S 1985 Resource availabilityand plant antiherbivore defense. Science 210, 895–899.

Conway W C, Smith L M and Bergan J F 2002 Potential allel-opathic interference by the exotic Chinese tallow tree (Sapiumsebiferum). Am. Midl. Nat. 148, 43–53.

Dalton B R 1999 The occurrence and behavior of plant phenolicacids in soil environment and their potential involvement in alle-lochemical interference interactions: methodological limitationsin establishing conclusive proof of allelopathy. In Principles andPractices in Plant Ecology: Allelochemical Interactions. Eds. In-derjit, K M M Dakshini and C L Foy. pp. 57–74. CRC Press,Boca Raton, FL.

Darwin C 1859 On the Origin of Species by Means of Natural Se-lection, or the Preservation of Favoured Races in the Struggle forLife. John Murray, London.

Davis M A, Grime J P and Thompson K 2000 Fluctuating resourcesin plant communities: a general theory of invasibility. J. Ecol. 88,528–534.

De Candolle A P 1832 Physiologie Vegetale. Bechet Jeune, Paris,France.

del Moral R 1972 On the variability of chlorogenic acid concentra-tion. Oecologia 9, 289–300.

Dietz H, Steinlein T, Winterhalter P and Ullmann I 1996 Role ofallelopathy as a possible factor associated with the rising of dom-inance of Bunias orientalis L. (Brassicaceae) in some native plantassemblages. J. Chem. Ecol. 22, 1797–1811.

Ehrlich P R and Raven P H 1964 Butterflies and plants: a study incoevolution. Evolution 18, 586–608.

El-Ghareeb R M 1991 Suppression of annuals by Tribulus terrestrisin an abandoned field in the sandy desert of Kuwait. J. Veg. Sci.2, 147–154.

Elmore C D 1985 Assessment of the allelopathic effects of weedson field crops in the humid midsouth. In The Chemistry of Al-lelopathy. Ed. A C Thompson. pp. 21–32. American ChemicalSociety, Washington.

Elton C S 1958 The Ecology of Invasions by Animals and Plants.Metheun, London.

Fletcher R A and Renney A J 1963 A growth inhibitor found inCentaurea spp. Can. J. Plant Sci. 43, 475–481.

Freeland W J and Janzen D H 1974 Strategies in herbivory by mam-mals: the role of plant secondary compounds. Am. Nat. 108,269–289.

Gersani M, Brown S, O’Brien E E, Maina G M and Abramsky Z2001 Tragedy of the commons as a result of root competition. J.Ecol. 89, 660–669.

Gershenzon J 1984 Changes in the levels of plant secondary meta-bolites under water and nutrient stress. Recent Adv. Phytochem.18, 273–320.

Gillet J B 1962 Pest pressure, an underestimated factor in evolution.In Systematics Association Publication No. 4, Taxonomy andGeography. pp. 37–46.

Goodnight C J 1990 Experimental studies of community evolutionII: the ecological basis of the response to community selection.Evolution 44, 1625–1636.

Goslee S C, Peters D P C and Beck K G 2001 Modeling invasiveweeds in grasslands: the role of allelopathy in Acroptilon repensinvasion. Ecol. Modell. 139, 31–45.

38

Grant D W 2000 Influence of Russian knapweed on four nativegrasses. M.S. Thesis, Colorado State University. Fort Collins,Colorado, US.

Grime J P 1977 Evidence for the existence of three primarystrategies in plants and its relevance to ecological and evolution-ary theory. Am. Nat. 111, 1169–1194.

Gurevitch J, Scheiner S M and G A Fox 2002 The Ecology of Plants.Sinauer Associates, Sunderland, MA.

Harper J L 1977 Population Biology of Plants. Academic Press,London.

Herms D A and Mattson W J 1992 The dilemma of plants: to growor defend. Q. Rev. Biol. 67, 283–335.

Huang P M, Wang M C and Wang M K 1999 Catalytic trans-formation of phenolic compounds in the soil. In Principles andPractices in Plant Ecology: Allelochemical Interactions. Eds. In-derjit, K M M Dakshini and C L Foy. pp. 287–306. CRC Press,Boca Raton, FL.

Kaminsky R 1981 The microbial origin of the allelopathic potentialof Adenostoma fasciculatum H. & A. Ecol. Monogr. 51, 365–382.

Kanchan S D and Jayachandra 1979 Allelopathic effects ofParthenium hysterophorus L. I. Exudation of inhibitors throughroots. Plant Soil 53, 27–35.

Kanchan S D and Jayachandra 1980 Allelopathic effects ofParthenium hysterophorus L. IV. Identification of inhibitors.Plant Soil 55, 67–75.

Karachi M and Pieper R D 1987 Allelopathic effects of kochia onblue grama. J. Range Manage. 40, 380–381.

Karban R, Baldwin I T, Baxter K J, Lane G and Felton G W2000 Communication between plants: induced resistance in wildtobacco plants following clipping of neighboring sagebrush.Oecologia 125, 66–71.

Keay J, Rogers W E, Lankau R and Siemann E 2000 The role ofallelopathy in the invasion of the Chinese tallow tree (Sapiumsebiferum). Tex. J. Sci. 52, 57–64.

Keeley J E 1988 Allelopathy. Ecology 69, 293–294.Keeley J E and Fotheringham C J 1997 Trace gas emissions in

smoke-induced seed germination. Science 276, 1248–1251.Kelsey R G and Bedunah D J 1989 Ecological significance of allel-

opathy for Centaurea species in the northwestern United States.In The Proceedings of the Knapweed Symposium. pp. 10–32.Montana State University, Bozeman, MT.

Kelsey R G and Locken L J 1987 Phytotoxic properties of cni-cin, a sesquiterpene lactone from Centaurea maculosa (spottedknapweed). J. Chem. Ecol. 13, 19–33.

Komai K and Tang C 1989 A chemotype of Cyperus rotundus inHawaii. Phytochemistry 28, 1883–1886.

Komai K, Tang C and Nishimoto R K 1991 Chemotypes of Cyperusrotundus in Pacific rim and basin: distribution and inhibitoryactivities of their essential oils. J. Chem. Ecol. 17, 1–8.

Korhammer S A and Haslinger E 1994 Isolation of a biologicallyactive substance from rhizomes of quackgrass [Elymus repens(L.) Gould]. J. Agric. Food Chem. 42, 2048–2050.

Krogmeier M J and Bremmer J M 1989 Effects of phenolic acids onseed germination and seedling growth in soil. Biol. Fertil. Soils8, 116–122.

Letourneau D and Heggeness H G 1957 Germination and growth in-hibitors in leafy spurge foliage and quackgrass rhizomes. Weeds5, 12–19.

Livingston B E, Britton J C and Reid F R 1905 Studies on the prop-erties of an unproductive soil. Bulletin 28, Bureau of Soils, U. S.Department of Agriculture.

Lonsdale W M 1999 Global patterns of plant invasions and theconcept of invasibility. Ecology 80, 1522–1536.

Mack R N, Simberloff D, Lonsdale W M, Evans H, Clout M andBazzaz F A 2000 Biotic invasions: causes, epidemiology, globalconsequences, and control. Ecol. Appl. 10, 689–710.

Maddox D M, Mayfield A and Poritz N H 1985 Distribution ofyellow starthistle (Centaurea solstitialis) and Russian knapweed(Centaurea repens). Weed Sci. 33, 315–327.

Mahall B E and Callaway R M 1991 Root communication amongdesert shrubs. Proc. Natl. Acad. Sci. USA 88, 874–876.

Mahall B E and Callaway R M 1992 Root communication mech-anisms and intracommunity distributions of two Mojave Desertshrubs. Ecology 73, 2145–2151.

Maron J L and Connor P G 1996 A native nitrogen-fixing shrubfacilitates weed invasion. Oecologia 105, 302–312.

Maron J L and Vilà M 2001 When do herbivores affect plant in-vasion? Evidence from the natural enemies and biotic resistancehypotheses. Oikos 95, 361–373.

McKey D, Waterman P G, Mbi C N, Gartlan J S and Struhsaker T T1978 Phenolic content of vegetation in two African rain forests:ecological implications. Science 202, 61–63.

Muir A D and Majak W 1983 Allelopathic potential of diffuseknapweed (Centaurea diffusa) extracts. Can. J. Plant Sci. 63,989–996.

Muller C H 1966 The role of chemical inhibition (allelopathy) invegetational composition. Bull. Torrey Bot. Club 93, 332–351.

Muller C H 1969 Allelopathy as a factor in ecological process.Vegetatio 18, 348–357.

Nath R 1988 Parthenium hysterophorus L. – a review. Agric. Rev.9, 171–179.

Osvald H 1948 Toxic exudates from the roots of Agropyron repens.J. Ecol. 39, 192–193.

Pandey D K 1994 Inhibition of salvinia (Salvinia molesta Mitchell)by parthenium (Parthenium hysterophorus L.). II. Relative effectof flower, leaf, stem, and root residue on salvinia and paddy. J.Chem. Ecol. 20, 3123–3131.

Pugnaire F I and F Valladares (Eds) 1999 Handbook of FunctionalPlant Ecology. Marcel Dekker, New York.

Quayyum H A, Mallik A U, Leach D M and Gottardo C 2000Growth inhibitory effects of nutgrass (Cyperus rotundus) on rice(Oryza sativa) seedlings. J. Chem. Ecol. 26, 2221–2231.

Rice E L 1964 Inhibition of nitrogen-fixing and nitrifying bacteriaby seed plants. Ecology 45, 824–837.

Rice E L 1974 Allelopathy. Academic Press, New York.Rice E L 1995 Biological Control of Weeds and Plant Diseases:

Advances in Applied Allelopathy. University of Oklahoma Press,Norman, OK.

Ridenour W M and Callaway R M 2001 The relative importance ofallelopathy in interference: the effects of an invasive weed on anative bunchgrass. Oecologia 126, 444–450.

Robinson T 1974 Metabolism and function of alkaloids in plants.Science 184, 430–435.

Sakai A K, Allendorf F W, Holt J S, Lodge D M, Molofsky J, WithK A, Baughman S, Cabin R J, Cohen J E, Ellstrand N C, Mc-Cauley D E, O’Neil P, Parker I M, Thompson J N and Weller SG 2001 The population biology of invasive species. Annu. Rev.Ecol. Syst. 32, 305–332.

Schenk H J, Callaway R M and Mahall B E 1999. Spatial rootsegregation: are plants territorial? Adv. Ecol. Res. 28, 146–180.

Schreiner O and Reed H S 1907 The production of deleteriousexcretions by roots. Bull. Torrey Bot. Club 34, 279–303.

Schreiner O and Reed H S 1908 The toxic action of certain organicplant constituents. Bot. Gaz. 45, 73–102.

Selleck G W 1972 The antibiotic effects of plants in laboratory andfield. Weed Sci. 20, 189–194.

39

Siemens D H, Garner S H, Mitchell-Olds T and Callaway R M 2002Cost of defense in the context of plant competition: Brassicarapa may grow and defend. Ecology 83, 505–517.

Spring O and Hager A 1982 Inhibition of elongation growth bytwo sesquiterpene lactones isolated from Helianthus annuus.Possible molecular mechanism. Planta 156, 433–440.

Stachon W J and Zimdahl R L 1980 Allelopathic activity of Canadathistle (Cirsium arvense) in Colorado. Weed Sci. 28, 83–86.

Steenhagen D A and Zimdahl R L 1979 Allelopathy of leafy spurge(Euphorbia esula). Weed Sci. 27, 1–3.

Steffens J C, Roark J L, Lynn D G and Riopel J L 1983 Host recog-nition in parasitic angiosperms: use of correlation spectroscopyto identify long-range coupling in an haustorial inducer. J. Am.Chem. Soc. 105, 1669–1671.

Steinlein T, Dietz H and Ullmann I 1996 Growth patterns of thealien perennial Bunias orientalis L. (Brassicaceae) underlying itsrising dominance in some native plant assemblages. Vegetatio125, 73–82.

Stevens K L 1986 Allelopathic polyacetylenes from Centaurearepens (Russian knapweed). J. Chem. Ecol. 12, 1205–1211.

Stevens G A and Tang C-S 1985 Inhibition of seedling growth ofcrop species by recirculating root exudates of Bidens pilosa L. J.Chem. Ecol. 11, 1411–1425.

Stowe L G 1979 Allelopathy and its influence on the distribution ofplants in an Illinois old field. J. Ecol. 67, 1065–1085.

Tang C-S, Cai W-F, Kohl K and Nishimoto R K 1995 Plant stressand allelopathy. In Allelopathy: Organisms, Processes, and Ap-plications. Eds. Inderjit, K M M Dakshini and F A Einhellig. pp.142–157. American Chemical Society, Washington.

Tinnin R O and Muller C H 1971 The allelopathic potential of Avenafatua: influence on herb distribution. Bull. Torrey Bot. Club 98,243–250.

Vaughn S F and Berhow M A 1999 Allelochemicals isolated fromtissues of the invasive weed garlic mustard (Alliaria petiolata). J.Chem. Ecol. 25, 2495–2504.

Wardle D A, Nilsson M-C, Gallet C and Zackrisson O 1998 Anecosystem-level perspective of allelopathy. Biol. Rev. 73, 301–319.

Welbank P J 1960 Toxin production from Agropyron repens. InThe Biology of Weeds. Ed. J L Harper. pp. 158–164. Blackwell,Oxford.

Weston L A, Burke B A and Putnam A R 1987 Isolation, character-ization and activity of phytotoxic compounds from quackgrass[Agropyron repens (L.) Beauv.]. J. Chem. Ecol. 13, 403–421.

Whittaker R H and Feeny P P 1971 Allelochemicals: chemicalinteractions between species. Science 171, 757–770.

Williams J R 1954 The biological control of weeds. In Report ofthe Sixth Commonwealth Entomological Congress, London, pp.95–98.

Williamson G B 1990 Allelopathy, Koch’s postulates, and the neckriddle. In Perspectives on Plant Competition. Eds. J B Grace andD Tilman. pp. 143–162. Academic Press, Inc., San Diego, CA.

Wilson D S 1997 Biological communities as functionally organizedunits. Ecology 78, 2018–2024.

Yamamoto Y 1995 Allelopathic potential of Anthoxanthumodoratum for invading Zoysia-grassland in Japan. J. Chem. Ecol.21, 1365–1373.