133

Chapter 14.

Understanding and Controlling

Nonnative Forest Pests in the South

Kerry O. Britton,Donald A. Duerr II,and James H. Miller1

Abstract—Invasive nonnative forest pests aremultiplying and spreading in every forest typein the Southern United States. The costs ofcontrolling these pests have become extremelyhigh, and the damage they cause to ecosystemcomposition, structure, and function continuesto increase. Plants imported for potential releasefor forage, crops, soil reclamation, and ornamentalpurposes are not evaluated for invasiveness.Insect pests and diseases arrive in infested nurserystock, wood products, pallets, and dunnage,in spite of our regulatory system, which has beenoverburdened by the rapid increase in internationaltrade. The biological basis for the invasivenessof nonnative pests and possible means for dealingwith them are discussed.

INTRODUCTION

Nonnative insects, pathogens, and plantscontinue to flow into the United States, asthey have for the past 500 years (Committee

on the Scientific Basis for Predicting the InvasivePotential of Nonindigenous Plants and PlantPests in the United States 2002). With globaltrade comes a mixing of once-separatedorganisms, often with harmful effects on theirnew natural systems and substantial costs formitigation. Invasive nonnative pests pose majorchallenges. We are challenged to (1) detect andminimize entries, (2) detect critical outbreaksand mobilize rapid responses, (3) monitor existinginvasive populations and apply integrated pestmanagement (IPM) programs, and (4) disseminateinformation about the nature of the problem ofinvasive pests and possible means of its solution.Executive Order 13112, issued in 1999, establishedthe National Invasive Species Council, comprisedof the heads of eight Federal Agencies. ThisExecutive order defined an invasive species asa species that is (1) nonnative (or alien) to theecosystem under consideration, and (2) whoseintroduction causes or is likely to cause economicor environmental harm or harm to humanhealth. The council finalized in 2001 a “NationalManagement Plan: Meeting the Invasive SpeciesChallenge,” which is aimed at coordinatingoffensive and defensive efforts among theGovernment Agencies, nongovernmentalorganizations, and the public. New nationalinitiatives in all elements of an IPM approach toinvasive species are planned and specified, withactual regulatory and policy changes anticipated,as appropriations become available. This chapteraddresses the biological and social bases for thecurrent predicament, identifies the most damaginginvasive pests, gives recommendations for theircontrol, and formulates initiatives required forthe defense of our native forests.

1 National Pathologist, U.S. Department of AgricultureForest Service, Forest Health Protection, Arlington, VA 22201;Entomologist, U.S. Department of Agriculture Forest Service,Forest Health Protection, Asheville, NC 28804; and PlantEcologist, U.S. Department of Agriculture Forest Service,Southern Research Station, Auburn, AL 36849, respectively.

Sout

hern

For

est

Scie

nce:

Past

, Pre

sent

, and

Fut

ure

Fore

st H

ealt

h

134

Economic and Ecological EffectsInvasive nonnative pests cost the United States

an estimated $137 billion per year (Pimenteland others 2000). This figure does not includethe costs of species extinctions. Of the 958 listedthreatened and endangered species, 57 percentare at risk primarily because of competition withand predation by invasive nonnatives (Reichardand White 2001). It is difficult or impossible toaccurately and objectively determine the costof species extinctions or of less severe damageto species and habitats. For this reason, naturalresource losses are more difficult to estimatethan agricultural losses. Forest product industries,although they represent only a small part of totalforest value, are easier to evaluate economically.National losses in traditional forest products dueto nonnative invasive insects and pathogens wereestimated at $4.2 billion per year (Pimentel andothers 2000). It has been estimated that 360nonnative insects have become established inAmerican forests (Liebhold and others 1995).

Data specific to southern forests are scarce,especially for invasive nonnative weeds. Althoughno comprehensive figures specific to forestrylosses due to nonnative weeds are available, theState of Florida has compiled some impressivestatistics for invasive nonnative weeds in wetlands.Their control costs for melaleuca (Melaleuca spp.L.) alone are $3 to $6 million per year and forpurple loosestrife (Lythrum salicaria L.) $45million per year. Florida spends $14.5 millionper year to control Hydrilla spp. L.C. Rich., andstill estimates losses in recreation values for justtwo lakes at $10 million per year (Pimentel andothers 2000).

Since European settlement, nonnativeforest pests have changed the composition andfunction of eastern forests in important ways.For example, as early as 1864, American chestnut[Castanea dentata (Marsh.) Borkh.] trees werebeing eliminated from the Southern AppalachianMountains, although the cause was notdiscovered until 1932. Ink disease, caused bythe nonnative pathogen Phytophthora cinnamomiRands, virtually eliminated American chestnutin valleys and coves and gradually was extendingupslope when chestnut blight [Cryphonectriaparasitica (Murrill) Barr] arrived and removedthe remaining trees, which occupied drier ridges(Crandall and others 1945, Hansen 1999). P.cinnamomi continues to impact southern forests,causing littleleaf disease of shortleaf pine (Pinusechinata Mill.), root rot on Fraser fir [Abies fraseri(Pursh) Poir.] Christmas trees, a decline syndrome

in loblolly pine (P. taeda L.), and hundreds of otherhosts. This same fungus killed 79 percent of theflora in the forests of Western Australia (Westeand Marks 1987) and was recently cited as causingan oak (Quercus spp. L.) mortality epicenter inMexico (Tainter and others 1999).

The oak component in Kentucky, Virginia, andNorth Carolina is under attack from the advancingfront of gypsy moth [Lymantria dispar (L.)].The same forests may soon be threatened by a newspecies of Phytophthora now causing sudden oakdeath (Phytophthora ramorum Werres, de Cock &Man in’t Veld) in parts of California. An outbreakof this disease in Oregon is being eradicated, butpathologists are conducting surveys to determinewhether other, undetected outbreaks may exist.Beech bark disease (Neonectria galligena),dogwood anthracnose (Discula destructivaRedlin), and butternut canker (Sirococcusclavigignenti-juglandacearum) have reducedhost populations as they spread through theunderstory. Adelgids [Adelges piceae (Ratzeburg)]attacking balsam fir [Abies balsamea (L.) Mill.]are causing losses of rare and threatened speciesdependent upon the special habitat associatedwith the fir (Alsop and Laughlin 1991). Similarlosses are anticipated in hemlock forest types(Tsuga spp. Carr.) as the hemlock woolly adelgid[Adelges tsugae (Annand)] spreads south.

The threats posed by diseases and insectpests have long been recognized by the forestrycommunity. In contrast, invasive nonnative forestplants are more insidious and have received farless attention from foresters. Although weedscause losses roughly equivalent to those causedby insects and diseases in agricultural systems(Pimentel 1993), the frequent reliance of nonnativeplants on disturbance as an entrée to invasionhas led to the expectation that such invasions,therefore, are less significant in forests. However,this expectation has proven to be false for tworeasons. First, a number of invasive weedsestablish successfully without disturbance.Among them are garlic mustard [Alliariapetiolata (Bieb.) Cavara & Grande], orientalbittersweet (Celastrus orbiculatus Thunb.),and melaleuca. Second, forests are subject tofrequent disturbances of various origins. Invasivenonnative plants often proliferate after harvests,fire, windthrow, or hurricanes, which create gapsof disturbed habitat. The increasing occupationof forests by nonnative plants has also beenlinked to increasing anthropogenic disturbance(Stapanian and others 1998). Such plants inhibitregeneration of native plants and reduce forest

135

growth and yield. Invasive nonnative weedscan alter ecosystems by changing nutrientcycling, geomorphology and physical structureof the site, drainage patterns and water flow,sedimentation rates, and disturbance regimes.They displace native flora by competition,and thus alter wildlife habitat (D’Antonio 2001,Reichard and White 2001).

PathwaysMany invasive forest plants were intentionally

introduced as ornamentals or forage crops (table14.1), often as a result of Government-sponsoredplant introduction programs (Mack and Lonsdale2001). Some of these plants are still being sold asnursery stock. Herbaceous weeds are more likelyto have been introduced as seed contaminants orin soil used as ballast (Reichard and White 2001).

In contrast, most nonnative insects andpathogens were introduced unintentionally ascontaminants on nursery stock (U.S. CongressOffice of Technology Assessment 1993). Thesudden oak death pathogen probably arrivedon infected rhododendron (Rhododendron spp.)nursery stock. Its origin is unknown. TheAmerican strains of this pathogen cause only smallleafspots and twig blight on rhododendron andmany other hosts, but cause lethal cankers on oaksin coastal regions surrounding the San Francisco

Bay (Rizzo and others 2002). Species killed bythe pathogen include coast live oak (Q. agrifoliaNee), tanoak [Lithocarpus densiflorus (Hook.& Arn.) Rehd.], and California black oak (Q.kelloggii Newb.). Nursery sanitation practicesand fungicide applications can sometimes maskinfection, particularly in the case of Phytophthoraspecies, and may allow infected material to passinspection. Sometimes an import host is onlyslightly susceptible to a disease but may harborthe nonnative pathogen, as infected Chinesechestnut (Castanea mollissima Blume) probablyharbored chestnut blight. The associated pathogenis unnoticed on the resistant host, but underparticularly favorable conditions may sporulateand spread to more susceptible native species.Nurseries with overhead irrigation systemsoften provide this ideal environment.

Another common source of nonnative insectsand pathogens has been the trade in woodand wood products (U.S. Congress Office ofTechnology Assessment 1993). In the UnitedStates, 35 percent of all softwood consumed isimported, and up to 70 percent of all internationalcargo arrives supported by solid wood packingmaterial. The recent arrival of the Asianlonghorned beetle [Anoplophora glabripennis(Motschulsky)] in solid wood packing materialhas focused attention on this previously looselyregulated pathway. In addition to establishedpopulations in New York and Chicago, the beetleshave been intercepted in 26 warehouse locationsin 12 other States. Solid wood packing materialis usually constructed of poor-quality wood,often from trees damaged or killed by pests.Bark remnants increase the likelihood of pestassociation, and boards with bark attached canbe hidden in middle layers of products such aswooden spools. One study found 2,500 live insectsin 29 short log bolts used to brace granite blocksin metal containers (Allen 2001).

The particularly invasive nature of manynonnative forest pests first became apparent nearthe close of the 19th century. Over the past 100years, plant pathologists, entomologists, and weedscientists have developed a broadly applicableconcept of IPM. In this chapter, we will describe afew important nonnative forest pathogens, insectpests, and invasive plants, and will discuss theirentry pathways, control strategies, and ecologicaland environmental impacts. We will apply thelessons learned from these examples to developrecommendations for a more proactive IPMapproach to preventing future invasions.

Cha

pter

14.

Non

nati

ve F

ores

t Pe

sts

in t

he S

outh

Table 14.1—Examples of intentionally introducedinvasive nonnative weeds

Common name Scientific name

Melaleuca MelaleucaAustralian pine Pinus nigra ArnoldJapanese climbing fern Lygodium japonicum (Thunb.

Ex Murr.) Sw.Old World climbing fern L. microphyllum (Car.) R. Br.Kudzu Pueraria montana (Lour.) Merr.Mile-a-minute weed Ipomoea cairica (L.) SweetTree-of-heaven Ailanthus altissima (P. Mill.)

SwingleOriental bittersweet Celastrus orbiculata Thunb.Silktree or mimosa Albizia julibrissin Durazz.Chinaberrytree Melia azedarach L.Winged burning bush Euonymus alata (Thunb.) Sieb.Bush honeysuckle Lonicera spp. L.Cogongrass Imperata cylindrica (L.) Beauv.Chinese silvergrass Miscanthus sinensis Anderss.Chinese privet Ligustrum sinense Lour.Tallowtree Triadica sebifera (L.) SmallChinese wisteria Wisteria sinensis (Sims) DC.Japanese honeysuckle Lonicera japonica Thunb.

Sout

hern

For

est

Scie

nce:

Past

, Pre

sent

, and

Fut

ure

Fore

st H

ealt

h

136

INVASIVE NONNATIVE FOREST PATHOGENS

Nonnative pathogens are presumably moredisruptive than native pathogens becausethey have not coevolved with their new host.

Therefore, the host lacks resistance genes, unlesssome generalized response to attack providesadequate protection against the new pest.Chestnut blight, dogwood anthracnose, andDutch elm disease [Ophiostoma ulmi (Buisman)Nannf.] will be used here to provide examplesof such “unnatural” interactions.

Chestnut BlightIn 1904, H.W. Merkel, Chief Forester of the

New York Zoological Society, noticed that chestnuttrees in the Bronx were dying. At first, recentdroughts were suspected as the cause, but latera fungus, now called Cryphonectria parasitica(Murrill) Barr, was discovered killing the barkand cambial layers of American chestnut. Orientalchestnuts (Castanea spp.) were unaffected, andasymptomatic nursery stock is believed to haveprovided the initial inoculum for this epidemic.Despite every effort to quarantine, remove, andburn infected trees and to protect the uninfectedtrees with fungicidal sprays, the fungus spreadwithin 40 years throughout the range of Americanchestnut. Because this is a nonsystemic barkdisease, the roots of chestnut survive and producecoppice, but the sprouts eventually becomediseased. The fungus is a weaker pathogen butcan survive on oak; e.g., live (Q. virginiana Mill.),post (Q. stellata Wangenh.), scarlet (Q. coccineaMunchh.), and white (Q. alba L.), as well asoriental chestnut. Thus there is no hope of thedisease ever dying out for lack of host material(Anagnostakis 1987, Liebhold and others 1995).

Two separate avenues of research have beentaken to reduce the impact of chestnut blight:(1) hypovirulence and (2) resistance breeding.Hypovirulence is a debilitating disease of thefungus, caused by infection by hypoviruses. Inthe 1950s, researchers in Italy noted that cankersappeared to be callusing over and healing dueto hypoviruses. Italian chestnut (C. sativa Miller)recovered and remains a viable crop today. In theUnited States, unfortunately, greater diversityexists in vegetative compatibility (v-c) groups ofthe fungus than is found in Europe. Cryphonectriaparasitica strains in the United States are lesslikely than European strains to fuse mycelium andexchange the virus. Much effort has been directedat getting the virus into the recalcitrant strains.Recently researchers succeeded in gettingsynthetic DNA coding for viruslike ribonucleic

acid particles into the DNA of uninfected strains.It is hoped that the synthetic genes will eventuallyspread through sexual reproduction into allv-c groups, thus promoting the spread ofthe hypovirulence.

Early attempts to incorporate Asian resistancegenes into American chestnut by crossbreedinggave disappointing results. The hybridsoften resembled the Asian species ratherthan the majestic American parent, because ofbackcrossing to the Asian parent. The AmericanChestnut Foundation (ACF’s) has selected third-generation backcrosses, containing 94 percentAmerican chestnut genes and possessing varyinglevels of resistance. Their results indicate thatsome individuals have resistance genes acquiredfrom the American parents as well. The time andcost required to identify resistant progeny couldbe reduced greatly by the use of marker-assistedselection for the resistance trait. The ACF hybridswere developed mainly from three Chinesecultivars. The ACF intention now is to broadentheir breeding program by incorporating moreChinese sources of resistance and outcrossing tolocally adapted American parents (Hebard andothers 2000).

Dogwood AnthracnoseThe cause of dogwood anthracnose is a fungus

named Discula destructiva Redlin. The details ofintroduction and origin are not precisely known,but the lack of genetic diversity in the pathogenpoints to a recent introduction (Daughtrey andothers 1996). The relative resistance of Chinesedogwood (Cornus kousa Hatch) suggests that thefungus has Asian origins. In addition, the diseasewas first detected in North America almostsimultaneously near two port cities, on oppositecoasts, shortly after trade with China wasreopened in 1975. Features of pathogen biology,forest history, and the silvical characteristicsof the tree all help explain the severe damagecaused by this disease.

The fungus produces only asexual spores,but these grow in great numbers in pustules witha slimy matrix, mostly on the underside of theleaf. They are well adapted to spread in splashingrain. The wet period necessary for infections isunusually long (24 to 48 hours), which partiallyexplains why the disease is more severe in themountains, at higher elevations, on north-facingslopes, and near streams and waterfalls wheremoist conditions are common. Wet periods within2 weeks of each other were needed to maintainepidemic development, whereas dry periods of a

137

month or more greatly reduced the infection rate(Britton 1993). These requirements greatly slowedthe spread of the fungus as it reached the southernedge of the Appalachians.

Eastern flowering dogwood (C. florida L.), themain host in southern forests, is a rapid colonizerof gaps, and its population probably expandedgreatly after the demise of chestnut and as aconsequence of logging activity in the early 20th

century. This shade-tolerant species persistedafter gap closure, surviving under as little as 2percent ambient light in the photosyntheticallyactive range (Chellemi and Britton 1992). Treesgrowing in these conditions had few carbohydratereserves and could not withstand the stress ofrepeated defoliation when a susceptible populationand environmental conditions favorable forepidemic disease development coincided.

Since it was first reported in the southernregion in 1986, anthracnose has spread into277 counties (Anderson and others 1994; U.S.Department of Agriculture, Forest Service 1999).The epidemic is now spreading West more thanSouth or East and is generating much concern inMichigan, Indiana, Ohio, and Missouri. Floweringdogwood impact plots in western North Carolina,where the climate is very favorable for thedisease, have incurred 56 percent mortality since1991 (http://fhpr8.srs.fs.fed.us/2001Conditions/index.html). Disease severity today is muchgreater at the epidemic front than behind it, forseveral reasons: (1) the dry weather experiencedrecently in the South has probably reduced thenumber of secondary disease cycles occurring eachyear, (2) the loss of so many dogwoods growingin microsites optimal for fungal developmentreduced the inoculum load for the surviving trees,(3) survivors are growing on sites less favorablefor fungal development, and (4) survivors maypossess some genetic resistance.

No economically feasible control measureshave been found to protect dogwood in forestenvironments. A 10-point program for reducingdisease severity was developed for landscapetrees. The main goal of the program is to improvetree vigor and thus reduce disease impact(Bailey and Brown 1991). The 10 points are:

1. Select healthy trees to plant.

2. Purchase trees from a reputable nursery;do not transplant trees from the wild.

3. Select good planting sites to promote rapidfoliage drying.

4. Use proper planting techniques.

5. Prune and destroy deadwood and leavesyearly, and prune trunk sprouts in the fall.

6. Water weekly in the morning during drought;do not wet foliage.

7. Maintain a 4- to 6-inch deep mulch aroundtrees; do not use dogwood chips as mulch.

8. Fertilize according to soil analysis.

9. Use proper insecticides and fungicideswhere appropriate.

10. Avoid mechanical and chemical injury to trees.

Hybrids of C. florida x C. kousa resistantto anthracnose were developed at RutgersUniversity. Selections from resistant C. floridasurvivors at Mt. Catoctin National Park werepropagated and tested by the University ofTennessee and entered the market in 2002.

Dutch Elm DiseaseThe story of Dutch elm disease [Ophiostoma

ulmi (Buisman) Nannf.] clearly illustrates a weaklink in the defensive cordon of our quarantineregulations. Current U.S. regulations prevententry only of pests that are (1) not present inthe United States; or (2) present, but of limiteddistribution, and subject to an active eradication/control program.

To be effective, inspectors must be able to findand identify new invaders before they enter andbecome established. Unfortunately, the necessarytaxonomic information did not exist in the case ofDutch elm disease. A new invader arrived and wasmistakenly assumed to be the original Dutch elmdisease fungus, which had become widespreadand consequently not subject to regulation.

The new invader was much more aggressivethan the first Dutch elm disease species. Thusthere have been two separate epidemics of thisvascular wilt in North America, Europe, and Asia.The original causal fungus, Ophiostoma ulmi(Buisman) Nannf., was probably of Himalayanorigin and reached the Netherlands by way ofthe Dutch East Indies (Brasier 1990). It wasintroduced from there into North Americain the 1930s.

The second, visually similar species, O. novo-ulmi Brasier, was not discovered in the AmericanMidwest until after it began killing elms inBritain that had survived the original epidemic.The second epidemic was traced to elm logsshipped from North America in the 1960s.

Cha

pter

14.

Non

nati

ve F

ores

t Pe

sts

in t

he S

outh

Sout

hern

For

est

Scie

nce:

Past

, Pre

sent

, and

Fut

ure

Fore

st H

ealt

h

138

In Britain alone, O. novo-ulmi killed 30 millionelm trees. Hundreds of millions of elms (Ulmusspp. L.) in the United States were lost to the newfungus (Brasier 2001). Gene flow between thetwo species has been demonstrated usingmolecular techniques, and this gene flowbrings advantageous O. ulmi genes forheterogeneity of v-c groups (and subsequentprotection from debilitating viruses) into themore pathogenic O. novo-ulmi (Brasier 2001).

All North American elm species, andparticularly the historically significant streettree U. americana L., are susceptible to Dutchelm disease. The spores are carried from treeto tree by Hylurgopinus rufipes (Eichhoff), anative elm bark beetle, in the northern tier of theUnited States and Canada. In the South, Scolytusmultistriatus (Marsham), the smaller Europeanelm bark beetle, is the more common vector. Thebeetles become infested with spores as they feedon dying elms, and when they emerge as adultsthey spread the spores to healthy trees whilefeeding in twig crotches. The fungus spreadswithin the tree by spores transported in the xylem,and by mycelial growth through other tissues.Leaves on infected branches wilt, curl, turn yellow,and die. Sometimes the tree dies within a fewweeks, its vascular tissue plugged with fungalmycelium, tyloses, and gums. This is particularlytrue in cases where the fungus has spread throughroot grafts. In other cases, the tree may die onelimb at a time over a period of a year or more(Haugen 2001). The cost of removal of dead elmsis estimated at $100 million per year (Pimenteland others 2000). Although U. americana was notplanted as widely in the South as in the NorthernUnited States, it is gradually losing its place insouthern landscapes, as well as in native forests.

Control measures for Dutch elm disease aremost successful when adopted communitywide.Rapid sanitation of dead branches and dyingtrees greatly reduces populations of the beetlevectors. Prunings must be destroyed prior tobeetle emergence. Insecticides can also be usedto reduce vector populations. Root grafts betweendiseased and healthy trees should be broken with avibratory plow or a trenching machine. Trenchingshould be done prior to the removal of diseasedtrees to prevent the drawing of inoculum acrossroot grafts from diseased roots to the transpiringhealthy tree (Haugen 2001). Santamour andBentz (1995) list five varieties of Dutch elmdisease-resistant U. americana: (1) Princetonelm, (2) American Liberty, (3) Independence,(4) Valley Forge, and (5) New Harmony. Other

nonnative Ulmus species and some hybridsare also resistant to Dutch elm disease.

Injection or infusion of fungicides is used asa preventive measure only for high-value trees.Since the treatment must be repeated every1 to 3 years, depending on the fungicide used,damage to the tree in creating injection portsis also a significant factor in overall tree health.Stipes and Fraedrich (2001) suggest that injectionsrise in priority relative to other control optionswhen other factors, such as poor sanitationpractices and community objections to insecticidalsprays, contribute to the development of plentifulinoculum. Fungicide injection improves the successof sanitation pruning and has the advantage oflocalizing control chemicals within the tree, asopposed to insecticidal sprays, which are subjectto drift and possible nontarget effects. Again thereare no economically feasible control measuressuitable for use in the forest environment.

INVASIVE NONNATIVE FOREST INSECTS

Nonnative insects have had a profoundeffect on southern forests. Over 70 speciesof nonnative forest insects are currently

established throughout the Southeastern UnitedStates. Because these pests have rapid dispersalrates and high reproductive capacities, it isnecessary to detect new ones quickly and thenapply effective eradication programs based onIPM before they become established and causefurther damage. This portion of the presentchapter will focus on several of the moredestructive nonnative insects which have past,present, or potential future impacts on SouthernU.S. forests.

Gypsy MothThe gypsy moth [Lymantria dispar (L.)] is

one of the most serious pests of hardwood trees inthe Eastern United States. In most years, millionsof acres are defoliated by the gypsy moth (fig.14.1), and the costs of damage and control run intotens of millions of dollars annually. Useful generalinformation about the gypsy moth can be foundin the “Forest Insect & Disease Leaflet 162” forgypsy moth (McManus and others 1989) andin the book “Insects of Eastern Forests” (U.S.Department of Agriculture, Forest Service 1985).

The gypsy moth is native to Europe andwas introduced into the United States in 1869by a French scientist living in Boston. The firstoutbreak occurred in 1889. The gypsy moth hasspread to all or parts of 17 States, mostly in theNortheast and the Great Lakes region, as well

1390

2

4

6

8

10

12

14

Year

Def

olia

tion

(mill

ion

acre

s)

19441949

19541959

19641969

19741979

19841989

19941999

19402003

as to the District of Columbia. In the Southeast,the current advancing front runs eastwestacross northern North Carolina then slantsnorthwest through southwestern Virginia andeastern Kentucky.

The gypsy moth life cycle has four stages:(1) egg, (2) larva, (3) pupa, and (4) adult (mothstage). Only the larvae damage trees and shrubs.Gypsy moth egg masses are most often laid onbranches and trunks of trees, but egg masses maybe found in any sheltered location. Egg massesare buff-colored when first laid, but may bleachout during the winter months. The hatching ofgypsy moth eggs coincides with the budding ofmost hardwood trees, from early spring throughmid-May. Larvae are dispersed naturally bythe wind and artificially on cars and recreationalvehicles, firewood, household goods, and otherpersonal possessions. The larvae feed until earlyJuly before pupating. Adult females do not fly.

Gypsy moth larvae prefer hardwoods, butmay feed on several hundred different speciesof trees and shrubs (for a list of host plants,see http://www.gypsymoth.ento.vt.edu/vagm/index.html). When gypsy moth populations aredense, the larvae feed on almost all vegetation.In the Eastern United States, the gypsy moth’smain ecological effect is on oaks and in oak-dominated hardwood forests.

The effects of defoliation depend primarilyon the amount of foliage removed, the conditionof the tree at the time it is defoliated, the numberof consecutive defoliations, available soil moisture,and the species of the host. If < 50 percent oftheir crown is defoliated, most hardwoods willexperience only a slight reduction in radial growth.If > 50 percent of their crown is defoliated, mosthardwoods will produce a second flush of foliage

by midsummer. Healthy trees can usuallywithstand one or two consecutive defoliationsof > 50 percent. Trees that have been weakenedby previous defoliation or that have been subjectedto other stresses, such as drought, frequently dieafter a single defoliation of > 50 percent.

Natural controls, including introduced insectparasites and predators, fungal and virus diseases,and adverse weather conditions, help control thegypsy moth. A number of tactics have the potentialto minimize damage by gypsy moth and to containgypsy moth populations at levels consideredtolerable. These tactics include monitoringgypsy moth populations, maintaining the healthand vigor of trees, discouraging gypsy mothsurvival, treating with Bacillus thuringiensisvar. kurstaki, disrupting mating with pheromoneflakes containing disparlure, treating with gypsymoth nuclear polyhedrosis virus, treating withdiflubenzuron, and mass trapping. The tacticor combination of tactics used depends on thecondition of the site and of the tree or stand andthe level of the gypsy moth population. Tacticssuggested for homeowners, such as removingegg masses, placing burlap bands around boles,or spraying individually affected trees, areusually too labor intensive for managers touse in forest stands.

The gypsy moth infestation spreads at anaverage rate of 21 km/year along its border tothe west and south. In 1999 following a successfulpilot project initiated in 1992, the U.S. Departmentof Agriculture Forest Service (Forest Service),along with State and Federal cooperators,implemented the National Gypsy Moth Slowthe Spread (STS) project across the 1,200-milegypsy moth frontier from North Carolina through

Figure 14.1—Amountof defoliation by theEuropean gypsymoth 1940–2003.

Cha

pter

14.

Non

nati

ve F

ores

t Pe

sts

in t

he S

outh

Sout

hern

For

est

Scie

nce:

Past

, Pre

sent

, and

Fut

ure

Fore

st H

ealt

h

140

northern Michigan. The goal of the STS project isto use novel IPM strategies to reduce the rate ofgypsy moth spread into uninfested areas. The STSproject significantly decreases the new territoryinvaded by the gypsy moth each year and protectsforests, forest-based industries, urban and ruralparks, and private property. Estimated spreadrates declined from 20 to 40 km/year to 5 to 14 km/year after STS control and eradication methodswere employed in an STS project in the centralAppalachians. The average rate of gypsy mothspread was 26.5 km/year before 1990 and 8.6 km/year after 1990 (Sharov and Liebhold 1998). Moreinformation on the spread of gypsy moth and theSTS project may be found on the STS Web site:http://www.gmsts.org/operations.

Although gypsy moth has been present inthe United States for > 100 years, it is difficultto explain and predict the extent of the changesit causes in forest vegetation. A major concernis the potential loss of economically significantand ecologically dominant oak species. Moststudies of forest compositional changes aftergypsy moth defoliation indicate that lesssusceptible species will dominate the forest.

Hemlock Woolly AdelgidThe hemlock woolly adelgid [Adelges tsugae

(Annand)] has been in the United States since1924 (McClure 1994). This serious pest ofeastern hemlock [T. canadensis (L.) Carriere]and Carolina hemlock (T. caroliniana Engelmann)is a native of Asia. Through 2001, hemlock woollyadelgid infestations have been found in > 150counties in Connecticut, Delaware, Massachusetts,Maryland, New Jersey, New York, North Carolina,Pennsylvania, Rhode Island, Virginia, and WestVirginia. In 2001 alone, 20 additional counties werefound to have infestations. At present, hemlockwoolly adelgid cannot be controlled in the vastmajority of forest settings.

Hemlock woolly adelgid is a sucking insectwith an extremely complicated life cycle. Fourforms each complete six life stages, some ofwhich develop wings and migrate to feed onspruce. Successful reproduction on spruce hasnot been observed in North America (Salom1996b). The forms most damaging to hemlock arewingless and remain on hemlock all year round.

White cottony sacks at the base of the needlesare good evidence of hemlock woolly adelgidinfestation. These sacks resemble the tips ofcotton swabs. They are present throughout theyear, but are most prominent in early spring.

When immature nymphs and adults suck sapfrom their twigs, trees lose vigor and drop needlesprematurely. If uncontrolled, the adelgid can killa tree in a single year. The widespread hemlockmortality that the hemlock woolly adelgid causesis alarming, in view of the importance of hemlocktrees to the ecosystems in which they occur.

Application of insecticides is currentlyrecommended for controlling hemlock woollyadelgid in areas where this is feasible (Salom1996b). Infested trees are drenched with botanicaloils and or soaps, or systemic insecticide(imidacloprid) is injected into the trees andor the soil beneath them.

Several native predators feed on the hemlockwoolly adelgid, but none of them reduces adelgidpopulations enough to help the current situation.Two nonnative predators, Pseudoscymnus tsugaeSasaji and McClure (a ladybird beetle native toJapan) and Laricobius nigrinus (Fender) (a beetlenative to the Pacific Northwest), hold promisefor biological control of hemlock woolly adelgidinfestations. Under certain circumstances, releasesof these predators are a feasible and effectivecontrol option (McClure and others 2001).

Balsam Woolly AdelgidIntroduced from Central Europe around

1900, the balsam woolly adelgid [Adelges piceae(Ratzeburg)] is considered a serious pest of forest,seed production, landscape, and Christmas trees(Salom 1996a). First discovered in Brunswick, ME,in 1908, the balsam woolly adelgid was found inthe Southern Appalachian Mountains in the 1950sand has spread to all fir stands in the region sincethat time. The pest has also found its way into thePacific Northwest. In the Eastern United States,the adelgid feeds on balsam fir and Fraser fir.Very extensive stands of Fraser and balsam firhave been killed throughout much of these species’range in the East. Because the adelgid does notattack Fraser fir until the trees approach maturity,and because some mature trees escape attack longenough to produce seeds, young Fraser fir treesstill exist in their natural range. However, by themid-1980s, this insect had significantly altered allof the mature Fraser fir-red spruce (Picea rubensSarg.) forest type in the Southern Appalachians.

The balsam woolly adelgid life stages includethe egg, three nymphal stages, and female adults.There are no males; females reproduce byparthenogenesis. They are wingless, oval,purplish-black insects about 0.8 mm in length,and are covered with secretions of waxy threads

141

that appear as a dense white wool mass. A femaleis capable of laying 200 eggs or more in a clusternear her body. The first-instar crawlers, reddishbrown and about 0.4 mm in length, are the onlystage of the insect capable of moving anddispersing. Once the crawler finds a suitablefeeding location, it inserts its tubelike mouthparts into the bark of the host and remains therefor the rest of its life. The second and third instarsare about 0.5 to 0.65 mm in length, respectively,and closely resemble the adult.

The balsam woolly adelgid generallyconcentrates either on the outer portions of treecrowns or on the main stem and large branches.Crown infestations are characterized by abnormaldrooping of the current shoots and gouting ofthe outer twigs. The crown becomes increasinglythin, and dieback may occur. Persistent crowninfestation can kill a tree over a number of years.Stem infestations usually cause greater damageand mortality. Conspicuous white woolly massescharacteristic of stem attack can give the lowerbole a whitewashed appearance in the most severecases. The tree responds to feeding by adelgidsin an allergic manner that causes swelling ofthe sapwood, gouting of the twigs, and increasedheartwood formation in the sapwood—a conditioncalled rotholz or redwood. This abnormalgrowth of sapwood tissue inhibits water flowwithin the tree.

In forest situations, silvicultural andmanagement techniques can be used to reducebalsam woolly adelgid populations and damage(Salom 1996a). Tree stress may be minimizedby thinning overstocked stands, by fertilizingnutrient-poor sites, and by replanting orencouraging more tolerant trees and varieties.There are many different varieties and crossesof Fraser fir, and some varieties are more tolerantof balsam woolly adelgid. A hazard-rating systemwas developed to aid in management decisions.The main variables used in the system are siteelevation, soil moisture regime, percent balsamfir by basal area, total basal area of balsam fir,and stand age. In general, lower elevation dry siteswith > 40 percent balsam fir at an older age (45years of age or more) are most susceptible. Treesbetween 25 and 45 years of age are moderatelysusceptible, and trees < 25 years old are leastsusceptible. In Christmas tree plantations in whichonly a few trees are infested, it should sufficeto rogue and burn those trees. Chemical controlcan be used effectively on ornamental trees, seedproduction trees, and Christmas trees (Day andothers 2001). Several insecticides are available for

use in spraying infested bark and foliage. Whenfeasible, the cutting and removal of infested treesis effective. Cut trees must be wrapped in tarps toensure that adelgids do not fall off the trees asthey are being removed.

Beech Scale and Beech Bark DiseaseBeech bark disease (Neonectria galligena)

is one of the more recent problems to plagueEastern U.S. forests. Beech bark disease refersto a complex consisting of a sap-feeding scaleinsect and at least two species of Nectria fungi(McCullough and others 2001). Beech scale(Cryptococcus fagisuga Lind. = C. fagi Baer.)was accidentally introduced into Nova Scotiain 1890 on ornamental beech trees from Europe.The scale and associated fungi have spread sincethat time, and the current range in the UnitedStates includes most of New England, northernPennsylvania, and northeastern West Virginia.Localized infestations of beech scale have beendiscovered in Virginia, North Carolina, Tennessee,and Ohio (McCullough and others 2001). Theoverall effect of this insect-disease complexis the mortality of roughly 50 percent of thebeech (Fagus spp.) trees > 8 inches in diameter(Houston and O’Brien 1983). The resulting foresthas a few residual large beech trees and standsof many small trees, often root sprouts fromsusceptible trees, which are frequently defective.

Beech scale insects are yellow, soft bodied,and 0.5 to 1.0 mm long as adults. They feedon American beech (F. grandifolia Ehrh.) andEuropean beech (F. sylvatica L.). Adult scales arelegless and wingless and have only rudimentaryantennae. Reproduction is parthenogenic. Thistype of reproduction allows for rapid populationgrowth. Beech scale has one generation per year.Immature scales, called crawlers, have functionalantennae and are mobile. Crawlers are spread bywind, birds, and people moving infested wood.When a crawler finds a suitable feeding locationon a host tree, it inserts its long, tubelike styletinto the bark and begins to suck sap. It then moltsto the second crawler stage, which has no legsand is immobile. These produce a white wax thateventually covers their bodies. Thus when treesare heavily infested with beech scale, they appearto be covered by white wool. The small woundsproduced by the beech scale’s feeding allow theNectria fungi to invade the infested trees(Houston 1994).

Crawlers that fall from trees or are washedoff by precipitation usually die. Severely coldweather (-35 °F) that persists for a few days

Cha

pter

14.

Non

nati

ve F

ores

t Pe

sts

in t

he S

outh

Sout

hern

For

est

Scie

nce:

Past

, Pre

sent

, and

Fut

ure

Fore

st H

ealt

h

142

may kill beech scale, but such weather conditionsprobably never occur in the Southeast. A smallladybird beetle [Chilocorus stigma (Say)] feedson this scale and is common throughout mostof the Eastern United States, but this predatordoes not reduce scale populations enough tocontrol infestations.

Although the scale feeding alone weakens trees,mortality usually does not occur until the treeshave been invaded by Nectria fungi. This invasiontypically occurs after 3 to 6 years of scale feeding.Most large-diameter beech trees in areas wherebeech bark disease becomes established are killed.Beech is a very important source of food andhabitat for many wildlife species and areas withlarge beech components may change dramaticallyas a result of beech bark disease. Some treesare partially resistant to beech bark disease,and a very few are completely resistant. Treeswith smoother bark appear to be more resistant,probably because the scales prefer to feed wherebark is rough (Houston 1997).

The only control is removal of the trees mostheavily infested with beech scale or Nectria fungi.Resistant trees should be identified and retained.After it is cut, beech often regenerates by prolificroot sprouting. This is undesirable because thesprouts form dense thickets, have little value towildlife, and eventually increase susceptibility tomore beech bark disease infestations. Herbicidecontrol of beech root sprouts is, therefore, oftennecessary. Increasing the diversity of forest standsin which beech is present will reduce the risks andspread rate of the disease. Care should also betaken to avoid transporting infested firewoodor logs to uninfested areas.

Asian Longhorned BeetleThe Asian longhorned beetle [Anoplophora

glabripennis (Motschulsky)] was discovered inNew York City in 1996 and in Chicago in 1998.Tunneling by the beetle larvae girdles treestems and branches, impeding water and nutrienttransport within the attacked tree. Repeatedattacks lead to dieback of the tree crown and,eventually, death of the tree (U.S. Departmentof Agriculture, Forest Service and Animal andPlant Health Inspection Service 1999). The Asianlonghorned beetle probably traveled to the UnitedStates inside solid wood packing material fromChina. The beetle has been intercepted at portsand found in warehouses throughout the UnitedStates, although New York City and Chicagoremain the only two areas where infestations of

live trees have been found. Since 1996, > 7,000trees in the two cities have been killed by thebeetle, or cut down and destroyed to stop thebeetle’s spread. Most of the trees lost were highlyvalued urban trees that provided shade, wildlifehabitat, aesthetic value, and benefits for cleanwater and air. The Asian longhorned beetle hashad an economic impact in the tens of millionsof dollars.

The Asian longhorned beetle is also a seriouspest in China where it kills hardwood trees. Inthe United States, the beetle prefers maple species(Acer spp.), including boxelder (A. negundo L.),Norway (A. platanoides L.), red (A. rubrum L.),silver (A. saccharinum L.), sugar (A. saccharumMarsh.), and sycamore (A. pseudoplatanus L.)maples. A complete list of host trees in the UnitedStates has not been determined. An updated listis available at http://www.na.fs.fed.us/spfo/alb/index.htm. Because not all hosts are known andbecause the beetle has been restricted to urbanforests thus far, it is difficult to predict its potentialeffects on natural forests. It appears, however, thatAsian longhorned beetle may have the potential toirrevocably alter many eastern forest ecosystems.

There is usually one generation of Asianlonghorned beetle per year, although the lifecycle may take as long as 2 years. Adult beetlesare usually present from May to October, but theycan be found earlier in spring or later in fall iftemperatures are warm. Adults typically stay onthe trees from which they emerge, but they maydisperse short distances to a new host to feed andreproduce. Adult females chew oval to round egg-laying sites in the bark of the tree and place asingle egg in each. Each female is capable of laying30 to 70 eggs. These hatch in 10 to 15 days, andthe larvae tunnel under the bark and deep intothe wood where they eventually pupate. Emergingadults create a perfectly round exit hole three-eighths inch in diameter. Adult beetles are 1 to 1.4inches long and have striking white marks againsta jet black body. The antennae are longer thanthe body and have black and white bands.

Currently the only effective means to eliminateAsian longhorned beetle is to remove infestedtrees and destroy them by chipping or burning.To prevent further spread of the insect,quarantines have been established to avoid thetransportation of infested trees, branches, andwood from the area. Early detection of infestationsand rapid treatment response are crucial tosuccessful eradication of the beetle. Early

143

Lag phase

Exponentialspread phase

Maximumoccupation

detection is difficult, time consuming, and costly,and to be effective, it must involve tree climbersand surveyors in bucket trucks. Since 2000,unattacked potential host trees have been injectedwith the systemic insecticide imidacloprid as apreventive treatment. Researchers are assessingthe biological control potential of a variety of thebeetles’ natural enemies in Asia.

INVASIVE NONNATIVE FOREST PLANTS

M illions of acres of forest land in the Southeastare being increasingly occupied by nonnativeinvasive plants, which are also termed exotic

weeds. Their range, infestations, and damage arecontinually expanding. All Federal parks andforest lands in the Southeast have nonnativeinfestations (Hamel and Shade 1985, Hester 1991).The actual infested acreage, spread rates, anddamage estimates are still unknown, although thisinformation is essential for planning containmentand eradication strategies and programs (U.S.Congress Office of Technology Assessment 1993).The Forest Service and State partners haveinitiated a cooperative survey of 42 invasivenonnative plants within the region and another20 species in Florida; however, it will takeseveral years to collect initial data (for a list,see “Nonnative Invasive Plants of SouthernForests” at http://www.srs.fs.usda.gov/fia/manual/Nonnative_Invasive_Plants_of_Southern_Forests.pdf).

Invasive plants are able to outcompete nativespecies. They reproduce rapidly because of theabsence of predators from their native ecosystems,and eventually form dense infestations thatexclude most other plants, except certain othernonnatives (Randall and Marinelli 1996). Otherreasons for their invasiveness are that they arenaturally robust plants or have been made sothrough plant breeding efforts; that most areperennials with tough roots or rhizomes; thatmany are still being sold as ornamentals and someare widely planted for wildlife use and soilstabilization; that most produce abundant seedsor spores that are spread widely by birds,wind, and water; and that their seeds or tuberspersist in the soil (Randall and Marinelli 1996).It remains unclear what percentage of nonnativeplants arriving in the Southeastern United Statesbecome invasive. One problem in determining thisis the nature of invasive plant spread, which canbe characterized by a short-to-lengthy lag phasepreceding an exponential spread phase (fig. 14.2).In many species, e.g., kudzu, tallowtree, wisterias,etc., the lag phase can be very protracted and can

mask eventual problems. This spread function alsoexplains why eradication is most possible duringthe early lag phase.

Occupation and infestations by nonnativepest plants decrease forest productivity,threaten forest health and sustainability, andlimit biodiversity and wildlife habitat in theSoutheast (Wear and Greis 2002). Alterationsto ecosystem structure, functions, and processesare occurring, but study of these effects hasjust begun (Ehrenfeld and others 2001). Someinvasives, such as cogongrass [Imperatacylindrica (L.) Beauv.], can alter natural fireregimes and increase risk of wildfire occurrenceand damage (Lippincott 2000). Nonnative plant“biological pollution” is one of the greatest threatsto biodiversity across the southern landscape,attacking our highly valued nature preserves andrecreational lands. Adjoining croplands, homesites, pastures, and wetlands contain invasive plantspecies that will eventually affect forests. Thesenonnative invaders (often called nonindigenous,alien, or noxious weeds) include trees, shrubs,vines, grasses, and forbs. In all there are about70 infestation-forming, terrestrial plant speciesinvading forests and their edges in the temperateparts of the Southeast. Thirty of these arediscussed briefly here to provide a generalsense of identifying characteristics, commonpathways of introduction, mechanisms ofinvasiveness, ecosystem effects, and rangeof current occupation. Not discussed here arethe approximately 70 tropical and subtropicalnonnative species currently invadingsouth Florida.

Figure14.2—Logistic spread model for invasive nonnative plants.

Cha

pter

14.

Non

nati

ve F

ores

t Pe

sts

in t

he S

outh

Sout

hern

For

est

Scie

nce:

Past

, Pre

sent

, and

Fut

ure

Fore

st H

ealt

h

144

Invasive Nonnative TreesNonnative tree species hinder management

of forests, rights-of-way, and natural areasby replacing native plants. This dramaticallyalters habitat and may alter important naturalprocesses. Almost all of the invasive nonnativetrees are hardwoods. Some presently occur asscattered trees, while others form dense stands.Most spread widely by prolific seed productionand animal dispersal, while existing infestationsincrease by abundant root sprouting.

Tree-of-heaven or ailanthus [Ailanthusaltissima (P. Mill.) Swingle] was introduced toNorth America as an ornamental in 1784 fromEurope, although it originates in Eastern China(Miller 1990). A short-lived species with no timbervalue, ailanthus grows up to 80 feet tall withlong, pinnately compound leaves, slightly fissuredgray bark, and large terminal clusters of greenishflowers in early summer. Flowers and otherparts of the plant have a strong odor. It is adioecious species and spreads by seeds fromfemale trees. It is shade intolerant, floodintolerant, and allelopathic. Ailanthus establishesafter disturbance and increases by root sprouts,often forming dense thickets that displace nativevegetation. It occurs throughout the Southeastand is most abundant in Kentucky, Virginia,and Tennessee.

Silktree or mimosa (Albizia julibrissin Durazz.)was introduced as an ornamental from Asia in1745. It is a leguminous tree, 30 to 50 feet tall.It has feathery, pinnately compound, deciduousleaves, smooth light brown bark, and showy pinkspring and summer blossoms, yielding abundantdangling seedpods that persist into winter. Theseedpods float, which aids in long-distance spreadalong waterways, and seeds remain viable formany years. Infestations are spreading alongrights-of-way, fencerows, and riparian zones,and are encroaching into adjoining forested areasafter disturbance, especially into pine plantations.Partially shade tolerant, mimosa invades theforest midstory and replaces native shrubs byroot sprouting. It is becoming increasinglycommon along roadsides throughout the Southeastand is most abundant in Mississippi, Alabama,and Georgia.

Princesstree or paulownia [Paulowniatomentosa (Thunb.) Sieb. & Zucc. ex Steud.]was introduced from East Asia in the early 1800s.It is grown as an ornamental and in scatteredplantations for speculative production of high-valued wood for export to Japan. It has large

heart-shaped leaves with fuzzy hairs on both sides,and in early spring produces showy pale violetflowers that yield clusters of pecan-shapedcapsules, each filled with thousands of tiny wingedseeds. Paulownia reproduces by abundant seedsand root sprouts, replacing native vegetation,including young trees that might otherwise reachthe overstory. It is shade intolerant and invadesafter disturbance. This deciduous tree growsto 60 feet tall. Because it sprouts rapidly, it oftenobscures scenic vistas along roadsides. It occursthroughout the Southeast and is presently mostabundant in central Tennessee and Virginia.

Chinaberrytree (Melia azedarach L.) is anotherAsian introduction. This traditional ornamentalis commonly found around old home sites. Itgrows to about 50 feet tall and is spread by birds,which disperse its seeds. It has lacy, bipinnatelycompound dark green leaves and produces paleblue flowers in spring. The flowers yield roundyellow fruit that persist during winter. Infestationsspread by means of abundant seeding and rootsprouting along rights-of-way to adjoining landthat has been disturbed. Because it is somewhatshade tolerant, it is increasing in the midstoryof pine plantations in parts of the South. The fruitare poisonous to humans and livestock but havepotential use as natural pesticides. Chinaberryis common throughout the Southeast and is mostabundant in Arkansas, Louisiana, Mississippi,Alabama, and Georgia.

Tallowtree or popcorn tree [Triadica sebifera(L.) Small, formerly Sapium sebiferum (L.) Roxb.]is a shade-tolerant tree that grows to 50 feet tall.It has light green heart-shaped leaves that turnscarlet in the fall, long drooping flowers in spring,and bundles of white, waxy, popcornlike seeds thatremain attached to the tree in fall and winter. Theabundant seeds are spread by birds and on water.Tallowtree is a prolific root sprouter and formsmonospecific stands (Bruce and others 1997).It was introduced from China to the U.S. gulf coastin the early 1900s, and the U.S. Department ofAgriculture encouraged its use as a seed oil cropfrom 1920 to 1940. Tallowtree is still being soldand planted and is thought to be the most rapidlyinvading tree species in the region. Tallowtreeseedlings are shade tolerant and yet grow rapidlyin full sun (Jones and McLeod 1990). Its waxyseeds were traditionally used to make candles,and it has current value as a honey plant forbeekeeping and limited pulpwood use. It formsdense stands, and because it tolerates flooding,tallowtree replaces bottomland hardwoodreproduction and understory plants in wetland

145

forests throughout the Coastal Plain (Jones andSharitz 1989). It is also spreading into uplandforests from widespread ornamental plantings.It occurs in all the Southern States exceptOklahoma, Kentucky, and Virginia, and thereare severe infestations in coastal areas of Texas,Louisiana, Mississippi, and Alabama.

Invasive Nonnative ShrubsInvasive nonnative shrubs often occur with

invasive tree species and present similar problems.Herbicide control options are similar to those fortrees, but foliar sprays are often more effectivelyused against shrubs than against trees. All of themost common invasive shrubs are abundant seedproducers, and their fruits are often consumedand spread by birds.

Chinese privet (Ligustrum sinense Lour.) andEuropean privet (L. vulgare L.) are shade-toleranttall shrubs or small trees growing to about 30 feetin height. These common southern ornamentalshrubs were introduced from China and Europein the early to mid-1800s and have already becomesome of the most severely invasive species. Theyform dense stands in the understory of bottomlandhardwood forests and exclude most native plantsand replacement reproduction. These privetsare also increasing in upland forests, fencerows,rights-of-way, and special habitats throughout theregion. They drastically alter habitat and criticalwetland processes. Both species have leafystems with opposite leaves < 1 inch long. Chineseprivet is semievergreen, and European privet isdeciduous, but the two species are nearly identicalin all other respects. Both have showy clusters ofsmall white flowers in spring that yield droopingclusters of small, spherical, dark purple berriesduring fall and winter. Birds spread seed veryeffectively, but privet stands also increase indensity by stem and root sprouts. Both speciesoccur throughout the Southeast.

Multiflora rose (Rosa multiflora Thunb. exMurr.) is an erect-to-arching shrubby rose growingto about 10 feet tall and taller when it climbs intotrees. The recurved thorny stems have pinnatelycompound leaves with 3 to 7 leaflets. White roseflowers are produced in many clusters in spring,and red rose hips, which are spread by birds,appear in fall to winter. Sprouts and runnersthat root consolidate and expand infestations.The species was introduced from Japan and Koreain the 1860s as an ornamental. Later, Governmentprograms encouraged its planting for use as livingfences for livestock containment and as wildlifehabitat. Infestations have been confined to

pastures but are now extending into forest edgesand interior forests, including wetlands. Thespecies occurs throughout the Southern andEastern United States.

Bush honeysuckles—Amur honeysuckle[Lonicera maackii (Rupr.) Herder], Morrow’shoneysuckle (L. morrowii Gray), tatarianhoneysuckle (L. tatarica L.), and sweet breathof spring (L. fragrantissima Lindl. and Paxton)—are generally deciduous multistemmed shrubs 6 to16 feet tall with arching branches. The leaves aredistinctly opposite, usually oval to oblong in shape,and range in length from 1 to 3 inches. Fragrant,tubular flowers occur in pairs from May to Juneand are creamy white in most species, but turnyellow or pink to crimson in varieties of tatarianhoneysuckle. Red-to-orange berries in pairs areabundant on plants in fall to winter, and seedsare long lived in the soil. All were introducedfrom Asia in the 1700s and 1800s as ornamentalsand wildlife plants. They are widely invading andforming exclusive understory layers in lowlandand upland forests, replacing most native plantsand preventing regeneration of native trees. Mostalarming is the increased occupation of wetlands.These invasive species occur everywhere in theSoutheast except Louisiana and Florida and aremost abundant in Kentucky and Virginia.

Autumn olive (Elaeagnus umbellata Thunb.)is a deciduous, bushy shrub growing to 20 feettall. It has alternate leaves that are dark greenabove and silvery beneath. It produces abundantspherical red berries with silvery scales in the fall.Introduced from China and Japan, and still widelyplanted for wildlife habitat, reclamation of stripmines, and shelterbelts, autumn olive is beingspread rapidly and widely by birds and otheranimals. It is becoming a scattered understoryshrub in open forests throughout the Southeast,to the detriment of native trees and shrubs.

Silverthorn or thorny olive (E. pungens Thunb.)is a popular ornamental evergreen bushy shrubwith long limber shoots projecting to 20 feet whensupported by tree limbs. It has alternate leaves,which in spring are silver and scaly on bothtop and bottom and which by midsummer havebecome dark green above and silvery beneath.Thorns are widely scattered on its branches andare subtended by brown-scaled red fruit thatappear in spring. The fruit are consumed andwidely dispersed by wildlife, which results inscattered infestations. This widely plantedornamental shrub was introduced from Chinaand Japan. A shade-tolerant species, it replaces

Cha

pter

14.

Non

nati

ve F

ores

t Pe

sts

in t

he S

outh

Sout

hern

For

est

Scie

nce:

Past

, Pre

sent

, and

Fut

ure

Fore

st H

ealt

h

146

native understory vegetation and prevents naturaltree regeneration. It occurs in all SoutheasternStates except Texas, Oklahoma, and Arkansas.

Winged burning bush [Euonymus alata(Thunb.) Sieb.] is a shade-tolerant, deciduous,bushy shrub up to 12 feet tall with opposite leavesalong stems with four corky wings. Introducedfrom Northeast Asia in the 1860s, it is still widelyplanted as an ornamental. In fall, the leaves turnbright red, while orange fruit appear as stemmedpairs in leaf axils. Birds and animals are attractedto the fruit and spread seed widely. E. alata isincreasingly invading forests, pastures, andprairies. It forms dense stands that excludenative plants and eventually stop native treeregeneration. This problem is spreading inOklahoma, Kentucky, Tennessee, Virginia,Georgia, North Carolina, and South Carolina.

Invasive Nonnative VinesNonnative vines are among the most

troublesome invaders because they often formthe densest infestations, making control effortsdifficult, especially the application of herbicide.Many of these vines overtop even mature forestsand often form mixed infestations with nonnativetrees and shrubs.

Japanese honeysuckle (L. japonica Thunb.),the most prevalent invasive nonnative vine, is ashade-tolerant, climbing and trailing woody vinewith semievergreen, opposite leaves. Paired whiteto yellow flowers in early summer yield blackishberries in fall and winter. Introduced from Japanin 1806, it is the most widespread and invasivenonnative plant species. It occupies multiple stratain lowland and upland forests, replaces nativevines, and alters habitat and ecosystem processes.Japanese honeysuckle is sold as an ornamentaland has some value for erosion control. It is alsoplanted and cultured in wildlife food plots andsustains deer herds during winter. It occursthroughout the Southeast and is spread bywidely rambling vines that root at nodes,as well as by bird-dispersed fruits.

Kudzu [Pueraria montana (Lour.) Merr.,formerly P. lobata (Willd.) Ohwi] is a woodyleguminous vine with lobed trifoliate leaves.It is spread by vines rooting at nodes and byanimal- and water-dispersed seeds. Introducedas an ornamental from Japan in 1876, kudzu wasplanted extensively for erosion control and foragein Government-sponsored programs from 1920to 1950. It forms dense infestations that exclude

native plants, halting forest productivity andchanging habitat on millions of acres of land.Kudzu is increasingly invading riparian habitatalong rivers and streams by means of floatingseedpods. Hydrologic impacts from this modeof spread are anticipated. Kudzu has become apopular southern icon and provides some rawmaterial for folk art. The Forest Service hasinitiated a biocontrol program for kudzu(Britton and others 2002).

Oriental or Asian bittersweet (Celastrusorbiculatus Thunb.) is an attractive but veryinvasive vine with elliptic to rounded deciduousleaves 2 to 3 inches broad and long, alternatingalong a woody vine with drooping branches.Clusters of scarlet fruit appear in fall and remainduring winter at most leaf axils. The fruits arewidely spread by birds. Oriental bittersweet wasintroduced from Asia in 1736. The showy berriesare used as home decorations in winter, and thesedecorations contribute to spread when discarded.Oriental bittersweet colonizes disturbed forestsand along forest edges, spreading into interiorforests, forming expanding thickets, anddecreasing plant diversity. It is invading fromthe Northeast and is not yet found in Oklahoma,Texas, Louisiana, or Mississippi. Americanbittersweet (C. scandens L.) has flowers andfruit only in terminal clusters and does notform extensive infestations.

Air yam (Dioscorea bulbifera L.) and Chineseyam (D. oppositifolia L., formerly D. batatasDcne.) are twining and sprawling vines with heart-shaped leaves and small dangling, yamlike tubers(bulbils) at leaf axils in mid-to-late summer. Thesetubers drop and form new plants. Although thevines are deciduous, they grow rapidly and cancover small trees in one growing season. NativeDioscorea species do not produce “air potatoes,”nor do they form infestations that cover trees.Chinese yam is from Asia, and air yam is fromAfrica. Both were introduced as possible foodsources in the 1800s, but are now cultured forornamental or medicinal use and are often spreadby unsuspecting gardeners. Once established,these vines colonize persistently because theprolific bulbils form new plants as they scatterdownslope. The vines expand throughout theunderstory to form exclusive infestations. Theirdistribution is scattered throughout the Southeast,with air yam occurring mostly in the southern GulfCoastal Plain and Chinese yam more common inthe Appalachians.

147

Wintercreeper or climbing euonymus[Euonymus fortunei (Turcz.) Hand.-Maz.] is atrailing, climbing, or shrubby evergreen plantwith opposite, thick, dark green or green-whitevariegated leaves. It is shade tolerant, spreads toform a dense ground cover, and climbs by aerialroots. Abundant reddish-hulled orange fruitappear in fall and are widely spread by birds.Introduced from Asia as an ornamental groundcover and still widely planted, E. fortuneicontinues to form dense exclusive infestationsthat decrease diversity, hinder access, and alterhabitat. It occurs in Kentucky, Virginia, Tennessee,Mississippi, Alabama, Georgia, North Carolina,and South Carolina.

Japanese climbing fern [Lygodium japonicum(Thunb. Ex Murr.) Sw.] is a viney deciduous fernwith lacy, finely divided leaves and green-to-orange-to-black wiry stems that climb and twineover shrubs and trees. Native to Asia and tropicalAustralia, it was introduced to North Americafrom Japan as an ornamental and is often spreadby unsuspecting gardeners. It is one of threespecies of climbing ferns in the Southeast.The American climbing fern [L. palmatum(Bernh.) Sw.] and Old World climbing fern [L.microphyllum (Cav.) R. Br.], another nonnativewhich grows in Florida, have once-divided leaves.All are perennial plants that grow from creepingrhizomes and are spread by wind-dispersedspores. Dispersal of spores from nonnative speciesresults in rapid spread and widely scattered denseinfestations that cover native herbs, shrubs, andeventually trees. L. japonicum is invading fromthe South to the North and has yet to arrive inOklahoma, Tennessee, or Kentucky.

Chinese wisteria [Wisteria sinensis (Sims) DC.]and Japanese wisteria [W. floribunda (Willd.) DC.]are woody, leguminous vines with long pinnatelycompound leaves and showy spring flowers.They spread by adventitious rooting and are lesscommonly dispersed by seeds. These traditionalsouthern porch vines were introduced from Asiain the early 1800s. They usually spread slowly,although more rapidly near rivers and streams.They form dense infestations mainly around oldhome sites, often in mixtures with other nonnativeplants. Both hinder reforestation and commonlyoccur as scattered patches throughout theSoutheast. The native or naturalized Americanwisteria [W. frutescens (L.) Poir.] does notform dense infestations.

Nonnative Invasive GrassesNonnative grasses spread along highway rights-

of-way and then into adjoining forest lands. Mostnonnative grasses are highly flammable andincrease fire intensity. Intense fires tend to killplants with which the grasses occur and thusfacilitate the spread of the grasses after wildfireor prescribed burns. Wildland firefighters andforest home sites are subjected to increased riskswhere nonnative grasses form heavy infestations.Repeated applications of herbicides are requiredfor control.

Cogongrass [Imperata cylindrica (L.)Beauv.] is a dense, erect perennial grass. Itswide yellowish green leaves have off-centermidveins and finely sawtoothed margins. It wasintroduced from Southeast Asia in the early 1900s,first accidentally and then intentionally for soilstabilization and use as forage. It has been ratedas the world’s seventh worst weed (Holm andothers 1979). It spreads by windblown seeds inearly summer and by rhizome movement infill dirt along highways, often yielding circularinfestations. This grass is highly flammable.It is mostly shade tolerant. Dense infestationsincreasingly occupy forest openings, open forests,and rights-of-way in the Southern Gulf CoastStates and eventually exclude most native plants.Forest regeneration is hampered and habitatdestroyed. This process is hastened by burning(Lippincott 2000). Cogongrass is spreadingnorthward from the Gulf Coast States and hadnot reached North Carolina, Tennessee, Virginia,Kentucky, Arkansas, or Oklahoma as of 2001.

Nepalese browntop [Microstegium vimineum(Trin.) A. Camus] is an annual grass. Stems arefrom 1 to 3 feet long with alternate, lanceolateleaves to 4 inches long. It forms dense mats andconsolidates occupation and spreads by prolificseed production in late summer. Seed remainviable for 1 to 5 years. This shade-tolerantweed is native to temperate and tropical Asiaand was first collected near Knoxville, TN, in1919. It increasingly occupies creek banks, floodplains, forest roadsides and trails, damp fields,and swamps. It spreads into adjoining forests,where it forms exclusive infestations and displacesmost, if not all, native understory plants. It occursthroughout the Southeast except in Oklahoma.

Chinese silvergrass (Miscanthus sinensisAnderss.) is a densely clumped perennial grasswith upright to arching long, slender leaves withwhitish upper midveins. It can grow to a height of5 to 10 feet. Silvery to pinkish loose plumes appear

Cha

pter

14.

Non

nati

ve F

ores

t Pe

sts

in t

he S

outh

Sout

hern

For

est

Scie

nce:

Past

, Pre

sent

, and

Fut

ure

Fore

st H

ealt

h

148

in fall. Viability of the seed is unpredictable.Native to Eastern Asia, M. sinensis has beenplanted in all States for landscaping, recentlyusing sterile cultivars. It is spreading fromolder fertile plants in all States except Arkansas,Oklahoma, and Texas. Still widely sold and plantedas an ornamental, it is highly flammable. It formsdense infestations along rights-of-way and indisturbed upland forests, excluding nativevegetation and altering habitat.

Invasive Nonnative Forbs and SubshrubsForbs are broadleaf herbaceous plants,

while subshrubs are semiwoody. They areusually treated with foliar herbicide spraysor pulled by hand.

Garlic mustard [Alliaria petiolata (Bieb.)Cavara & Grande] is an aptly named biennialherb; all parts of the plant have a garlic odor. Itgrows in small-to-extensive colonies under forestcanopies. In the first year, the plant appears as abasal rosette of leaves that remain green duringwinter. In the second year, stems emerge andgrow, becoming 2 to 4 feet tall. Leaves are broadlyarrow-point shaped with wavy margins. Theflowers form in terminal clusters, and each flowerhas four white petals. Introduced originally as amedicinal herb from Europe in the 1800s, garlicmustard is displacing native forest understoryplants and drastically altering habitat. This speciesproduces prolific seed that can lie dormant in thesoil for 2 to 6 years, building large seed banks.Germination occurs only in spring under favorableconditions. A biocontrol program has been startedat Cornell University (Blossey and others 2001).Garlic mustard is invading from the Northeast andhas yet to arrive in Florida, Alabama, Mississippi,Louisiana, or Texas.

Shrubby lespedeza (Lespedeza bicolorTurcz.) and Chinese lespedeza [L. cuneata (Dum.-Cours.) G. Don] were both introduced from Japan.Shrubby lespedeza is a shade-tolerant bushylegume that grows up to 10 feet tall. It has threeleaflets and produces small purple-pink peatypeflowers with white centers. Chinese lespedeza is asemiwoody plant up to 3 feet tall with many small,three-leaflet leaves feathered along erect, whitishstems. It forms tiny cream-colored flowers duringsummer. Both species produce abundant single-seeded legumes, but dispersal mechanisms arepoorly understood. They have been plantedextensively for wildlife food and soil stabilization.They are still planted for quail food, and plantsoften invade surrounding forests, replacingnative plants throughout the Southeast.

Invasive Plant ControlThe most effective and efficient strategy

for control is early detection and effective earlytreatment of initial invaders. Any successful effortto combat and contain invasive nonnative plantsrequires an integrated vegetation managementapproach (Miller 2003, Tennessee Exotic PestPlant Council 1996). Integrated programsincorporate all effective control methods, whichmay include (1) preventive measures, i.e., legalcontrols such as quarantines, border inspections,and embargoes; (2) biocontrol by means of naturalpredators and diseases; (3) herbicide technology;(4) prescribed fire; (5) livestock overgrazing; and(6) mechanical and manual removal. Preventivemeasures and biocontrol programs are bestorganized on a regional basis. Biocontrol agentsare largely unavailable now, and although projectsto identify such agents are underway, it will takeyears to develop them (Simberloff and Stilling1996). Only through careful and precise researchand development can effective biocontrol agentsthat minimize impacts on nontarget organismsbe identified.

Current treatment options for specific areasusually involve herbicides, prescribed fire, grazing,and mechanical or manual removal. Fire, grazing,and mechanical cutting treatments usually controlonly the aboveground plant parts, reducing theirheight but suppressing the plants only temporarily.Manual treatment usually involves grubbing orpulling plants. This is very labor intensive andis practical only where plants and infestationsare small. Thus manual treatment has limitedbut effective application in special habitats,such as recreational trails or nature preserves,and as a rapid means of first-sight elimination.Mowers, chain saws, and brush cutters removeaboveground plant parts, while leaving roots andrhizomes. Tree shears, root rakes, and harrowscan cut and dislodge woody and rhizomatousplants, but leave soil bare for probable reinvasionand possible erosion if it is not rapidly stabilizedwith native plants. Nonetheless, these soil-disturbing techniques can start reclamationprograms when multispecies infestations ofinvasive woody plants are encountered.

Herbicide treatments often can be moreeasily and effectively applied following theseother treatments. Herbicide treatments alsominimize soil disturbance and leave the soilseed bank in place to reestablish native plants.Carefully planned and executed herbicideapplications can specifically target nonnativeplants and minimize impacts to native plants

149

(Miller 2003) (http://www.invasive.org/weeds.cfmand http://www.srs.fs.usda.gov/pubs/gtr/gtr_srs062). Well-developed applicator-directedtechniques for selective control of nonnativetrees and shrubs are tree injection and girdletreatments, basal sprays and wipes, cut-stemapplications, and foliar-directed sprays (U.S.Department of Agriculture, Forest Service 1994).Directed treatments of nonnative vines and forbsusually involve foliar sprays applied with backpacksprayers. For treating extensive inaccessibleinfestations, broadcast applications of sprays andpellets using helicopter and tractor-mountedsystems may be required. Yet even in broadcasttreatments, the use of carefully timed selectiveherbicides can safeguard native plants. If thetreatment is to be safe and effective, herbicideapplicators must read, understand, and follow theherbicide label and its prohibitions before andduring use. Continued surveillance and followuptreatments are often required to control nonnativeplant infestations.

Site Rehabilitation after NonnativePlant Control

The rehabilitation phase is the most essentialfinal part of an eradication and reclamationprogram. Fast-growing native plants that willoutcompete any surviving nonnative plants mustbe planted or released. Native plant seeds andseedlings are becoming increasingly available(http://plant-materials.nrcs.usda.gov/). If thesoil seed bank remains intact, native plantcommunities may naturally reclaim many areasafter nonnative plants are controlled. Constantsurveillance, treatment of new unwantedarrivals, and rehabilitation of current infestationsare the necessary steps in managing nonnativeplant invasions.

CONCLUSIONS

We have learned much that can help uscontrol invasive nonnatives in the future.An important point is that the cost of

controlling nonnative invasives increases greatlythe longer control measures are deferred. Thissuggests that the best approach might be to findways to improve our ability to prevent invasions orto control invasions before they become crises.

PreventionThe entry and spread of invasive organisms

could be stopped by effective legal and policybarriers. Such barriers could range from Federal,State, and county laws that prohibit importationto sanitation of equipment and vehicles beforethey leave infested zones.

It is helpful to examine opportunities toprevent intentional and unintentional introductionsseparately. Most invasive nonnative plants havebeen imported intentionally, in ignorance of theirpotential invasiveness. Yet, plant exploration andinternational seed exchange continues. Presentregulations only examine incoming plant materialfor the presence of insect pests and pathogens orcontamination with listed noxious weed seed. Asystem to test invasiveness of plant introductionswas developed in Australia in the 1990s and hasbeen helpful in addressing the problem (Mackand others 2000). Several such systems havebeen proposed (Reichard 2001).

Prevention of spread also requires examiningthe Internet sales of nonnative plants and animals.This remote means of mail order shipments ofnonnative organisms will only increase the globalproblem. Retail sales within the United Statesof even federally listed noxious weeds likeI. cylindrica persist with unproven sterilityof cultivars being sold. Only a rapid phasing outof the sale of known invasive nonnative plants willhalt the spread through commercial networks.

Unintentional introductions require a differentapproach. Inspection processes developed foragricultural products have inherent weaknessesin preventing the importation of forest pests.International trade agreements specify thatimport regulations will only address pests knownto be present on the commodity in the exportingcountry, and for which a risk assessment has beenperformed. Provisional regulations can be adoptedwhen insufficient data about the pest exist, butthe risk assessment process must be initiated.The mitigation measures must be those thatprotect our resources with the minimumdisruption of trade. Crop plants are similar theworld over, and it is generally known which pestspose problems. When pests of natural ecosystemsare considered, the major difficulty is in knowingwhich ones might prove invasive.

Biological and ecological characteristics of thepests themselves may render them particularlyeffective as nonnative invasives. Among thesehigh-risk characteristics are a cryptic nature,which helps them avoid early detection, andextended diapause or dormant periods, whichhelp them survive transit and quarantine. Othercharacteristics can also increase the probabilityof pest establishment. Asexual reproduction,for example, reduces the minimum populationsize needed to establish the pest in a new land.The presence of related hosts, usually at least

Cha

pter

14.

Non

nati

ve F

ores

t Pe

sts

in t

he S

outh

Sout

hern

For

est

Scie

nce:

Past

, Pre

sent

, and

Fut

ure

Fore

st H

ealt

h

150

in the same genus as the original host, increasesthe risk that a pest will be successful. Importationin association with host material, such as nurserystock or seed, makes establishment much morelikely. Additional factors suggested by Pimenteland others (2000) as contributing to pestinvasiveness include a lack of natural enemies,an ability to switch to a new host, an abilityto be an effective predator in the new ecosystem,the availability of suitable habitats, and highadaptability to novel conditions.

Unfortunately, the supposition that we willknow or should know in advance which peststo study, assess as risks, and quarantine hasnot been borne out by historical experience withany introduced forest pest. Information aboutthe biology and distribution of known pests couldpossibly be shared more effectively acrossinternational borders. However, only a smallpercentage of the insects and microbes thatinhabit forest ecosystems have even beendescribed to date (Campbell 2001). A differentapproach may be needed to regulate importationof articles likely to contain forest pests.

The present policy of the United States isthat imported articles are “innocent until provenguilty.” This has also been called the dirty listapproach; it requires study of particular articlesto prove that they pose an unacceptable risk.In contrast, the inverse policy of “when in doubt,keep it out,” or clean list approach, requires studyof particular articles to prove they are safe,prior to importation. This is a more conservativeapproach, but for all the reasons given above,it may be more appropriate to introductionpathways for forest pests. Studies to developenvironmentally friendly and economically feasiblestandard treatments for major import pathwaysmight prove a better investment than continuingto develop regulations on a country-by-countryand pest-by-pest basis.

Detection and MonitoringDetecting early entry is a main defense against

unintentionally introduced harmful organisms.Improved detection technology is needed to reducerisk, as the sheer volume of international tradehas overwhelmed the present regulatory system.Advances in molecular technology, such as real-time microarrays, which can test for the presenceof up to 30,000 organisms in 15 minutes, need tobe adapted for implementation on a broad scale.The expense of installing such systems at all portsof entry may seem exorbitant today because this

technology is new. But as this technology becomesmore widely used, its application to this criticalinterface may become economically feasible.Again, such technology is only effective againstknown pests. Monitoring is the basis for effectivecontrol and containment programs, both fortargeting efforts where the organisms arelocated and for judging the effectiveness ofcontrol measures.

Control, Containment, or ManagementEarly detection can make it possible to

eradicate invasive pests in specific circumstances.If eradication efforts prove ineffective, the nextcontrol efforts should be an attempt to providecontainment; i.e., to stop the spread. Containmentefforts can protect adjoining forests, counties, andStates. At present, individual landowners mustdefend their properties through their own controlactivities. Sometimes interagency cooperationcould be useful. An example of this is theinteragency weed team concept U.S. Departmentof Agriculture, Animal and Plant HealthInspection Service developed to promote prompteradication across land ownerships. Controlmethods include cultural methods, pesticideapplications, sanitation, physical and mechanicalcontrol, and biological control. When invasiveorganisms cannot be completely controlled oreradicated, then cost:benefit or similar analysesare used to choose which infestations should bemanaged to minimize ecological degradation,human hazards, and economic loss.

RestorationUnless affected forest ecosystems can be

made more resistant, they will probably bereinvaded. It may be impossible to restore anaffected ecosystem to its prior condition becauseof the residual influence of the pest infestationand because the ecosystem lacks resiliency. Atpresent, it appears feasible only to establishplant components that are resistant to nonnativeinvasive organisms and leave it to naturalprocesses, such as plant succession, to completethe process.

ResearchThe current situation with nonnative invasive

organisms shows clearly that inadequate researchhas been applied and applied too late. The recentdiscovery that interspecific hybridization canoccur when nonnative pathogens or nonnativeand native pathogens meet (Spiers and Hopcroft1994), highlights the urgency of further research.

151

Sometimes such interactions can result in newhost ranges (Brasier and others 1999, Newcomband others 2000) or increased aggressiveness(Brasier 2001). Only through research andtechnology development for each of the keyelements of IPM and successful implementationof proven strategies may current invasions behalted and future invasions be prevented. Becauseour resources are limited, and the supply ofinvasive pests is virtually unlimited, landscape-level analyses should be used to learn whichecosystems are most at risk and to prioritizecontrol efforts. Also, methods for screening plantintroductions must be developed (Committee onthe Scientific Basis for Predicting the InvasivePotential of Nonindigenous Plants and PlantPests in the United States 2002).

Education and ExtensionInformed individuals are needed to combat

the invasive nonnative problem. Much of theproblem from invasive organisms is perpetuatedand exacerbated by an unaware and poorlyinformed populace. Our Federal Government wasdesigned to react slowly to broad swells of concernraised by the constituency to the attention of itsleaders. Managers can only react when theyperceive the threat and have the resources, andthe citizen consumer will stop spreading nonnativeorganisms when they are made aware of thedangers. Public education programs might bemore successful if we inform the traveling public,in advance of their foreign travel, of the threat toour natural resources from smuggling forbiddenproducts. Once they have made their purchasesand packed them away in their suitcases, theoption to ignore this issue is much more tempting.

Similarly, a proactive “plant natives”program (http://plant-materials.nrcs.usda.gov)might be easier to promote than the negativemessage “Don’t buy nonnative pest plants.”Beneficial characteristics of native plants, such asbetter adaptation to local climate, less irrigationrequirements, and the joys of restoring naturalecosystems in your own backyard should bestressed in homeowner education programs.In fact, many Government land managementagencies could set a good example by makingimprovements in their own landscape designsin this regard. The problem of fighting invasivenonnative pests seems overwhelming, but thewar must be won one battle at a time.

ACKNOWLEDGMENTS

The authors gratefully acknowledge theeditorial reviews provided by JohnnyRandall, Steve Oak, and Bill McDonald.

We also appreciate the assistance of ErwinChambliss and Sherry Trickel, who helpedprepare the manuscript.

RELEVANT WEB SITESAsian Longhorned Beetle

h ttp://www.na.fs.fed.us/spfo/alb/index.htm

http://www.aphis.usda.gov/lpa/issues/alb/alb.htm

http://www.uvm.edu/albeetle/

Balsam Woolly Adelgidhttp://fhpr8.srs.fs.fed.us/idotis/insects/bwa.html

http://www.ext.vt.edu/departments/entomology/factsheets/balwoade.html

http://fhpr8.srs.fs.fed.us/hosf/bwa.htm

Beech Bark Diseasehttp://www.na.fs.fed.us/spfo/pubs/fidls/beechbark/fidl-beech.htm

http://www.invasive.org/symposium/houston.html

Chestnut Blightwww.ppws.vt.edu/griffin/accf.html

http://www.apsnet.org/online/feature/chestnut

http://www.forestpests.org/southern/Diseases/chsntblt.htm

http://www.forestpests.org/southern/

http://www.srs.fs.fed.us/pubs/rpc/1999-03/rpc_99mar_33.htm

Dogwood Anthracnosehttp://www.na.fs.fed.us/spfo/pubs/howtos/ht_dogwd/ht_dog.htm

http://fhpr8.srs.fs.fed.us/pubs.html

Dutch Elm Diseasehttp://www.na.fs.fed.us/spfo/pubs/howtos/ht_ded/ht_ded.htm

http://www.fs.fed.us/na/morgantown/fhp/palerts/ded/elm.htm

http://www.ext.nodak.edu/extnews/askext/treeshr/1423.htm

Cha

pter

14.

Non

nati

ve F

ores

t Pe

sts

in t

he S

outh

Sout

hern

For

est

Scie

nce:

Past

, Pre

sent

, and

Fut

ure

Fore

st H

ealt

h

152

Nonnative Plantshttp://www.se-eppc.org/

http://www.aphis.usda.gov/ppq/weeds/(Federal Noxious Weed Program)

http://www.nrcs.usda.gov/technical/invasive.html(Natural Resources Conservation Service’s Websites related to invasive plants)

http://tncweeds.ucdavis.edu/handbook.html(The Nature Conservancy’s Weed MethodsControl Handbook)

General Nonnative Forest Species Informationhttp://www.pestalert.org/

http://spfnic.fs.fed.us/exfor/

http://www.forestryimages.org/ (for foresthealth images)

http://www.ceris.purdue.edu/napis/ (NationalAgricultural Pest Information System Web site)

http://www.invasivespecies.gov/

http://www.invasive.org (photos of invasivenonnative species)

General Web Sitehttp://www.issg.org/database/welcome/(Global Invasive Species Database)

Gypsy Mothhttp://na.fs.fed.us/wv/gmdigest/

http://www.gmsts.org/operations (Slow-the-SpreadWeb site)

http://www.gypsymoth.ento.vt.edu/vagm/index.html

http://www.fs.fed.us/ne/morgantown/4557/gmoth/

Hemlock Woolly Adelgidhttp://www.fs.fed.us/na/morgantown/fhp/hwa/hwasite.html

http://www.ento.vt.edu/~sharov/hwa/

Sources of Native Plants for Reclamationhttp://www.plant-materials.nrcs.usda.gov/

LITERATURE CITEDAllen, E. 2001. Solid wood packing material as a pathway

for nonindigenous species. In: Risks of exotic forestpests and their impact on trade: an online workshop[CD-ROM]. [Place of publication unknown]: APS Press.www.exoticpests.apsnet.org. [Date accessed unknown].

Alsop, F.J.; Langhlin, T.F. 1991. Changes in the spruce-firavifauna of Mt. Guyot, TN, 1967–1985. Journal of theTennessee Academy of Science. 66: 207–209.

Anagnostakis, S.L. 1987. Chestnut blight: the classicalproblem of an introduced pathogen. Mycologia. 79: 23–37.

Anderson, R.L.; Knighten, J.L.; Windham, M. [andothers].1994. Dogwood anthracnose and its spreadin the South. For. Serv. Prot. Rep. 26. [Place ofpublication unknown]: [Publisher unknown].[Number of pages unknown].

Bailey, K.R.; Brown, E.A., II. 1991. Growing andmaintaining healthy dogwoods. For. Rep. R8–FR14.[Place of publication unknown]: [Publisher unknown].[Number of pages unknown].

Blossey, B.; Nuzzo, V.; Hinz, H.; Gerber, E. 2001. Developingbiological control of Alliaria petiolata (M. Bieb.) Cavaraand Grande (garlic mustard). Natural Areas Journal.21: 357–368.

Brasier, C.M. 1990. China and the origins of Dutch elmdisease: an appraisal. Plant Pathology. 39: 5–16.

Brasier, C.M. 2001. Rapid evolution of introduced plantpathogens via interspecific hybridization. BioScience.51: 123–133.

Brasier, C.M.; Cooke, D.E.L.; Duncan, J.M. 1999. Originof a new Phytophthora pathogen through interspecifichybridization. Proceedings of the National Academyof Sciences, U.S.A. [Place of publication unknown]:[Publisher unknown]. 96: 5878–5883.

Britton, K.O. 1993. Anthracnose infection of dogwood seedlingsexposed to natural inoculum in western North Carolina.Plant Disease. 77: 34–37.

Britton, K.O.; Orr, D.; Sun, J. 2002. Kudzu. In: Van Driesche,R.V.; Blossey, B.; Hoddle, M.; [and others], eds. Biologicalcontrol of invasive plants in the Eastern United States.FHTET-2—2—04. [Place of publication unknown]: U.S.Department of Agriculture, Forest Service: 325–330.

Bruce, K.A.; Cameron, G.N.; Harcombe, P.A.; Jubinsky, G. 1997.Introduction, impact on native habitats, and management ofa woody invader, the Chinese tallow tree, Sapium sebiferum(L.) Robx. Natural Areas Journal. 17: 225–260.

Campbell, F.T. 2001. The science of risk assessment forphytosanitary regulations and the impact of changingtrade regulations. BioScience. 51: 148–153.

Chellemi, D.O.; Britton, K.O. 1992. Influence of canopymicroclimate on incidence and severity of dogwoodanthracnose. Canadian Journal of Botany: 1093–1096.

Committee on the Scientific Basis for Predicting the InvasivePotential of Nonindigenous Plants and Plant Pests in theUnited States. 2002. Predicting invasions of nonindigenousplants and plant pests. Washington, DC: National ResearchCouncil. 194 p.

153

Crandall, B.S.; Gravatt, G.F.; Ryan, M.M. 1945. Root diseaseof Castanea species and some coniferous and broadleafnursery stocks, caused by Phytophthora cinnamomi.Phytopathology. 35: 162–180.

D’Antonio, D.M. 2001. Ecosystem impacts of invasiveexotic plants in wildlands—characterizing the culprits.In: Proceedings of the U.S. Department of Agricultureinteragency research forum on gypsy moth and otherinvasive species. Annapolis, MD: [Publisher unknown]:41–45.

Daughtrey, M.L.; Hibben, C.R.; Britton, K.O. [and others].1996. Dogwood anthracnose: understanding a diseasenew to North America. Plant Disease. 80: 349–358.

Day, E.R.; Salom, S.M.; Schultz, P.B.; Tigner, T.C. 2001. Forestinsects. In: 2001 pest management guide: horticultural andforest crops volume. Blacksburg, VA: Virginia PolytechnicInstitute and State University, Virginia CooperativeExtension: 174–177.

Ehrenfeld, J.G.; Kourtev, P.; Huang, W. 2001. Changes in soilfunctions following invasions of exotic understory plants indeciduous forests. Ecological Applications. 11: 1287–1300.

Hamel, D.R.; Shade, C.I. 1985. Weeds, trees, and herbicides:a public forest and rangeland survey. Washington, DC: U.S.Department of Agriculture, Forest Service, Forest PestManagement. 52 p.

Hansen, E.M. 1999. Phytophthora in the Americas. In:Proceedings: IUFRO working party 7.02.09: Phytophthoradiseases of forest trees. Grants Pass, OR: [Publisherunknown]: 23–27.

Haugen, L. 2001. How to identify and manage Dutch elmdisease. In: Ash, C.L., ed. Shade tree wilt diseases. St. Paul,MN: APS Press: 37–52.

Hebard, F.V.; Sisco, P.H.; Wood, P.A. 2000. Meadowview notes1999–2000. Journal of the American Chestnut Foundation.14: 7–15.

Hester, F.E. 1991. The U.S. National Park Service experiencewith exotic species. Natural Areas Journal. 11: 127–128.

Holm, L.; Pancho, J.V.; Herberger, J.P.; Plucknett, D.L. 1979.A geographical atlas of world weeds. New York: John Wiley.632 p.

Houston, D.R. 1994. Major new tree disease epidemics: beechbark disease. Annual Review of Phytopathology. 32: 75–87.

Houston, D.R. 1997. Beech bark disease. In: Britton, K.O., ed.Exotic pests of eastern forests. Proceedings of the exoticpests of eastern forests conference. Nashville, TN:[Publisher unknown]: 29–41.

Houston, D.R.; O’Brien, J.T. 1983. Beech bark disease. For.Insect and Dis. Leafl. 75. [Radnor, PA]: U.S. Departmentof Agriculture, Forest Service, Northeastern Area State andPrivate Forestry. 1 p.

Jones, R.H.; McLeod, K.W. 1990. Growth and photosyntheticresponses to a range of light environment in Chinese tallowtree and Carolina ash seedlings. Forest Science. 36: 851–862.

Jones, R.H.; Sharitz, R.R. 1989. Effects of root competition andflooding on growth of Chinese tallow tree seedlings.Canadian Journal of Forest Research. 20: 573–578.

Liebhold, A.M.; McDonald, W.L.; Bergdahl, D.; Mastro, V.C.1995. Invasion by exotic forest pests: a threat to forestecosystems. Forest Science Monographs. 30: 1–49.

Lippincott, C.L. 2000. Effects of Imperata cylindrica(cogongrass) invasion on fire regime in Florida sandhills(U.S.A.). Natural Areas Journal. 20: 140–149.

Mack, R.N.; Lonsdale, W.M. 2001. Humans as global plantdispersers: getting more than we bargained for. BioScience.51: 95–102.

Mack, R.N.; Simberloff, D.; Lonsdale, W.M. [and others]. 2000.Biotic invasions: causes, epidemiology, global consequencesand control. Ecological Applications. 10: 689–710.

McClure, M. 1994. Hemlock woolly adelgid. Pest Alert NA–PR–03–94. [Radnor, PA]: U.S. Department of Agriculture, ForestService, Northeastern Area State and Private Forestry. 1 p.

McClure, M.S.; Salom, S.M.; Shields, K.S. 2001. Hemlockwoolly adelgid. FHTET–2001–03. Morgantown, WV: U.S.Department of Agriculture, Forest Service, Forest HealthTechnology Enterprise Team. 19 p.

McCullough, D.G.; Heyd, R.L.; O’Brien, J.G. 2001. Biology andmanagement of beech bark disease: Michigan’s newest exoticforest pest. Ext. Bull. E–2746. East Lansing, MI: MichiganState University, Michigan State University Extension. 11 p.

McManus, M.; Schneeberger, N.; Reardon, R.; Mason, G.1989. Gypsy moth. For. Insect and Dis. Leafl. 162. [Place ofpublication unknown]: U.S. Department of Agriculture,Forest Service. 13 p.

Miller, J.H. 1990. Ailanthus altissima (Mill.) Swingle:Ailanthus. In: Burns, R.M.; Honkala, B.H., eds. Silvics ofNorth America: hardwoods. Agric. Handb. 654. Washington,DC: U.S. Department of Agriculture, Forest Service. 877 p.Vol. 2.

Miller, J.H. 1998. Exotic invasive plants in southeastern forests.In: Britton, K.O., ed. Proceedings of the exotic pests ofeastern forests. Nashville, TN: Tennessee Exotic Pest PlantCouncil: 97–105.

Miller, J.H. 1999. Controlling exotic plants in your forest.Forest Landowner. 58: 60–64.

Newcomb, G.; Stirling, B.; McDonald, S.; Bradshaw, J.R. 2000.Melampsora x columbiana, a natural hybrid of M. medusaeand M. occidentalis. Mycological Research. 104: 261–274.

Pimentel, D. 1993. Habitat factors in new pest invasions. In:Kim, K.C.; McPherson, B.A., eds. Evolution of insect pests—patterns in variation. New York: John Wiley: 165–181.

Pimentel, D.; Lach, L.; Zuniga, R.; Morrison, D. 2000.Environmental and economic costs of nonindigenous speciesin the United States. BioScience. 50: 53–65.

Randall, J.M.; Marinelli, J., ed. 1996. Invasive plants: weedsof the global garden. Handb. 149. [Place of publicationunknown]: Brooklyn Botanic Garden. 111 p.

Reichard, S.H.; White, P. 2001. Horticulture as a pathway ofinvasive plant introductions in the United States. BioScience.51: 103–113.

Rizzo, D.M.; Garbelleto, M.; Davidson, J.M. [and others]. 2002.Phytophthora ramorum as the cause of extensive mortalityof Quercus spp. and Lithocarpus densiflorus in California.Plant Disease. 86: 205–214.

Salom, S.M. 1996a. Balsam woolly adelgid. Entomol. Publ. 444–233. Blacksburg, VA: Virginia Polytechnic Institute and StateUniversity, Virginia Cooperative Extension. 2 p.

Cha

pter

14.

Non

nati

ve F

ores

t Pe

sts

in t

he S

outh

Sout

hern

For

est

Scie

nce:

Past

, Pre

sent

, and

Fut

ure

Fore

st H

ealt

h

154

Salom, S.M. 1996b. Hemlock woolly adelgid. Entomol. Publ.444–244. Blacksburg, VA: Virginia Polytechnic Instituteand State University, Virginia Cooperative Extension. 2 p.

Santamour, F.S., Jr.; Bentz, S.E. 1995. Updated checklistof elm (Ulmus) cultivars for use in North America. Journalof Arboriculture. 21: 122–131.

Sharov, A.A.; Leibhold, A.M. 1998. Model of slowing the spreadof gypsy moth (Lepidoptera: Lymantriidae) with a barrierzone. Ecological Applications. 8: 1170–1179.

Simberloff, D.; Stilling, P. 1996. How risky is biologicalcontrol? Ecology. 77: 1965–1975.

Spiers, A.G.; Hopcroft, D.H. 1994. Comparative studies of thepoplar rusts Melampsora medusae, M. larici-populina, andtheir interspecific hybrid M. medusae-populina. MycologicalResearch. 98: 889–903.

Stapanian, M.A.; Sundberg, S.D.; Baumgardner, G.A.; Liston,A. 1998. Alien plant species composition and associationswith anthropogenic disturbance in North American forests.Plant Ecology. 139: 49–62.

Stipes, R.J.; Fraedrich, B.R. 2001. The management of Dutchelm disease with emphasis on fungicides. In: Ash, C.L., ed.Shade tree wilt diseases. St. Paul, MN: APS Press: 53–64.

Tainter, F.H.; Jolley, L.; Hernandez, A. [and others]. 1999.Histology of the zone line in secondary phloem of Mexicanoak trees infected with Phytophthora cinnamoni. In:Proceedings, IUFRO working party 7.02.09: Phytophthoradiseases of forest trees. Grants Pass, OR: [Publisherunknown]: 71-74.

Tennessee Exotic Pest Plant Council. 1996. Tennessee exoticplant management manual. Nashville, TN: Tennessee ExoticPest Plant Council. 119 p.

U.S. Congress, Office of Technology Assessment. 1993.Harmful nonindigenous species in the United States.OTA-F–565. [Place of publication unknown]. 391 p.

U.S. Department of Agriculture, Forest Service. 1985. Insectsof eastern forests. Misc. Publ. 1426. Washington, DC. 608 p.

U.S. Department of Agriculture, Forest Service. 1994. Pestand pesticide management on southern forests. Man. Bull.R8–MB 60. Atlanta: U.S. Department of Agriculture. 46 p.www.srs.fs.fed.us. [Date accessed: May 18, 2003].

U.S. Department of Agriculture, Forest Service; Animal andPlant Health Inspection Service. 1999. Asian longhornedbeetle (Anoplophora glabripennis): a new introduction. PestAlert NA–PR–01–99GEN. Radnor, PA: U.S. Department ofAgriculture, Forest Service, Northeastern Area, State andPrivate Forestry. [Not paged].

U.S. Department of Agriculture, Forest Service. 1999.Southern forest health atlas of insects and diseases. R8–MR35. [Atlanta]: [U.S. Department of Agriculture, ForestService, Southern Region, Forest Health Protection]. 20 p.

Wear, David N.; Greis, John G., eds. 2002. Southern forestresource assessment. Gen. Tech. Rep. SRS–53. Asheville,NC: U.S. Department of Agriculture, Forest Service,Southern Research Station. 635 p.

Weste, G.; Marks, G.C. 1987. The biology of Phytophthoracinnamomi in Australian forests. Annual Review ofPhytopathology. 25: 207–229.