A Fungal Symbiont of the Redbay Ambrosia Causes a Lethal Wilt in Redbay and Other in the Southeastern United States

S. W. Fraedrich, Southern Research Station, USDA Forest Service, Athens, GA 30602; T. C. Harrington, Depart- ment of Plant Pathology, Iowa State University, Ames 50011; R. J. Rabaglia, Forest Health Protection, USDA Forest Service, Arlington, VA 22209; M. D. Ulyshen, Southern Research Station, USDA Forest Service, Athens, GA 30602; A. E. Mayfield, III, Florida Department of Agriculture and Consumer Services, Division of Forestry, Gainesville 32608; J. L. Hanula, Southern Research Station, USDA Forest Service Athens, GA 30602; J. M. Eickwort, Florida Department of Agriculture and Consumer Services, Division of Forestry, Gainesville 32608; and D. R. Miller, Southern Research Station, USDA Forest Service, Athens, GA 30602

fungi have been assigned to the anamor- ABSTRACT phic genera Ambrosiella or Fraedrich, S. W., Harrington, T. C., Rabaglia, R. J., Ulyshen, M. D., Mayfield, A. E., III, Hanula, (3,22), and subsequent phylogenetic stud- J. L., Eickwort, J. M., and Miller, D. R. 2008. A fungal symbiont of the redbay ies have found close relationships between causes a lethal wilt in redbay and other Lauraceae in the southeastern United States. Plant Dis. these anamorphs and the ascomycetaceous 92:215-224. genera Ceratocystis and Ophiostoma (7,24,40). Many bark , including Extensive mortality of redbay has been observed in the coastal plain counties of Georgia and ambrosia beetles, have mutualistic associa- southeastern South Carolina since 2003 and northeastern Florida since 2005. We show that the tions with species in these genera of asco- redbay mortality is due to a vascular wilt disease caused by an undescribed Raffaelea sp. that is a fungal symbiont of glabratus, an exotic ambrosia beetle. Trees affected by the disease mycetes, and it is thought that the ambro- exhibit wilt symptoms that include a black discoloration of the sapwood. Redbay trees and con- sia beetle symbionts are asexual tainerized seedlings died within 5 to 12 weeks after inoculation with the Raffaelea sp. When derivatives of Ceratocystis and Ophio- redbay seedlings were challenged with X. glabratus, the beetles tunneled into 96% of the plants, stoma (22). 70% died, and the Raffaelea sp. was recovered from 91%. X. glabratus and the Raffaelea sp. Redbay is an attractive, aromatic, ever- have also been associated with mortality of sassafras, and the Raffaelea sp. has been isolated green tree that is common in forests of the from wilted pondberry and pondspice. Additional inoculation studies have shown that the Raf- Atlantic and Gulf Coastal Plains of the faelea sp. is pathogenic to sassafras, spicebush, and , but not to red maple. Female adults southeastern United States (5). Confusion of X. glabratus have paired mycangia near the mandibles, and the Raffaelea sp. is routinely iso- currently exists regarding the of lated from the heads of beetles. The is apparently introduced into healthy redbay during P. borbonia and other possible taxa in the beetle attacks on stems and branches. The wilt currently affecting redbay and sassafras repre- southeastern United States (6). Some bota- sents a major threat to other members of the Lauraceae indigenous to the Americas, including nists recognize swampbay (P. palustris avocado in commercial production. (Raf) Sarg.) as an additional Persea spe- cies (10,41), but others consider swampbay (P. borbonia var. pubescens (Pursh) Little) Extensive mortality of redbay (Persea stems and branches of many affected trees, to be a variety of redbay (8,29) or do not borbonia (L.) Spreng.) has been observed and several species of ambrosia beetles recognize it as a distinct species or variety in the lower coastal plains of South Caro- (Coleoptera: , Scolytinae) (5,14,38). In this paper, we refer to redbay lina and locations around Savannah, GA were also found in symptomatic redbay as P. borbonia sensu lato without attempt- since 2003. Initial reports of the mortality samples, including Xyleborinus gracilis ing to discern species or varietal differ- suggested that the problem might be Eichhoff, Ambrosiodmus obliquus Le- ences, except in a pathogenicity test where drought related, but direct evidence for this Conte, and Xyleborus glabratus Eichhoff. we obtained seedlings from a commercial was lacking and no explanation was pro- The two former beetles are native to the nursery that listed the seedlings as swamp- vided for the localized nature of the mor- southeastern United States, but X. glabra- bay (P. palustris). Species in the genus tality. During a site visit to Hilton Head tus is an exotic beetle reported only once Persea are members of the family Lau- Island, SC in November 2004, many dead before in the United States at Port Wen- raceae, which includes many important redbay trees were found, as well as live tworth, GA in 2002 (37). The beetle is species of plants worldwide, including trees exhibiting varying levels of branch native to Southeast Asia (e.g., India, Japan, avocado (P. americana Mill.). dieback. Stems and branches of affected and Taiwan), where it is often associated Redbay in the broadest sense can toler- trees had streaks of black discoloration in with plant species in the family Lauraceae ate a wide variety of sites ranging from the sapwood, and a fungus was consis- (e.g., Lindera latifolia Hk.f and Litsea xeric to mesic to swampy, and the species tently isolated from the discolored sap- elongata (Nees) Hk.f.) (50). is suitable for many landscape applications wood. Beetle entrance holes were noted in Little is currently known about the biol- (16). Redbay is regarded as a minor-use ogy of X. glabratus, although we assume hardwood, and the wood is occasionally that it is typical of other ambrosia beetles, used for interior finishing, cabinets, and Corresponding author: Stephen Fraedrich which do not consume wood but build furniture (5). Trees are small to medium E-mail: [email protected] gallery systems in wood to cultivate sym- size and are reported to grow as much as Accepted for publication 14 September 2007. biotic fungi as food for their larvae (28). 18 to 21 m high and 60 to 90 cm in diame- Many xylem-inhabiting ambrosia beetles ter (5). The fruit is eaten by songbirds, and some bark beetles carry their symbi- wild turkey, and other and deer doi:10.1094/ PDIS-92-2-0215 otic fungal associates in spore-bearing sacs browse on the foliage (5). Redbay also This article is in the public domain and not copy- serves as a primary larval host for the pala- rightable. It may be freely reprinted with custom- called mycangia, which can vary greatly in ary crediting of the source. The American Phyto- morphology and location among beetle medes swallowtail (Papilio palamedes pathological Society, 2008. species (22,45). The majority of ambrosia Drury) (43).

Plant Disease / February 2008 215 and diseases have not caused se- July 2005 at the Timucuan Ecological and of sterile water was placed into the wound. rious damage to redbay in the past (5,16). Historic Preserve (National Park Service) Inoculation points on all seedlings were Since our initial evaluations on Hilton on Fort George Island (Duval County), FL. wrapped with Parafilm M (Pechiney Plas- Head Island, many additional sites with Redbay mortality on the site was first ob- tic Packaging, Menasha, WI). Seedlings redbay mortality have been found and served during the fall of 2004, and ap- were placed in a growth chamber with a evaluated in Georgia, Florida, and South proximately one-third of the largest diame- 16-h photoperiod and day and night tem- Carolina. The vascular discoloration, X. ter redbay were dead by July 2005. Five peratures were set at 28 and 25°C, respec- glabratus, and the unidentified fungus (0.08-ha) circular plots were established tively. After 9 weeks, disease symptoms have been associated with dead and dying along a 1.1-km section of park road where were evaluated and pieces of stem tissue redbay trees in coastal areas of South dead and wilting redbays were observed. from 10 to 20 cm above the inoculation Carolina, Georgia, and northeast Florida. Plot centers were established approxi- points were surface sterilized and plated on In addition, mortality of sassafras (Sassa- mately 20 m from the road and at least 70 CSMA. fras albidum (Nuttall) Nees), another m apart and all were located within the A second test was established in June member of the Lauraceae, has been ob- same hardwood hammock forest type 2006 to assess possible variation in patho- served in several coastal counties of Geor- dominated by live oak (Quercus virginiana genicity among isolates. Ten isolates of the gia and X. glabratus and the unidentified Mill.), redbay, red cedar (Juniperus virgin- unknown fungus were obtained from fungus were associated with these trees. iana L.), southern magnolia (Magnolia symptomatic redbay and sassafras trees at The objectives of this study were to: i) grandiflora L.), and cabbage palm (Sabal various locations within the known range assess the distribution and determine the palmetto (Walter) Lodd. ex Schult. & of the disease. Seedlings were 2 to 3 years cause of the mortality of redbay in the Schult.f.). In each plot, every redbay with a old and averaged 82 cm high and 12 mm in southeastern United States; ii) evaluate the diameter at breast height (dbh; diameter at diameter at groundline. Inoculum for each pathogenicity of the unidentified fungus to 1.4 m high) greater than 2.5 cm was tagged isolate was produced as previously de- redbay and other members of the Lau- and visually rated for crown condition (i.e., scribed, and inoculum concentrations for raceae found in the southeastern United foliage primarily green, wilting, or brown). the isolates ranged from 2.6 × 106 to 2.8 × States; and iii) determine the association of Trees with primarily brown or wilting 107 conidia/ml. Fifty seedlings were the unidentified fungus with X. glabratus. foliage were considered dead. Ratings wounded with a drill bit as previously were made every 3 months through De- described and five seedlings were inocu- MATERIALS AND METHODS cember 2006. A total of 132 trees with a lated with 0.1 ml of the spore suspension Field survey and assessment of dis- greater than 2.5-cm dbh were monitored in of each fungal isolate. An additional five ease development. Natural forests as well the five plots. In July 2006, a 0.01 ha cir- seedlings were wounded and inoculated as urban and residential areas with dead cular subplot was added to the center of with 0.1 ml of sterile deionized water. and symptomatic redbay trees were located each plot, and all redbay stems less than Seedlings were placed in a growth cham- on the basis of information from state for- 2.5-cm dbh and at least 15 cm tall were ber with the photoperiod and temperature estry and pest management personnel, counted and rated as alive or dead. conditions previously described. After 5 managers of state and federal parks, land- Pathogenicity tests, growth chamber. weeks, seedlings were evaluated for symp- owners, and USDA Forest Service person- The initial pathogenicity test in May 2005 toms and pieces of wood were plated on nel. Representative locations with redbay utilized 30 redbay seedlings that were 2 to CSMA to determine the presence of the that exhibited symptoms of wilt were se- 3 years old and grown in 3.8- or 11.4-liter inoculated fungus. lected and samples were collected, placed (1- or 3-gallon) containers with a commer- Pathogenicity test, field. A pathogenic- in plastic bags, and transported to the cial nursery mix. The mean plant height ity test was conducted at a field site in the USDA Forest Service Laboratory in Ath- and diameter at the groundline were 63 cm Victoria Bluff Heritage Preserve near ens, GA. In the laboratory, samples of and 10 mm, respectively. Before inocula- Bluffton, SC, where mortality of redbay branch and stem sections were dissected to tion, seedlings were grown in a greenhouse due to the wilt had been previously ob- locate beetle galleries and identify the for a minimum of 1 month with tempera- served. Twenty-four redbay trees that species of beetle when present. Sections of tures generally between 20 and 35°C and showed no evidence of wilt were selected branches and stems with sapwood discol- no supplemental lighting. and paired on the basis of size and prox- oration were surface sterilized by dipping The inoculation study utilized two sin- imity to one another. Tree heights ranged the sections in 95% ethyl alcohol and gle-spore isolates (HH2 = C2266 and HH5 from 1.8 to 4.9 m and tree diameters flaming. Wood chips were removed from = C2428) of the unknown fungus isolated ranged from 2.1 to 6.2 cm at groundline. samples and placed on malt extract agar from symptomatic redbay trees on Hilton One tree of each pair was inoculated with (MEA; 2.5% malt extract and 2.0% agar) Head Island, SC. The isolates were grown one of two isolates (HH1 = C2265 or HH2 or cycloheximide-streptomycin malt agar on MEA for 14 days. Conidia were col- = C2266) of the unknown fungus on 3 or (CSMA; 1% malt extract, 1.5% agar, and lected by flooding plates with approxi- 17 March 2005; the other tree of each pair 200 ppm of cycloheximide and 100 ppm of mately 20 ml of sterile deionized water, was treated as a control. A hole (5 mm in streptomycin sulfate added after autoclav- loosening conidia with a glass rod, and diameter and 1 to 2 cm deep) was drilled ing), a medium selective for Ophiostoma passing the suspension through 3-ply ster- into each tree between 0.6 to 1.2 m above spp. and their related anamorphs (19). ile gauze. Spore concentrations were deter- groundline. The isolates were grown on Evaluations of isolation plates were typi- mined with a haemacytometer and ranged MEA and an agar plug with the fungus was cally performed 7 to 10 days after plating. from 1.5 × 106 to 2.6 × 106 conidia/ml. inserted into the drilled holes for all fun- Morphological and molecular characteriza- Fifteen redbay seedlings were wounded gus-inoculated trees. Seven of the control tions of an unidentified fungus associated by drilling holes (2 mm in diameter and 3 trees received a sterile agar plug and the with the sapwood discoloration were per- to 7 mm deep) into the stem at 4 to 7 cm other five control trees received no agar formed in laboratories at Iowa State Uni- above groundline and another 15 seedlings plug. Inoculation points were wrapped versity. Amplification and sequencing of the were wounded by slitting stems to a depth with Parafilm. Trees were evaluated on 9 small subunit (SSU) rDNA and the large of approximately one-third the diameter of June 2006 and representative samples were subunit (LSU) of the rDNA were performed the stem. Five seedlings of each wound obtained from the stems of all trees, placed as described in Cassar and Blackwell (7) type were inoculated with 0.1 ml of conid- in plastic bags, transported in coolers with and Paulin and Harrington (36). ial suspension from one of the two fungal ice packs, and then stored at 4 to 6°C for 1 An assessment of the rate of tree mortal- isolates. The other five seedlings of each week before isolations were attempted. ity in an affected forest was initiated in wound type served as controls, and 0.1 ml Pieces of wood from all trees were surface

216 Plant Disease / Vol. 92 No. 2 sterilized, plated on CSMA, and incubated two isolates or sterile water as previously nique by Schedl (42), to soften the cuticle. as previously described. Plates were evalu- described. Three redbay seedlings were A rotary microtome was used to cut 8- to ated after 7 to 10 days for the presence of also inoculated with the inoculum from 12-µm transverse and longitudinal sec- the inoculated fungus. each isolate to confirm pathogencity. Seed- tions. The sections were stained with Har- Pathogenicity to other species. The lings were placed in a growth chamber ris-hematoxylin and eosin-phloxine (30) susceptibility of sassafras, spicebush (Lin- with a 16-h photoperiod and day and night and observed with a light microscope. dera benzoin (L.) Blume), and swampbay temperatures at 28 and 25°C, respectively. Ten additional, recently emerged female to the wilt fungus was compared with that After 8 weeks, the seedlings were evalu- beetles from infested bolts were used to of redbay in a growth chamber test in Au- ated for symptoms and representative stem determine the association of fungal gust 2005. All seedlings were grown in samples from the seedlings were plated on propagules of the unknown fungus with 3.8- or 11.4-liter pots. The mean diameters CSMA as previously described. the head, thorax, and abdomen. Body parts at groundline and heights of plant species Mycangia of X. glabratus. Ten female were aseptically separated and individually ranged from 8.5 to 12 mm and 62 to 100 X. glabratus adults, freshly emerged from placed in a sterile ceramic mortar with 0.1 cm, respectively. Inoculum of isolates HH1 infested bolts, were examined for the pres- ml of sterile deionized water and thor- and HH2 was prepared as previously de- ence of mycangia. Beetles were fixed in oughly ground. An additional 1 ml of ster- scribed, and inoculum concentrations Carnoy’s solution and then soaked in phe- ile deionized water was then added to the ranged from 4.8 to 6.4 × 106 condia/ml. nol for 1 week, a modification of a tech- mortar. The suspension was mixed and 0.1 For each plant species, 15 seedlings were wounded with a 2.8-mm-diameter drill bit to a depth approximately one-half of the diameter of the stems, and five seedlings were inoculated with one of the two fungal isolates or sterile water as previously de- scribed. Seedlings were placed in a growth chamber with a 16-h photoperiod and day and night temperatures at 28 and 25°C, respectively. After 5 weeks, the seedlings were evaluated for symptoms and repre- sentative stem samples were plated on CSMA and evaluated for the presence of the fungus as previously described. The susceptibility of avocado (P. a me ri - cana) to wilt caused by the unknown fun- gus was evaluated in a growth-chamber test in February 2006. Seedlings were ob- tained from a nursery in south Florida and were grown in 3.8-liter pots. Seedlings were less than 1 year old with a mean di- ameter of 10.8 mm and height of 47 cm. Inoculum of isolates HH1 and HH5 was prepared as previously described and con- centrations ranged from 2.0 to 3.5 × 106 conidia/ml. Thirty seedlings were wounded with a drill bit (2.8 mm in diameter) and 10 seedlings were inoculated with one of Fig. 1. Locations (O) in Georgia, South Carolina, and Florida at which the Raffaelea sp. has been iso- the two fungal isolates or sterile water as lated from dead and dying redbay trees. Counties where the disease has been observed are shaded. previously described. Seedlings were placed in a growth chamber with a 16-h photoperiod and day and night tempera- tures at 28 and 25°C, respectively. After 4 weeks, seedlings were evaluated for exter- nal symptoms on foliage and shoots. After 8 weeks, the seedlings were evaluated again for symptoms and representative stem samples from seedlings were plated on CSMA as previously described. The susceptibility of red maple (Acer rubrum L.) to wilt caused by the unknown fungus was evaluated in a growth chamber test in April 2006. Seedlings were grown in 11.4-liter pots and were 2 years of age when inoculated. The mean diameter of seedlings was 14.5 mm and the mean height was 90 cm. Inoculum of isolates HH1 and HH2 was prepared as previously described and inoculum concentrations ranged from 5.7 to 6.6 × 106 conidia/ml. Fifteen seedlings were wounded with a drill bit (2.8 mm in diameter) and five Fig. 2. Cumulative percent redbay mortality by diameter class (diameter at breast height) on five 0.08- seedlings were inoculated with one of the ha plots at Fort George Island, FL between July 2005 and January 2007.

Plant Disease / February 2008 217 ml was placed on each of three petri plates mean diameter at groundline was 10.2 mm. lings were observed for symptoms over the containing CSMA. The suspension was Beetles were reared from bolts of naturally course of the next 8 weeks. At the end of spread over the plates with a sterile glass infested, diseased redbay collected from the experiment, seedlings were evaluated rod and plates were incubated at 25°C. the field. Individual female beetles that for tunnel development and symptoms and Plates were examined after 7 to 8 days and emerged from bolts were placed in gel representative samples of wood from all the number of fungal colonies was deter- capsules, and the capsules were attached to seedlings were plated on CSMA medium mined. Numbers of colony forming units healthy seedlings by lightly scoring the and subsequently evaluated for the pres- per beetle body part were expressed as a stems with a cork borer approximately 4 to ence of the fungus. mean of the three plates. 5 cm above the groundline and inserting A second experiment was initiated in Plants challenged with X. glabratus. the edge of one-half of the capsule with the April 2006. Redbay, sassafras, spicebush, Two experiments were conducted to evalu- beetle into the resulting groove. Beetles red maple, and white oak (Quercus alba ate the potential for X. glabratus to tunnel that died within 24 h after attachment of L.) were challenged with X. glabratus into plants and introduce the pathogen. In the capsules to stems were replaced. Six adults reared from bolts of infested and the first experiment, 23 redbay seedlings additional redbay seedlings were scored diseased trees. All seedlings were grown in were challenged with X. glabratus in Octo- with a cork borer and capsules without 3.8- or 11.4-liter containers, and the mean ber 2005. The mean height of seedlings beetles were attached; these seedlings diameters at groundline varied among used in the experiment was 48 cm and the served as nonchallenged controls. All seed- species from 11.2 to 14.0 mm and mean

Fig. 3. Examples of wilted redbay trees in the southeastern United States caused by a Raffaelea sp. A, Redbay in the initial stages of wilt at Hunting Island State Park, SC in March 2006. B, Wilted redbay in a residential area of Jekyll Island, GA in October 2006. C, Wilted and nonsymptomatic redbay at Hunting Island State Park, SC in October 2005. D, Wilted redbay trees at the Timucuan Ecological and Historic Preserve on Ft. George Island, FL in November 2005.

218 Plant Disease / Vol. 92 No. 2 heights ranged from 87 to 129 cm. For four locations in Georgia, and the unidenti- gia. The fungus was also isolated from each species, 9 to 11 seedlings were scored fied fungus and X. glabratus have been wilted pondberry (Lindera melissifolia with a cork borer and challenged with X. associated with the dead and dying trees. (Walt.) Blume) at a site in Georgia. Thus glabratus as previously described. Beetles The unidentified fungus has also been far, we have not found X. glabratus associ- that died within 24 h after placement on isolated from wilted pondspice (Litsea ated with the dead pondspice or pondberry. stems were replaced. Seedlings were ob- aestivalis (L.) Fern.) at a location in South Disease and fungus description. In the served over the course of the next 6 weeks. Carolina and at another location in Geor- early stages of the disease, some trees At the end of the experiment, seedlings were evaluated for development of tunnels and disease symptoms and representative samples of wood from all seedlings were plated on CSMA medium.

RESULTS Field survey. Sample locations and the known range of the new disease on redbay as of January 2007 are presented in Figure 1. Sampled redbay ranged from seedlings to trees approximately 15 m high and 30- cm dbh. Diseased trees exhibited branch mortality and wilt over a portion of the tree crown or the entire crown. The characteris- tic black discoloration of the sapwood of stems or branches was also observed. The disease was common in urban, residential, and natural areas on a range of sites in- cluding swamps, mesic flatwoods, and xeric dune and upland coastal plain forests. An unknown fungus was isolated from Fig. 4. A, Sapwood discoloration and streaking in wilted redbay (Fort George Island, FL). B, Close-up symptomatic trees at all locations sampled. of sapwood discoloration with tunnel entrance holes (arrows) made by Xyleborus glabratus. The presence of X. glabratus in sympto- matic trees was also confirmed at many of these locations. Another Asian ambrosia beetle, Xylosandrus crassiusculus Mot- schulsky, was also observed in stems of dead and dying redbay at many locations. At some locations, dieback of twigs and small branches of redbay was associated with attacks by a third Asian ambrosia beetle, Xylosandrus compactus (Eichhoff). The mortality of redbay trees increased dramatically at the Timucuan Ecological and Historic Preserve over a 16-month period. The cumulative percentage of mor- Fig. 5. Microscopic characteristics of the Raffaelea sp. on malt extract agar. A, Conidia, conidio- tality among the 132 monitored redbays phores, and budding yeast-like spores. B, Conidia and budding yeast-like spores. (2.5-cm dbh and greater) increased from 9.8 to 92.4% between July 2005 and Octo- ber 2006 (Fig. 2). Larger diameter classes died more quickly than smaller classes, and by December 2006, all redbays greater than 10.3-cm dbh were dead. Of the 222 redbay stems less than 2.5-cm dbh exam- ined in the five 0.01 ha subplots estab- lished in July 2006, only one was dead in January 2007. The diseased Persea trees that we exam- ined were considered P. borbonia sensu lato. However, the symptomatic trees var- ied in traits that are used by some taxono- mists to differentiate redbay varieties or separate species. The trees had varying degrees of pubescence on twigs and the undersides of leaves that ranged from sparse, short, appressed trichomes to abun- dant, long, erect, reddish or golden brown trichomes that gave the appearance of dense pubescence. Other species of Lauraceae appear to be affected by the wilt. Wilted sassafras with sapwood discoloration has been found at Fig. 6. Redbay seedlings 5 weeks after inoculation with the Raffaelea sp.

Plant Disease / February 2008 219 exhibited wilting of branch tips throughout gated conidia at the tips (Fig. 5). Conidia (1,007 bp in common with AY858658), O. the crown, or dieback and wilting of indi- in turn produced blastospores and yeast- auraucariae (1,007 bp in common with vidual branches (Fig. 3A). During the like colonies developed at the tips of the AY497510), O. stenoceras (1,007 bp in spring and summer months, trees fre- conidiophores. The conidia were highly common with AY858663), and Sporothrix quently exhibited uniform wilting of foli- variable in shape, mostly 4 to 9 × 2 to 3 schenckii (1,007 bp in common with age throughout the entire crown and the µm, ovate to pyriform to elongate. The M85053). A portion of the LSU rDNA was trees died rapidly (Fig. 3B–D). Dying conidiophores were small and hyaline and also sequenced, and a BLAST search with leaves developed a red brown-to-purple the conidiogenous cells did not have con- a 533-bp segment of the LSU (Accession brown coloration and often persisted on spicuous scars or had inconspicuous anne- No. EU123077) placed the fungus in the branches for more than 1 year after the tree lations at the point of conidial dehiscence. genus Ophiostoma: 498 bp of the 533-bp died. A dark black discoloration was pres- The conidia and conidiophores are consis- fragment matched with the sequence of O. ent in the sapwood of branches and stems tent with the generic concept of Hyalorhi- huntii (AY707208), Leptographium yun- of diseased trees (Fig. 4). In the early nocladiella, a common anamorph of Ophi- nanense (AY707207), L. wingfieldii stages of wilt, the discoloration was fre- ostoma spp. However, an anamorph genus (AY707205), and Fragosphaeria purpurea quently noted in the outermost sapwood of ambrosia beetle symbionts, Raffaelea, (AF096191). and occurred as localized streaks in por- also produces similar conidiophores, but in Thirty-seven isolates of the Raffaelea sp. tions of branches and stems. As the disease sporodochia (3), and the unknown fungus were sequenced for the 5′ end of the LSU. progressed, the discoloration developed could be accurately placed in this genus. Each of these sequences was identical to more extensively through the cross-sec- The fungus is herein referred to as Raf- Accession No. EU123077. The sequenced tional area of stems or branches. Small faelea sp. Like other Ophiostoma species isolates included those from diseased red- beetle entrance holes (approximately 0.75 and their anamorphs (18), the Raffaelea sp. bay in South Carolina, Georgia, and Flor- mm in diameter) were frequently found in tolerates high concentrations of cyclo- ida, and isolates from lauraceous hosts branches and stems with sapwood discol- heximide and selective media containing , Lindera melissifolia, oration, but the beetle holes were some- cycloheximide have been routinely used in Litsea aestivalis, and from the ambrosia times rare and found only after an inten- isolations. beetle, X. glabratus. A representative iso- sive search in the early stages of the Attempts to amplify the internal tran- late was deposited in the Centraalbureau disease. After trees have wilted and died, scribed spacer region of the rDNA were voor Schimmelcultures (CBS 121567). stems and large branches were repeatedly not successful, so a portion of the SSU was Pathogenicity tests, growth chamber. attacked by X. glabratus and other species amplified and sequenced (36). A BLAST In the first inoculation experiment, all of ambrosia beetles during the spring, (NCBI, BLASTN 2.2.1.5) search of a redbay seedlings inoculated with the Raf- summer, and fall months. 1,026-bp portion of the SSU (exons only; faelea sp. by either the drill hole or slit A hyaline, fast-growing fungus was con- Accession No. EU123076) matched most treatments exhibited sapwood discolora- sistently isolated from discolored sapwood closely with the SSU sequence of a num- tion and 19 of 20 inoculated seedlings of diseased trees. Colonies on MEA were ber of ambrosia beetle symbionts and other wilted or died within 8 weeks after inocu- initially yeast like, but then a submerged species in the genus Ophiostoma: Am- lation (Fig. 6). One of the seedlings inocu- mycelium grew at the margins. The older brosiella brunnea (1,019 bp in common lated by the slit method exhibited sapwood portions of the colonies were very muci- with AY858654), Raffaelea ambrosiae discoloration but did not wilt. Leaf discol- laginous with abundant yeast-like growth. (1,011 bp in common with AY497518), R. oration or wilting of shoots or leaves was The advancing margin of the colonies had albimanens (1,009 bp in common with observed in some seedlings within 10 days sparse, mostly submerged hyphae with AY858664), R. canadensis (1,007 bp in after inoculation. Often, one of the first many small conidiophores producing elon- common with AY858665), A. sulcati external symptoms of infection was a dark discoloration of the midvein of some leaves. The leaves of inoculated seedlings subsequently developed a dark red brown- to-purple brown discoloration and wilted. Foliar symptoms progressed to include entire branches and then entire seedlings. The Raffaelea sp. was reisolated from the discolored sapwood of all symptomatic seedlings at the end of the experiment. The 10 control seedlings remained healthy and exhibited no symptoms. Each of the 10 isolates of the Raffaelea sp. was pathogenic to redbay in the second experiment. All but one of the 50 inocu- lated seedlings wilted or died, and the pathogen was recovered from the 49 symp- tomatic seedlings. One seedling inoculated with an isolate from Bryan County, Geor- gia survived and showed no sapwood dis- coloration and the pathogen was not recov- ered from this seedling. All control seedlings were healthy and did not exhibit symptoms at the end of the experiment and the pathogen was not isolated from the five control seedlings. Pathogenicity test, field. Each of the 12 redbay trees inoculated with the Raffaelea Fig. 7. Transverse section through the head of Xyleborus glabratus showing paired mandibular my- sp. wilted within 12 to 14 weeks after in- cangia (myc) containing fungal spores. oculation and sapwood discoloration was

220 Plant Disease / Vol. 92 No. 2 observed in the stems and branches of each Mycangia of X. glabratus. Mycangia periments. Living or dead adults were tree. The Raffaelea sp. was recovered from were found at the base of each mandible in found in only a few seedlings and no eggs, each of the 12 inoculated trees. Three of each of the female X. glabratus examined larvae, or pupae were observed in tunnels. the 12 control trees also wilted, but subse- (Fig. 7). The paired mycangia appeared to quent examinations of these trees revealed be packed with small fungal spores, and DISCUSSION that they had beetle entrance holes charac- the spores were similar in size and shape to The extensive mortality of redbay in teristic of X. glabratus as well as sapwood the blastospores (yeast cells) of the Raf- coastal plain counties of South Carolina, discoloration. The Raffaelea sp. was recov- faelea sp. Georgia, and northeastern Florida is due to ered from those three control trees. The The Raffaelea sp. was routinely recov- a new wilt disease caused by an unnamed other nine control trees were healthy with ered from the heads of X. glabratus adults species of Raffaelea. Other members of the no evidence of sapwood discoloration, wilt reared from diseased bolts but was not Lauraceae are also affected by the disease, symptoms, or beetle entrance holes and the isolated from the thoracic or abdominal and we propose that the new disease be Raffaelea sp. was not isolated from these sections of the same beetles. The Raffaelea named “” to reflect the breadth control trees. sp. was recovered from heads of 9 of the of susceptible plant species. Females of an Susceptibility of other species. All 10 beetles, with a range of 5 to 182 CFU exotic ambrosia beetle, X. glabratus, carry spicebush, sassafras, swampbay, and red- per plate (mean = 98.4 CFU) and as much the Raffaelea sp. in their paired mandibular bay seedlings (10 inoculated seedlings of as 2,000 CFU per head. The fungus was mycangia. We hypothesize that the patho- each species) wilted and died following not isolated from one of the 10 beetles. gen was introduced to the United States inoculation with the Raffaelea sp. All in- Plants challenged with X. glabratus. from Asia with X. glabratus, and the oculated seedlings exhibited sapwood dis- Adult beetles initiated tunnels into most of pathogen and its vector pose a major threat coloration and the pathogen was reisolated the redbay plants within 48 h after place- to members of the Lauraceae in the Ameri- from each of the inoculated seedlings. In ment on stems. At the end of the first ex- cas. spicebush, leaf chlorosis and wilting was periment, 22 of the 23 seedlings chal- The fungus responsible for the wilt falls evident in some seedlings within 5 days lenged with beetles had evidence of within the genus Ophiostoma based on its after inoculation and most spicebush seed- tunnels, 17 of those plants had discolored tolerance to cycloheximide and rDNA lings died within 3 weeks. All inoculated xylem, 16 of the plants wilted or died, and sequences. A few other species in the ge- sassafras seedlings died within 5 weeks the Raffaelea sp. was recovered from 21 of nus are capable of causing a vascular wilt and leaves turned red and abscised from the plants including the 16 plants that disease, that is, a disease in which the the stems. In redbay and swampbay, new wilted or died. The pathogen was also pathogen initially moves systemically shoots at branch terminals wilted and ma- isolated from five of the nonsymptomatic, through the sapwood in the nonliving ves- ture leaves developed a red brown-to-pur- beetle-attacked plants, but only from the sel elements. The sapwood discoloration ple brown discoloration and persisted on vicinity of the beetle tunnels. Control caused by the laurel wilt fungus is believed the branches. All redbay and swampbay plants did not exhibit symptoms of wilt, to be a host response to colonization of the seedlings died within 5 weeks after inocu- and the Raffaelea sp. was not isolated from pathogen, and the pathogen is readily iso- lation. Control seedlings of all species those plants. lated from all discolored vascular tissue remained healthy and exhibited no symp- In the second experiment, beetle tunnels including discolored leaf veins. The streak- toms and the pathogen was not recovered. were observed in 73 to 90% of sassafras, ing pattern found in the sapwood of af- In the test of the avocado seedlings, spicebush, and redbay plants challenged fected trees is similar in appearance to that three of the fungus-inoculated seedlings with X. glabratus (Table 1). Each seedling in elm trees infected with Ophiostoma ulmi were dead after 4 weeks, and 12 other of these species that had beetle tunnels or O. novo-ulmi (44). Conifers do not have seedlings exhibited some symptoms that wilted, and the Raffaelea sp. was isolated vessel elements, but Leptographium wag- included a brown discoloration and wilting from discolored sapwood of all sympto- eneri (teleomorph = O. wageneri) moves of some leaves, leading to branch dieback matic plants. Tunnel development was slowly through the nonliving tracheids in some of these seedlings. After 8 weeks, observed in 40% of the red maples chal- (20,44), and the sapwood discoloration in the 12 fungus-inoculated seedlings that lenged with adults of X. glabratus, but pines and Douglas-fir with black stain root had previously exhibited symptoms were there was neither sapwood discoloration disease is also similar in appearance to the still alive and there was no evidence that nor wilt symptoms in this species. The sapwood discoloration in diseased redbay. the disease had progressed much beyond Raffaelea sp. was not isolated from any of The unnamed agent of laurel wilt appears the level observed at 4 weeks. Five of the the challenged red maple seedlings. There to reside in a third lineage within Ophio- fungus-inoculated seedlings and all of the was no evidence of tunnel development in stoma that is capable of moving systemi- control seedlings remained asymptomatic any of the white oak seedlings, and these cally through their hosts in the nonliving after 8 weeks. At the termination of the seedlings appeared healthy at the end of xylem elements. experiment, 19 of the 20 fungus- the experiment with no evidence of sap- On the basis of rDNA sequences, the inoculated seedlings had a brown discol- wood discoloration. new pathogen is most closely related to oration in the xylem that frequently ex- Beetles were usually not present in the Ambrosiella and Raffaelea species that are tended throughout the stems of seedlings tunnels on attacked seedlings of any spe- thought to be asexual members of the ge- and the Raffaelea sp. was isolated from cies when examined at the end of the ex- nus Ophiostoma (7,24). These ambrosia the discolored xylem of all 19 seedlings. There was no vascular discoloration in Table 1. Development of beetle tunnels, sapwood discoloration, wilt symptoms, and isolation of Raffaelea the control seedlings and the Raffaelea sp. in redbay (), sassafras (Sassafras albidum), northern spicebush (Lindera benzoin), red sp. was not isolated. maple (Acer rubrum), and white oak (Quercus alba) challenged with Xyleborus glabratus In the test of red maple, all fungus-in- oculated and control seedlings were as- Number of plants ymptomatic with no evidence of wilting or Tunnel Sapwood Raffaelea sp. sapwood discoloration at the end of 8 Species Challenged development discoloration Wilted isolated weeks. The Raffaelea sp. was not reiso- Redbay 10 9 9 9 9 lated from any of the seedlings. In con- Sassafras 11 8 8 8 8 trast, all fungus-inoculated redbay seed- Spicebush 9 7 7 7 7 lings wilted and the Raffaelea sp. was Red maple 10 4 0 0 0 reisolated from the discolored sapwood. White oak 10 0 0 0 0

Plant Disease / February 2008 221 beetle symbionts produce small conidio- observed in these tunnels, but we have ont appears to be the death of Quercus phores in tight clusters (sporodochia), isolated the pathogen from discolored sap- crispula Blume and Q. serrata Thumb (ex which are thought to be adaptations for wood traced back to the tunnels. On the Murray) in Japan associated with Raf- grazing by ambrosia beetle larvae or by basis of our growth chamber studies, X. faelea quercivora, a fungal associate of the some bark beetles (3,22). Ambrosiella and glabratus is capable of tunneling into ambrosia beetle Platypus quercivorus Raffaelea are distinguished by annelidic healthy redbay seedlings as well as seed- (26,27). This ambrosia beetle is thought to versus sympodial proliferation of the co- lings of other species in the Lauraceae and be indigenous to Asia and the decline of nidiogenous cells, respectively (3), but the these test seedlings became infected. It oaks in Japan has been documented since use of this character in delineating anamor- appears that X. glabratus females tunnel the 1930s (25). The attacked trees appear phic genera of Ophiostoma species has into stems and branches, even those too to be weakened by drought or other agents, been questioned (21) and this would ap- small for brood development, and then and mass attacks by P. quercivorus are pear to be an artificial distinction in the abandon the tunnels without egg-laying. necessary for the oak trees to die (12,26). case of ambrosia beetle symbionts. The We hypothesize that the pathogen first R. quercivora is thought to facilitate tree type species of Ambrosiella (A. xylebori infects healthy trees during such tunneling death through localized colonization of the Brader ex Von Arx & Hennebert) is known by X. glabratus and the trees later become tree in the vicinity of the ambrosia beetle to be closely related to Ceratocystis rather suitable for brood production after the galleries, but the fungus does not colonize than Ophiostoma, leaving Raffaelea as a fungus has moved systemically throughout the host systemically (12,32,33). In con- better anamorphic genus for the asexual the main stem. It is likely that the larvae trast, the Raffaelea sp. that causes laurel symbionts of ambrosia beetles related to develop and feed on the pathogen in dis- wilt is a very aggressive vascular wilt Ophiostoma (7,24). Thus, the laurel wilt eased trees, and mycangia of young female pathogen that is capable of colonizing the pathogen will be described in a forthcom- adults are colonized before emergence. entire tree from a single introduction into ing study as a new species of Raffaelea. Thus, the laurel wilt pathogen may be small branches or stems, and the fungus The types of mycangia found in bark introduced into stems and branches of appears to predispose the trees to egg-lay- and ambrosia beetles vary greatly, from healthy trees by tunneling without egg- ing attacks by X. glabratus. simple external pits to complex invagina- laying, in a manner similar to the transmis- Laurel wilt has the potential to move tions of the cuticle lined with special ex- sion of the Dutch elm disease fungi by very rapidly within stands of redbay as cretory glands (4,22,45). Three types of their vectors during maturation illustrated in the mortality data collected at mycangia have been described for species feeding in elm twigs (23,47). However, to the Timucuan Ecological and Historic of Xyleborus: pouches at the base of each our knowledge, this behavior of tunneling Preserve. These data are consistent with elytron, intersegmental sacs between the without egg-laying has not been demon- observations and reports from other areas pronotum and mesonotum, and paired strated for an ambrosia beetle. Little is where the disease has occurred. For in- pouches at the base of the mandibles (34). known about X. glabratus in its native stance, reports of dying redbay on Hilton The latter type is most common in Xyle- habitat, and there is no indication that this Head Island, SC and surrounding areas borus (4) and consistent with our observa- beetle and its fungal associate are capable began in 2003, 1 year after the first detec- tions of X. glabratus. Isolations of the of killing trees in Asian forests. Clearly, tion of the beetle at nearby Port Wen- Raffaelea sp. from the heads of X. glabra- more work is needed on this critical stage tworth, GA. According to local newspaper tus, as well as experiments that challenged in the disease cycle of laurel wilt. accounts, the first reports of redbay mor- plants with the beetles, provide evidence There are currently about 3,400 species tality on Hilton Head Island may have that the Raffaelea sp. is reproducing within of ambrosia beetles worldwide that are occurred as early as 2002, but by the fall of the mycangia of X. glabratus. The budding taxonomically assigned to two subfamilies 2004, officials on Hilton Head Island were spores of the Raffaelea sp. are thought to of the Curculionidae (13). Most species of estimating that they had lost approximately ooze from the paired mycangia, and thus wood-tunneling ambrosia beetles are re- 80% of the redbay population. Likewise, are introduced to the host xylem during garded as opportunists that attack stressed park officials at Hunting Island State Park, beetle tunneling. Males of Xyleborus spe- or dying trees (15) and live only in re- SC reported that they had not observed cies are much smaller than the females, are cently cut, wounded, or weakened portions redbay mortality in the fall of 2004, but in generally flightless, and mate with siblings of woody plants (49). Kuhnholz et al. (28) the spring of 2005, the first losses of red- before young females emerge (37). We indicate that in recent years there have bay were observed, and as of October assume that only female X. glabratus have been reports of secondary ambrosia beetles 2006, approximately 80% of the island’s mycangia and can serve as vectors of the attacking apparently healthy trees and redbay have died of the disease (S. W. Raffaelea sp., but we have not studied possibly causing tree death. In some in- Fraedrich, unpublished data). At other males of X. glabratus. stances, these beetles are exotics and ap- sites in Georgia and South Carolina where Ambrosia beetles have not been consid- pear to be more aggressive in their new the disease has been observed since 2004, ered effective vectors of plant pathogens habitats. Numerous, nonnative ambrosia almost all of the redbay trees have suc- because they generally lay their eggs in beetles have been introduced into the cumbed to the disease (S. W. Fraedrich, dead or dying trees. In the case of X. United States over the last 100 years, but personal observation). Compared with glabratus, successful egg-laying galleries these beetles have not caused widespread larger diameter trees, seedlings appear have been commonly observed in the main mortality in native plant populations much less affected by the disease, pre- stems of redbay trees in the advanced (2,17,48). Some ambrosia beetles, such as sumably because they are not as readily stages of laurel wilt or after trees have Xylosandrus crassiusculus, have been re- attacked by X. glabratus. In addition, died. However, we have also observed sponsible for mortality of apparently sprouting of redbay at the groundline of isolated and infrequent beetle tunnels with healthy plants in nurseries, urban land- dead trees has been observed at numerous X. glabratus on stems and branches of scapes, and orchards (1,11,31,35). Others, locations. Some sprouts have died, but redbay trees in the early stages of wilt such as X. compactus, have caused death many others continue to grow. when sapwood discoloration is localized of twigs on healthy trees in natural and On the basis of the results of these stud- and wilting of foliage may be evident in urban landscapes (9). ies and the association of X. glabratus with only one or a few branches. In addition, we The association of a wilt-causing fungus species in the family Lauraceae in Asia have observed aborted tunnels similar in with an ambrosia beetle is unprecedented. and the United States, there is reason to be diameter to those produced by X. glabratus The only previously documented case of concerned that the wilt associated with in stems and branches as small as 1 cm in extensive mortality of forest trees caused redbay could affect additional members of diameter. No eggs or larvae have been by an ambrosia beetle and a fungal symbi- this family in the Americas. Recent obser-

222 Plant Disease / Vol. 92 No. 2 vations of the wilt on sassafras, pondberry ing Office, Washington, DC. warming on deciduous oak dieback caused by (a federally endangered species), and 6. Brown, C. L., and Kirkman, L. K. 1990. Trees ambrosia fungus Raffaelea sp. carried by am- of Georgia and Adjacent States. Timber Press, brosia beetle Platypus quercivorus (Coleop- pondspice (a species endangered in Florida Portland, OR. tera: Platypodidae) in Japan. Bull. Entomol. and threatened in Georgia), as well as re- 7. Cassar, S., and Blackwell, M. 1996. Conver- Res. 92:119-126. sults of pathogenicity tests presented in gent origins of ambrosia fungi. Mycologia 26. Kinuura, H., and Kobayashi, M. 2006. Death this study, provide support for this con- 88:596-601. of Quercus crispula by inoculation with adult cern. Various species in the Lauraceae are 8. Coker, W. C., and Totten, H. R. 1945. Trees of Platypus quercivorus (Coleoptera: Platypodi- Southeastern States. The University of North dae). Appl. Entomol. Zool. 41:123-128. common components of forests in other Carolina Press, Chapel Hill. 27. Kubono, T., and Ito, S. 2002. Raffaelea quer- regions of the United States and other 9. Dixon, W. N., and Woodruff, R. E. 1982. The civora sp. nov. associated with mass mortality countries in the Western Hemisphere. For black twig borer, Xylosandrus compactus of Japanese oak, and the ambrosia beetle example, California laurel, Umbellularia (Eichhoff) (Coleoptera: Scolytidae). Fla. Dep. (Platypus quercivorus). Mycoscience 43:255- californica (Hook and Arn) Nutt., is com- Agric. Consum. Serv. Div. Plant Ind. Entomol. 260. Circ. 250. Gainesville, FL. 28. Kuhnholz, S., Borden, J. H., and Uzunovic, A. mon in California and Oregon (46). Addi- 10. Dunkin W. H., and Dunkin, M. B. 1988. Trees 2001. Secondary ambrosia beetles in appar- tionally, there is great diversification of the of the Southeastern United States. The Univer- ently healthy trees: Adaptions, potential causes Lauraceae in Central and South America, sity of Georgia Press. Athens, GA. and suggested research. Integr. Pest Manage- and numerous species of this family are 11. Dute, R. R., Miller, M. E., Davis, M. A., ment Rev. 6:209-219. major components of tropical forests there Woods, F. M., and McLean, K. S. 2002. Ef- 29. Little, E. L., Jr. 1979. Checklist of United fects of ambrosia beetle attack on Cercis cana- States Trees (Native and Naturalized). U.S. (39). Last, avocado is a species indigenous densis. IAWA J. 23:143-160. Department of Agriculture, Forest Service, Ag- to Central America and is commercially 12. Esaki, K., Kato, K., and Kamata, N. 2004. ric. Handb. 541. U.S. Government Printing Of- grown in Florida and California. Our Stand-level distribution and movement of fice, Washington, DC. evaluation of avocado indicates that it is Platypus quercivorus adults and patterns of in- 30. Luna, L. G. 1968. Manual of Histologic Stain- also susceptible to laurel wilt, although the cidence of new infestation. Agric. For. Ento- ing Methods of the Armed Forces Institute of mol. 6:71-82. Pathology. 3rd ed. McGraw-Hill Book Com- response was less consistent compared 13. Farrell, B. D., Sequeira, A. S., O’Meara, B. C., pany, New York. with other species in the Lauraceae that we Normark, B. B., Chung, J. H., and Jordal, B. 31. Mizell, R., Braman, S. K., Sparks, B., and tested. We are continuing to assess the risk H. 2001. The evolution of agriculture in bee- Hudson, W. 1994. Outbreak of the Asian am- and impact of X. glabratus and the Raf- tles (Curculionidae: Scolytinae and Platypodi- brosia beetle, Xylosandrus crassiusculus faelea sp. to this valuable crop species as nae). Evolution 55:2011-2027. (Motschulsky), is cause for concern. Proc. 14. Fernald, M. L. 1945. Botanical specialties of South. Nursery Assoc. Res. Conf. 39:191-193. well as other species in the Lauraceae. the Seward Forest and adjacent areas of south- 32. Murata, M., Yamada, T., and Ito, S. 2005. eastern Virginia. Rhodora 47:149-182. Changes in water status in seedlings of six spe- ACKNOWLEDGMENTS 15. Francke-Grosmann, H. 1967. Ectosymbiosis in cies in the Fagaceae after inoculation with Raf- We thank James Johnson and Chip Bates (Forest wood-inhabiting insects. Pages 142-206 in: faelea quercivora Kubono et Shin-Ito. J. For. Health Specialists, Georgia Forestry Commission) Symbiosis. Vol. 11, S. M. Henry, ed. Academic Res. 10:251-255. and Laurie Reid and Andy Boone (Forest Health Press, New York. 33. Murata, M., Yamada, T., Matsuda, Y., and Ito, Specialists, South Carolina Forestry Commission) 16. Gilman, E. F., and Watson, D. G. 1994. Persea S. 2007. Discolored and non-conducive sap- for their help in locating sites with wilted redbay, borbonia: Redbay. Fla. Coop. Ext. Serv. Inst. wood among six Fagaceae species inoculated sassafras, pondspice, and pondberry and sharing Food Agric. Sci. Fact Sheet St-436. University with Raffaelea quercivora. For. Pathol. 37:73- their observations and thoughts. We also thank Don of Florida. Gainesville. 79. Duerr (Forest Health Protection, USDA Forest 17. Haack, R. A. 2006. Exotic bark- and wood- 34. Norris, D. M. 1979. The mutualistic fungi of Service, Atlanta, GA) for securing funding to sup- boring Coleoptera in the United States: Recent Xyleborini beetles. Pages 53-63 in: -Fun- port portions of this work. Richard Bryant (Na- establishments and interceptions. Can. J. For. gus Symbiosis: Nutrition, Mutualism, and tional Park Service, Timucuan Ecological and Res. 36:269-288. Commensalism, L. R. Batra, ed. Allanheld, Historic Preserve, Jacksonville, FL) is gratefully 18. Harrington, T. C. 1981. Cycloheximide sensi- Osmun and Co., Montclair, NJ. acknowledged for providing site access and field tivity as a taxonomic character in Ceratocystis. 35. Oliver, J. B., and Mannion, C. M. 2001. Am- assistance. Laboratory assistance from Susan Best Mycologia 73:1123-1129. brosia beetle (Coleoptera: Scolytidae) species (USDA Forest Service, Athens, GA) and Joseph 19. Harrington, T. C. 1992. Leptographium. Pages attacking chestnut and captured in ethanol- Steimel (Iowa State University, Ames) is greatly 129-133 in: Methods for Research on Soil- baited traps in middle Tennessee. Environ. En- appreciated. Last, the senior author is grateful to borne Phytopathogenic Fungi. L. L. Singleton, tomol. 30:909-918. Kathyrn Kromroy (Minnesota Department of Agri- J. D. Mihail, and C. M. Rush, eds. American 36. Paulin, A. E., and Harrington, T. C. 2000. culture, St. Paul) who observed extensive redbay Phytopathological Society, St. Paul, MN. Phylogenetic placement of anamorphic species mortality during a trip to Hilton Head Island, SC in 20. Harrington, T. C. 1993. Diseases of conifers of Chalara among Ceratocystis species and August of 2004 and thought there was more to the caused by Ophiostoma and Leptographium. other ascomycetes. Stud. Mycol. 45:209-222. problem than just drought. Pages 161-172 in: Ceratocystis and Ophio- 37. Rabaglia, R. J., Dole, S. A., and Cognato, A. I. stoma, Taxonomy, Ecology and Pathogenicity. 2006. Review of American Xyleborina (Col- LITERATURE CITED M. J. Wingfield, K. A. Seifert, and J. F. Web- eoptera: Curculionidae: Scolytinae) occurring 1. Atkinson, T. H., Foltz, J. L., and Wilkinson, R. ber, eds. American Phytopathological Society, north of Mexico, with an illustrated key. Ann. C. 1988. Xylosandrus crassiusculus (Mot- St. Paul, MN. Entomol. Soc. Am. 99:1034-1056. schulsky), an Asian ambrosia beetle recently 21. Harrington, T. C. 1993. Biology and taxonomy 38. Radford, A. E., Ahles, H. E., and Bell, C. R. introduced into Florida (Coleoptera: Scolyti- of fungi associated with bark beetles. Pages 1968. Manual of the Vascular Flora of the dae). Fla. Dep. Agric. Consum. Serv. Div. 37-58 in: Beetle-Pathogen Interactions in Carolinas. The University of North Carolina Plant Ind. Entomol. Circ. 310. Gainesville, FL. Conifer Forests. T. D. Schowalter and G. M. Press. Chapel Hill. 2. Atkinson, T. H., Rabaglia, R. J., and Bright, D. Filip, eds. Academic Press, New York. 39. Rohwer, J. G. 1993. Lauraceae. Pages 366-391 E. 1990. Newly detected exotic species of Xy- 22. Harrington, T. C. 2005. Ecology and evolution in: The Families and Genera of Vascular Plants leborus (Coleoptera: Scolytidae) with a revised of mycophagous bark beetles and their fungal II. K. Kubitzki, J. G. Rohwer, and V. Bittrich, key to species in eastern North America. Can. partners. Pages 257-292 in: Insect-Fungal As- eds. Springer-Verlag, Berlin. Entomol. 122:93-104. sociations: Ecology and Evolution. F. E. Vega 40. Rollins, F., Jones, K. G., Krokene, P., Solheim, 3. Batra, L. R. 1967. Ambrosia fungi: A taxo- and M. Blackwell, eds. Oxford University H., and Blackwell, M. 2001. Phylogeny of nomic revision and nutritional studies of some Press, Inc. New York. asexual fungi associated with bark and ambro- species. Mycologia 59:976-1017. 23. Jacobi, W. R, Koski, R. D., Harrington, T. C., sia beetles. Mycologia 93:991-996. 4. Beaver, R. A. 1989. Insect-fungus relationships and Witcosky, J. J. 2007. Association of Ophi- 41. Sargent, C. S. 1922. Manual of the Trees of in the bark and ambrosia beetles. Pages 121- ostoma novo-ulmi with Scolytus schevyrewi North America. Dover Publications, Inc. New 143 in: Insect-Fungus Interactions. N. Wilding, (Scolytidae) in Colorado. Plant Dis. 91:245- York. N. M. Collins, P. M. Hammond, and J. F. Web- 247. 42. Schedl, W. 1962. Ein Beitrag zur Kenntnis der ber, eds. Academic Press, London. 24. Jones, K. G., and Blackwell, M. 1998. Phy- Pilzubertragungsweise bei xylomycetophagen 5. Brendemuehl, R. H. 1990. Persea borbonia logenetic analysis of ambrosial species in the Scolytiden (Coleoptera). Sitzungsber. Oesterr. (L.) Spreng. Redbay. Pages 503-506 in: Silvics genus Raffaelea based on 18S rDNA se- Akad. Wiss. ABT. I. 171:363-387. of North America. Vol. 2, Hardwoods. R. M. quences. Mycol. Res. 102:661-665. 43. Scott, J. A. 1986. The Butterflies of North Burns and L. H. Honkala, Technical Coordina- 25. Kamata, N., Esaki, K., Kato, K., Igeta, Y., and America: A Natural History and Field Guide. tors. Agric. Handb. 654. US Government Print- Wada, K. 2002. Potential impact of global Stanford University Press. Stanford, CA.

Plant Disease / February 2008 223 44. Sinclair, W. A., and Lyon, H. H. 2005. Dis- M. Burns and L. H. Honkala, Technical Coor- American Scolytidae (Coleoptera). Great Ba- eases of Trees and Shrubs. 2nd ed. Cornell dinators. Agric. Handb. 654. US Government sin Nat. 37:67-74. University Press, Ithaca, N.Y. Printing Office. Washington, DC. 49. Wood, S. L. 1982. The bark and ambrosia 45. Six, D. L. 2003. Bark beetle-fungus symbio- 47. Webber, J. F., and Gibbs, J. N. 1989. Insect beetles of North and Central America (Coleop- ses. Pages 99-116 in: Insect Symbiosis. K. dissemination of fungal pathogens of trees. tera: Scolytidae), a taxonomic monograph. Bourtzis and T. Miller, eds. CRC Press, Boca Pages 161-193 in: Insect-Fungus Interactions. Great Basin Nat. Mem. 6:1-1359. Raton, FL. N. Wilding, N. M. Collins, P. M. Hammond, 50. Wood, S. L., and Bright, D. E. 1992. A catalog 46. Stein, W. I. 1990. Umbellularia californica and J. F. Webber, eds. Academic Press, Lon- of Scolytidae and Platypodidae (Coleoptera), (Hook. And Orn) Nutt. Pages 826-834 in: Sil- don. Part 2: Taxonomic index. Vol. A and B. Great vics of North America. Vol. 2. Hardwoods. R. 48. Wood, S. L. 1977. Introduced and exported Basin Nat. Mem. No. 13.

224 Plant Disease / Vol. 92 No. 2