827

Two species in the Ceratocystis coerulescenscomplex from conifers in western North America

Michael J. Wingfield, Thomas C. Harrington, and Halvor Solheim

Ah"tntd: Two spec-irs ('Ifrrmfnrysfi'l (lfC dC"Cfiheo fwm \I..'('<.;t('-rnNMlh A.mcrica raarncysri<: rl~(i'I(,lIlIi \1/inrfic1d.Harrington, & Solheim is associated with the North American spruce bark beetleDelldroctonus rufiPCllflis infesting both

Engelmann spruce(Picea enge/mannjjParry) and white spruce (Picea glauca (Moench) Voss) in British Columbia. Thisfungus is a primary component of the bark beetlc mycota and has a relatively high degree of virulence to Engelmannspruce. Ceralocystis douglasii (Davidson) Wingfield & Harrington was previously described asEndoconidiophora

coerulescellsf. douglas;;. It is common on Douglas-fir lumber in western North Americawhere it causes blue stain.Both fungi grow optimally at low temperaturesand produce perithecia only after extendedincubation underrefrigeration. Ceratoc)'sris rufipenniproduces conidiophores mostly in association withperithecia, whereas conidia andconidiophores of C. douglasii are produced abundantly on wood and agar.

Key words: Cerarocysris, Chalara, DefJdrocrofJus, Ophiostoma,bark beetle - fungus interactions, symbiosis.

Resume: Les auteurs decrivent deux especes deCeratocysris de J'ouest de l'Amerique du Nord. LcCeralOc)'srisrufipenni Wingfield, Harrington & Solheim est associe al'insecte de I'ecorce du pin, IeDendrocrol1us rufipennis,quiinfcste I'epinette d'Engelmann(Picea engelmanniiParry) aussibien que I'epinette blanche (Picea glauca (Moench)Voss), enColombie Canadienne. Ce champignon cst une composante principaledu mycota a dcndroctoneset possedeundegre relativemcnt cleve de virulence sur I'epinette d'Engelmann. LcCeralOcysris douglasi; Wingfield & Harrington a

cte precedcmment dccrit comme IeEndoconidiophora coerulescensf. douglasii. II est commun sur les billes de sapindouglas dans I'ouest de l'Amerique du Nord, ou iI cause latache bleue. La croissance optimale des deuxchampignonsse situc aux basses temperatures et ils produisent des perithcces seulement apres de longues periodcs d'incubation sousdes conditions de refrigeration. Le Ceraroc)'sris rufipenni produit des conidiophores surtout en association avec lesperitheces, alors que les ~onidies etconidiophorcs du C. doug/asii sc forment enabondance sur Ie bois et surI'agar.

Mors cIes : Ceratoc)'sris, Cha/ara, Dcndrocronus, OphioslOma, interactions insectes des ecorceset champignons,symbiose.

ITraduit par la redaction]

Introduction

For many years, the genusCeratocystisEll. & HaIst. hasbeen considered part of a group of genera that has becomebroadly known as the Ophiostomaloid fungi in the orderOphiostomatales. This group jncludesCeratocystis sensustriclo, Ophiostoma H.& P. Sydow andCeratocystiopsisUpadhyay & Kendrick. Species within these genera sharevarious morphological characters that facilitate insect disper-sal. For example, most species have elongate ascomatalnecks from which masses of sticky ascospores exude. Inmany cases, their conidial states arc also typified by sticky

ReceivedJuty 8,1996.

M.J. \\'ingfield.1 Tree Pathology Cooperative Programme,Depanment of Microbiology and Biochemistry, University ofthe Free State, P.O. Box 339, Bloemfontein, 9300, SouthAfrica.T.e. Harrington. Departmentof PlantPathology, Iowa StateUniversity, Ames, Iowa 50010, U.S.A.II. Solheim. Norwegian Forest Research Institute, Divisionof Forest Pathology, N-1423, As-NLH, Norway.

Author to whom all correspondence should be addressed.e.mail: mike@wwg3.uovs.ac.za

Can. J. Bot. 75: 827-834 (1997)

I10....-

conidial massesthat arc acquired byinsects (Wingfield et al.1993).

A contemporary viewof the Ophiostomatoid fungi is thatthey represent at least two distinct phylogenetic entities.These are typified byCeratoc)"stis sensustricto and Ophio-stoma. Ceratocystisspecies have anamorphs. inChalaraRabenhorst, with conidia produced in chains by ring wallbuilding (Minter et al. 1982). This is in cpntrastto Ophio-stoma, in which conidia arc produced byapical wall buildingand the most common anamorphs arc treated in genera suchasGraphiumCorda,LeptographillmLagerberg& MelinandSporothrix Hektoen & Perkins (Upadhyay 1981; Moutonet al. 1994).SpeciesofOphiostomaare also unusualamongtrue fungi and unlike Ceratoc)'stissensu stricto in that theycontain celluloseand rhamnose in their cell walls (Rosinskiand Campana 1964; Jewell 1974; Weijman and Dc Hoog1975). In culture, the two groups can easilybe distinguishedby the fact thatOphiostoma spp. tolerate high concentrationsof cycloheximide, whereas Ceratocystis spp. arc extremelysensitive to this antibiotic (Harrington 1981). Recent studiesbased on rDNA sequence data have also confirmed that thesetwo groups arc distinct and phylogenetically unrelated(Hausner et al. 1992, 1993; Blackwell 1994; SpataforaandBlackwell 1994).

@ 1997 NRC Canada

828 Can. J. Bot. Vol.

Figs. 1-6. Perithecia, ascospores, and conidiophores ofCeratocystis rufipenni(holotype). Fig. 1. Pcr.ithecium. Scale bar= 200 Itffi.

Fig. 2. Base of perithecium showing hyphal ornamentation. Scale bar= 200 JLm.Fig. 3. Ostiolar hypae at apex of perithecial neck.Scale bar = 30JLm.Fig. 4. Sheathed ascospores with distinct guttulcs at either end. Scale bar= 14 11m.Fig. ? Elongate tubularconidiogenous cell. Scale bar= 40 Ilffi. Fig. 6. Short conidiogenous cell showing flared apex. Scale bar = 20iLm.

One of the best known species ofCeratocystis is C. coern-leseens (Munch) Bakshi, first described asEndoeonidio-phora coeru/escensMunch from blue stained spruce(Piceaspp.) ane! pine (Pinus spp.) in Germany (Munch 1907). Itwas subsequently recorded in northern Europe, Great Britainand the U.S.A. (Lagerberg et a1. 1927; Davidson 1935,1944; Siemaszko 1939; Bakshi 1950,1951; Roll-Hansen andRoll-Hansen 1980) as a saprophyte or wound colonizer ofsapwood. Ascomata have long necks terminating in ostiolarhyphae and distinct spines ornamenting the bases. Asco-spores are elongate and are sheathed (with an outer and innerwall). The species has aChalara anamorph and tends to pro-duce a strong fruity aroma, presumably for insect attraction.

Considerable confusion has surrounded the taxonomy ofC. coerulescens. For example, isolates resembling thisspecies from hardwoods were described asEndoconidio-phora vireseensDavidson (Davidson 1944). This fungus,which was later transferred toCeratocystisas C. virescens(Davidson) Moreau (Moreau 1952), is the causal agent of animportant disease of maple known as sapstreak (Kile 1993).Subsequent authors, however, treated C.'virescensas a syno-nym of C.eoeru/eseens(Hunt 1956; Grimn 1968; Upadhyay1981). Another species in the C.eoeru/eseenscomplex, con-sidered to be unique by Davidson (1953), was found to occurspecifically on Douglas-fir (Pseudo/suga menziesii(Mirb.)Franco) in the western United States. It was designatedEndoconidiophora coerulescensf. douglash'. This funguswas also treated as a synonym of C.coerulescensby Hunt(1956) and subsequently by Upadhyay (1981). Unfortu-nately, the re-description of C.eoeruleseellsby Upadhyay(1981), which is the most recent treatment of this fungus, isconfused by the fact that data were taken from an amalga-mated group of isolates that probably represent a number ofdistinct taxa. Davidson (1955) reported the presence of anEndoconidiophorasp. similar toE. coerulescens(includingE. coerulescensf. douglasii) among fungi associated withEngelmann spruce(Picea engelmanniiParry), infested withthe bark beetleDelldroe/onlls rufipenllis (Kirby) (D. engel-mannil Hopk.) in the Rocky Mountain area of the UnitedStates. In a more recent study of the fungi associated withD. rllfipenllis in British Columbia, Solheim (1995) com-monly isolated a species ofCeratocystis.

A recent examination of isozyme variation in the C.coeru-/escenscomplex identified many distinct taxa, includingE. coerulescensf. douglasii and theD. rujipennis associate(Harrington et a1. 1996). The aim of the present study wasto describe the latter fungus as a new taxon and to elevate thestatus of the former to that of a distinct species.

Materials and methods

Engelmann spruce trees at Caribou Creek, East Nelson, B.C.,Canada, and white spruce(Picea glauca (Moench) Voss) atMcGregor Rivcr, Prince George, B.C., Canada, infested withD. rujipeflnis, were examined during October and November 1992.

Isolations were made from the sapwood of beetle-infested Engle:mann spruce trees. In the case of white spruce, isolates originatedt"!

from ascospore masses on peritheciaproduced on wood that had<.

heen stored in 11cold room.A siugle isolate (CBS 142.53) of /::.cuerulesU'lls L Jouglasij

from the holotype specimen (FP 70703) collected by Dr. R. W.Davidson from Douglas fir lumber shipped from Oregon was avail-able for study. A second strain (CBS 269.53) proved to be ofanother unidentified Chalara species. Various herbarium specimensof the fungus collected by Davidson and later examincd by Hunt(1956), including the type, were obtained from the Herbarium of.~.

the State University of New York, College of Forestry, Syracuse,'~'New York and the Commonwealth Agricu1tural Bureau, Interna.",tional Mycological Institute (1M!). Six isolates of theCerarocysrissp. from D. rujipcflf/is were examincd in detail, one of which wasinduced to produce perithecia by storing cultures at lOoC. Theseperithecia were then lIsed for morphological comparisons. ~

Temperature requirements for growth in culture were comparedin two isolates of theCerarocYJtissp. fromD. rujipcflflis, the one 1.isolate ofE. coerult'seensf. douglasii and an isolate considered 10bc typicalof C.coerllfesccf/sfrom Scotspine (Pinussyh'e!itrisL.)in England (Harrington et al. 1996). The lalter isolate (C 488) was PIcollected by J. Gibbs on Windsor Estate, Berkshire. England in1988. Mycelial plugs of each isolate were placed Ht the centre ofthree Petri dishes containing 20 mL malt extract agar (MEA; 20 g

>~1DifcD malt extract; 20 g Difco Bacto agar/L w<.lter).The disheswere then placedupside down in incubatorsrangingin temperaturefrom 10 to 35°C at 5° intervals and the fungi allowed to grow inthe dark for 5 days. Two diameter measurements of each colonywere taken at right angles, and avcrJges of the six measurementsfor each fungus and tcmperature were computed.

Results

Detailed examination of theCerato(."ystissp. from D. rufi-pennis and E. coerulescensf. doug/asHhave led us to con-elude that both fungi represent distinct taxa. These taxaappear to grow best at low temperatures and are morphologi-c'

cally similar, but occur in distinct niches. They Canbe distin-'f~guished from each other as well as from otherCeratocystis

:ispp. occurring on conifers on the basis of a number of mor-phological characteristics. Isozymes also distinguish thesetwo new taxa (Harrington et al. 1996).We provide the fol-lowing description for the newCeratocystissp. associated \with D. rufipemds on spruce in western North America:

Ceratoeystis rufipetlni Wingfield, Harrington, & Solheimsp.nov. Figs. 1-8

Segregata in malti extracti agaro variabilia ratione incre-menti, ratione optimaad 20'C attingente30 usque60 mmdiam. quinque diebus. Incrementum ad 25°C mu\tum restric- :1turn, nullo incremento cveniente ad 30°C. Coloniae atro-viridule olivaceae (Ridgeway 1912; imago XXX) colore,

"initiopallide coloratae et atricolorescentes dum senescunt nullopaene aerio mycelia. Perithecia formantur in porticibus

'f_

scarabei Dendoctroni rufipennis, et aliquando in segregatis;,coacervatis demissis temperaturis (15-20°C) et praesertim

@ 1997 NRC Canada

.1'

Wingfield et al. 829

,..~, 'I!:\'!''''''"''f!..~~_:,' C':"-".'...~1'

-.

-.

@ 1997 NRC Canada

.?~(

830 Can. J. Bot. Vol. 75. 19974

Figs. 9-15. Perithecium, ascospores, conidiophores, and canidia ofCeratoc)'stis douglasii(ex holotype cuhurc CBS 142.53). Fig. 9..t{~{

Perithecium. Scale bar = 80 Jtffi. Fig. 10. Apex of perithecial neck with well developed ostiolar hyphae. Scale bar= 180 11m. Fig. 11.'c

Base of perithecium with hyphal ornamentation. Scale bar = 50Jtm. Fig. 12. Elongate ascospores with sheaths. Scale bar = 10Jtm.Fig 13. Long flexuous, slightly tapering conidiophore formed on wood. Scale bar = 90Jtm. Fig. 14. Short brush-like conidiophorcs.Scale bar = 90 Jtm. Fig. 15. Young, cylindrical, hyaline conidia of variable size. Scale bar= 20 11m.

.,

Figs. 7 and 8. Conidiogcnous cells ofCeratocystis rufipenni (holotypc). Fig. 7. Apex of conidiogcnous cell slightly flared with tornouter wall tissue and cylindrical conidium. Scale bar = 20 Jtffi. Fig. 8. Conidiogenous cell that has undergone pcrcurrent proliferation.SCdk b'lf = 20 Idn.

in vitro in summis penariis ampullis ad 10°C diutius conscr-vatis. Ascosporae solae gignentes colonias ct autofcrtilcs ctautosterilesproportionibusquasiaequalibus.Peritheciarobustaatris basibus 224 -480 ~m (286 ~m) diamctro et ornatabrevibus pigmentiferis angustatis hyphis et collis exiguepallescentibus colore ad apicem versus, 640-1800 ~m(926 ~m) longis, 14-18 ~m (16 ~m) latis ad apicem et24-38 ~m (30 ~m) latis in medio. Colla terminantia incristam ostiolarium hypha rum 23 -44 ~m (34 ~m) longarumet 2 - 3 ~m (2 ~m) latarum. Ascosporae productae vel exiguecurvataefinibus rotundatiset cinctae distincto et comparatelata translucido strato externo vel vagina 6-8 ~m (7 ~m)longae et 2 - 3 ~m (2 ~m) latae vagina exclusa, quae variatab I ,urn usque 2 ,urn longitudine ad extremum utrumquesporarum.Conidiophoratypicaspeciei Chalarae, brunnea,erecta. et quasi tubularia sed aliquot eorum sunt exigueattenuata ad apices apertos versus qui aliquando ex textolacerato constituti sunt et quoque aliquando prolificant gig-nentes novum locum conidiogenum 64-190 ~m (133 ~m)longum et 5-6 ~m (6 ~m) latum. Conidia hyalina, rectan-gula vel exique doliiformia et valde variabilia longitudine,7 -17 ~m (10 ~m) longa et 5 -10 ~m (7 ~m) lata, manifestiusdoliiformescentia dum maturescuntet saepe melaizascentia.

HOLOTYPUS:Caribou Creek, East Nelson, ColumbiaBritannica, Canada, ex Piceo engelmamlii infesta D. rufi-penni, Novembris, 1992. Coil. H. Solheim, C 608, 25218,DAOM 222349.

Isolates on malt extract agar variable in growth rate withan optimum at 200C, 30 -60 mm diam. in 5 days. Growth

at 25°C slow, no growth at 300C. Colonies dark greenisholive (Ridgeway 1912, plate XXX), light coloured at firstand becoming darker with age, with virtually no aerialmycelium on MEA. Perithecia forming in galleries of thebark beetle Dendroc/onus rufipemlis, occasionally in isolates.stored at low temperatures (15-20°C), and especially on thei~,glass near the edge of MEA in vials maintained at 10°C..Single ascospores giving rise to both self-fertile and self-sterile colonies in approximately equal proportions. Peri-thecia (Fig. I) robust with black (olive ,green) bases 224-480 ~m (x = 286 ~m) in diameter, ornamented (Fig. 2) withshort, pigmented and tapered hyphae, and with necks thatbecome slightly lighter in color towards the apex, 640-1140~m(x= 926~m)long, 14-18~m(x= 16~m)wideat the apex and 24- 38 ~m (x = 30 ~m) wide at the middle.Necks terminating in ostiolar hyphae 23-44 ~m(x =34 ~m) long and 2 - 3 ~m(x = 2 ~m) wide (Fig. 3). Aseo-spores (Fig. 4) elongate to slightly curved, with roundedends and surrounded by a distinct and relatively broad trans-lucent outer layer or sheath 6-8 ~m(x = 7 ~m) long and,2 - 3 ~m (x = 2 ~m) wide excluding sheaths that vary from',I to 2 ~m in length at each end of the spores. Conidiophorestypical of Chalara species, brown, erect and approximately'tubular (Figs. 5 and 6) but in some cases slightly taperedtowards theopenapices,whicharesometimescomposedoftorn tissue (Figs. 6 and 7) and also occasionally proliferate,;,(Fig. 8) to give rise to a new conidiogenous locus, 64-

,190 ~m(x = 133 ~m) long and 5-6 ~m(x = 6 ~m) wide,.Conidia at first hyaline, rectangular to slightly barrel-shaped~

@ 1997 NRC Canad3.~

Wingfield ef al. 831

'.

"

"

-:-- -@ 1997 NRC Canada

I~

832 Can. J. Bot. Vol. 75.199.i

Figs. 16-18. Conidiophore and conidia of Cerarocystis douglasii. Fig. 16. Apex of conidiophore showing percurrent extensions. Scale..bar = 20 ~m. Fig. 17. Chainsof elongate conidia becoming slightly barrel shaped. Scale bar = 20 f(ffi. Fig. 18. Older, barrel-shapedto round conidia, in some cases~ith thickened and darkly pigmented walls. Scale bar= 20 pm.

and very variable in length, 7- 17 pm (x = 10 pm) long and5-10 pm (x = 7 pm) wide, becoming more obviouslybarrel-shaped with maturity and often developing pigmented'walls.

MATERIAL EXAMINED: HOLOTYPE: Caribou Creek, EastNelson, British Columbia, Canada, fromPicea ellgelmarmiiinfested withD. rufipellllis, November 1992. Coil. H. Sol-heim, C 608,252/8, DAOM 222349. PARATYPE. CaribouCreek, East Nelson, British Columbia, Canada, fromPiceaellge/mallnii infested with D. "'fipellllis, November 1992.ColI. H. Solheim, C 609, 252/8, DAOM 222350; CaribouCreek, East Nelson, British Columbia, Canada, fromPieeaellge/malllli; infested with D. rufipellnis, November 1992.Coil. H. Solheim, C 610, 28017, DAOM 222351; CaribouCreek, East Nelson, British Columbia, Canada, fromPiceacligeimamlii infested with D. rujipcnnis, November 1992.ColI. H. Solheim, C 611,258/8, DAOM 222352; McGregorRiver, Prince George, British Columbia, Canada,fromPiCl'a g/auca infested with D. "'fipell/lis, February 1993.Coil. H. Solheim, C 613, 404/2, DAOM 222353.

Morphological characters as well as isozyme data (Harring-ton et 'al. 1996) for the fungus known asE. cocru/eseellsf.doug/asii are quite distinct. It appears to occur commonly onDouglas-fir lumber in western North America. We elect toraisethe status of this fungus from form to specieslevel andredescribe it as follows:

Ceralocyslis doug/asii(Davidson) Wingfiel~ & Harringtoncomb. et stat. nov. Figs. 9 - 18

== Endoconidiophora coerulescens Munch f. douglasii.Mycologia 45: 584. 1953. (Basionym)

Emended description: Relatively slow growing on MEA,with a growth optimum between 20 and 25°C reaching

24 mm diameter in 5 days. No growth occurring at 30°C.Colony light coloured initially and becoming dark greenisholive (Ridgeway 1912; plate XXX) with age, with virtuallyno aerial mycelium. Perithecia (Fig. 9) only observed in cul-ture after extended incubation at 10°C but abundant onwood, robust, with black (olive green) bases 160-320 I'm,(x = 245 pm) wide and ornamented (Fig. 11) with distinct~short, pigmented and tapered hyphae 60-140 pm(x =138 pm) long. Necks black at the base, 610-1080 pm(x = 788 pm) long, tapering slightly to a somewhat lighterapex, 10-13 pm(x = 12 I'm) wide at the apex, 17-29 pm.(x = 23 pm) wide at the middle and terminating in a crestof ostiolar hyphae (Fig. 10) 14-25 pm(x = 19 pm) longand 1-2 I'm (x = 2 pm) wide. Ascospores .(Fig. 12) elon-gate to slightly curved with rounded ends and surrounded bya distinct outer layer or sheath, 5-9 pm(x = 7 I'm) longand 4-6 pm (x = 3 pm) wide excluding sheaths, whichaverage I pm at each end of the spores. Conidiophorestypical of Cha/ara species, brown, erect and of two distinctkinds. Long, flexuous (Fig. 13) and narrow conidiophorestending to be slightly curved and tapering gradually to a rela-tively fine apex 264-560 pm(x = 362 pm) long and 5-7 pm (x = 6 pm) wide, usually seen only on the naturalsubstrate. Short brush-like conidiophores (Fig. 14) abundantboth on wood and in culture, 72- 320 pm (.f = 183 I'm) longand 5-8 I'm (x = 6 I'm) wide, often with torn apertures(Fig. 14), indicative of the production of conidia (Fig. 15)larger in diameter than the conidiogenous cell apex. Conidio-genous cells often extending (Fig. 16) at least once 10 giverise to a new conidiogenous locus:Conidia (Figs. 15 and 17)very variable in length, probably correlated with the lengthof conidiophores from which they originate, cylindrical, hya-line. with two attachment points and produced in chains;

@ 1997 NRC Canada_~'j

I

I

,.

Wingfield et al.

6-21 ~m(x = 13 ~m) long and 5-7 ~m(x = 6 ~m). Withage, conidia becoming barrel-shaped to round and oftenthick-walled and dark in colour (Fig. 18).

HOLOTYPE:BPI 59527, also known as F.P. 70703,Ft. Collins, Colorado, USA, from lumber ofPseudo/sugomenziesiishipped from Oregon, collected by R.W. David-son, 1953. PARATYPES:IMI 56862 aod 1. Hunt 150 (Herbar-ium of the State University of New York, College ofForestry, Syracuse, New York, USA). OTHERMATERIALEXAMINED:BPI 595726. also known as 1. Hunt 144. Bellin~-ham, \\'ashington, USA, from wood of p.\'eUdolsugamellzi-esii, collected by R.W. Davidson, 1953; IMI 56851, alsoknown as J. Hunt 25 (Herbarium of the State University ofNew York), from wood of Pseudotsugamenz;esii; IMI47571, on wood ofPseudo/suga menziesii,collected byD. Wells, 13.10, 1949. Culture from the holotype depositedat the Centraalbureau voor Schimmelcultures as CBS 142.53and in the collection of T.e. Harrington as C 324; DAOM222362 is a dried culture of CBS 142.53.

Discussion

In this study, we have characterized two species ofCerato-cys/is belonging to a group of fungi that we consider torepresent a complex of morphologically similar but distincttaxa, theCeratocystiscoerulescenscomplex. \Ve recognisethat other taxa in the group have yet to be studied anddescribed (Harrington e! al. 1996). Davidson (1953) wasaware of the fact that C. doug/asii was distinct fromC. coerulesceflsand described it as a new form. Similarly,Davidson (1955) noted unique features of the form ofC. coerulescensassociated withD. rufipennis. With betterunderstanding of the ecology ofCera/ocys/isand with theisozyme analyses of the C.coerulescenscomplex (Harring-ton et al. 1996), we can now recognize Davidsoo's fungi asdistinct species.

The two species described here have various features incommonthat distinguishthemfrom othermembersofCera-tocystison conifers. Both species have very robustascomatawith bases wider than those found in other isolates of theC. coerulescenscomplex. The perithecia also have neckslonger than in other species and the new species grow poorlyand fail to produce perithecia at room temperature.

Despite theirsimilarities, C. rufipennican be distinguishedfrom C. douglasii on the basis of numerous morphologicalcharacteristics.Ceratocystis rufipennihas ostiolarhyphaethat are almost twice the length of those found in C.douglasii.It also has distinctive ascospores with a well defined guttuleat either end. Ascospore sheaths in C.rufipenni are betterdefined and about twice as long as those in C.douglasii.Conidiophores are common in cultures of C.douglasii andalso develop profusely on the surface of colonised wood. Incontrast, C. rufipenni produces conidiophores sparingly,oftenonly in associationwith ascomata.Another importantfeature of C.douglasii is that it produces two distinct formsof conidiophore, long flexuous conidiophores that are usuallyonly seen oncolonized wood andshort,brush-likeconidio-phores found both in culture and on the natural substrate.

Although neither species is well studied,Cera/ocys/isrufipellni and C.douglasii appear to differ in their ecology.Ceratocystisdouglasiiis reported as a common saprophyte,specifically on Douglas-fir timber. In contrast, C.rnfipelilli

833

is a pathogen on spruce that is closely associated with animportant bark beetle (Davidson 1955; Solheim 1995).Cera-tocystis douglasiiproduces a strong,fruity aroma in culture,which may be importantin attracting insects vectors innature. This feature is absent in our cultures of C. rujipenni,and it would probably be unnecessary for aCera/oc)'stisspecies that sporulates in bark beetle galleries (Harringtonet at. 1996). Abundant conidiophores appear to be a welldeveloped feature in C.douglasii, and conidia dispersed byinsects arc prohably important for spcrmatization. Develop-ment of ascomata III C. rujipclIlli appears to be lriggered bywounding which may be another derived feature of barkbeetle associates (Harrington et at.1996).

The description of C. rufipenni has provided us with athird example of a Ceratoc)'stis sp. that is closely associatedwith a specificbark beetlevector. The overwhelming majorityof Ophiostomatoid fungi that are carried by Scolytid barkbeetles are species of Ophiostoma and related anamorphs(Upadhyay 1981; Harrington and Cobb 1988; Wingfieldet at. 1993). Until recently, the only recognised exampleofa Ceratocystis carried by a bark beetle was C. laricicola,which is associatedwith Ips cembraeHeer (Redfern et at.1987). More recently, it has been discovered that O.pololli-cum,which is closely associatedwith Ips /ypographus (Furnisset at. 1990; Solheim I992a), is a speciesof Ceratocys/is(Visser et al. 1995) aod also occurs togetherwith a numberof other Ophiostoma spp. (Solheim 1986; Solheim 1992b).This fungus is the only Cerafocystis sp., and also the mostvirulent pathogen among the fungi carriedby I. typographus(Horntvedt et al. 1993; Solheim 1988, 1993).Ceratoeystislaricicola, which is very similar to C. p%nica (Visser et a!.1995; Harrington et at. 1996), also appears to be a highlyvirulent pathogen (Redfern et at. 1987). Similarly, C.rufi-pellni appearsto be considerably more virulent than otherOphiostomatoid fungi occurring in association with D. rufi-pellnis (Solheim and Safranyik 1997). We therefore believethat these bark beetle associatedCeratocys/is spp. and thewood-staining Ceratoc)'sfis sp. on conifers are importantfrom the ecological, biogeographical, and evolutionary biol-ogy standpoints.

Acknowledgements

We thank Joe Steimel and Doug McNew for laboratoryassistance, Christa Visser for assistance in preparing themaouscript, Louis van Ryneveld for preparing the Latindescriptions, and various colleagues and herbaria for supply-ing cultures and specimens used in this study. An anonymous

reviewer also improvedonthemanuscriptconsiderably. Thefinancial assistance of the Foundation for Research Develop-ment (South Africa), the Ernest Oppenheimer Trust, aodMondi Paper Company to M.J.W. and the AgriculturalResearch Council of Norway to H.S. is also acknowledged.

ReferencesBakshi, H.K. 1950. Fungi associated with ambrosia beetles in Great

Britain. Trans. Hr. Myco!. Soc. 33: 111-120.Bakshi, B.K. 1951. Studies on four species of Ceratoc)'stis, with a

discussion of fungi causing sapstain in Britain. Commonw.Myco!. Inst. Myco!. Pap. No. 35. pp. 1-16.

Blackwell, M. 1994. Minute mycological mysteries: the influenceof arthropods on the lives of fungi. Mycologia. 86: 1-17.

@ 1997 NRC Canada

834

Davidson, R.W. 1935. Fungi causing stain in logs and lumber inthe Southern States, including five new species. J. Agric. Res.50: 789-807.

Davidson, R.W. 1944. Two American hardwood species ofEndo.conidiophora described as new. Mycologia, 3-1: 650-662.

Davidson, R.W. 1953. Two common lumber staining fungi in thewestern United States. Mycologia, 45; 579-586.

Davidson, R.W. 1955. Wood-staining fungi associated with barkbeetles in Engelmann spruce in Colorado. Mycologia, 47:58-67.

Fumi",s, M.!\.L, Solheim, II., ,dId Chrislia]I:-CII, E. 1990.1'1'411:-.111;:;.-sion of blue st.ain fungi byIps t)'pographus(Coleoptera: Scolyti-dae) in Norway spruce. Ann. Enlomal. Soc. Am. 83: 712-716.

Griffin, H.D. 1968. The genusCeraloC)'stisin Ontario. Can. J.Bot. 46: 689-718.

Harrington, T.C. 1981. Cycloheximide sensitivity as a taxonomiccharacter inCeratocystis. Mycologia, 73: 625-633.

Harrington, T.C. 1993. Biology and taxonomy of fungi associatedwith bark beetles.In Beetle-pathogen interactions in coniferforests. Edited byT.D. Schowalter and G.M. Filip. AcademicPress. pp. 37 - 58.

Harrington,T.C., and Cobb, F.W., Jr. 1988.Leptographiumrootdiseases on conifers. American Phytopathological Society Press,St. Paul, Minn.

Harrington, T.C., Steimel, J.P., Wingfield, M.J., and Kile, G.1996. Isozyme variation and species delimitation in theCeraro.cystis coerulescenscomplex. Mycologia, 88: 104-113.

Hausner, G., Reid, J., and Klasscn, G.R. 1992. Do galeate asco-spore members of the Cephaloascaceae, Endomycetaceae andOphiostomataceae share a common phylogeny? Mycologia, 84:870-881. .

Hausner, G., Reid, J., and Klassen, G.R. 1993. On the subdivisionof Ceratocystiss.I., bascd on partial ribosomal DNA sequences.Can. 1. Bot. 71: 52-63.

Horntvedt, R., Christiansen, E., Solheim, H., and Wang, S. 1993.Artificial inoculation with Ips t}pographus-associatcdblue stainfungi can kill healthy Norway spruce trees. Medd. Nor. Inst.Skogforsk. No. 38. pp. 1-20.

Hunt, 1. 1956. Taxonomy of the genusCeratocystis. Lloydia, 19:I-58.

Jewell, T.R. 1974. A quantitative study of cellulose distribution inCeratocystisand Europhium. Mycologia, 66: 139-146.

Kile, G.A. 1993. Plant diseases caused by species ofCeratoc)'stissensu strictoand Chalara. In Ceratoc)'stisand Ophiostoma.Taxonomy, ecology and pathogenicity.Edited by M.J. Wing.field, K.A. Seifert, and J.F. Webber. American Phytopathologi-cal Society Press, St. Paul, Minn. pp. 173-183.

Lagerberg, T., Lundberg, G., and Melin E. 1927. Biological andpractical researches into blueing in pine and spruce. Sven.Skogsvardsfoeren. Tidsk. 25: 145-272.

Minter, D.W., Kirk, P.M., and Sutton, B.C. 1982. Holoblasticphialides. Trans. Bf. Mycol. Soc. 79: 75 -93.

Moreau, C. 1952. Coexistcnce des formesThiela~'iopsis etGraphiumchez une souche deCeratocystis major(van Beyma)novo com. Rev. Mycol. 17(Suppl.): 17-25.

Mouton, M., Wingfield, M.J., and Van Wyk, P.S. 1994. Conidiumdevelopment in anamorphs ofCeratoc)'sris sensu lato.S.Afr.J. Sci. 90: 293-298.

Can. J. Bot. Vol. 75, 1997i

Munch, E. 1907. Die blaufaule des Nadelholzes. I-II. Naturwiss-.:

Z. Forst. Landwirtsch. 5: 531-573.~.-If

Redfern, D.B., Stoakley, J.T., Steele, H., and Minter, D.W. 1987.'..>Dieback and deathof larch caused by CeraroC)'stis laricicola sp. (:;~

nov. following attack byIps cembrae.Plant Pathot. 36: 467-480.

Ridgeway, R. 1912. Color standards and color nomenclature. Pub-lished by the author, Washington D.C.

Roll-Hansen, F., and RolI.Hansen, H. 1980. Microorganismswhich invadePicen abies in seasonal stem wounds. II. Asco-JII)'CI.::h'::::.,r'uJlgl impl.:rfl.:di, aud Ua..:ll.:ri"l0GCIII.:rai di:<>..:u:>sion,Hymenomycetes included. Eur. J. For.Pathot. 10; 396-410.

Rosinski, M.A., and Campana, R.J. 1964. Chemical analysis of thecell wall of Ceratoc)'stis ulmi. Mycologia, 56: 738-744.

Siemaszko, W. 1939. Zespoly grzybow towarzyszacych kornikompolskim. (Summary: Fungi associated with bark beetles inPoland). Planta Pol. 7: I-52. .

Solheim, H. 1986. Speciesof Ophiostomataceae isolatedfromPicea abies infestcd by the bark beetleIps t)'pographus.Nord.J. Bot. 46: 199-207.

J

Solheim, H. 1988. Pathogenicity of someIps t}pographus-associated blue stain fungi to Norway spruce. Medd. Nor. Inst.Skogforsk. No. 40. pp. 1-11.

Solheim, H. 1992a.The early stages of fungal invasion in Norwayspruce infested by the bark beclleIps t)1)ographus.Can. J. Bot.70: 1-5.

I

Solheim, H. 1992b. Fungal succession in sapwoodof Norwayspruce infestedby the bark beetleIps t}pographus.Eur. J. For.Pathol. 22: 136-148..

Solheim, H. 1993.Ecological aspects of fungi associated with thespruce bark beetle II's t)pograpJws in Norway. III CeralOc)'stis

and Ophiostof1la. Taxonomy, ecology and pathogenicity.Editedby M.J. Wingfield, K.A. Seifert, and J.P. Webber. AmericanPhytopathological Society Prcss, St. Paul, Minn. pp.235-242.

Solheim, H. 1995.A comparison of blue stain fungi associated withthe North American spruce beetleDt'1lllroClcmus mjipenllis andthe Eurasian sprucebark beetleIps t)1JOgral'hus. Itl ForestPathology researchin the Nordic Countries, 1994. Proceedingsof the SNS-Meeting in Forest Pathology, Skogbrukets Kurs.senter, Bid. Norway, 9-12 August 1994.Edited byD. Aamlid.Aktuelt Sogfosk No. 4/95. pp. 61-67.

Solheim, H., and Safranyik, L. 1997. Pathogenicity of the sprucebeetle associated blue stain fungi, CerarocJstis rufipellllj andLeptographium abietillllm,to Sitka spruce. Can. J. For. Res. 27.In press.

Spatafora, J .W., and Blackwell. M. 1994. The polyphyletic originsof Ophiostomatoid fungi. Mycol. Res. 98: 1-.9.

Upadhyay, H.P. 1981. A monograph of Ceratoc)'stisand Cerdlo-cystiopsis. University of Georgia Press, Athens.

Visser, c., Wingfield, M.J., \Vingfield, B.D., and Yamaoka,Y.1995. Ophiostoma polonicum is a species ofCeratocystis sensustricto. Syst. Appl. Microbiol. 18: 403 -409.

Weijman, A.C.M., and De Hoog,G.S. 1975. On the subdivisionof the genus Ceratocystis.Antonie van Lceuwenhoek. 41: 353-360.

Wingfield, M.J., Seifcrt, K.A., and Webber, J.F. 1993.Ceratocys-tis and Ophiostoma. Taxonomy. ecology and pathogenicity.American Phytopathological SocietyPress. St. Paul, Minn.

@ 1997 NRC Canada