Carcinomycetaceae: a New Famity in the Heterobasidiomycetes

carcinomycetaceae: a new famity in the Heterobasidiomycetes

Franz Oberwinkler and Robcrt Bandoni

Qbeminkler, F. & Bandoni, R. 1982. Carcinomycetaceae: a new famity in the Heterobasidiomycetes. - Nord. J. Bor. 2: 501-516. Copenhagen. tSStt OIOT-OSSX.

A group of holobasidiate mycoparasites, growing on Basidiomycetes, was studied by light- and electron microscopy. The parasites develop haustoria which penetrate thä host-cells. Often a hypertrophic, gall-like growth of the host is induced by this parasitic interaction. Ana- and tel€omorph stages and functions are highly diverse ffi and not yet fully understood. Two species produce unique zygoconidia. The mechanism of spore abstriction on apically sterigmate holobaaidia is hymenomycet- ous or gasteromycetous, and the basidiospores usually germinate by budding. Doli- pores without parenthesomes were found in three species. A heterobasidiomycetous relationship is proposed for the three genera, Christiansenia, Syzygospora, aid Car- cinomyces gen. nov. Taxonomic considerations led to the proposal of a new family, the Carcinomycetaceae. This is Part 11 of a series, "Studies in Heterobasidiomycetes" of the Institut für Biologie _I, Lehrstuhl Spezielle Botanik, Universität Tübingen and the Department of Botany, University of British Columbia.

{ Qbgywlnkler, Lehrstuhl Spezielle Botanik, lJniv. Tübingen, Auf der Morgenstelle l, D-7!q0 !ü!\Cel, Federal Republic of Germany. R. Bandoni, Dept of Boiany, Univ. of Bitish Columbia, Vancouver, Canada V6T 2Bl

Introduction third species, Christiansenia mycophaga (M. P. Christ.) Boidin, a taxon described in the genus Ceratobasidium The proposal of the genus Christiansenia by Hauerslev by M. P. Christiansen (1959). When Ginns & Sunhede (1969) drew attention to species with highly uncommon (1978) investigated mycotrophic fungi on Collybia characters for Basidiomycetes. Christiansenia pallida dryophila (Fr.) Kummer they found characters reminis- Hauerslev grows parasitically on Phanerochaete cremea cent of those ol Chrßtiansenia and consequently de- (Bres.) Parm. and interacts wilh the host by special cided to emend this genus for incorporation of peck,s haustoria. The parasite is holobasidiate, has Tremella mycetophila and two undescribed species. Our hymenomycetous spore abstriction, and a curious con- taxonomic investigations of mycoparasitic Heterobasi- idium development with fusion of monokaryotic con- diomycetes were extended to this group several years idium-initials to form a dikaryotic propagule. These ago and included cultural experiments and ultrastruc- zygoconidia were already known from Syzy6 ospora alba tural studies. Martin (1937), who, however, originally interpreted rhe structures as special heterobasidia. Additional material which later became available enabled Kao (1956) to discover the true basidial stage and to unravel the ana- Melerials and methods morph-teleomorph connections. Boidin (1970) re- For descriptions and illustrations of the species, the studied and reinterpreted Christiansenia and Syzygos- following collections were used: pora, concluding that they are congeneric and that the name Christiansenra should be accepted and Syzygos- C.arc.ingmycgs eflibularu & Sunhede) Oberu. pora used only for the lCinns & Ban- anamorph. He also included a doni; Canada. British Columbia. Vancouver, campus of Simon

@ NORDIC JOURNAI- OF BOTANY

NORD. J. BOT. 2: 501-516, MyCoL 022 33 Nord. J. Bot. 2 (5) 1982 501 Fraser University, l0 Jul 1980, lcg. J. Hammerbcrg, cult. R. J. Carcinomyces, the type of the family is described in Bandoni. - Holotype: Sweden, Bohuslän, Ucklum par., the present report. Presslorp, 7 Aug 1977, Sonja Hansson (GB). - C. mycetophilus (Peck) Oberu. & Bandoni; as Tremella mycetophila Peck; USA, New York, Essex County, North EIba, Peck. Lectotype (NYS). - Chrisriarcenia pallida Key lo genem Hauerslev; Ostalpen, Osterrich, Kärnten, Karnische Alpen, ca. 1. With zygoconidia; parasitic on corticioid Basidiomycetes 6 km S von Mauthen im Gailtal, etwa 2.5 km NE vom Plöck- enpass, südliche Abhänge des Polinik, WeB zur Unteren Spiel- boden-Alm; ca. 1400-1500 m, auf Fruchtkörpern von Phanerochaete uemea (Bres.) Pam. (on Ä/nm viridr), 18 Jul 1978; leg. S. Michelitsch 'GZU, Herb. Michelitsch). - C. tumefaciens Ginns & Sunhede. - Holotype: Denmark, Sjael- land, Vemmelofte skov. 23 Sept 1977. Karin Toft (GB). - Syzygospora alba Mzrtio; Fungi of Panama, Prov. Chiriqui: Valley of the upper Rio Chiriqui Viejo, alt. 1600-1800 m, I Jul 1935; G.W. Martin 2167; co-type (Herb. State Univ. Iowa, now BPI). - Mexico: Entre los Guayabos y las Cabanas, 15 km al SW de Mazamitla, Carretera a Tamazula, Jalisco; Christiansenia Hauerslev 1969 bosque de Prnas-Quercß, mvy perturbado, en transaccidn con vegetaciön subtropical; alt. 1700-1800 m, 24 Aug 1974; col. Basidiocarps pustulate (Fig. 1), up to I mm diam, rarely G. Guzman 11843 (Herbario de la Escuela Nacional de Cien- larger, 50-200-(300) pm thick, confluent with closely cias Biologicas, Instituto Politecnico Nacional, Mexico, D.F.). neighbouring pustules;, soft gelatinous and wat- ery-hyaline when fresh, ochraceous and horny when For transmission electron microscopy, material was dry, shrinking to an inconspicuous film. Hyphae (Figs fixed in glutaraldehyde and osmium tetroxide, washed 24, l0) thin-walled, hyaline, with clamps, 1.5-3 pm with distilled water, stained in aqueous uranyl acetate, diam, gelatinizing. Hyphal branching by outgrowth of, dehydrated in an ethanol series, and embedded in or from the opposite side of, the clamps. Short side- epoxid resin according to Spurr (1969). Ultrathin sec- branches often globose and with a filamentous and bent tions were mounted on unsupported mesh copper grids, appendage to form a haustorium (Figs 2-8) of the Tre- and examined in a Zeiss EM 9 S-2 transmission electron mella-tyry (Olive 1946). Early developmental stages microscope. usually with zygoconidia (Boidin 1970): terminal, paired short-cylindrical cells simultaneously form bud- like outgrowths (Fig. 2); after further enlargement the outgrowths fuse to form one cell. This cell receives two nuclei, one passing through each of the two connections Carcinomycetaceae Oberwinkler & Bandoni fam. nov. from the original mother cells. The dicaryotic Heterobasidiomycetes in Homobasidiomycetum car- zygoconidium breaks free from lhe mother cells (Figs 2, posomatis parasiticae, tumefactae, raro pustulis distinc- 6, 12); mature zygoconidia are 7-10 x 3-4.5 pm, with tis erumpentes. Hyphae hyalinae, tenuitunicatae, inter slightly thickened cell walls (Figs 2, 1 2). Apparently this cellulas hospitalium crescult saepe cum haustoriis spore stage is important for propagation of the parasite propriis. Ultrastructura septorum hypharum doliporis on appropriate hosts. Later, terminal cells elongate sine parenthesomatis notata est. Cystidia desunt. Hy- considerably (Figs 5, a) with swellings close to the basal phae fertiles in consuetudine ex substrato prodeuntes et clamp and to the apex. These probasidia develop into conidia vel holobasidia basidiosporasque gerentes. Ba- basidia,35-100-(140) x 6-9-(12) pm, with predomin- sidiosporae hyalinae, tenuitunicatae tunicisque levibus antly more than 4 and up to 6 horn-shaped, terminal plerumque cellulas singulas germinant. - Typus fami- sterigmata. The basidiospores, 7.5-1 I x 5-9 pm, grow liae: Carcinomyces Oberwinkler & Bandoni gen. nov. asymmetrically and mature as subglobose cells with a distinct apiculus, thin, unsculptured, hyaline, non-amy- The new family Carcinomycelaceae includes heteroba- loid walls. Spore germination has not yet been observed. sidiomycetous mycoparasites growing on Homobasi- None of the examined fungi had budding basidiospores diomycetes, often causing hypertrophic swellings of the present, and it is thus unclear whether this characteristic host; rarely, the parasites develop insconspicuous pus- is present or not. tules. All hyphae are hyaline and thin-walled, intercel- lular, sometimes penetrating the host cells with charac- All known collections of Christiansenia pallida gtew on teristic haustoria. Cystidia are lacking. Usually the fer- Phanerochaete cremea (Btes.) Parm. (Figs 3, 4, 6). The tile hyphae are produced on the outer surface of the species is reported from Austria, Denmark, France and host and the parasite. Sometimes, however, anamorphs Sweden (Michelitsch 1980, Hauerslev 1969, Boidin are present in internal holes of host fruitbodies. The 1970, Larsson 1972). holobasidiate meiosporangia produce spores on apical sterigmata. The hyaline, thin-walled, smooth basidios- Hauerslev (1969) proposed the genus with one species, pores bud in a yeast-like manner. Christiansenia pallida, and described the most impor-

502 Nord. J. Bot. 2 (5) 1982 T A5mm

I

('hrbtianseniq Iifl:-lj plllida. -.F.ig. .l: The.parasite growing on Phanerochaete cremea, tsually on the underside of the sxbstratum (illustration of inverted.basidiocarp). - Fig. 2. developmental stages of zygäconidia. Fig. 3. young developmental Pifferent - stage of the pmsite showing the-associat-ion with the hosr. - Fig. +. r'ully deüeloped,'äa- and teleom6rph bearin! paraslte in iost hymenium. Fig. 5. Ana- and teleomorphs - descending from the same generative- hyphae; basidiospore's. - Fig. 6l zygmontdiophores asrcciated with, and protruding from the hymenium of the host. tant characters, viz. cells conjugating to form conidia, transferred to Christiansenia by Boidin (1970). Con- basidial morphology, and the haustorial type. Judging idium formation is not well understood in C. from our own investigations and the descriptions of mycophagum, probably because of the poor condition Christiansen (1959) and Eriksson & Ryvarden (1973), of the scanty type material. However, zygoconidia are these characters seem not to be present in not present in the type material; the basidia are not Ceratobasidiwn mycophagum M. P. Christ., a species urniform and do not produce supernumerary sterig-

33' Nord. J. Bot.2 (5) 1982 503 Figs7-_l2 TEM-microsranh\if chris-tia.nseniapaltidu.Allbarsequal 2pm.-Fig.7.Hausroriumpenerratingahosr-cell.-Fis.g. Hauslorial interaction iitü a host-cell..showiny'"; .bliqr;jy '- ;;;;'"tipore. - Fig.-9. Basidiospores. Fig. 10. Short-cytindricat iart ;lä,ii:::T'Jf_-iiliä:t:,r"#i.lli$ii,?ia.-Ed i'i. r,i.ai"; secrion är a ooripore'*iirr'u,nä.d po,. pi,s',;;; i;;ri;; 504 Nord. J. Bor.2 (s) 1982 mata. According to Eriksson & Ryvarden (1973) the tion of Filobasidiella neoformanr cannot be excluded. basidiospores germinate "with globose yeastlike sprout- Both Filobasidium and Filobasidietta cells". floriforme We conclude, therefore, that Ceratobasidium neoformans produce yeast stages during mycophagum their life_ belongs neither in the genus cycles, but such a stage is unknown fot Chrßtiansenia Ceratobasidium nor in Christiansenia. This fungus is pallida. Therefore it would be important to study the highly remarkable because of holobasidia and yeas-t-like developmental stages of this species under culturaicon- budding by basidiospores. To our knowledge, a com- ditions- parable situation (but not consequently a relationship) occurs to some extent in Paullicorticium pearsonii (Bourd.) J. Eriksson, basidiospores of which germinate Syzygospora Martin 1937 with microconidia. Further developmental stages in both species are unknown and therefore a better in- Basidiocarps tremelloid, cerebriform with reticulate terpretation possible is not at the moment. Boidin holes, up to 5 cm broad and 10 cm long or in diameter (1970) was also of the opinion that Syzygospora alba (Figs 13, 14); soft gelatinous and pure white when fresh Martin should be transferred to Christiansenin We ag- (Martin 1937), horny and daik brown when dry. ree that there is a close relationship between these Hyphae thin-walled, hyaline, with clamps, 2-4_(5) pm species, as indicated by the tremelloid haustoria and diam, gelatinizing. Hyphal branching by outgrowth of, zygoconidia. Different characteristics in basidiocarp or from the opposite side of the clamps. §hort side construction, basidial morphology and basidiospore branches often subglobose to short cylindrical with germination make Sy gos po generically zy ra different, filamentous appendages which form hiustoria of the however, as will be explained in more detail below. Tremella-type (Figs 15, 16, 22, 23) and which are at- Christiansenia pallida is true a mycoparasite that de- tached to the hosr cells (Fig. 16). Zygoconidia of the velops haustoria (Figs 7, 8). Similar haustoria are Christiansenia-type (Fig. 15), 5-6 x 2.54 pm, present known to us only in some species of Tremella and re- on outer and inner surfaces of the basidiocarps and fated taxa, species of the Filobasidiella grotp, in often intermixed in the peripheral layers with basidia. Platygloea (Schum. pers.) fimetaria ex v. Höhnei (un- Sometimes several zygoconidia are formed on the same publ. data), as well as in Syzygospora (Oberwinkler a/äc conidiophore (Kao 1956; presenr conrribution Fig. 15). & Lowy l98l ). Parts of the outer surface of the basidiocarp covered Another striking character is found in the septal pore with thickening hymenia (Fig. ZZ) built up ofbasidia of type (Fig. 11), a dolipore lacking parenthesomejbr,t different developmental stages. Very young probasidia with electrondense banding between the central pore- hyphalike, then long, slender, gradualiy swelliog, swellings. Comparable septal pores are present in cystidia-like cells (Fig. 22), mostly with 4 terminal, Filobasidium florifurme Olive (Moore & Kreger-van horn_-shaped sterigmata, the basidia appearing apically Rü_1972) and in species ofFitobasidiella (Kwon-Chung quadripartite (Figs 19, 20). Occasionally 2- anä & Popkin 1976, Khan et al. 1981). This mighr well bä 3-sterigmate basidia can be found (Käo 1956). an indication of a taxonomic relationship between these Basidiospores asymmetrically attached to the sterig- taxa. Tremelloid haustoria are known in Filobasidium mata (Figs 19, 20), subglobose to laterally depresseä, (Olive floriforme 1968) and Filobasidiella neoformans dropshaped,9-11 x 6-8 pm when abstricted, (Fig. 17), (Kwon-Chung 1976). However, these species strongly thin- and smooth-walled, hyaline, non-amyloid. Germi- differ in having gasteroid basidia which produce sessile nation mostly by budding (Figs 18, 2l) with a scar-like spores. rupturing of the mother cell wall, sometimes with Though the type specimen of Fitobasidium floriforme simultaneous multiple budding (Fig. 2l), rarely ger- grew on dead inflorescences of Erianthus giganteus minating by hyphae (Fig. 21). Repetitive ,econJary (Walt.) Muhl. (Poaceae), it is likely thar it was as- spores not observed. sociated with other fungi. There is nothing known, how- ever, concerning a parasitic interaction. The anamorph Syzygospora alba Martin is a parasitic species (Fig. 16) of Filobasidiella neoformans, Cryptococcus neoformais, growing on basidiomycetous hosts (Oberwinkler & has been known for more than half a century as a hu- Lowy 1981) which may belong to corticioid Aphyl- man pathogen, but it was first isolated from peach-juice lophorales. The fungus is known only from a few colläc- (Sanfelice 1894). Ir has been reported by many *oik"., tions from Panama and a single specimen from Mexico. from pigeon droppings and was later isolated from soil (Emmons 1951), plants and plant debris (Staib 1971, The species was first described as an unusual au- Staib et al. 1972), and peach fruit (Staib et al. 1973). To ricularioid Heterobasidiomycete by Martin (1937) who our knowledge, mycoparasitic behavior of this fungus interpreted the zygoconidiophores as basidia. Well de- has not yet been reported. It must be considered that veloped specimens, collected later on, were studied by the various natural habitats from which Cryptococcus Kao (1956) and enabled her to reinterprete the speciej: neoformans is known, also are the ecologicainiches of Conidial and holobasidial stages were found to originate various other fungi. Therefore a mycopaiasitic interac- from the very same hyphae and she concluded that

J. (5) Nord. Bot. 2 1982 50s

I (/ ocDO I ,D Cö 5mm C I 17 I 20pn o3o o I öo )q Da ö

23

Figs.l3-23. Syzyg ospora alba. Figs 13-16.and 19-23 reproduced from Oberwinkler & Lowy ( 1981). - Fig. 13. Fruirbody in a dry condition. - Fig. 14. Section through a fruitbody showing the intemal lacunar system. - Fig.'ts. zygoconiäiophores withiiffereni developmental slages zy8ocont{t, Parasiteiyphae atrached wirh haustoiia -of - ^!ig..19. to ho'sr hyphae. - Fig. 17. Mature basidiospores. - Fig. 1 8. Yeast budding of basidiosporeg.,- !!ei I 9, 20. Upper parts of basidia, differeni irages of on"togeny. - Fig. 2 l. Diverse types-of basidiospore germination. - Fig. 22. Thickening hynieniüm with basidia of different-ages and h"yp(ae wiä haustoria. - Fig. 23. Haustoria. "Syzygospora is probably only a conidial stage of the which make the type responsible for the application of homobasidiomycete". tsoidin (1970), who investigated the name. We also accept this interpretation of Art. 59 cotype material (Martin 2517), detected some basidio- and consequently use the name Syzygospora alba Mar- spores of the shape and measurements (taken from his tin for the teleomorph. illustration) reported by Kao (1956) and intermixed with innumerable zygoconidia. Finally, Oberwinkler & Lowy (1981) studied a collection made in Mexico which is in a fairly good developmental condition. 'I'his speci- Carcinomyces Oberwinkler & Bandoni gen. nov. men was again used for the present, comparative study. In addition to the known characters, several other im- Fungi mycoparasitici, in carposomatis Agaricalium portant features were elucidated. These include gyrose crescunt, cellulas hospitum luxurari adducunt. Carpo- fruitbodies composed of a heteromorphic hyphal sys- somata propria cystidiaque desunt. Hyphae hyalinae, tem, i.e. host and parasite hyphae intermixed, host- tenui-tunicatae cellulis conidia gerentes vel holobasidiis parasite interaction involving haustoria of the tremel- terminant. Basidiosporae hyalinae, ex sterigmatibus in loid type, thickening hymenium, partially com- apice basidiorum ortae, eiectae vel delatae sunt, cel- partmented basidial apex, and dolipores with and with- lulasque singulas germinantes plerumque producunt. - out parenthesomes. The latter finding has to be inter- Typus generis: Chrktiansenia effibulata Ginns & preted very carefully because only herbarium material Sunhede, Bot. Notiser 131: 168 (1978). was available for TEM-studies. However, it appears that dolipores without parenthesomes belong to septa of Etymologia: Kaqx(vog - cancer; prixlE - fungus. the parasite hyphae and those with a homobasidiomy- cetous dolipore to the host. On the basis of all these Carcinomyces is a heterobasidiomycetous genus with data, Oberwinkler & Lowy (198 l) concluded that there species parasitic on Agaricales, presently known only on is obviously a close relationship between Christiansenia Collybia dryophila (Fr.) Kummer. The parasites do not pallida and Syzygospora alba. However, they did not produce basidiocarps by themselves but cause a hyper- accept the decision of Boidin (1970) to include both trophic, cancer-like growth of the hosts, i.e. galls. The species in a single genus and to use the name Christian- hyaline and thin-walled hyphae develop terminal coni- senia alba Boidin ex Martin as the nomenclatorically diophores (Figs 24, 25,26,32) and holobasidia (Figs correct one for the teleo- and holomorph of Syzygos- 24,25,26,30, 31) with apical sterigmata from which pora alba. the basidiospores break off passively or can be forcibly It is obvious that uncommon morphologically and discharged. The basidia of the type species (Figs 24, 25, functionally identical characteristics, e.g. tremelloid 26) are functionally gasteroid and the basidiospores haustoria, zygoconidiophores and conidia, are indi- germinate by yeast-like budding (Figs 24, 26), a beha- cators of a close relationship. However, the other dif- vior which is assumed to be present also in the other fering characteristics of Syzygospora cannot be neg- species of the genus (Fig. 30). lected, e.g. a) the gyrose fruitbody composed of inter- Though generically different, we assume Christianse- mixed host and parasite hyphae, b) morphology of the nia pallida and Syzygospora alba lo be closely related mature basidial apex, including the 4 sterigmata, and c) because of the presence of tremelloid haustoria and zy- basidiospore germination by budding. The latter feature goconidiophores. Conidia in Carcinomyces spp. seem is uncertain for Christiansenia pallida, basidiospores of to be of two types, one of which is aleuriospore-like and which have never been seen to produce buds by differ- is here referred to as chlamydospore (Figs 25,26: ar- ent workers (Hauerslev 1969, Boidin 1970, Eriksson & rows), the second being blastoconidia (Fig. 24: arrow; Ryvarden 1973, Michelitsch 1980) and by ourselves. It Fig. 32). These types of conidia differ from zygoconidia, is our opinion that these differences should be accepted although the latter is certainly blastogenously produced. not only for a specific, but for generic separation of the Further, tremelloid haustoria could not be found in spe- two taxa. cies of Carcinomyces but haustorial hyphae were obser- Boidin (1970) gave comprehensive arguments on the ved (Figs 25,28,31,33: double arrows). Also, the basi- nomenclatorial problem concerning the correct generic dial morphology is quite distinct, at least in those spe- name of the teleo- and holomorph of Syzygospora alba. cies which lack horn-shaped sterigmata and have basi- We also agree, as did Oberwinkler & Lowy (1981), with diospores which are almost symmetrically attached Boidin's reinterpretation of Martin's Fig. I a ( 1937) as a (Figs 24, 25,26,30,31). Finally, there is a peculiar young basidium and not a cystidium. This clearly indi- yeast type growth in the generic lype, Carcinomyces cates that the type material does contain the teleo- effibulatus (Figs 27,29), with slender cells similar to morph, as well as masses of conidiophores and conidia. those in Exobasidium (Blaaz 1917,1978) and thus dif- Despite Martin's misinterpretation of the conidiophores fering markedly from the tremelloid budding in Syzy- and failure to note the true basidia, Oberwinkler & gospora alba (Figs 18, 2l). It has yet to be determined Lowy (1981) accepted an interpretation of Art. 59 of whether or not Carcinomyces mycetophilus and Christi- the Nomenclatorial Code by Weresub et al. (1974) ansenia tumefacrezs possess this characteristic or not.

Nord. J. gor.2 (5) 1982 507 _ t-1 N.l M -: (t)/ I \1 I eum rN Ry\ tht/ llD r N\I \H 0\tl. v /\ll i 'ilW ,.\\J k \, \\ \ \rA J____- / a= ('YA', ütr r/ - \\ / li I F/, (t rlrr)\k,"i' ulllJ/ il n il,,ll Nl I u o!50 29\! 5"

J/W,, Figs 2.1-29. Carci.nomyces effibulatu. Figs 24-26. Generarive hyphae Arrow: blastoconidiophore: budding basidiospore. - Fig. 25. Arruw: t Upper left corner: budding basidiosporel arrow: chlamyäospore? _ Fig. attacked by haustoria (double arrow) of the parasite. _'Fis.'29. yeast-s lgs 30-33. Carcinomyces mycerophilx. Fig. :0. Oasidiäspore atract Uenerative. hyphae with different developmental stages of bäsidia; basi hyphae with attached hausrorium (doublä arrow). However, the above features clearly indicate that Chri- lection of Christiansenia tumefociens is known. Though stiusenid pallida and Syzygr.tspora alba are sufficiently this material is well developed, further specimens distinct as to be placed in separate genera. should be available for determining the characteristics On the other hand, some highly important similarities of this species. Therefore we are not yet prepared to are evident among the three genera. All species are transfer the taxon to another genus. mycoparasites of Basidiomycetes. The host-parasite- The following new combinations are proposed: interaction is established by haustoria penetrating into the cells (Figs 7, 8,38). A hypertrophic growth, often Carcinomyces (Ginns macroscopically visible as galls, is the peculiar host re- effrbuletm & Sunhede) Oberu. & Bandoni sponse to species of Syzygospora and Carcinomyces. All comb. nov. species bear conidial stages though these are distinct in Basionym: Chrßtiansenia effibulata Ginns & Sunhede, Bot. ontogeny and morphology. Notiser 131: 168 (1978). Another significant element which was examined du- ring our investigations was the septal pore apparatus. Dolipores without parenthesomes (Figs 11, 34-37) Carcinomyces mycetophilus (Peck) Obem. & Bendoni were found in Christiansenia pallida and Carcinomyces comb. nov. effibulatus and can be assumed for Syzygospora alba Basionym: Tremella (Oberwinkler Lowy these mycetophila Peck, Rep. New York State & 1981). AII characteristics Museum 28: 53 (1879). together illustrate and justi§ a natural grouping of the taxa presented in. this report. Carcinomyces effibulanu is easily separated from C. According to Ginns & Sunhede (1978) 3 species can mycetophilus and Christiansenin tumefaciens by its lack be recognized as parasites on Collybia dryophila. All of clamps (Figs 2a-26) and the mainly two-sterigmare cause identical tumors on the host. However, the basidia (Figs 24-26). This species is further charac- authors found adequate microscopic characters which terized by the morphologically and functionally gas- permit sufficient specific distinctions. teroid basidia with symmetrically placed basidiospores. In 1879 Peck described Tremella mycetophila as a This unusual variation was correctly interpreted by parasitic fungus growing on Collybia dryophila. Burt Ginns & Sunhede (1978) and confirmed in the present (1901) recognized some similarities of the parasite with contribution by the study of living fungi. However, we species of the genus Exobasidium and consequently were not able to elucidate the nuclear behavior of the transferred the parasite to this genus. However, he basidia which we interpret as such because of their gen- (Burt 1915) later interpreted the gatls of Collybia dry- eral morphology. ophila as ateratological growth of the mushroom. Ginns We had no living specimens of. Carcinomyces & Sunhede (1978) proposed the transfer of this species mycetophilus available; therefore we can only refer to to the genus Cänstiansenia. In their treatment the latter some morphological features of the basidiospores and authors described two more species, Christiansenia ef- sterigmata (Figs 30, 31). We are inclined to agree wirh fibulata Ginns & Sunhede and Christiansenia tumefa- Ginns & Sunhede (1978) that active spore liberation clens Ginns & Sunhede; they were of the opinion that does not occur in this species. The spore attachment to all 3 taxa are closely related. We agree with this in- the sterigma favors such an interpretation. terpretation based upon important identical features, e.g. mycoparasitism, lack of basidiocarps, hyphal struc- tures and measurements, characters of basidiospores, Mycoparasilism and ontogeny and characteristics of conidiophores and conidia. Chrßtiansenia tumefaciens is like Carcinomyces Though there are mycoparasitic species known from the mycetophilus in being fibulate (Figs 40, 41, 43, 45). Homobasidiomycetes, many more such fungi are found However, this is the only species in the genus which in the Auriculariales and Tremellales. Many species of shows a definite asymmetrical attachment of the the genus Tremella are parasites of other fungi and sev- basidiospores on sterigmata (Ginns & Sunhede 1978, eral grow on Basidiomycetes. There are conspicuous and Figs 4M3); therefore hymenomycetous spore species, like Tremella aurantia Schw. and T. tremel- abstriction is indicated. loidea (Berk.) Mass. on Stereum hirsutum (Willd. ex To the best of our knowledge, no other basidiomy- Fr.) S. F. Gray s. lat. and T. encephala Pers. ex Pers. on cetous genus contains both hymenomycetous and ga- Stereum sanguinolentum (A. & S. ex Fr.) Fr. (Bandoni steroid species, i.e., with abstricted or passively released 1967). Aleurodiscus amorphru (Fr.) Schroet. is infected basidiospores. It must be considered, however, that by the small, pustulate species, Tremella mycophaga such characteristics are not commonly carefully studied. Martin; the closely related species, Z. simplex Jacks. & A breakdown of the abstriction mechanism in hymeno- Martin (Martin 1940, Reid 1970) occurs olAleurodis- mycetous species can be occasionally demonstraled in cus tsugae Yasuda. T. parasitica Fr. was reported from fruiting cultures, an example of which was treated in Clavariaceae (Saccardo 1888) and T. versicolor Berk. & detail by Watling (1971). At present only the type col- Broome (1854) as growing on Corticium species. Fi

14 Nord. I. Bot.2 (5) 1982 509 te

Figs 3't-38.-TEM_ti9r9g1aphg olfar1in_omyces effibulatu. All bars equal 1 pm. Bar of Fig. 34 corresponds also for Fig. 35 and thalofFig.36toFig.3T Figs3zl-37.Dolipores.-'Fig.38.Haustorium(ärro*) ana"h"dror"tost".u.-'rig.:l.rrrr,r-.i'cro$aph of a median section of Collybia dryophild; bar equals I pm.

Nord. J. Bot. 2 (5) 1982 nally, there are intrahymenial parasites, such as 7. tremelloid haustorial morphology. Also, in Car- obscura (Olive) Christ. in Dacrymyces species (Olive cinomyces mycetophilus a clear morphological interac- i946, Christiansen 1959, McNabb 1964) and Sebacina tion of species could be seen in spite of the old her- penetrans Hauerslev (1979), with tremelloid atTinites, barium material (Fig. 33: double arrow). With the aid of also in Dacrymyces species. Tetlagoniomyces uliginosus phase-contrast we were able to detect haustorial organs (Karst.) Oberw. & Bandoni (1981), the type and only in Christiansenia tumefaciens (Figs 43-.15). These species of the genus (Tetragoniomycetaceae; Tremel- structures resemble the tremelloid haustorial type to lales) develops in sclerotia of Rhizoctonit. some extent. The host-parasite-interaction of these species is Short terminal cells which are rich in cytoplasmic realized by haustoria which may penetrate the host contents and which are formed as hyphal branches (Figs cells. Morphologically the haustoriaI type is highly 25, 3l: double arrows) are interpreted by Ginns & characteristic: small subglobose cells subtended by basal Sunhede (1978) as possible organs for obtaining clamps form apical, thread-like and irregularly bent nourishment by the parasite. outgrowths which are capable of penetrating the host More detailed studies are certainly necessary to un- cells. ln Tremella rhytidhysterii Bezerra & Kimbrough derstand whether the haustorial organs of the Christ (1978), a parasite ofRhytidhysterium rufulum (Spreng.) iansenia-Syzygospora-groüp and those of the Car- Speg., the close interaction with the host has been de- cinomyces species are homologous or not. monstrated by electron microscopy. ln Tetragoniomyces uliginosus, haustorial filaments, closely attached to the host cells have been illustrated (Oberwinkler & Ban- Morpholory including ultraslructure doni 1981), and the penetration of the haustorium into Hyphae, sepll and septd pores the host was demonstrated by TEM (unpubl. data). Morphologically identical haustoria are present in In all species investigated in this study, uniform hyphal Chrßtiansenia pallida (Hawrslev 1969, Boidin 1970, systems with thin-walled cells were found. Most cell Michelitsch 1980, and Figs 2-8) and in Syzygospora walls are partly gelatinized, a fairly widespread charac- a/äa (Oberwinkler & Lowy 1981, and Figs 15-18). It is ter among heterobasidiomycetous taxa. This gelatiniza- striking that Filobasidium Jlor{orme (Olive 1968) and tion is responsible also for the consistency of the fruit- Filob as idie lla neofor mans (Kwon-Chung 1 976) produce bodies which are soft gelatinous in living condition and similar tremelloid haustoria. We know from our own hard, horny when dried. There is a continuous inter- studies that Platygloea fimetaria (Schum. ex Pers.) v. grading series from scattered hyphae growing in host Höhnel also possesses this haustorial type. fruitbodies and forming inconspicuous layers on host In Figs 7 and 8 the penetration ofthe haustorium into surfaces (Figs 3,6, 24,25,31,4O,42)to pustulate (Fig. the host cell is shown by TEM-micrographs of median 1) and even cerebriform, multilacunar, tremelloid sections. It appears that a comparatively narrow penet- basidiocarps (Figs 13, 14). ration neck is formed and that inside the host cell the Typically, the septa are fibulate. Only one species, parasitic hypha expands again. Furthermore, in this Carcinomyces effibulatus (Figs 24-26), lacks clamps case, the host plasmalemma surrounds the intracellular and its basidia are typically two-sterigmate, the cells part of the haustorium. appearing to be dikaryotic (Figs 25, 26). Ginns & We conclude from these findings that the specialized Sunhede (1978) understandably compared this with a cells with the tremelloid haustorium morphology may similar situation in Tremella mycophaga (fibulate, function likewise in other species where they are pre- basidia four-sterigmate) and T. simplex (efibulate, sent. It can be seen, nevertheless, that many haustoria basidia two-sterigmate). are not attached to host cells (see Fig. 22) and, addi- We studied the ultrastructure of the septal pore ap- tionally, that haustoria are formed in pure cultures paratus in Chrßtiansenia pallida, Syzygospora alba and without the host. We believe that species with such Carcinomyces effibulatus. Only the latter species could haustoria may at least be potentially parasitic, even be fixed in a living condition; fot Christiansenia pallida, though a host is not known, e.g. Tremella mesenterica approximately one year old herbarium material was Fr. used and the collection ofSyzygospora a/ba was already The unique haustorial type is certainly of high 5 years old when it was fixed for TEM-studies. The taxonomic importance and may also be used for sup- latter species yielded only poor results which are not rageneric classification. reproduced by illustrations here. The parasitic interactions of Carcinomyces species Christiansenia pallida and Carcinomyces effibulatus with Collybia dryophila are not well understood. Ginns agree almost completely in ultrastructural differentia- & Sunhede (1978) found "no evidence thar the para- tion of the septal pore apparatus (Figs ll, 34-37); it ;ites penetrate the host cells". In Carcinomyces ef- appears to us that .§yzygospora alba also shares the cibulatus, we found narrow, ramified parasite hyphae same features. Most characteristic is the lack of with pointed ends which were attached to the host cells parenthesomes in these pores. Median sections show iFig. 38). However, these structures did not show the also striated bandings which occlude the pore channel

14. Nord. J. Bot. 2 (5) 1982 511

I öQop D.Cf{r ildd? ,"((o()i fig!jO-+0. Ch,tistiansenia tumefaciens. Figs 4O,42, 43. Generative hyphae with different developmental stages of basidia. - Fig. 40. B-ig hostcells surro.ulded by generative hyphae olthe parasite. - Fig. 41. Four-sterigmate and-two-sterigriate basidium. - Fi!. 44. Host cells attacked by hyphae of the p_arasite._--Fig- 45. Haustoria. - Fig. 46. Basidiospores; budding 6asidiospores (above); young basidiospore attached ro sterigma (to the left); budding yeasts (below).

Nord. J. Bot.2 (5.) 1982 orifices. Comparable dolipores are already known in totrema. In both, Syzygospora and Christiansenia, an F;lobasidium floriforme (Moore & Kreger-van Rij active spore discharge mechanism can be assumed be- 1972), Filobasidiella neolormans (Kwon-Chung & cause of the asymmetric spore attachment on the Popkin I 976), and in Ä arachnophila Khan et al. 1 98 1 ). sterigmata. Similar structures in septal pores are apparenlly present The Carcinomyces basidium (Figs 24-26,30, 31) is also in Tric hos po ro no ide s oedocepfr a/ls (Haskins 1 975). also long clavate, with a slight basal swelling. Car- Arches of endoplasmatic reticula often cover the cinomyces effibulatus and C. mycetopftilas appear to outer dolipore openings (Figs 34, 36, 37). These struc- possess gasteroid basidia, though they are as yet only tures resemble to some extent those which were found poorly understood and have to be studied in more detail by Martinez (1979) in Moniliella acetoabutens Stolk & in living specimens. Christiansenia tumefaciens has Dakin and M. suaveolens (Lindn. ex Lindn.) v. Arx. An basidiospores which are obliquely attached to the essentially similar dolipore construction is present in sterigmata and probably abstricted (Ginns & Sunhede Itersonilia perplexans Derx, as first shown by Kreger- 1978, and Figs 40-43). This is a unique situation, at van Rü & Veenhuis (1971) and confirmed by our own least within the Heterobasidiomycetes, in which unpublished investigations. It is likely that, in some de- hymenomycetous and gasteroid taxa are so closely con- velopmental stages, the dolipores of species of some nected. Tremellas, e.g. T. mesenterica and T. encephala, show similar structures. Normally, parenthesomes of isolated U-cups are formed in Tremella (Khan 1976, Moore Dispersal 1,978a, Bezerra & Kimbrough 1978) and in species of Basidiospores germination Sirobasidium (Moore 1978b), Tetragoniomyces (Ober- and spore winkler & Bandoni 1981), and Filobasidium incon- Like basidia, basidiospores are also reliable indicators spicuum (unpubl. data). of taxonomic relationships. Hyaline, thin- and smooth- There are other reports of dolipores without walled, non-amyloid basidiospores, as present in species parenthesomes as in Tilletiaria anomala (Moore 1980) of Christiansenia, Syzygospora, and Carcinomyces, are or irregular endoplasmatic reticulum arches as in En- highly characteristic for Heterobasidiomycetes. torrhiza casparyanu (Deml & Oberwinkler 1981). The Homobasidiomycetes, however, have a broad range of latter examples demonstrate the inadequacy of our variation in spore markings, color, and iodine reactions. present knowledge of septal pore characteristics for The typical heterobasidiomycetous repetitive or secon- making well-founded taxonomic interpretations. dary spores have not been found in the group under consideration, but all species of Carcinomyces (Figs 24, Basidia 26) and Syzygospora (Figs 18, 2l) have yeast-like budding of the basidiospores, a germination type not Sexual reproduction plays a major role in the propaga- known in the Homobasidiomycetes. The only living tion of Basidiomycetes, and at least three essential steps material (Carcinomyces effibulatus) which we could in this process, karyogamy, meiosis, and spore produc- examine and culture showed yeast-like growth com- tion are bound to one specialized cell, the basidium. The parable to that found in species ol Exobasidium (Blanz diversity in basidial morphology of extant Basidiomy- l9'77, 1978). ln Syzygospora, germination of the spores cetes certainly expresses a phylogenetically meaningful could occasionally be observed in preserved specimens variation pattern, a subject that has been discussed (Fig. 21), whereas in Chrbtiansenia pallida, basidio- several times (e.g. Rogers 1934,1971, Lindner 1940) spore germination has not been observed. and reinterpreted again at some length by Oberwinkler We are convinced that the principal types of basidio- (1964, 1e77, 1982). spore germination (Oberwinkler 1978), i.e. budding, The species studied in the present report are repetition, conidial formation, and germ tubes, have holobasidiate, though in Syzygospora alba (Figs 19, 20) phylogenetic significance and are therefore also the basidial apex is often partly longitudinally septate, taxonomically important. However, these features must strongly resembling the apex of the basidium in Tre- be interpreted as parts of the whole life cycle of each mellodendropsis tuberosum (Grev.) Craw. (Crawford species and öannot be expected to be absolutely invari- 1954, Corner 1966, Oberwinkler 7972). Two other able criteria. It is noteworthy that secondary spores genera, Metabourdotia (Olive 1957) and Pseudotulas- and/or budding of basidiospores is widely spread in the nella (Lowy 1964), have apically septate basidia that Heterobasidiomycetes and lacking in rhe distinguish these genera from similar taxa that have de- Homobasidiomycetes. Further, it is significant that the finite phragmobasidia (Bourdotia) or holobasidia earliest ontogenetic stages show these fundamentally (Tulasnella). different growth patterns. Certainly, one of the major The long clavate Syzygosporc basidium is predomin- evolutionary problems of the higher fungi has been to antly foursterigmate (Figs 19, 20, 22); in Christiansenia, establish hyphal growth, the basis for progressive fruit- supernumerary sterigmata (mostly 6) are frequent (Fig. body development as well as for substrate utilization. 4) on an urniform basidium resembling that of Sis- Therefore, it is not surprising that only among the

Nord. J. Aot.2 (5) 1982 s13

I Heterobasidiomycetes are there numerous species with Further, spore germination by budding occurs in both very tiny basidiocarps or with no basidiocarps at all. groups and conidial formation is also commonly pre- sent. The meiosporangia are holobasidiatei the mode of basidiospore development, however, is highly diverse, Conidia reaching from typical hymenomycetous to characteristic Recently, increasing interest has been paid to gasteroid patterns. anamorphs of Basidiomycetes and Kendrick & Watl- Because of the obviously different basidial mor- ing's survey (1979) witl certainly draw further attention phology, basidiospore development and, in part, dis- to those important aspects of basidiomycetous mitotic similar spore discharge, we see no justification for in- dispersal mechanisms. It can be expected that the cluding the Carcinomyces group in the Filobasidiaceae. taxonomy of Basidiomycetes will become strongly in- The Exobasidiaceae is another taxon with several fluenced when anamorph stages can be adequately comparable characteristics: they are parasitic, but on compared, i.e. ontogenetic and comparative data are higher plants, often inducing gall formation by the host, more comprehensively available. develop haustoria, and have long, clavate holobasidia In the fungi discussed here, three systems of conidium with a variable number of short, stout sterigmata. The formation are found, and these mitotic propagules often basidiospores are capable of producing a yeast phase, are produced simultaneously with the meiospores. and (micro-) conidia are frequently present. Yet, there Blastic conidium ontogeny is to be expected in Car- are important differences in the morphology of the cinomycetaceae because basidiospore yeast budding in haustoria, the simple pore type, the incurved position of Syzygospora and Carcinomyces occurs in this mode. the basidiospores on the sterigmata, the mature trans- The unique zygoconidia in Syzygospora alba and versely septate spores, and their germination by conidia. Christiansenia pallida also develop by blastic outgrowth However, this group is still far from being well under- from adjacent mononucleate conidiogenous cells (Figs stood, especially in regard to the tropical relatives, e.g. 2, 6, 15). Fusion of these blastic initials forms the the Brachybasidiaceae. binucleate zygoconidium (tsoidin 1970), a propagule In addition to holobasidiate taxa, Phragmobasidio- which finally is released from the supporting bridges. mycetes should also be considered for this comparison Kendrick & Watling (1979) used the term "isthmos- There are significant similarities of several important pore" for this unusual conidium, also described, but not characteristics in the Carcinomycetaceae and Tremella- illustrated, for the auricularioid fungus, Cystobasidium ceae. These include mycoparasitism, often of Basidio- sebaceum Martin & Couch (Martin 1939). mycetes, identical haustoria and a similar range in basi- Hyphae which disarticulate at the septa and thus diocarp complexity, i.e. no basidiocarp to conspicuous perform a thallic-arthric conidial formation were re- cerebriform types. The basidiocarp may have a similar ported by Ginns & Sunhede (1978) for Carcinomyces gelatinous consistency, the basidiospores germinate by species, but Kendrick & Watling (1979) suspect a blas- budding, and anamorphs are commonly present. How- tic development by interpreting the authors' illustra- ever, the Tremella septal pore has several U-shaped tions. Dikaryotic cells, still attached to the supporting parenthesome cups and the meiosporangia usually are hyphae (Figs 25, 26: arrows) apparently are released cruciately septate. Nevertheless, the dolipore channel from the conidiophore by lysis along a layer between with electron dense banding is present in some species the two cell walls. When released, these conidia have a of lheTremella group, or at least it can be found in some broad, truncate base and thus differ from the blas- developmental stages. We also know that taxa closely toconidia. related to Tremella have holobasidia, but these are commonly one-sterigmate. It can be expected that a number of fruitbody-less, intrahymenial parasitic representatives are still to be Taxonomic conclusions discovered, and that they may again extend the varia- -I'he available dala on Christiansenia, Syzygospora and tion of the characters discussed above. Carcinomyces species can be used to some extent for a We conclude that the Carcinomycetaceae are appar- taxonomic interpretation of these taxa. The most im- ently related to the Filobasidiaceae. lt might be ap- portant features are: a) mycoparasitism, b) haustoria, c) propriate to include both families in the Cryptococcales. dolipores without parenthesomes, d) holobasidia, e) On the other hand, obvious relations with the Tremel- budrJing of basidiospores, f) conidial propagation. lales cannot be ignored; several important characteris- Species of the family Filobasidiaceae have several tics also are shared with the Exobasidiales. significant characters in common with the members of Finally, we conclude that the Carcinomycetaceae is a the Carcinomycas complex. Both contain parasites family of Heterobasidiomycetes, rather than of the which mostly have morphologically and certainly also Aphyllophorales. Likewise, we regard the Filobasidia- functionally similar haustorial organs. The septal pores ceae as a heterobasidiomycetous taxon. in taxa of the two groups are ultrastructurally identical, and thus represent, so far as known, a unique type.

514 Nord. J. Bol. 2 (5) 1982 Morphogenesis of. Filobasidiella Acknowledgements We thank Mr S. Michelitsch, Graz, Prof Kwon-Chung, K. J. 1976. - sexual slate of Cryplococcus neoformans. Dr lrr F. Saaib, Berlin, and Dr J Hammerberg, Vancouver' for neoformuns, lhe * 821-833. most valuable fungus specimens which we used for this study' Mycologia 68: of septal complex in We are also indebted to the lat€ Dr G. w. Martin, to Dr B. - & Popkin, T. J. 1976. Ultrastructure (Cryptococcus Lowy. and lhe curatol\ of OB and N YS. who kindly have sent Filobusidiella neoformans neoformans). - J. us imnortunt collectiuns. Miss S. Dinkelmeyer skillfully as- Bacteriol. 126: 524-528. pallida för Sverige ny si\redihe invesligalions with the transmission electron micros- Larsson. K.-H. 1972. Christiansenia - et Svampklubbs Arsskrift 1972: cooe. Porlions oithis study were supp()rled by grants from the basidiomycet. - Göteborgs Nätural Sciences and Engineering Council of Canada (Grant A 4449. of the Basidiomycetes and its 801), and from the De-utsche Flrschungsgemeinschaft (SFB Lindner, D. H. 1940. Evolution of the basidium Mycologia 87, Project N). relation to the terminology - 32: 419477. Lowy, B. 1964. A new genus of the Tulasnellaceae. - Mycologia 56: 696-700. References Martiri, G. W. 1937. A new type of heterobasidiomycete. - J. Bandoni, R. J. 1961. The genus Naemctelia. - Am. Midl. Nat. Washinglon Acad. Sci. 27:112-114. 66: 319-i28. - 1939. New or noteworthy fungi from Panama and Colom- Berkelev. J. M. & Broome, C. E. 1854. Notices on British bia. IV. - Mycologia 3l: 507-518. fun[i. - Rnn. Mag. Nat. Hist. Ser. 2. I3: 396-456. - 1940. Some heterobasidiomycetes ftom eastern Canada. - Bezerra, J. L. & Kimbrough, J. W. 1978. A new species of Mycologia 32:683495. Tremella parasitic o, Rhytidhyslerium rufulum. - Can J Martinez, Ä. f. tsZS. Ultrastructure of Moniliella, Trichos' Bot.56:3021-3033. poronoides and Hyalodendron. - Stud. Mycology 19: Blanz, P. 1977. Vergleichende Merkmalsanalysen an 50-57. Exobosidium-Arten und verwandten Basidiomyceten. - McNabb, R. F. R. 1964. New Zealand Tremellales. - I. N. Z. J. Diss. Univ. Tübingen. Bol. 2: 403--414. pallida ein Erstfund in - I 978. Über die systematische Stellung der Exobasidiales. - Michelitsch, S. 1980. Christiansenia - Z. Mykol. 14:91-107. den Alpen. - Z. Mykol. 46:173-176. resuspin6s et Helero- Moore, R. T. 1978a. Taxonomic significance of §eptal ultras- Boidin, J. 1970. HomobasidiomycÖtes jelly basidiomycötes saprophytes: XII. - Le genre Christiailse- tructure with particular reference to the fungi. - nia Hauerslev 1969. - Bull. Soc. Linn. Lyon 39: 132-137. Mycologia 70: 1007-1024. Burt. E. A. 1901. Structure and nature of Tremella '- 19i8b. Septal ultrastructure in Sirobasidium magnum and mycetophila Peck. - Bull. Torrey Bot. Club 28: 285-287. its taxonomic implications. - Ant. Leeuwenh.45: - 1915. The Theleophoraceae of North America IV. I l3-1 18, Exobasidium. - Ann. Missouri Bot. Gard. 2:627456. - 1980. Taxonomic proposals for the classification of marine Christiansen, M. P. 1959. Danish resupinate fungi' Part I. - yeasts and other yeast-like fungi including lhe smuts - Dansk Bot. Arkiv 19,1: 1-55. botanica marina 23: 361-373. of Corner, E. J. H. 1966. The clavarioid complex of. Aphelaria - & Kreger-van Rij, N. J. W. 1972. Ultraslructure atd Tremellodendropsis. - Trans. Br. Mycol. Soc. 49: Filobasidium Olive. - Can. J. Microbiol. 18: 1949-1951' 205-211. Oberuinkler, F. 1964. Basidientypen niederer Basidiomyce- Crawford, D. A. 1954. Studies on New Zealand Clavariaceae. ten. - Ber. Deutsch. Bot. Ges.77: 114-117. I. - Trans. R. Soc. N. 2.87: 617431. - 1972.The relationships between the Tremellales and the Deml. G. & Oberwinkler, F. 1981. Investigations on Entor- Aphyllophorales. - Persoonia 7: 1-16. rhiza caspatyana by light and electron microscopy. - - §li.fie morphology of the basidium and its significance Mycologia 73: 392-398. for the phyloginy ofihe Basidiomycetes, (Abstr.) - Sec- Emmons, C. W. 1951. Isolation of Cryptococcus neoformans ond lnt. Mycol. Congress Tampa. Fla. from soil. J. Bacteriol. 62: 685-690. - 1978. wasist ein Baiidiomvceti - Z. Mykol. 44: 1l-29' - of the Eriksson, J. & Ryvarden, L. 1973. The Corticiaceae of North - 19E2. The significance of the morphology -baridium Europe 2. - Oslo. in the phylo--geny of Basidiomycetes. - In: Wells' K- & Ginns, J. & Sunhede, S. I 978. Three species of Cärisliazsenia Wells. E. k.lc,tit. Basidium and basidiocarp Evolution' (Corticiaceae) and the teratological galls on Collybia cytology, function, and development, Springcr Verl New dryophifu. - Bot. Notiser l3l, 167-173. York, pp. 9-35. Haskins, R. H.1975. Septal ultrastructure and hyphal - & Bandoni. R. J. 1981. Tetragoniomycetaceae: a new branching in the pleomorphic imperfect fungus Trichos- family in the Tremellales. - Can J. Bot. 59: 1034-1040' poronok)es oed.ocephalis. - Can. J. Bot.53: ll39-1148. - & Lowy, B. 1981 . Syzygosp ora alba Marlin, a mycoparasi- I l5' Hauerslev, K. 1969. Christianseniu pallida gen. nov. A new tic Het'erobasidiomycete - Mycologia'hetercbasidiomycetes 7l: I108-l parasitic Homobasidiomycete from Denmark. - Friesia 9: Olive. L. S. 194ö. New or rare from 43-45. Nonh Carolina. Il. - J. Elisha Mitchell Sci. Soc. 62: 65-71' - 1979. New or rare resupinate fungi from Denmark. - - 1957. Two new genera of the Ceratobasidiaceae and lheir Friesia 1 l: 272-280. phylogenetic significance. - Am. J. Bot. 44: 429435. Kao, C. J. 1956. The cytology of Syzygospora alba. - - i S6A.-An unus-ual Heterobasidiomycete with Tilletia-Like Mycologia 48: ö77-684. basidia. - J. Elisha Mitchell Sci. Soc. 84: 261-?66. Kendrick, B. & Watling, R. 1979. Mitospores in Basidiomy- Peck, C. H. 1879. Report of the botanist. - Rep. New York cetes. - In: Kendrick, B. (ed.), The whole fungus, vol. 2, State Museum 28: 53. Kanaskis II, Nat. Mus. Nat. Sci. Canada, Ottawa, pp. Reid, D. A. 1970. New or interesting records of British 473-545. hymenomycetes. IV. - Trans Br. Mycol. Soc 55: Khan, S. R. 1976. Ultrastructure of the septal pore apparatus 41341r. of Tremella. - J. gen. Microbiol. 97: 339-342. Rogers, D. P. 1934. The basidium - Univ. Iowa Stud Nat' -, Kimbrough, J. W. & Kwon-Chung, J.J. 1981. Ultras- Hist.16: 160-183. tructure of .lci/oäasidiella arachnophih. - Can. J. Bot. 59: - 1971. Patterns of evolution to the homobasidium' - In: 893-897. Petersen, R. H. (ed.), Evolution in the higher Basid]omy- Kreger-van Ri.i, N. J. W. & Veenhuis, M. 1971. Septal pores in cetes, Knoxville,'Univ. Tennessee Press, pp 24l-253 Trichosporon cutaneum. - Sabouraudia 9: 36-38.

515 Nord. J. 8o1.2 (5) 1982

I