hfycologza, 93(6), 2001, pp. 1081-1101.
0 2001 by The Mycological Society of America, Lawrence, KS 660448897
A phylogenetic redefinition of anamorph genera in Mycosphaerella based on
ITS rDNA sequence and morphology
Pedro W. Crousl
Ji-Chuan Kang
Dtpartmmt of Plant Pathology, University of Stellenbosch,
I! Bag XI, Matieland 7602, South Afica
Uwe Braun
Martin-Luther-Universitat,FB. Biologze, Institut fur
Geobotanik und Botanlscher Garten, Neuwerk 21, D06099 Halle (Saale), Germany
Abstract: Anamorph genera associated with Mycosphaerella are separated on a combination of characters such as conidiomatal structure, the nature and
arrangement of conidiophores, conidiogenesis, dehiscence scars and pigmentation. To date 27 genera
have been reported as anamorphs of Mycosphaerella,
19 of which were included in this study. Based on the
phylogeny of the ITS1, 5.8S, and ITS2 rDNA sequence data, Mycosphaerella proved to be monophyletic. The major group consisted of several small clusters correlating to specific anamorph genera, as well
as other clusters containing an assemblage of anamorph genera. Although Mycosphaerella is monophyletic, it appears that many of the morphological characters defining different anamorph genera evolved
more than once and for the most part do not represent true groups within Mycosphaerella. Nevertheless, the anamorph generic concepts contain valuable
information of practical use to plant pathologists and
mycologists working with these organisms. Based on
these results, as well as a re-evaluation of the criteria
upon which anamorph genera are distinguished in
this complex, a reduced set of informative criteria
and genera are proposed. The degree of scar thickening, darkening and refraction, as well as the presence or absence of pigmentation in conidiophores
and conidia still appear to be useful features delimiting anamorph genera of Mycosphaerella.
Key Words: Cercospora, Cladosporium, ITS phylogeny, Mycosphaerella, Mycovellosiella, Passalora, Pseudocercospora
INTRODUCTION
Bisby and Ainsworth (1942) stated that 'Nature may
make species, but man has made the genera'. To this
Accepted for publication April 19, 2001.
Corresponding author, Email: pwc@maties.sun.ac.za
'
end the genus Mycosphaerella Johanson (Dothideales,
Mycosphaerellaceae) has been linked to at least 27
different coelomycete or hyphomycete anamorph
genera (Kendrick and DiCosmo 1979), 23 of which
were accepted by Crous et al (2000). The anamorphic genera are separated using a combination of
characters including conidiomatal structure (pycnidia, acervuli, sporodochia, etc.), mycelium (presence
or absence of superficial mycelium and texture thereof), conidiophores (arrangement, branching, pigmentation and ornamentation), conidiogenous
cells
(placement, proliferation and scar type) and conidia
(formation, shape, septation, ornamentation, pigmentation and catenulation). One of the anamorph
genera for which more than 3000 names have already
been published is Cercospora Fresen. (Pollack 1987).
Cercospora was last monographed by Chupp (1954),
who proposed a broad concept for the genus, simply
stating if hila were thickened or not, and if conidia
were pigmented, single or in chains. Deighton, Sutton, Braun and several other workers divided the Cercospora-complex into smaller, more morphologically
similar units based on different combinations of the
characters discussed above (Braun 1995a, 1998). The
abandonment of the "Chupp concept" has resulted
in close to 50 genera being recognized in this complex (Braun 1995a). One of the reasons for this was
the strict interpretation of the numerous conidiogenous events as defined by Sutton and Hennebert
(1994), as well as the additional characters discussed
above. In the ophiostomatoid fungi, however, Wingfield et al (1993) found that the number of anamorph genera attributed to Ceratoqstis Ellis & Halst.
and Ophiostoma Syd. & P. Syd. could be reduced from
sixteen to seven based on a re-evaluation of conidium
ontogeny (Wingfield et al 1993). Several anamorph
genera have been found to have species with conidiomata varying from mononematous, scattered conidiophores to sporodochia with a basal stroma, or
from pycnidia to sporodochia to synnemata (Samuels
and Seifert 1987, Seifert and Okada 1990). Based on
similar observations Sutton (1980) and Nag Raj
(1981) saw the need to abandon the distinction between hyphomycetes and coelomycetes, as acervuli
were frequently found to form a continuum with
more stromatic, sporodochial forms. If this plasticity
is taken into consideration when examining the 23
anamorph genera accepted by Crous et a1 (2000),
many appear superfluous.
Crous (1998) was of the opinion that iVycosphaerella is polyphvletic, and that it could possibly be split
along the lines as defined by its different anamorphs.
However, in recent phylogenetic studies done on this
group (Stewart et a1 1999, C:rous et a1 1999, 2000),
~Mjcosphnerellaproved to be mo~lophyletic,having
several well-defined clades correlating with specific
anamorph genera, and other clades corlsisting of a
motley assemblage of species from diverse anamorph
genera. The aims of the present study were thus to
derive a molecular phylogeny of selected anamorph
genera linked to ~Mjcosphner~lla,
and to compare
these data with morphology, re-evaluating the generic concepts of the respective genera studied. A further aim was to derive a new set of characters and
concepts that would enable us to propose a simplified, but workable classificatiorl system, without the
loss of too much taxonomic information.
DLVAamfilijcatio?7 and sequencing.-Genomic DNA Iva, isolated from fungal mycelium (TABLEI) collected directly
from malt extract agar (MEX) (Biolab, Midrand, South .Mrica) plates using a modification of the isolation protocol
of Lee and Taylor (1990). DNA quantification was done by
LT7 spectroscopy using a Beckman DU Series 7500 Spectrophotometer. Buffers and sequencing conditioris were as state d in Crous e t a1 (2000). The purified PCR products were
sequenced using an M I Prism 377 DNA Sequencer (PerkillElmer, Norlvalk, Connecticut), lvith an AABIPRIShI1" Dye
Terminator Cycle sequencing Ready Reaction Kit (Perkin
Elmer, Thrrington, U.K.).
Phjlogenetic a17a!ssis.-The
nucleotide sequences of the
3.XS rRVA gene and the flanking internal transcribed spacers (ITS1 and ITS2) were assembled using Sequence
NavigatorT\' version 1.0.1. (Perkin Elmer, Applied Biosysterns, Inc., Foster City California). Alignment of the sequence files was conducted using the CLUSTAL W software
(Thompson et a1 1994). Adjustments for improvement were
made by eye where necessary. .Uignment gaps were coded
as a fifth base in the parsimony analysis. T h e parsimony
analyses were performed with PAUP* version 4.0b2a (Swoffbrd 1999). The neighbor-joining distance lrlethod was also
used to infer a gene tree. Clade stability waa assessed using
1000 neighbor-joining bootstrap replications. The parsimony tree scores including tree length, consistenc~index,
retention index, rescaled consiste~icyindex and homoplasy
index (TL, CI, RI, RC and HI) were also calcl~lated.
Xlorphology-Isolates were either obtained from other culture collection^ or isolated fi-om infected plant material as
explained in Crous (1998), and deposited in the culture
collection of the Departrnerlt of Plant Patholo,?; Uni1-ersity
of Stellenbosch. South ,Vi-ica (STE-U. T ~ B LIE) , and at the
Centraalbureau voor Schimmelcultures, Utrecht, Nether-
lands. Isolates were cultured on divided plates containing
MEA and carnation-leaf agar ( C U ) (Fisher et a1 1982,
Crous et a1 1992). (:ulture, were incubated at 25 C ullder
near-llltraviolet light, and exarnined tceekll- for 1-2 mo.
Mounts were prepared in lactophenol, examined usitig Nomarski interferewe phase contrast and hright-field phase
contrasL ruicroscopy, and nieasure~ne~lts
made at XI000
magnification.
~Molrcularplzy1ogenj.-Maximum parsimony analysis
was c o n d ~ c t e dfor 383 parsimony-informative characters in the alignment using the heuristic search option with 1000 random sequence input orders for exact solution. The other 265 sites were excluded from
the analysis to avoid fragmentary ambiguities observed in the ITS1 and ITS2 regions of the aligllment
(data not shown). A total of 208 equally most parsi~nonioustrees (MPT) were obtained and compared
fbr the best topolo'q using the Kishino-Hasegawa test
(FIG. 1 ) . The 383 parsimony-informative characters
in the alignment were also analyzed by means of the
neighborjoining (NJ) rrlethod using u~lcorrected
pair frequencies, and evaluated with 1000 bootstrap
replications, which produced a similar tree t o p o l o g ~
(FIG. 2 ) , ~vithhigh bootstrap values for the relevant
clades. Fusnriu?n prolzferntum (Matsush.) Nirenberg
(GenBank X94l'il) was used as outgroup. Trvo species of Didj~nellnSacc. as well as Ran~z~lis$or.cr
Miura
were also included. The latter genus has been lirlkecl
to teleornorphs in Taprsicc (Pers.) Fuckel (Crous et a1
2000). Sequence data of this stud)' were deposited at
GenBank ('2F362048-3620'70), and the alignment in
TreeBase (S.390, h1889).
Based on the ITS rDNA phylogeny obtained for
the taxa investigated in the present study (FIGS.1, 2 ) ,
five clusters were identified. These include a major
morlophyletic ~Mjcosphaerrllacluster (NJ bootstrap =
84, FIG. 2) (with subclusters for cercosporoicl and
coelomycete anamorphs), and two closely related
clusters for Dissocorziu?n de Hoog, Oorschot & Hijrvegen (NJ bootstrap = 100, FIG. 2) and Clado.~$oriz~nz
L,ink anamorphs (NJ bootstrap = 100, FIG. 2)).X
fourth cluster represents Kamulzs;hora spp. rt.ith ;\.Iollzsin (Fr.) P. Karst. ( = 7apesin) teleomorphs (NJ bootstrap = 100, FIG.2). Species of Didymellci formed the
fifth cluster (NJ bootstrap = 88, FIG. 2 ) .
The present data support :l/Ijcosphn~rellcias monophyletic, with the I~issoc.oniurnand Cladosfio~iutnclusters representing sections or subgenera in 1L!~cosphnn-ella. ITS sequences of other related taxa in the
Dothideccles, namely Botr;l;osphneria Ces. & De Not.
(AF 195774),1)othidea Fr. (AF027764), Leptosphnrrin
Ces. & De Not. (U04203), Phaeos$lzurrin I Mivake
(U77359) and Guignardia Viala & Ravaz (AF312008AF312015) from GenBank were found to be nonalignable with those of Mycosphaerella.
i2lorphology.-Several of the c~llturesobtained from
other collections were sterile, thus making it impossible to confirm their identity (indicated as ? in TABLE I). Although cultures of only 19 anamorph genera were available for inclusion in the present study,
numerous generic type and other specimens have
been examined. The genera dealt with in this study
are discussed below.
DISCUSSION
Cladosporium.-Cladospom'um Link (1816) is the oldest generic name for this genus. No monographic
treatment of this important, but complicated genus
has been attempted. Hughes (1958), who examined
numerous types of hyphomycetes described in the
lgth century, reassessed various old generic names
and reduced the following genera to synonymy with
Cladospom'um: Spar-ocladium Chev., 1826; Myxocladium
Coi-da, 1837; Didymotrichum Bonord., 1851 and Heterospom'um Klotzsch ex Cooke, 1877. A review of the
history of Cladosporium has been given by David
(1997). The status of the genus Heterosponum has resulted in much controversy. De Vries (1952) reduced
it to synonymy with Cladospom'um, a conclusion supported by Hughes (1958) and Ellis (1971, 1976).Arx
(1981, 1983) re-introduced Heterospom'um, and McKemy and Morgan-Jones (1990) as well as Braun
(1995,) agreed with this taxonomic concept. David
(1997) examined Cladosponum and Heterospom'um by
means of SEM and clearly demonstrated that the
scars and hila in the two genera are similar, and of
the 'Cladosporium-type', i.e., protuberant with a central dome surrounded by a raised rim. Based on these
results, David (1997) placed Heterosporium in Cladospom'um and proposed the combination Cludospom'um
subgen. Heterospom'um (Klotzsch ex Cooke) J.C. David, a treatment that should be followed.
Arx (1983) considered Arroconidiella J.C. Lindq. &
Alippi as synonym of Heterospom'um. This treatment
is, however, not tenable since Acroconidiella is characterized by having tretic conidiogenous cells (Ellis
1971, David 1997). Furthermore, Arx (1983) also reduced Stenella Syd. to synonymy under Cladospo?ium.
The scars in the former genus are, however, quite
distinct from those of the latter genus (pileate, sensu
David 1993), and these two genera should be retained as separate entities.
Cladospom'um (incl. Heterospom'um) is characterized
and distinguished from other Mycosphaerella anamorphs by its unique type of scars and conidial hila.
FIG. 1. One of 208 equally most parsinlonious trees inferred from 383 parsimony-informative characters of aligned sequences of the 3.8s rRNA gene and flanking ITS1 and ITS2 regions using maximum parsimony a~lalysiswith heuristic search
option and 1000 random sequence input orders (TL = 3873 steps, (:I = 0.312, RI = 0.348, RC = 0.171, HI = 0.688). The
sequence of Fz~sariumprol@utum was used as outgroup to root the tree. Branch lengths are indicated.
The peculiarity and separate position of Cladospo~iurn has also been confirmed in a previous molecular
study (Crous et a1 2000). Descriptions and an expanded key to the species of Cladosporiz~rnavailable
in culture have recently beer1 published by Ho et a1
(1999). The taxonomic concept applied in the latter
paper is rather wide and not solely based on the
structure of loci and conidial hila, so that a few species are included which belong in other genera, viz.,
.M~rouellosielln fulvn (Cooke) Arx [= Clado.sporiurrr
I
I
1 94.47 Phaeoisariopsis 1
Paracercospora
I
I
I
I
2552
462'75 336.33 Stigmina
1 P~eudocercospom
I
I
I
I
I
I
100
,.
,
1453
Mycovellosiella
221 Phaeoramularia
'efogloeopsis
I
I
1 346 Phaeophleospora
I
.,
31 3.76 Scolecosrigmina 1 Phaeophleospora
C'webraunia
~ 9 4 1 7 1 Fusar~umprol?firafum outgroup
I
-.
-
0.05 changes
FIG. 2 . Neighbor-joining gene tree derived from 383 parsimony-informative characters in the alignment of ITS sequences
using uncorrected pair frequencies. The bootstrap values representing 1000 bootstrap replicatiolls are indicated at the nodes
of the tree. The sequence of Fusam'um prolzferatztm was used as outgroup to root the tree.
fulvum Cooke, = Fuluza fulua (Cooke) Cif.], (,'ladorponunz malorum Ruehle [ = Pseudocladosponum k ~ l lermanzanum (Maraqas & I.H. Bredell) U. Braun] and
Fuszcladzum @sum G. Winter [= CZ(~dosporzt~m
carjzgenum (Ellis & Langl.)] Gottwald (the taxonomy of
C. rnalorum and l? effuutum will be treated in separate
monograph of Vtntuna Sacc. anamorphs by Braun
and collaborators).
Human pathogenic " Cladospomurn spp." must be
excluded since they are anamorphs of ascomycetes
belonging to the Herpotnrhzaceae, which has been
confirmed by molecular studies (de Hoog et a1 1995,
Masclaux et a1 1995, Takeo et a1 1995, Untereiner
2000). Some morphologically similar saprobic hyphomycetes are anamorphs of Caproventuria U. Braun
[Wnturia, according to Untereiner (1997)], (Wnturiaceae) and have been placed in Pseudocladosporium
U. Braun (1998).
In the present study iZI. astmma (Fr.) Lindau was
found to cluster in the Cludosporium clade (FIGS.1,
2). As the culture proved to be sterile, further research would be required to elucidate its possible relationship with C. polygonati M.B. Ellis, which also
occurs on this host. Cladosporium (incl. Heterosporium) should be confined to saprobic and phytopathogenic (rarely hyperparasitic) hyphomycetes with cladosporium-like scars and hila and i%fycosphaerella teleomorphs (Mycosphaerellacene). I11 our analyses Clndosporium clustered adjacent to the main
Mycosphaerella cluster, which suggests that it represents a separate section of iMycosphaerella. The issue
of recognizing sections in iLfycosphaerella has been
dealt with by several workers (Arx 1949, Barr 1972,
David 1997), the correct nomenclature of which
awaits further type and molecular studies.
Ramularia and allied genera with colorless corzidiophore~
and conidia.-The genus Ramularia was described by
Unger (1833). Ramularia and allied genera are characterized by having colorless conidiophores and conidia. A monograph of this complex of genera of
colorless cercosporoid hyphomycetes ( Cercosporella
Sacc., Ramularia, 1Veoramularia U. Braun, Neooz)ularin U. Braun, Phacellium Bonord., PseudocPrcosporella
Deighton) has been published by Braun (1995a,
1998). Ouularia Sacc. (aseptate conidia) was reduced
to synomymy with Ramularia sensu Saccardo (multiseptate conidia) (Hughes 1949, Sutton and W7aller
1988, Braun 1998). Ophiocladium Cavara was distinguished from this complex because of its curled conidiophores, and conidia that have somewhat eccentrically positioned scars. This genus was, howlever, reduced to synonymy with Ramularia (Braun 1988a,
Sutton and M7aller 1988), as was supported by molecular data presented by Crous et a1 (2000). Conidium
septation and unusual conidiophore shape with eccentric scars are thus characters to be considered at
the species level.
Passalora Fx, Cercospora, and allied genern with thickened, darkened conidiogenous loci and conidial hila.The genera in this complex are all very similar, and
as more taxa were described, many intermediates
were found that could not easily be allocated to any
one of these. For the purpose of this study these genera are divided into five groups and discussed as such
below.
Asperisporium il/lnubl., Cercosporidium Earl; Dis-
tocercospora Pons & R. Sutton, Passalora, Phaeoisariopsis Ferraris, Prathigada Subrum., Pseudocercosporidium Deighton, Quasiphloeospora B. Sutton,
Crous & Shamoun. The separation of Passalora and
Phaeoisariopsis is rather tenuous, and Braun (1995a)
accepted Phaeoisnriopsis as the sjmnernatous counterpart of Pnssalom.
Passalora was introduced by Fries (1849). Passalora
bacilligera (Mont. & Fr.) Mont. & Fr. (- Cladosporium
bacilligerum Mont. & Fr.), the type species, is characterized by having pigmented conidiophores and ellipsoid-fusiform, obclavate-subcylindric, (0-) 1(-3)septate, pigmented conidia, formed singly. Deighton
(1967) emended the genus Passalora, confined it to
three species occurrirlg on Alnus, and discussed the
generic history in detail.
Earle (1901) described the genus Cmosporidium
[type species: Scolecotrichuv2 euphorbiar Tracy & Earle
= Cercosporidium churtomium (Cooke) Deighton] .
Earle (1902) reduced Cercosporidium to synonymy
with Passalora. Deighton (1967) considered C;mcosporidium "a useful and definable genus, distinct from
Passalorc~",but failed to give clear morphological differences between the two genera. Based on the descriptions in Deighton (1967), Cmosporidium was
only distinguished from Passalora by forming welldeveloped stromata. The taxonomic value of stronlatal formation for the differentiation of genera in
this complex is, however, very d o u b t f ~ ~Arx
l . (1983)
stated that even in Pu.ssalora b a r i l l i p , the type species of this genus, small substomatal stromata are often present. The size of the stromata is often influenced by the size and structure of the substomatal
cavities, which are relatively small in hosts of the latter genus. rlrx (1983) followed the wider concept of
P(~ssalora,which was generally used before Deighton
(196'7) reintroduced Crrcosporidium, and maintained
the latter genus in synonymy with Pas.saloru. Castaiieda and Braun (1989),Deighton ( I 990), and Braun
(1995,) followed Arx's (1983) concept of Passalora
(incl. Cercosporidium). Braun (1995b) discussed the
differentiation of Pa.ssalora and allied genera in detail, provided a re-description of Passaloru ernend.,
and a key to the related genera.
The following additional genera have been reduced to synonymy
. . with Passalora (incl. Cercosfioridium) : Berteromjces Cif. (Deighton 1967) [type ipecies: B. aeneus Cif. = Passalhm cassiae (Henn.) U.
Braun] and Orrophyllum Cif. (Braun et al 1999) [type
species: 0. angelaemariae Cif. = Passalora gliricidiasis
(Gonz. Frag. & Cif.) Castaiieda & U. Braun].
The genus Aspem's@orium,introduced by Maublanc
(1913a, b), resembles Passaloru, but differs in having
verrucose conidia (Ellis 1971, 1976, Arx 1983).
Maublanc (1919a, h) considered Asperisporium cari-
cae (Speg.) Maubl., the type species of this genus, to
be the anamorph of Sphaerella caricae Maubl. ( = Mycosphaerella caricae Syd. & P. Syd.), but this connection has never been proven. The conidiogenous cells
in Aspe-risporium spp. are usually only slightly geniculate and the scars, which agree well with those of
Passalora spp., are usually more or less terminal and
lateral, flat, and little protruding. Asperisporium spp.
are usually easily distinguishable from Passalora spp.,
though there are a few intermediate taxa, e.g., Passalora scariolae Syd. (Deighton 1967) and l? milii
(Syd.) de Vries (1952) with finely verruculose-rugose
conidia. Although no molecular data are available to
assess the validity of this genus, its separation seems
doubtful.
Pons and Sutton (1988) described Distocercospora
to accommodate Cercos@orapachyderma Syd. & P. Syd.
The slightly thickened conidial scars and pigmented
conidia in Distocercospora resemble those of Passalora
spp., and the frequently branched conidiophores in
D. pachyderma are reminiscent of Mycovellosiella Rangel and some Phaeoramularia Munt.-Cvetk. species,
but the conidia are predominantly distoseptate. As
mentioned for taxa in the Pseudocercospora cluster,
distoseptation appears to be important at species level, which questions the separation of Distocercospora
from this complex.
Pmthigada Subram. (Subramanian and Ramakrishnan 1956) is very close to Passalora (Braun 1995a),
but differs, based on the present interpretation of
this genus (Sutton 1994, Braun 1996, Braun and Melnik 1997), in having scolecosporous, pluriseptate,
thick-walled, usually more or less rostrate conidia. Although molecular data are not yet available, this separation also appears uncertain.
Quasiphloeospora (Sutton et a1 1996) was introduced to accommodate Cercospora saximontanensis
Deighton, an unusual cercosporoid species characterized by having large subepidermal sporodochia
(described as acerwli by Sutton et a1 1996), filiform
conidiophores with percurrent as well as sympodial
conidiogenous cells, and pale, but not quite colorless
scolecospores. Sutton et a1 (1996) explained the differences between Quasiphloeospora and Cercospora as
well as Pseudocercospora Speg., but failed to contrast
it with Passalora, which is, indeed, quite similar.
Quasiphloeospora differs from Passalora in having frequently percurrently proliferating conidiogenous
cells and more or less cylindrical, rather cercosporalike conidia. It appears, therefore, that this genus
may also eventually be reduced to synonymy with either the Passalora, or the Pseudocercospora complex.
Given the mode of conidiogenesis, as well as the
slightly thickened conidial scars, it may well prove a
synonym of the latter.
Pseudocercosporidium Deighton (1973) was described to accommodate Cercosporidium venezuelanu m Syd., a dematiaceous hyphomycete with long,
erect, frequently branched conidiophores and cicatrized conidiogenous cells. This species is close to the
Passalora/Mycovellosie22a/Phaeoramulariacomplex,
but the structure of the conidiogenous loci, which
are rather cercosporella-like, is quite distinct. Its status remains unclear.
Phaeoisariopsis Ferraris (1909) was introduced to
accommodate Isam'opsis gm'seola Sacc., the type species, which is a synnematous hyphomycete distinguished from Isariopsis Fresen. s.str. ( = Phacellium
Bonord., see Braun 1998) by having pigmented conidiophores and conidia. Ferraris (1909) included
three additional North American species, which were
later excluded: Phaeoisariopsis grayiana (Ellis) Ferraris [E Fusicladium grayianum (Ellis) Deighton], P
mexicana (Ellis & Everh.) Ferraris [ E Exosporium mexicanum (Ellis & Everh.) M.B. Ellis] and P pilosa
(Earle) Ferraris [ = Morrisog-raphiurn pilosum
(Schwein.) Deighton] (see Deighton 1990). Ellis
(1971, 1976) and Arx (1983) recognized the genus
Phaeoisan'opsis for synnematous hyphomycetes close
to Cercospora, but included some species with large,
dense, non-synnematous fascicles. Deighton (1990)
re-examined and reassessed the genus Phaeoisariopsis. He considered the synnematous arrangement of
conidiophores to be a feature unsuitable for generic
differentiation. Phaeoisariopsis was therefore confined to a few species similar to l? griseola, characterized by non-geniculate conidiogenous cells with scars
lying more or less flat against the conidiogenous
cells. Species with conspicuously geniculate conidiogenous cells were placed in Passalora. Braun (1992,
1995a, b) disagreed and preferred to maintain Phaeoisariopsis as the synnematous counterpart to Passalora. The formation of svnnemata is, as the only feature
for generic delimitation, very questionable. For instance, the genus Pseudocercospora also includes some
synnematous species. Several species originally
placed in Phaeoisariopsis, but with inconspicuous conidial scars, have been reallocated to Pseudocercospora
(Deighton 1990). There are also some other genera
of hyphomycetes with synnematous as well as nonsynnematous species, e.g., Spiropes Cif. (Ellis 1971).
The differentiation between Ramularia and Phacellium (Braun 1998) is, however, not comparable since
the two genera are also distinguished based on other
features. Phaeoisariopsis is undoubtedly heterogenous
and not tenable in the present sense as a synnematous counterpart to Passalora. The key for a future
treatment of this genus is connected with the status
of Phaeoisa~iopsis griseola, the type species. Recent
molecular data (Crous et a1 2000), placed P gn'seola
in the Pseudocmcospora complex, which is surprising
at first sight. The close affinity of I? gmseola and Pseudocercospora spp. is, however, not totally unexpected,
since the scars in this species are very thin, unthickened or almost so, and only very slightly darkened.
In many collections, the conidiogenous loci range
from being inconspicuous to subconspicuous, so that
it is not surprising that Yen (Yen and Lim 1980), referred this species to Pseudocercosfiora. Yen introduced, however, the wrong combination [Pseudocercospora columnam's (Ellis & Everh.) J.M. Yen] and cited Isariopsis gn'seola, which is the oldest, valid name
for this species, as synonym. However, he did not take
into consideration that the genus Phaeoisum'opsis Ferraris (1909) antedates Pseudocerco.sporc~ Speg. (Spegazzini 1910), and that the inclusion of Plzaeoisariopsis gm'seola in the latter genus would reduce Psrudocmcosporn to synonymy with Phaeoisa?-iopsi.~,an action
which was already indirectly done by Sawada (1922)
who transferred Septonema vitis Lev., the type species
of Pseudocercospora, to Phaeoisariopsis. Kecently, type
material of Cercospora soli~naniSpeg. (LPS 918) has
been re-examined. This species was reduced to synonymy with Phaeoisariopsis griseoln by Deighton
(1990). The type material of C. .sol+mnni is also a collection with rather inconspicuous scars. The close affinity of Phaeoisam'opsis and Pseudocercospora was also
recognized by Arx (1983). It appears, therefore, that
I? griseolq the type species of Phaeoisam'opsis, is congeneric with Pseudocercospora. As in the case for Paracercospora Deighton (Stewart et a1 1999), it appears
that genera with pigmented conidia and thin, unthickened or almost so, slightly darkened scars belong in Pseudocercospora. Species with consistent, conspicuously thickened and darkened scars should be
assigned to Passalora. h formal merging of Plzaeoisariopsis and Pseudocercospora should, ho~vever,coincide
with a proposal to conserve Pseudocercospora over
Phaeoisariopsis.
Cercospora. The genus Cercospora was introduced
by Fresenius (in Fuckel 1863) for passalora-like fungi
with pluriseptate conidia. The concept and circumscription of Cercospora was subsequently widened,
and culminated in the treatment of the genus by
Chupp (1954), which included almost all cercosporoid hyphomycetes. In other treatments the circumscription of this genus was redefined and narro\ved.
Saccardo (1880) and Spegazzini (1910) considered
Cercosporn ferruganea Fuckel with pigmented conidia
to be typical for Cercospora, so that Spegazzini (1910)
introduced Cercosporina Speg. for cercosporoid hyphomycetes with hpaline conidia. Deighton (1979)
transferred C. ferruginea to ~M~covellosiella.Sutton
and Pons (1980) dealt with Cercosf~orinain detail, and
reduced this genus to synonymy with (,'rrcospora s.str.
Other taxonomic treatments restricted the genus to
hyaline-spored, scolecosporous species with conspicuous conidial scars, agreeing with Cercospo7-a apii,
which was proposed by Clements and Shear (1931)
to serve as lectotype (Muntaiiola 1960, Deighton
1976, Ellis 1976, Sutton and Pons 1980). This taxonoinic concept is now generally recognized and has
been confirmed by molecular studies (Crous et al
2000), in which species of Cercospora s.str. formed a
well-defined group.
C~rrcosporapen icillata (Ces.) Fresen. [ = C. depazroides (Desm.) Sacc.] is, however, the true type species
of this genus. The complicated history of its typification has been discussed in detail by Braun (1995a).
Braun (1993) referred some species with colorless
or subhyaline conidiophores to Cercospora, and pror ~ Braun for
posed the subgenus H ~ a l o c r ~ c o s p oU.
these taxa. The differentiation between almost or
quite colorless C:ercospora spp. (subgeii. Hjc~locerc.osporn) and Cercosporella spp. has been discussed in detail by Braun (199.3a). Braun (1993) discussed the
differentiation between C~rcospomand Passalom. C:rreospora s.str. has also been circumscribed by Pons and
Sutton (1988), Braun (1993a, h) , and Braun and Melnik (1997).
The structure of' the conspicuously thickened and
darkened conidiogenous loci (scars) and conidial
hila is very important for the identification of C;~rco.cporn spp. Pons et a1 (1985) provided ultrastructural
details of these structures in C,'ercospora brticoln Sacc.
David (1993) examined scar structures of cercosporoid hyphomycetes by means of SEM ant1 proposed
the term " ~ h o s p o r a - t y p e " for taxa ~vith planate
scars. I11 our analysis two strains with S r p t o ~ i nanamorphs (CBS 632.85, 183.97) clustered in the f~rrcosporn clade, apart from Septo7ia tm'tici (STE-U 658). In
other analyses, however, where more Srjf~toriclisolates
Jvere included (G. I'erkley, CBS, the Netherlands,
pers cornm), these specific isolates grouped separate
from the (~ercosporaclade, and have thus n o t been
included in FIG. 3. Although not the focus of the
present study, it appears that Septorin anainorphs
have e\~olveclfrom several lineages within I\.lyco,~phaerella, and a separate study will be required to address
this.
Elletevera Lleighton, Eriocercospora Deighton, Fulvia C$, Mycovellosiella, Sirosporium Bubcik & Seraiwinn., Stenella (gen0-u with supe~jficialserondnr~l112'phne 07- Prert, extensiuelj branched conicliophore.r). IWJcovellosiella was established by Rarigel (1917) to replace I+losielln Range1 (1915) (nor1 Baillon 1887,
.Scroph~ilnriacene).Muntariola (1960) and Deighton
(1974) reintroduced and redefined Rangel's genus.
~ j c o u e ~ ~ o s i rhas
l l a been generally recognized as a segregate of (,'ercospo?-a s.lat. for taxa with conspicuous
conidial scars and assurgent to repent external secondary hyphae on which the conidiophores are
borne terminally and as lateral branches. The conidia
are formed singly or in chains. Arx (1974, 1981) reduced Mycovellosiella to synonymy with Cladosporium,
but later (Arx 1983) maintained it as a separate genus. Scars of ~Uycovellosiellaare quite distinct from
those of Cladospom'um, belonging to the Cercosporatype (David 1993). Ragnhildiana Solheim (1929) and
Cercodeuterospora Curzi (1932) were reduced to synonymy with Mycovellosiella by Muntaiiola (1960) and
Walkmmyces Thaung (1976) by Deighton (1979).
The genus Ormathodium, described by Sydow (1928),
was considered a synonym of Mycovellosiella by Muntaiiola (1960), but, unfortunately, type material of 0.
stjracis Syd., the type species of this genus, is not preserved, so that this synonymy cannot be confirmed.
Muntaiiola (1960) and Deighton (1974) separated
Fulvia from Mycovellosiella by differences in habit
(loosely fasciculate, without any tendency to form
ropes or to ascend leaf hairs in Fulvia), although they
emphasized the close resemblance of the two genera.
Arx (1974, 1981) reduced Fulvia together with Mycovellosiella to synonymy with Cladosporium, but in
1983, he recognized Mycovellosiella as a separate genus and considered Fulvia a synonym of the latter,
since the differences in habit, discussed by Deighton
(1974), are not tenable. Several intermediate taxa exist, e.g., iUycovellosiella brachycarpa (Syd.) Deighton
(1974) and M . trichostemmatis (Henn.) U . Braun
(1999). Braun (1995a) followed Arx's (1983) concept and introduced the combination Mycovellosiella
subgen. Fulvia (Cif.) U. Braun. David (1993) examined the structure of the scars in Fulvia and 1Vfjcoz1ellosiella and found some differences. The scars of
Fulvia have been referred to the Stenella-type (pileate) and those of Mjcovellosiella to the Cercospora-type
(planate). The number of species examined was,
however, too small for a final conclusion. Further
study of a wider range of taxa are necessary to resolve
their scar types. Although Braun (1995a) recognized
Fulvia as a subgenus of Mycovellosiella, it should for
the interim be retained as synonym of Mycovellosiella.
Deighton (1969) described Elletmera and Em'ocercospora for hyperparasitic ~Vycovellosiella(incl. fulvialike) hyphomycetes. Elletmera is characterized by having erect, fasciculate, frequently branched, pigmented conidiophores, slightly thickened and darkened
conidial scars, and usually solitary (rarely catenate)
conidia. Deighton (1969) compared Elletevera only
with Fulvia and Mycovellosiella, and stated that it differs from the latter genus in the fasciculate arrangement of the conidiophores and from the former in
the absence of regular catenation of the conidia.
However, ~M~covellosiella
and Fzclvia are connected by
numerous intermediate taxa and have been united
in one genus. Elletmera is only distinguished from
Mycovellosiella by being hyperparasitic, but there is no
clear basis for a morphological differentiation. The
hyperparasitic habit alone, without any morphological basis for a differentiation, is not sufficient to introduce or maintain separate genera in the complex
of cercosporoid hyphomycetes. Hyperparasitic species are, for instance, also known in Ramularia
(Braun 1998) and Cladospom'um (Ellis 1976). Hence,
Elletmera should not be maintained as a separate genus.
Species of Eriocercospora possess superficial mycelium with solitary, lateral conidiophores, slightly
thickened and darkened conidial scars, solitary mycovellosiella/passalora-like conidia, and are indistinguishable from ~Uycovellosiella.Deighton (1969) compared Eriocercospora with the genus Cercospora, and
stated that it only differs from the latter genus in
having only slightly thickened conidial scars and in
its hyperparasitic habit. Therefore, Em'ocercospora cannot be maintained as a separate genus.
Sirospom'um (BubBk 1912) is morphologically close
to Mycovellosiella and only distinguished by having
thick-walled, dictyosporous conidia. Siros~oriumis
tentatively maintained as a separate genus until some
molecular data become available to establish the value of oblique septa in this complex. Species with thin
walls and without dictyospores should be excluded
since they are undoubtedly congeneric with Mycovellosiella.
Stenella was described by Sydow (1930) and recognized by Ellis (1971, 1976), who reduced Biharia
Thirum. & Mishra (Thirumalachar and Mishra 1953)
to synonymy with this genus. Deighton (1979) followed this concept of Stenella and differentiated it
from il/lvcovellosiella based on the formation of verruculose superficial hyphae and usually roughwalled, catenate conidia. However, many species with
conidia formed singly have been assigned to Stenella,
so that the verruculose creeping hyphae remain the
onlv reliable basis for the differentiation of the two
genera. According to David (1993), the scars in Stenella are pileate and differ from the planate Cercospora-type scars. However, too few specimens have
been examined to determine if this is a good distinguishing character. Molecular data also showed Stenella parkza Crous and Alfenas (anamorph of Mycosphaerella parkzz Crous & M.J. Wingf.), Stenella marasaszz (Crous & M.J. M'ingf.) B. Sutton & Crous (anamorph of Mycosphaerella marasaszz Crous & M.J.
Wingf.) and Mycosphaerella markszz Carnegie & Keane (anamorph unknown) to cluster separately from
the M~covelloszella/Passalora/Phaeoramulanaclades
(Crous et al 2000, 2001). Stenella a r a p a t a Svd. clusi
FIG.3. Clades of Mjcosphaerella spp. identified in the neighbor-joining gene tree derived in FIG. 2, indicating the overlap
in conidiomata and conidiophore arrangement (A), conidium shape, septation and arrangement (B), proliferation of the
conidiogenous cells (C) , and conidial hila and pigmentation (D) . Clade 1 = Pseudocercospora (incorporating Paracercospora,
Phaeoisan'opsis, S t i p i n a , a passalora-like species with slightly thickened scars and Pseudophaeoramulan'a); clade 2 = Pseudocercospora; clade 3 = Passalora (incorporating Mjcovellosiella and Psc.udophaeoramula?-za);clade 4 = Cercospora (presently
ters separately from other species of Stenella, however, suggesting that this genus is polyphyletic in Mycosphaerella. The relationship between Stenella and
the Mycovellosiella/Passalora/Phaeoramularia complex deserves further investigation. Several intermediates between Stenella and Passalora exist, as there
are some Stenella spp. with smooth conidia, etc. It
appears therefore that Stenella should also be seen as
part of the Passalora complex. The molecular data
available in this study, however, are insufficient to resolve this issue.
The relationship between Stenella and Stenellopsis
Huguenin (1966) is also unclear. Stenellopsis is morphologically similar to Stenella. It has single, conspicuously verrucose conidia with scars that are somewhat thickened and darkened, but lacks verruculose
superficial hyphae. However, the formation of superficial hyphae has now been recognized to be of lesser
importance at the generic rank. For instance, it is not
accepted at generic rank in Pseudocercospora (Deighton 1976) or Ramularia (Braun 1998), and must also
be rejected in the iVlycovellosiella/Passalora/Phaeoramularia/Stenella complex. It appears, therefore, that
Stenellopsis should also be seen as synonym. Verrucisporota, introduced by Shaw and Alcorn (1993),is similar to Stenellopsis. The scars are broad and of the
Cercospora-type, and the conidia are rugose (David
1997). Parastenella J.C. David (David 1991, MorganJones 1998, = Stenellopsis Morgan-Jones 1980, non
Huguenin 1966) resembles Stenella and Verrucisporota, but the conidial scars are unthickened, not darkened, and the conidiogenous cells are frequently intercalary and lateral. The fate of the latter two genera
remains to be determined.
Phaeoramularia and Tandonella S.S. Prasad &
R.A.B. Vmma. Phaeoramularia, described by Muntaiiola (1960), was introduced for Ramulam'a-like dematiaceous hyphomycetes with fasciculate, simple or
branched conidiophores and has been widely used
since (Ellis 1971, 1976, Deighton 1979, Liu and Guo
1982, Pons and Sutton 1988, Hsieh and Goh 1990).
Arx (1974) did not recognize the genus Phaeoramularia, and stated that he was not able to distinguish
it from M~~covellosiella,
Stenella, and other allied genera. In 1983 Arx treated Pizaeoramulam'a as a separate
genus, but emphasized that this genus is morphologically not clearly separable from Mycovellosiella.
Braun (in Braun and Melnik 1997) reallocated Fusicladium leviem' Magnus ( = E: kaki Hori & Yoshino) to
Phaeoramularia and reduced Hormocladium Hohn.
(type species: Fusicladium kaki; Hohnel 1919) to synonymy with Phaeoramularia. Phaeoramularia (1960),
would be a later facultative synonym of Hormocladium, but monographic studies in Fusicladium, including SEM examinations, recently carried out by U.
Braun and collaborators, showed that F: lmiem' is a
true Fusicladium belonging to a group of species with
catenate conidia. Hence, Hormocladium must be reduced to synonymy with Fusicladium and does not
pertain to Phaeoramularia.
Tandonella Prasad & Verma (1970) was introduced
for a synnematous hyphomycete with thickened,
darkened scars and hila and pigmented, catenate conidia. Sutton and Pascoe (1987) added a new species,
compared Tandonella with Sclerographiopsis Deighton, and emended the circumscription of the two
genera. Tandonella is a synnematous counterpart of
Phaeoramularia, also having superficial mycelium.
However, as stated earlier, both these features should
not be employed at generic level, so that Tandonella
is not tenable any longer as a separate genus.
Reassessment of Mycovellosiella, Passalora, Phaeoramularia, and allied genera. Passalora and Phaeoramularia are only differentiated by the mode of conidial formation, either formed singly or in chains.
Conidial formation is, however, a weak feature to employ at the generic level in this complex. Ramularia
comprises, for instance, species with solitary as well
as catenate conidia. Braun (1998) clearly demonstrated that the conidial formation (solitary vs catenate)
is not applicable within Ramularin slat. for a separation of the genus into smaller taxonomic units. In
Pseudocercospora, the conidia are usually formed singly, but a few species with catenate conidia are also
known (Deighton 1976). Conidia in short chains are,
for instance, known for Pseudocercospora nq'ctnnthis
(A.K. Roy) U. Braun, Bagyan. & Jagad. (Braun et a1
1992), I? millettine Goh & W.H. Hsieh and l? noveboracensis Goh & W.H. Hsieh (Hsieh and Goh 1990).
Pseudocercosporella Deighton is also a genus that usually forms solitary conidia, but a few taxa with catenate ones have also been included (Deighton 1973).
Even in Cercospora s.str. the conidia may occasionally
be formed in short chains [e.g., in C:Prcospora lartu-
incorporating two septoria-like isolates that separate in other analyses, and thus do not belong in this clade, see text); clade
5 = Passalora (incorporating M~couellosiellaand Phaeoramularia, and one isolate of Thedgonia; clade 6 = Ramularia (incorclade 10 = Lecanosticta;
porating Ouularia and Ophiocladium); clade 7 = unknown; clade 8 = Septona; clade 9 = Uwebraz~nia;
clade I1 = Stenella; clade 12 = Phaeophleospora (incorporating Colletogloeopsisand Scokcostigmina) ; clade 13 = CT7uehaunia;
clade 14 = Stenella; clade 15 = Di.rsoconium; clade 16 = Cladosporium (including Hetm-osporium).
cue-satiuae Sawada (Hsieh and Goh 1990, as C. longissima Sacc.) and in C. ricinella Sacc. & Berl. (Braun
et a1 1992)l. The formation of short conidial chains
in Cercospora s.str. is often connected with the germination of conidia and microcyclic conidiation.
This phenomenon is rather common and has been
described and illustrated for many species, e.g., C.
sesamigenaJ.M. Yen & Lim (1973), C. nurantia Heald
& F.A. Wolf (Pons 1988) and Cercospora kikuchii T.
Matsumoto & Tomoy. (Fernandez et a1 1991).
Both 1W~covellosiellaand Stenella include species
with solitary and catenate conidia. There are even
several intermediate taxa in Pnssalora which occasionally form short conidial chains, e.g., Passcclo'ra
janseana (Racib.) U. Brawl ( = C;ercospora orjzae Miyake) [see Constarltinescu 1975, under C;ercospora
jun,seana (Racib.) Constant.], P heterospora (Hohn.)
Hohn. (= Phaeoramularia hoehnelii S. Petzoldt, in
Braun 1992) and R aratai (Speg.) U. Braun, Delhey
& Kiehr (2000). Pons and Sutton (1996) described
Cercosporidium dezghtonii, which was referred to Cerco.spom'dium ( = Passalora) although the conidia in
this species are fomed in chains. In conclusion, it
must be stated that among cercosporoid hyphomycetes the formati011 of single or catenate conidia is
not tenable as a distinguishing character at generic
rank. Phaeoramularia has to be reduced to synonymy
with Passalora, which has also been confirmed by our
molecular data, in which Passalora, Phaeoramularia
and iVycovellosiella form mixed clusters.
The differentiation between Passalora (incl. Phaeoramularia) and 1Wlycovellosiella (incl. Fulvia) , which
comprises taxa with solitary as well as catenate conidia, is also difficult (Arx 1983, Pons and Sutton 1988).
Various intermediate taxa exist, e.g., iVl~couellosiella
Drachycarpa (Syd.) Deighton (Arx 1983), iM. solanitorvi (Gonz. Frag. & Cif.) Deighton (1974) and kf.
tm'chostemmatis (Henn.) U . Braun (1999). Fuluia
forms transitions from iMycovellosielln to Pussalora
(incl. Phaeoramuluria) .
The development of superficial secondary mycelium with solitary conidiophores is also a variable character. Deighton (in Ellis 1976) assigned, for instance,
CPrcosp~ratithonine Baker & Dale to Phaeoramularia,
but Hsieh and Goh (1990) found external hyphae
with solitary conidiophores in this species. C:~rcospora
eupatorii-odorati J.M. Yen has been placed in 1M~co7)ello.riella (Yen 1981) as well as Plzaeoramulam'a (Liu
and Guo 1982). Threads with terminal, intercalary,
and pleurogenous conidiogerlous cells may range in
this group of genera from being decumbent, developing into creeping secondary hyphae, to erect, fasciculate, forming ropes or climbing leaf hairs, but all
kinds of transitions mav exist. In culture, all differences in habit disappear. Therefore, the develop-
ment of creeping superficial hyphae is, as in Pseudocercospora (Deighton 1976), Cercosporella (Braun
1995a), and Ramulnvia (Braun 1998), not tenable for
the separation of cercosporoid genera. Even in Cercospora s.str., creeping hyphae with solitary conidiophores may be found [e.g., in Cercospora canescens
Ellis & G. Martin (Yen and Lim 1969) and C. hjalofilispora J . M . Yen (1966)l. The main character that
connects ~WjcovellosielLaand Fulvia is the formati011
of conidiogenous cells, which are integrated, terminal, intercalary pleurogenous, and often formed as
short lateral branches or shoulders immediately below transverse septa. However, even in Phaeommz~laria g o m p h r ~ ~ z a(Speg.)
e
Munt.-Cvetk., the type species
of Phaeoramula~ia[holotype (LPS 914), isotype (IMI
7706) examined], the fLll range of rnyco\~ellosiella/
fulvia-like conidiogenous threads have been found,
with loosely fasciculate to decumbent, frequently
branched conidiophores having terminal, intercalary
as well as pleuroge~lousconidiogenous cells (Muntaiiola 1960, Deighton 1979, Pons and Sutton 1988).
Hence, the type species of Phaeomrnularia is indistinguishable from LWl~cov~llosi~lla
(incl. Fuluicc). Similar
Phaeoramularia spp. with frequently branched conidiophores are rather common, e.g., Ph. acanthicoln
(Hansf.) Deighton (1987), Plz. isotomae Deighton
(1986), Ph. cz~curbiticola(Henn.) Deighton, Ph. lom ~ n s iD
s eighton, E'h. sudanensis Deighton (Deighton
1979), Ph. parccdoxci Munt.-Cvetk. and I'h. iresines
Munt.-Cvetk. (Muntaiiola 1960). Branched erect or
decumbent threads are not useful as a character for
taxoiiomic purposes on a generic level in cercosporoid hyphomycetes. Branched conidiophores are
even present in Passalom baeillig~ra,the type species
of Passalora, in P microsper~naFuckel and f? chaptomi u m (Cooke) , b x (Deighton 1967). There is not a
true basis for the separation of Passalorn (incl. Plzaeommularia) and i~fycovellosiella(incl. F u k ~ i n,) so that
the latter genus must also be included in Pu.rsalora
slat., which is in agreement with our molecular data.
Pa.rsalora emend. reveals the same range of morphological variation as observed in Ramularia and P.\eudocercospora. I-'assalo.ra may therefore be circumscribed as f o l l o ~ ~ s :
Passalora Fr., Summa \reg Scand p. 500. 1849.
emend.
Cercosporidiun~Earle, Muhlenbergia 1: 16. 1901.
lhllosi~llaRangel, Bolm Agric S Paulo, Ser 16 A, 2: 151.
1915, homonynl.
~~lycovellosiella
Rangel, Arch Jard Bot Rio de Janeiro 2:
71. 1917.
= C~rcodeuterospornCurzi, Boll Staz Patol \reg Roma, Ser
2. 12: 149. 1932.
= R ~ g n l ~ l l d z c c nSolheirn,
o
hf~l\colog~a
23: 365. 1931.
=
=
-
=
=
Berteromyces Cif., Sydowia 8: 267. 1945.
Fulvia Cif., Atti 1st Bot Univ Lab Critt Pavia, Ser 5, 10:
245-246. 1954.
= Mycovellosiella subgen.Fulvia (Cif.)U. Braun, A monograph of Ccrcosporella, Rumularia and allied genera
(phytopathogenic hyphomycetes),Vol. 1: 39. 1995.
= Oreophylla Cif., Sydowia 8: 259. 1954.
= Phaeoramularia Munt.-Cvetk.,Lilloa 30: 182. 1960.
= Eriocercospora Deighton, Mycol Pap 118: 3. 1969.
= Elleteueru Deighton, Mycol Pap 118: 17. 1969.
= Tandonella Prasad and Verma, Indian Phytopathol 23:
111. 1970.
= Phaeoisariopsis p.p.
Anamorphs of Mycosphaerella spp., usually on living hosts, phytopathogenic, often causing leaf spots,
occasionally hyperparasitic, rarely saprobic. Primary
mycelium internal, secondary mycelium lacking to
well-developed, external, superficial; hyphae
branched, septate, smooth, hyaline to pigmented.
Stromata absent to well-developed, substomatal to intraepidermal, rarely deeply immersed, subglobose to
applanate, subhyaline to pigmented, composed of
loosely to densely aggregated swollen hyphal cells.
Conidiophores solitary, loosely to densely fasciculate
or in sporodochial to synnematous conidiomata, arising from internal or superficial hyphae or substomatal to intraepidermal stromata, emerging through
stomata, erumpent through the cuticle or arising
from creeping hyphae, terminal or as lateral branches, conidiophores macronematous, unbranched or
branched, continuous to pluriseptate, subhyaline to
pigmented, smooth to finely verruculose; conidiogenous cells integrated, terminal, intercalary to
pleurogenous or conidiophores reduced to conidiogenous cells, conidiogenous loci conspicuous, conidiogenous cells cicatrized, scars somewhat thickened
and darkened-refractive, more or less applanate. Conidia solitary to catenate, in simple or branched
chains, amerosporous to scolecosporous, aseptate to
pluriseptate, euseptate, rarely with a few additional
distosepta, pale to distinctly pigmented (if subhyaline, conidia non-scolecosporous, broad, 4-15 km
diam., and with few septa, usually 0-4), smooth to
finely verruculose, hila somewhat thickened and
darkened-refractive, more or less truncate.
Pseudocercospora and allied genera.-Pseudocercospora was introduced by Spegazzini (1910). Deighton
(1976) re-introduced this forgotten name and widened the concept of this genus considerably to include a wide range of cercosporoids with inconspicuous scars. He reduced Helicominia L.S. Olive, Ancylospora Sawada and Cercocladospora G.P. Agarwal &
S.M. Singh to synonymy with Pseudocercospora. Deighton (1976) distinguished Cercoseptom'a Petr. from
Pseudocercospora by having narrow, acicular conidia,
but both genera cannot be properly differentiated
(Deighton 1987, Braun 198%). Arx (1983) merged
Cercoseptom'a, which is characterized by having pigmented conidiophores, with the colorless genera
Pseudocercosporella and Thedgonia B. Sutton. Presently
Cercoseptoria is accepted as a synonym of Pseudocercospora (Hsieh and Goh 1990, Guo and Hsieh 1995,
Crous and Braun 1996, Braun and Melnik 1997),
which is also supported by the molecular data derived
by Crous et a1 (2000) and in the present study.
Pseudocercospora is morphologically highly variable,
and accommodates a wide range of cercosporoid hyphomycetes with pigmented conidiophores and inconspicuous, unthickened, not darkened conidiogenous loci. Braun (1995a) demonstrated that all
other characters are ~ariableand unsuitable for consideration at generic rank. Braun (1998) proposed a
formal division of Pseudocercospora into five sections,
based on morphological features of the conidiomata
and conidia, but emphasised that these "units"
should only be regarded as morphological groups,
which are formally treated as sections, independent
of their taxonomic value. The molecular data derived
by Crous et a1 (2000) and in the present study show
Pseudocercospora to be polyphyletic within Mycosphaerella, having evolved more than once from different
Mq'cosphaerella holomorphs, and in several occasions
having lost the teleomorph. This complex includes
species with solitary or catenulate conidia, having euor distosepta, and scars which are inconspicuous to
denticle-like, unthickened or almost so, not to slightly pigmented-refractive.
Paracercospora Deighton (1979) was introduced for
some cercosporoid hyphomycetes with subconspicuous conidial scars (thickened slightly along the
rim). Paracercospora fijiensis (Morelet) Deighton was
originally referred to as a species of Pseudocercospora
(Deighton 1976). The differentiation of Pseudocercospora and Paracercospora is often difficult, and there
are numerous intermediate taxa in Pseudocercospora
that occasionally form paracercospora-like scars. In
molecular analyses (Crous et a1 1999, Stewart et a1
1999, Crous et a1 2000), Paracercospora spp. grouped
with Pseudocercospora spp. These results supported
the synonymy of these two genera.
Braun and Melnik (1997) included the description
of the new genus Pseudophaeoramulam'a U. Braun,
which contained cercosporoid hyphomycetes intermediate between Pseudocercospora and Phaeoramularia. The conidia are catenate, but the conidial scars
are subconspicuous, flat, truncate, minutely thickened, not darkened to somewhat darkened and refractive. Pseudophaeoramulam'n geranii (M7.B. Cooke &
C.G. Shaw) U. Braun, the type species of this genus,
was originally placed in Pseudocercospora by Braun
(1991). Based on our molecular data, Pseudopha~oramularia should be reduced to synonymy with Pseudocercospora. These results suggest, therefore, that the
separation of taxa with unthickened scars, and minutely thickened scars, are not informative at the generic level. The "type" of scar thickening (inconspicuous, planate, pileate or protruding) could therefore
be more important than previously accepted. These
results suggest that the separation of taxa with unthickened, non-pigmented scars and almost unthickened, slightly pigmented-refractive scars are not informative at the generic level.
Denticularia Deighton (1972) and Semipseudocercosporn J.M. Yen (1983) seem to be closely allied to Pseudocercospora, but the conidiogenous cells are distinctly denticulate and the conidia are non-scolecosporous. Because some species of Pseudocercospora also
form denticles (Crous 1998), it may mean that these
genera will eventually also be reduced to synonymy.
Presently no cultures are available for study and
hence this remains unresolved.
Deighton (1976) recognized Pantospora Cif. (Ciferri 1938) as a separate genus close to Pseudocerco~porn, distinguished by synnematous conidiomata,
percurrent as well as sympodial conidiogenous cells,
and frequently dictyosporous conidia. These differences are, however, rather vague. The formation of
dictyospores is the only clear character which distinguishes Pantosporn from Pse~~docercospo'ra,
although
conidia of Ps~udocercosporauitis (Ltv.) Speg., the type
species of this genus, occasionally possess a few
oblique or transverse septa. Molecular data on the
species concerned are not available and the taxonomic value of dictyospores as a generic feature is
not quite clear. As this feature can occur in the type
of Ps~udocercosporn,however, it appears that Pant0.sporn may also represent a synonym of Pseudocercospora. Stigrninn plntani (Fuckel) Sacc., the type species
of Stigminn, which has phragmo- and dictyospores, as
well as eu- and distoseptate conidia, also clustered in
the larger Pseudoem-cosporu clade in the present study.
Furthermore, Ellis (1971) also lists several genera
such as Annellophora S. Hughes, P ~ ~ z ~ d o s p i r o M.B.
pes
Ellis, Spiro@esCif. and S p o r i d ~ s m i u mLink, where taxa
occur that have a mixtue of eu- and distoseptate conidia.
Braun (1993) introduced Cerco.r.ticpinnfor some
cercosporoid hyphomycetes that have been segregated from Stigrnina Sacc. Species of this genus are morphologically close to Pseudocercospora spp. (scars inconspicuous), but distinct by having consistently percurrently proliferating conidiogenous cells. However,
the combination of sympodial and percurrent conidiogenous cells is not uncommon in P.~eudoc~rro.~po-
ra. Some Cercostigmina spp. have ~Mycosplzaerellateleomorphs. One such an example is (=ercostigminn
punctata (N7akef.) Crous (ibl. syzjp'i Crous) (Crous
1999). In a recent ~nolecularstudy comparing the
taxa occurring on Eucaljptus, Crous et a1 (2001) included one isolate of this fungus (STE-U 1124),
which only produces conidia on percurrently proliferating conidiogenous cells in culture, though also
sy~npodiallyon the host (Sutton and Crous 1997).
This species, belonging to Cercostigmina s.str., clustered with 97% bootstrap support to Pseudorercos~om,
suggesting that C~rco.stigminnshould be reduced to
synonymy M Jth
~ J-'Sez~do(~rcospom.
Stigminn Sacc. (Saccardo 1880) was recently circumscribed by Sutton and Pascoe (1989) to represent
foliicolous species allied to S. platani with pigmented
structures, percurrently proliferating conidiogenous
cells, and transversely, occasionally longitudinally distoseptate conidia. Genera that have since been segregated are Cercostigrnina ( = Pseudocrrco.sj)orr, see
above), X e n o s t i p i n a Crous (19981, with euseptate
conidia and sympodial and percurrently proliferating
conidiogenous cells and Scoleco.stipinn LT. Braun.
Scolerostigminn (Braun et a1 1999) was introduced for
stigrnina-like species with thick-walled structures, and
conidia having transverse eusepta, occasionally with
a few longitudinal septa. One isolate of Stigmirta plrtani (CBS 336.33) that was included in this study clustered in the Pseudocerro.sporn clade. As with Cmostigmina, it supports the contention that the strictly percurrent proliferation of conidiogenous cells should
be seen as a feature of value at species level. Furthermore, it also suggests that distoseptatiorl and
oblique conidial septa are not evolutionarily informative in this group. Presently no species of S t i p i n a
s.str. has a known L\f~cosphaerellu teleomorph. These
findings suggest, however, that the teleornorphs, if
found, ivould reside in lL1ycosphaerelln. One problem
with placing Stigrnina in the Pseudocer.co.~porrgroup
is that Stignintl (1880) predates Psrudoc~rco.sporu
(l910), so this synonymy will have to be accompanied
with a proposal to retain P.r.~zrdocercospomover Stignzina.
Sutton and Hennebert (1994) commented that
sporodochia are not fundamentally different from
acer\uli and thus S l i p r ~ i n ais not that distinct from
Leranosticta Syd. The fact that Lecnnosticta rricolc~
(Thiim.) Syd. clustered with other anamorphs of ~\{ycosphccrrella, casts doubt on the validity of Eruptio
M.E. Barr as proposed by Barr (1996). More species
will have to be included to resolve the fate of this
genus, and to determine if it is homogenous.
Sonderhenia H,]. Swart and J . Walker (1988) was
established for pycnidial ananlorphs of ~\.Ijro~phaereI/n with pigmented structures, percurrently prolifer-
ating conidiogenous cells and distoseptate conidia.
Although similar to Phaeophleospora Rangel, it is chiefly distinguished by its distoseptate conidia. In a study
comparing sequence data of Sonderhenia, it was
shown to cluster between a Pseudocercospora and Mycovellosiella clade (Crous et a1 2001), further casting
doubt on the value of distoseptation as character in
this complex.
Phaeophleospora was established by Rangel ( 1916)
for pycnidial anamorphs of Mycosphaerella that have
pigmented non-multi-euseptate, pigmented, smooth
to rough conidia that form on pigmented, percurrently proliferating conidiogenous cells (Taylor and
Crous 1999). Recently, however, collections were obtained of a species with conidiogenous cells that
could also proliferate sympodially (Crous unpubl) .
Colletogloeopsis Crous & M.J. Wingf. (Crous and M'ingfield 1997) represents acervular anamorphs of Mycosphaerella with brown, 0(-1)-septate, verruculose
conidia that form on brown, percurrently or sympodially proliferating conidiogenous cells. The fact that
species of these genera clustered together with high
bootstrap support in the phylogenies derived from
both ITS and LSU data (Crous et a1 2001) suggests
that conidiomatal structure and conidial septation
are of lesser importance in this complex than conidial pigmentation and conidiogenesis.
Presently no cultures are available of Xenosti,pina
Crous, thus its position cannot be clarified. However,
given the flexibility of percurrent and sympodial proliferation occurring in some of the Mjcosphaer~llaanamorphs, the separation of this genus needs to be
reconsidered. Crous and Corlett (1998) also recently
described a Cercostipina synanamorph for Xenostigmzna wolJii Crous & Corlett. Given the variation in
mode of proliferation, and the fact that oblique septa
can also occur in Pseudocercospora, it appears probable that Xenostipina, like Cercostipina, could be a
synonym of the Pseudocercospora/Stipina complex.
A species that has to be placed in Scolecostipina
U. Braun (sporodochia, pigmented structures, percurrently proliferating conidiogenous cells, conidia
subcylindric-obclavate, transversely multi-euseptate)
is S t i p i n a eucalypti Alcorn (CBS 313.76). This species, however, clustered with Colletogloeopsis (acervuli,
pigmented structures, sympodial and percurrently
proliferating conidiogenous cells, conidia subcylindrical, 0-1-euseptate) and Phaeophleospora (pycnidia,
pigmented structures, sympodial and percurrent proliferating conidiogenous cells, conidia subcylindrical
to obclavate, 0-multi-euseptate). If these genera were
to be considered objectively the main distinguishing
factor is their conidiomatal structure. To merge them
into the oldest genus name, Phaeophleospora, would
be to disregard conidial septation and conidiomatal
structure. The abandonment of these features has already been argued elsewhere in this paper. These
preliminary data also support this synonymy. How
ever, to complicate matters, the type of Phaeophleospora, P eugeniae Range1 clustered apart from this
group in a study by Crous et a1 (2001).It would seem,
therefore, that although the synonymy proposed
above could be feasible, it may have to await the inclusion of more species and other data sets to be finalized. In the interim, the following genera can be
seen as part of the Pseudocercospora complex (with
Phaeoisariopsis and Stigmina awaiting proposals to retain Pseudocercospora):
Pseudocercospora Speg., Ann Mus Nac Hist B Aires
20: 438. 1910; emend Deighton (1976).
Cercospo~iopszsMiura, Flora of Manchuria and East
Mongolia, 3, Cryptogams 527. 1928 (nom. illegit.).
= Cercoseptona Petr., Ann Mycol 23: 69. 1925.
= Pantospora Cif., Ann Mycol 36: 242. 1938.
= Anglospora Sawada, Rep Govt Res Inst Formosa 87: 78.
1944 (nom. illegit.).
= Helicomzna L.S. Olive, Mycologia 40: 17. 1948.
= Cercocladospora G.P. Agarwal & S.M. Singh, Proc Nat
Acad Sci India, B, 42: 439. 1972 (1974), fide Deighton
(1976).
= Paracercospora Deighton, Mycol Pap 144: 47-48. 1979.
=
=
=
Cercost~grnznaU . Braun, Crjpt Bot 4: 107. 1993.
Ps~udophaeoian~ulana
U. Braun, Proc K o m a r o ~Bot Int
20: 18. 1997.
The genera Dzssoconium de Hoog, van Oorschot &
Hijwegen (1983) and Uwebraunia Crous & MJ.
Wingf. (Crous and Wingfield 1996) are two morphologically indistinguishable anamorph genera linked
to Mycosphaerella. Crous et a1 (1999) compared these
taxa on a molecular basis, and concluded that they
should be retained as separate entities, as they were
phylogenetically distinct. It is interesting to note,
however, that Dissoconium has a hyperparasitic habit,
compared to the phytopathogenic one of Uwebmunia. As mentioned above for Elletevma and Mycovellosiella, several cercosporoid genera are now known
that include saprobic, phytopathogenic, and hyperparasitic species. Furthermore, the two species of
Uwebraunia also cluster separately (Crous et a1 1999,
2000), suggesting that this form has evolved more
than once in Mycosphaerella. However, species of Dissoconium always cluster together, forming a clade adjacent to the main Mycosphaerella clade, and is an
example of convergent evolution in the anamorph
form.
Conclusions.-Based on sequence data derived from
the ITS and LSU regions in the present and previous
studies (Stewart et a1 1999, Crous et a1 1999, 2000,
2001), the monophyly of Mycosphaerella is supported.
Presently these data also support the separation of
1Mycosphaerella into three sections, namely the main
Mjcosphaerella section, a section for species with Dissoconium anamorphs, and a section for taxa with
Cladosporium anamorphs. Approximately 23 anamorph genera have thus far been linked to iVycosphaerella based on cultural data. Of the 19 anamorph
genera investigated here, it appears that many are
polyphyletic, and do not represent natural groups
within iVycosphaerella (FIG. 3 ) . Furthermore, it
should be renlembered that anamorphs are form
genera. The only groups that were found to be clearCladospom'um
ly defined were Cercospora, Ram~~lam'a,
and llissoconium. As more species are added to the
phylogeny, however, this could possibly also result in
more polyphyletic groups, as is the case with the other anamorph genera. To revisit the genus-for-genus
concept, one would assume that iVycosphaerella, being monophyletic, should have one anamorph genus.
However, this approach would lead to a gross loss of
valuable taxononlic inforrnation that will not serve
mycology or plant pathology, and will have serious
implications. The solution, therefore, is to propose a
simplified but workable system, using the characters
shown here to remain informative at the generic level. From our preliminary data we conclude that the
following features are not useful for generic separation of 1Vjcosphaer~llaanamorphs:
1) Superficial mycelium with solitary conidiophores, and the surface texture thereof (it should be
noted, however, that the separation of Stenella and
Passalorn is still not clearly resolved);
2) The formation of stromata;
3) Conidiomatal structure (conidiophores solitary,
fasciculate to synnematous, sporodochia to pycnidia
and acermli) (FIG.3) ;
4) Conidial shape, size, and septation (even eusepta
vs distosepta), as well as solitary vs catenate conidia;
5) Saprobic, hyperparasitic, and phytopathogenic
habit.
Characters of value at the generic level include:
1) Structure of conidiogenous loci (scars) and hila
[unthickened or almost so, but slightly darkened appears to have the same value as being unthickened;
note a Passalol-a sp. with such scars (STE-U 2554) in
the Ps~udocercosporagroup]. Pigmentation and thickening of scars form a continuum between genera.
The distinction should thus be between the degree
of thickening, darkening, and refraction, building on
the concepts developed by David (1997);
2) Presence or absence of pigmentation in conidiophores and conidia [some overlap exists here, and
needs clarification, i.e., Passalora and C;wcosporu. Also
note Thedgonia (STE-U 1661) in the Passalora
group].
These conclusions are still preliminary, and as
more taxa are sequenced and other genes also investigated, we hope that this will become clearer. The
philosophy follo~iedhere thus supports a reduction
in genera, and a widening of anamorph generic concepts within iWycosphaer~lla,which presently still appears to be monophyletic. The next steps would be
to focus on specific anamorph genera of 1Vljcos~hnuella that could represent single groups, obtain more
isolates of these taxa, and investigate other areas of
their genomes.
LITERATURE CITED
h x Jhvon. 1949. Beitrage zur Kenntnis der Gattung Mycosphaerella. Sydowia 3:28-100.
. 1974. The genera of fungi sporulating in pure culture. 2nd ed. Vaduz: Cramer. 315 p.
. 1981. The genera of fungi sporulating in pure culture, 3rd ed. Vaduz: Cramer. 424 p.
. 1983. i4fycosphaerella and its anamorphs. Proc K
Ned Akad Wet Ser C Biol Med Sci 86(1):1.3-54.
Barr ME. 1996. Planistromellaceae, a new family in the
Dothideales. Mycotaxon 60:433-442.
. 1972. Preliminary studies on the Dothideales i11
temperate North Anlerica. Contr Univ Mich Herb 9:
523-638.
Bisby GR, iiinsworth GC. 1942. The rlulnbers of fungi.
Trans Br Mycol Soc 26:16-19.
Braun U. 1988a. Studies on Ra?nulam'a and allied genera
(I). Int J Mycol Lichenol 3:271-285.
. 1988b. Studies on Ra~nulariaand allied genera (11).
Nova Hedwigia 45:333-349.
. 1991. St~ldieson Ramularia and allied genera (K).
Nova Hedwigia 53:291-305.
. 1992. Taxonomic notes on some species of the CB-cospora-complex. Nova Hedwigia 55:211-221.
. 1993. Kew genera of phytopathogenic Deuteromycetes. Cryptogam Bot 4:107-114.
. 1995a. A nlonograph of Cm-cospo~~lla,
Rnmz~laria
and allied genera (phytopathogenic hyphomycetes).
Vol. 1. Eching: IH\V Verlag. 333 p.
. 1995b. Miscellaneous notes on phytopathogenic hyphomycetes (11). hlpcotaxon ,35223-241.
. 1996. Taxonomic notes on some species of the Cur-cospora-complex (n'). Sydowia 48:205-2 17.
. 1998. A monograph of Crr.cosporella, Ramularia and
allied genera (phytopathogenic hyphomycetes). Vol. 2.
Eching: IHW Verlag. 493 p.
. 1999. Taxonomic notes on some species of the Curcospora complex (W). Cryptogam Mycol 20:155-177.
, Bagayanarayana G, Jagadeeswar P. 1992. Notes on
Indian Cercosporae and allied genera (11). Int J Mycol
Lichenol 4:361-374.
, Delhey R, Kiehr M.2000. Some new cercosporoid
hyphomycetes from Argentina. Fungal Div 6: 18-33.
, Melnik VA. 1997. Cercosporoid fungi from Russia
and adjacent countries. Trudy Bot Inst Im VL Komarova (St. Petersburg) 20:l-130.
, Mouchacca J, McKenzie EHC. 1999. Cercosporoid
hyphomycetes from New Caledonia and some other
South Pacific islands. N Z J Bot 37:297-327.
BubLk F. 1912. Einige neue Pilze aus RuBIand. Hedwigia 52:
265-273.
Castaiieda RF,Braun U. 1989. Cercospora and allied genera
of Cuba (1). Cryptogam Bot 1:42-55.
Chupp C. 1954. A monograph of the fungus genus Cercospora.
Ithaca, New York: Published by the author. 667 p.
Ciferri R. 1938. Mycoflora domingensis exsiccata. Ann Mycol 36:198-245.
Clements FE, Shear, CL. 1931. Genera of Fungi. New York:
Wilson. 496 p.
Constantinescu, 0 . 1975. Studies on Cercospo'ra and similar
fungi 11. New combinations in Cercospora and Mycouellosiella. Cryptogam Mycol 263-70.
Crous PW. 1998. Mycosphaerella spp. and their anan~orphs
associated with leaf spot diseases of Eucalyptus. Mycol
Mem 21:l-170.
. 1999. Species of n/Iycosphaerella (Ascomycetes) and
related anamorphs occurring on Myrtaceae (excluding
Eucalyptus). Mycol Res 103:607-621.
, Aptroot A, Kang JC, Braun LT,Wingfield hfJ. 2000.
The genus 1~cosphaerellaand its anamorphs. Stud Mycol 45: 107-121.
, Braun U. 1996. Cercosporoid fungi from South Africa. Mycotaxon 57:233-321.
, Corlett M. 1998. Reassessment of iMycosphaerella
spp. and their anamorphs occurring on Platanus. Can
J Rot 76:1523-1532.
, Hong L, Wingfield BD, Wingfield MJ. 2001. ITS
rDNA phylogeny of selected 1Mycosphaerella spp. and
their anamorphs occurring on Myrtaceae. Mycol Res
105:425-431.
--, Wingfield MJ, &ng JC. 1999. Uwebrnunia
and Dissoconium, nvo morphologicaIly similar anamorph genera with distinct teleomorph affinity. Sydorvia 52:155-166.
, Phillips AJL, Wingfield MJ. 1992. Effects of ctlltural
conditions on vesicle and conidium morphology in species of Cylindrocladium and Cylindrocladiella. Mycologia 84:497-504.
, Wingfield MJ. 1996. Species of Mycosphaerella and
their anamorphs associated with leaf blotch disease of
Eucalyptus in South Africa. Mycologia 88:441-458.
--. 1997. Colletogloeopsis, a new coelomycete
genus to accommodate anamorphs of two species of
iMycosphaerella occurring on Eucahptus. Can J Bot 75:
667-674.
Curzi M. 1932. De fungis et morbis africanis. I. De quibusdam hyphomycetibus parasitis Somalie. Bot R Staz Pato1 Veg N S 12:149-168.
David JC. 1991. Parastenella, a new generic name for Heterosporium map~oliae.Mycol Res 95: 123-1 28.
. 1993. A revision of taxa referred to ffetmosporium
Wotzsche ex Cooke (Mitosporic Fungi) (PhD Thesis).
UK: Universi? of Reading. 290 p.
. 1997. contribution to the systematics of Cladosporium. Revision of the fungi preciously referred to Heterosporium. Mycol Pap 172:1-157.
Deighton FC. 1967. Studies on Cercospora and allied genera.
11. Passalora, Cercosporidium and some species of Fusicladium on Euphorbia. Mycol Pap 112:l-80.
. 1969. Microfungi. nr:some hyperparasitic hyphomycetes, and a note on Cercosporella uredinicola Sacc.
Mycol Pap 188:l-41.
. 1972. Four leaf-spotting hyphomycetes from Africa.
Trans Br Myco Soc 59:419-427.
. 1973. Studies on Cercospora and allied genera. n'.
Cercosporella Sacc., Pseudocmosporella gen. no% and
Pseudocercosporidium gen. nov. Mycol Pap 133:l-62.
. 1974. Studies on Cercospora and allied genera. V.
rMycouellosiella Rangel, and a new species of Ramulariopsis. Mycol Pap 137:l-75.
. 1976. Studies on Cercospora and allied genera. W.
Pseudocercospora Speg., Pantospora Cif. and Cercoseptoria Petr. Mycol Pap 140:l-168.
. 1979. Sudies on Cercospora and allied genera. WI.
New species and redispositions. Mycol Pap 144:l-56.
. 1986. Misidentification of Cercospora effusa. Trans
Br Mycol Soc 86:637-641.
. 1987. New species of Psez~docercosporaand Mycoriellosiella, and new combinations into Pseudocercospora
and Phaeornmularia. Trans Br Mycol Soc 88:365-391.
. 1990. Observations on Phaeoisanopsis. Mycol Res
94:1096-1102.
Earle FS. 1901. Some fungi from I'uerto Rico. Muhlenbergia 1:lO-17.
. 1902. A much-named fungus. Torreya 2:159-160.
Ellis MB. 1971. Dematiaceous hyphomycetes. Kew: CMI.
608 p.
. 1976. More dematiaceous hyphomycetes. Kew: CMI.
507 p.
Fernandez FA, Glawe DA, SinclairJB. 1991. Microcycle conidiation and nuclear beha~iorduring conidiogenesis in
Cercospora kikuchii. Mycologia 83:752-757.
Ferraris T. 1909. Osservazioni mycologische. Su specie del
gruppo Hyphales (Hyphomycetae) . Ann Mycol 7:273286.
Fisher NL, Burgess LW, Toussoun TA, Nelson PE. 1982. Carnation leaves as substrate and for preserving cultures
of Fusarium species. Phytopathologv 72: 151-153.
Fries EM. 1849. Summa vegetabilium Scandinaliae. Uppsala: Sectio posterior. 313 p.
Fuckel KWGL. 1863. Fungi Rhenani exsiccati, Fasc. I-IV.
Hedwigia 2:132-136.
Guo YL, Hsieh WH. 1995. The genus Pseudocercospora in
China. Mycosystema Monographicum 2:l-388.
Ho MH, Castafieda RF, Dugan FM, Jong SC. 1999. Cladosporium and Cladophialophora in culture: descriptions
and an expended key. Mycotaxon 72:113-157.
Hohnel F \on. 1919. Fiinfte vorlaufige Mitteilung mycologischer Ergebnisse (Nr. 399-500). Ber Dt Bot Ges 37:
153-1 61.
Hoog GS de, Gukho E, Masclaux F, Gerrits AHG van den,
Kwon-Chung EJ, McGinnis MR. 1995. Nutritional phys-
iology and taxonomy of human pathogenic Cladosporium-Xylohypha species. J Med Vet Mycol 33:339-347.
, Oorschot CAN van, Hijwegen T. 1983. Taxonomy
of the Dactylaria complex. 11. Dissoconium gen. nov.
and Cordana Preuss. Proc K Ned Akad Wet Ser C Biol
Med Sci 86(2):197-206.
Hsieh N'H, Goh TK. 1990. Cwcospora species and similar
fungi from Taiwan. Taiwan: Maw Chang Book Company. 376 p.
Hughes SJ. 1949. Studies on some diseases of sainfoin
(Onobrychis satiua) 11. The life history of Ramularia
onobrychidis Allescher. Trans Br Mycol Soc 32:3439.
. 1958. Revisiones hyphomycetum aliquot cum appendice de nominibus rejiciendis. Can J Bot 36:727836.
Huguenin B. 1966. Micromycetes du Pacifique Sud (Troisieme contribution). (1) D6matii.e~de Nouvelle-Caledonie. Bull Soc Mycol Fr 81:686-698.
Kendrick NB,DiCosmo F. 1979. Teleornorph-anamorph
contlections in ascomycetes. In: Kendrick W'B, ed. The
Whole Fungus. Ottawa: National bluseum of Natural
Sciences. 1:283-310.
Lee SB, Taylor<PV. 1990. Isolatiotl of DNA from fi~ngalmycelia and single spores. In: Innia MA, Gelfand DH,
Sninsky JJ, MXite 'rJ, eds. PCR protocols: a guide to
methods and applications. New York: Academic Press.
p 282-287.
Link HF. 1816. Observationes in ordines plantarum naturales. Dissertatio 11, sisteils nuperas de M u c e d i n ~ ~ et
m
Gastromycorurn ordinibus observationes. Mag Ges Naturf Freunde Berlin 7:25-45.
Liu XJ, Guo 11. 1982. Studies on some species of the genus
Pfzaeoramularia in China. Acta Phytopathol Sinica 12:
1-13,
Masclaux F, Gukho E, Hoog GS de, Christen R. 1995. Phylogenetic relationships of human-pathogenic Cludosporium (Xylohypha) species inferred from partial LSL
rRNA sequences. J Med Vet hfycol 33327-338.
Maublanc A. 1913a. Sur une maladie de feuilles du papayer
"C;c~rirapapaya". Lavoura 16:208-212.
. 1913b. Sur une maladie de feuilles du papayer (Carica papaya). Bull Soc Mycol Fr 29353-338.
McKemp JM, Morgan-Jones G. 1990. Studies in the genus
Cladosporium sensu lato. 11. Collcerning fIetrrosporiurn
gmcile, the causal agent organism of leaf spot disease
of Iris species. Mycotaxon 39:425-440.
Morgan-Jones G. 1980. Notes on hyphomycetes. XXXI'. Stenrllopsis gen. no\: Mycotaxon 10:403-408.
. 1998. Notes on hyphoniycetes. LXXK. Concerning
Parastenella magnoliae, its phylogenetic afffinic, and
the relationship between it and its host, Magnolia p a n dz$!om. Mycotaxon 66:421-428.
Muntaiiola M. 1960. rUgunos Hyphomycetes criticos. Notas
y descripciones. 1,illoa 30: 183-21 3.
Nag Raj TR. 1981. C;oelomycete systematics. In: Cole GT,
Kendrick B, eds. The b i o l o ~of conidial fungi. New
York: Academic Press. p 33-84.
Pollack FG. 1987. An annotated cornpilation of Cmospora
names. hlycol Mern 12:l-212.
Pons N. 1988. Ocurrencia de Crrcospora aurantza sobre Citrus spp. en Venezuela. Fitopatol Venezol 1:8-13.
Pons N, Sutton BC. 1988. Cercospora and similar fungi on
yams (Dioscorea spp.). Mycol Pap 160:l-78.
-. 1996. Cercospora and similar fungi on Heliotropium weeds. Mycol Res 100:813-820.
--, Gay JL. 1985. Ultrastructure of conidiogenesis in Cwcospora beticola. Trans Br Mycol Soc 85:405416.
Prasad SS, Verma RAB. 1970. A new genus of Moniliales
from India. Indian Phytopathol 23: 111-1 13.
Rangel E. 1915. Fungos parasitos do guando ( C a j a n u ~ijtdicus Spreng.). Bol Xgric S Paulo, Ser 16(2):145-136.
. 1916. Fungos do Brasil, novo ou ma1 conhecidos
por Ezigenia Rangel. Arch hIus Nac Rio de Janeiro 18:
1.55-164.
. 1917. Algunos fungos rlobas d o Brasil. Xrch Jard
Bot Rio de Ja~leiro259-74.
Saccardo PA. 1880. Conspectus generum fungorum Italiae
inferiorum, nempe ad Sphaeropsideas, Melanconicas
et Hyphomycetas pertinentium, systemate sporologico
dispositorum. Michelia 2:l-38.
Samuels GJ, Seifert IL4. 1987. Taxonomic implications of
variation arnong hypocrealeail anamorphs. In: Sugiya
ma J , ed. Pleomorphic fungi: the diversity and its taxonon~icimplications. Tokyo, Japan: Kodansha Ltd. and
Arnsterdan~:Elsevier. p 29-56.
Sarvada K.1922. Descriptive catalogue of the Forrllosan fungi 11. Dep L4gricGoy Res Inst Taiwan Rep 2:139-164.
Seifert KA, Okada G. 1990. Taxonomic implications of conidiomatal anatomy in synnernatous Hyphomycetes.
Stud Mycol 32:29-40.
Shaw DE, Alcorn JL. 1993. New names for Vwrztcispora and
its species. Aust Syst Bot 6:273-276.
Solheim LVG. 1929. Mol-phological studies of the genus Crr-cospom. I11 Biol Monogr 12:l-85.
Spegazzini C. 1910. Mycetes Argentinenses, Ser. V. h r i hlus
Nac Hist Nat B Ares 20329-467.
Stewart EL, Liu Z, Crous PLY, Smb6 L. 1999. Phylogenetic
relationships among some cercosporoid anamorphs of
L~l~cosphaerelia
based o n rDNA sequence analysis. Mycol Res 103:1491-1399.
Subranlanian CV, Rarnakrishnan R. 1956. List of Indian fungi 1952-36. J Madras Univ Sect B 26:327-421.
Sutton BC. 1980. The Coelomycetes. Fungi imperfecti with
pycnidia, acervuli and stromata. England: CMI. 696 p.
. 1994. Prathigadci twminaliae. IMI Descriptions of
Fungi and Bacteria, No. 1181. Mycopathol 125:45-46.
, Crous PM'. 1997. Lecano.rtictoj~sisg en. nov. and similar f~lngifrom Syzygiztm species. Mycol Res 101:21322.5.
. Hennebert GL. 1994. Interconnections amongst anamorphs and their possible contribution to Ascomycete
systematics. In: Hawksu-orth DL, ed. Ascornycete systematics. Problems and perspectives in the nineties.
New York: Plenuni Press. p 77-98.
, Pascoe I. 1987. Tandonella olenrine sp. nov. and generic separation of Tandonr~llaand Sclrro~graphiopsis.
Aust J Bot 33:183-191.
-. 1989. Reassessment of Pelto~oma,S t i p z l / n
and Batcheloromyces and description of Hyphothyn'um
gen. nov. Mycol Res 92:210-222.
, Pons N. 1980. Notes on the original species of Cercosporina. Mycotaxon 12:201-218.
, Shamoun SF, Crous PW. 1996. Two leaf pathogens
of Ribes spp. in North America, Quasiphloeospora saximontanensis (Deighton) comb. nov. and Phloeosporella
ribis (J.J. Davis) comb. nov. Mycol Res 100:979-983.
, Waller JM. 1988. Taxonomy of Ophiocladium hordei,
causing leaf lesions on Triticale and other Gramineae.
Trans Br Mycol Soc 90:55-61.
Swart HJ, Walker J. 1988. Australian leaf-inhabiting fungi
XXVIII. Hendersonia on Eucalyptus. Trans Br Mycol
SOC90:633-641.
Swofford DL. 1999. PAUP* (Phylogenetic analysis using parsimony* and other methods), 4.0b2a. Sunderland, Massachusetts: Sinauer Associates.
Sydow H. 1928. Novae fungorum species XIX. Ann Mycol
26:132-139.
. 1930. Fungi venezuelani. Ann Mycol 28:29-224.
Takeo K, Hoog GS de, Miyaji M, Nishimura K. 1995. Conidial surface ultrastructure of human-pathogenic and saprobic Cladosporium species. Antonie Leewenhoek 68:
51-55.
Taylor JE, Crous PW. 1999. Phaeophleospora faureae comb.
nov. associated with leaf spots on Faurea saligna (Proteaceae), with a key to known species of Phaeophleospora. Fungal Div 3:153-158.
Thaung MM. 1976. New hyphomycetes from Burma. Trans
Br Mycol Soc 66:211-215.
Thirumalachar MJ, Mishra JN. 1953. Contribution to the
study of fungi of Bihar, India-I. Sydowia 7:29-83.
Thompson JD, Higgins DG, Gibson TJ. 1994. Clustal W: improving the sensitivity of progressive multiple sequence
alignment through sequence weighting, position spe-
cific gap penalties and weight matrix choice. Nucl Acids Res 22:4673-4680.
Unger F. 1933. Die Exantheme der Pflanzen. Wien. 423 p.
Untereiner WA. 1997. Taxonomy of selected members of
the ascomycete genus Capronia with notes on anamorph-teleomorph connections. Mycologia 89:120131.
. 2000. Capronia and its anamorphs: exploring the
value of morphological and molecular characters in
the systematics of the Herpotrichiellaceae. Stud Mycol
45:141-149.
Vries GA de. 1952. Contribution to the knowledge of the
genus Cladosporium Link ex Fr. CBS, Baarn. 121 p.
Wingfield MJ, Seifert KA, Webber JN, eds. 1993. Ceratoqstis
and Ophiostoma: taxonomy, ecology and pathology. St.
Paul, Minnesota: APS Press. 293 p.
Yen JM. 1966. Etude sur les champignons parasites du SudEst asiatique. nr.Troisieme note sur quelques nouvelles
especes de Cercospora de Singapour. Rev Mycol31: 109149.
. 1981. ~ t u d esur les champignons parasites de SudEst asiatique, 42. Champignons parasites de Malaisie,
21. Bull Soc Mycol Fr 97:129-133.
. 1983. Studies on parasitic fungi from South East
Asia, 49. Parasitic fungi from malaysia, 25: Semipseudocercospora gen. nov. Mycotaxon 17:361-363.
, Lim G. 1969. Etude sur les champignons parasites
du Sud-Est asiatique. XIV. Huitieme note sur les Cercospora de Malaisie. Bull Soc Mycol Fr 85:459-474.
-. 1973. Etude sur les champignons parasites
du Sud-Est asiatique. XX. Dixikme note sur les Cercospora de Malaisie. Cah Pac 17:95-114.
--. 1980. Cercospora and allied genera of Singapore and the Malay Peninsula. Gardens Bull Singapore 33:151-263.