Additions to the rust fungi of South Africa

Abstract

This paper presents new species, combinations, national reports and host records for the South African rust fungi (Uredinales/Pucciniales).
Endophyllum mpenjatiense on cf. Hibiscus sp. (Malvaceae), Phakopsora combretorum (anamorph Uredo combreticola) on the new host Combretum apiculatum (Combretaceae) and Uredo sekhukhunensis on Ziziphus mucronata (Rhamnaceae) are described as new species. Dietelia cardiospermi and E. metalasiae are proposed as new combinations to replace Aecidium cardiospermi on Cardiospermum halicacabum (Sapindaceae) and A. metalasiae on Metalasia spp. (Asteraceae), respectively. Four species are new records for South Africa: Crossopsora antidesmae-dioicae on Antidesma venosum (Euphorbiaceae), Phakopsora ziziphi-vulgaris on Z. mucronata, and Uromyces cypericola and Puccinia subcoronata, both on a new host, Cyperus albostriatus (Cyperaceae). The record of P. subcoronata is the first one from outside the New World. Puccinia scirpi is reported as a possible addition to the South African rust fungi. New host records and observations are presented for Pucciniastrum agrimoniae that is recorded on two new host genera and species, Cliffortia odorata and Leucosidea sericea (Rosaceae), Uromyces cypericola whose urediniospores are described for the first time, Phakopsora stratosa in that spermogonia and Uredo-like aecia were discovered, and for Sphaerophragmium dalbergiae in that characters of the urediniospores are re-evaluated. A lectotype is selected for Aecidium garckeanum and spermogonia are reported for this rust for the first time. The rust fungi of Ehrharta (Poaceae) are discussed and critically evaluated in the light of spore morphology and host species.

Full text

ORIGINAL ARTICLE
Additions to the rust fungi of South Africa
Reinhard Berndt & Alan R. Wood
Received: 7 February 2011 /Revised: 15 April 2011 /Accepted: 19 April 2011 /Published online: 28 May 2011
#
German Mycological Society and Springer 2011
Abstract This paper presents new species, combinations,
national reports and host records for the South African rust
fungi (Uredinales/Pucciniales). Endophyllum mpenjatiense
on cf. Hibiscus sp. (Malvaceae), Phakopsora combretorum
(anamorph Uredo combreticola) on the new host Combretum
apiculatum (Combretaceae) and Ur edo sekhukhunensis on
Ziziphus mucronata (Rhamnaceae) are described as new
species. Dietelia cardiospermi and E. metalasiae are
proposed as new combinations to replace Aecidium cardio-
spermi on Cardiospermum halicacabum (Sapindaceae) and
A. metalasiae on Metalasia spp. (Asteraceae), respectively.
Four species are new records for South Africa: Crossopsora
antidesmae-dioicae on Antidesma venosum (Euphorbiaceae),
Phakopsora ziziphi-vulgaris on Z. mucronata,andUromyces
cypericola and Puccinia subcoronata, both on a new host,
Cyperus albostriatus (Cyperaceae). The record of P. sub-
coronata is the first one from outside the New World.
Puccinia scirpi is reported as a possible addition to the South
African rust fungi. New host records and observations are
presented for Pucciniastrum agrimoniae that is recorded on
two new host genera and species, Cliffortia odorata and
Leucosidea sericea (Rosaceae), Uromyces cypericola whose
urediniospores are described for the first time, Phakopsora
stratosa in that spermogonia and Uredo-like aecia were
discovered, and for Sphaerophragmium dalbergiae in that
characters of the urediniospores are re-evaluated. A lectotype
is selected for Aecidium garckeanum and spermogonia are
reported for this rust for the first time. The rust fungi of
Ehrharta (Poaceae) are discussed and critically evaluated in
the light of spore morphology and host species.
Keywords Combretum
.
Dietelia
.
Ehrharta
.
Endophyllum
.
Ziziphus
Introduction
Today, we know of some 530 species of rust fungi
(Uredinales/Pucciniales) from South Africa (Berndt 2008a,
and unpublished results). The bulk of them were recorded
or described by E.M. Doidge who enumerated 474 species
in her checklist of South African fungi and lichens (Doidge
1950). A total of 49 of these recorded taxa have been
recognized meanwhile to represent anamorph states or
synonyms
of already known species (Crous et al. 2006).
Thus, about 110 species were added to the South African
rust mycobiota during the last 60 years, many of them only
recently (e.g., Berndt 2008b, 2010; Berndt and Uhlmann
2006; Mennicken and Oberwinkler 2004; Wood 2006,
2007) indicating that South Africa remains under-collected.
An important aspect is that many of the described species have
been collected only once or a few times and are therefore
incompletely known with regard to morphological characters
and variability, host range and geographical distribution.
The present paper adds novel info rmation on both
mentioned aspects. It describes three species as new to science,
Taxonomical novelties: Dietelia cardiospermi (Cooke) R. Berndt & A.
R. Wood. Endophyllum metalasiae (Syd. & P. Syd.) A. R. Wood & R.
Berndt. Endophyllum mpenjatiense A. R. Wood & R. Berndt.
Phakopsora combretorum A. R. Wood & R. Berndt. Uredo
sekhukhunensis R. Berndt & A. R. Wood.
R. Berndt (*)
Institute of Integrative Biology (IBZ), ETH Zurich,
CHN D-37, Universitätstr. 16,
8092 Zürich, Switzerland
e-mail: reinhard.berndt@env.ethz.ch
A. R. Wood
Weed Pathology Unit, ARC-Plant Protection Research Institute,
P. Bag X5017, Stellenbosch 7599, South Africa
Mycol Progress (2012) 11:483–497
DOI 10.1007/s11557-011-0764-z

reports four known rust fungi as new for South Africa and
contributes a number of observations on the life cycle, host
range and morphology of already known rust species.
Materials and methods
The investigated rust species are dealt with by host families
which are arranged alphabetically. The studied specimens
are listed under the respective rust species.
Spores and hand sections obtained from herbarium
material were mounted in lactophenol or lactic acid and
gently heated to facilitate soaking of the fungal structures
and host tissue. Urediniospores of Sphaerophragmium
dalbergiae Dietel and Phakopsora stratosa (Cooke) Arth.
were stained according to the method of Jennings et al.
(1989). Spores of Aecidium spp. were germinated on water
agar (1.5%) in 6-cm-diameter Petri dishes at room
temperature, or incubated at 15 and 20°C and checked
after 24 or 48 h with a microscope for basidium formation.
Slides were exami ned with an Olympus BX51 compound
microscope equipped with a ColorView IIIu camera. The
Cell^B software package (Software Imaging System) was
used to capture micrographs and to measure details of spore
walls and their ornament. Such measurements are given to
0.1 μm while measurements made by the use of an ocular
micrometer scale are given to 0.5 μm. Normally, 40–50, but
at least 30 spores were measured for each spore state. If,
occasionally, fewer spores were studied this is indicated in
the text. The arithmetic means are given in parentheses.
Names of herbaria are abbreviated by their acronyms
according to Index Herbariorum (Holmgren et al. 1990).
HeRB stands for the first author’s herbarium located at ZT,
RSA numbers are field numbers of collections made by the
first author. Spore states are designated according to the
ontogenic terminology (Hiratsuka 1973). For brevity, the
spore states present in herbarium specimens were indicated
by the Roman numerals, I for aecia, II for uredinia, III for
telia (or the respective spores) and by 0 for spermogonia. It
should be noted that the application of these numerals to
states of the life cycle is in contrast to their original use for
morphological types of sori.
Results and discussion
Asteraceae
Endophyllum met alasiae (Syd. & P. Syd.) A. R. Wood & R.
Berndt, comb. nov. (Fig. 1)
Basionym: Aecidium metalasiae Syd. & P. Syd., Annls.
Mycol. 10: 35 (1912) . Type on Metalasia muricata R. Br.
[probably M. densa (Lam.) P. O. Karis, fide Wood and
Crous ( 2005)], Pole-Evans no. 686.
Material examined: South Africa: Western Cape:
Cedarberg, along R303 between Citrusdal and Ceres,
32°37′S19°9′E, on Metalasia dregeana DC., leg. AR
Wood no. 778, 3 Aug 2010 (PREM III); Cedarberg, along
R303 between Citrusdal and Ceres, 32°37′S19°7′E, on
Metalasia albescens Karis, leg. AR Wood no. 779, 3 Aug
2010 (PREM III).
Aecidium metalasiae was considered to be likely one of
a complex of Endophyllum species (Wood and Crous
2005), but aecidiospores have not so far been germinated.
Recently, fresh specimens were collected, and the spores
germinated at 15 and 20°C. At both temperatures, two septa
were produced in each basidium (5–6 μ
m in width), and the
two
distal
cells each germinated to form a further germ tube
(Fig. 1). This confirms that this species should be
transferred to Endophyllum, and also distinguishes it from
the other species in this complex that produce two (E.
osteospermi (Doidge) A.R. Wood, E. dimorphothecae A.R.
Wood & Crous) or four (E. elytropappi A.R. Wood &
Crous) mod ified vesicle-like basidiospores (Wood and
Crous 2005).
Combretaceae
Phakopsora combretorum A. R. Wood & R. Berndt, spec.
nov. (Fig. 2)
Anamorph: Uredo combreticola Doidge. Bothalia 3:506
(1939).
Etymology: Occurring on Combretum species.
Spermogonia et aecia ignota. Uredinia amphigena –
plerumque magis numerosa in pagina abaxiali – dense vel
laxe sparsa vel in gregibus parvis disposita, minuta
subepidermalia cupulata, peridio unistratoso (typi Milesiae)
et nonnunquam paraphysibus clavatis simplicibus tenue
tunicatis hymenii praedita; copia sporarum cinnam omea;
urediniosporae obovoideae 19–25×14–18.5 μm(21.5×
16.1 μm), pariete ca. 1 μm crasso pallide stramineo plus
minusve regulariter echinulato s pini s deli catis inter se ca.
1.5–2.5 μm distantibus, poris germinationis obscuris
verosimiliter 4–5 plus minusve sparsa. Telia in foliis
abaxialia vel adaxialia sparsa parva subplana ceracea
pallide brunnea vel straminea semper urediniis arcte
consociata; teliosporae dense et irregulariter aggregatae
3–5 superpositae, in sectione rectangulares subrhombicae
vel cuneatae 12.5–27×6.5–10 μm (18.7×8.2 μm), pariete
levi vix 1 μm crasso subhyalino vel pallidissime aureo in
teliosporis exterioribus, poro germinationis apicali lenti-
fomi; germinatio sporarum non visa ut videtur post
quiescentiam. In foliis Combreti apiculati Sond. et
Combretorum spp.
484 Mycol Progress (2012) 11:483–497

Fig. 1 Germinating teliospores
of Endophyllum metalasiae and
basidia (AR Wood no. 779).
Bars 100 μm (upper one
for upper row, lower one for
lower row)
Mycol Progress (2012) 11:483–497 485

Spermogonia and aecia unknown. Uredinia amphige-
nous on leaves, usually more numerous on abaxial surface,
densely to loosely scattered o r in small groups, tiny,
subepidermal, dome-shaped, provided with a one-layered
peridium (Milesia-type) and – sometimes – clavate, thin-
walled, simple, hymenial paraphyses, spore mass cinna-
mon; urediniospores obovoid, 19–25×14–18.5 μm (21.5×
16.1 μm), spore wall about 1 μm thic k, pale straw-colored,
more or less evenly covered by fine spines that are ca. 1.5–
2.5 μm apart, germ pores obscure, probably 4–5 more or
less scattered. Telia abaxial on leaves (always?) as small,
flat, wax-like, pale brownish crusts, scattered or often
closely associated with uredinia, composed of 3–5 super-
imposed layers of teliospores; teliospores in section
rectangular, rhomb oid or wedge-shape d, 12.5–27×6.5–
10 μm (18.7×8.2 μm), spore wal l smooth, hardly 1 μm
thick, subhyaline or very pallid golden in uppermost spore
layer, with an indistinct, apical lens-shaped germ pore;
germination not observed, apparently after resting period.
On leaves of Combretum apiculatum So nd. and Combretum
spp.
South Africa, Mpumalanga: S of Penge, near entrance
to Serowe gorge, 24°25′S 30°26′E, on C. apiculatum, leg.
AR Wood no. 734, 27 Jul 2008 (PREM 60081, holotype.
ZT Myc 951, isotype. II, III).
Additional material studied: South Africa, KwaZulu-
Natal: Silverglen Nature Reserve, Chatsworth, Durban, on
Combretum zeyheri Sond., leg AR Wood no. 714, 27 May
2008 (PREM 60118. II). Durban, Queensburgh, Roosfontein
Nature Reserve, on C. molle R. Br. ex G. Don, leg. AR Wood
no. 752, 15 Jun 2009 (ZT Myc 2597. II, III). Mpumalanga: S
of Penge, near entrance to Serowe gorge, 24°25′S30°26′E,
on Combretum hereroense Schinz, leg. AR Wood no. 735,
27 Jul 2008 (PREM 60132. II).
The present rust has been reported before as Uredo
combreticola from South Africa on C. zeyheri Sond.
(Doidge 1939) and from Namibia on C. cf. engleri Schinz
and C. hereroense Schinz (Mennicken et al. 2005
).
Combretum
apiculatum appears to be a new host. The only
other Phakopsora species described on Combretaceae is P.
chavesii Dianese, Santos & Medeiros on Terminalia in
Brazil. The morphology of that species is unclear. Accord-
ing to the Latin description (Dianese et al. 1993) it has 2–3
layers of teliospores and uredinia lacking a peridium but
with hymenial paraphyses. In the English description, 2–4
layers of teliospores are described and uredinia without
paraphyses but with a hyphal peridium. P. chavesii differs
from the present species in thicker-walled fulvous telio-
spores. Cerotelium combreti Cumm. from West Africa and
Uganda has uredinia with dorsally and apically thick-walled
peripheral paraphyses (Cummins 1952; Viennot -Bourgin
1958). According to the description given by Viennot-
Bourgin (1959) the telia resemble those of the present
species quite closely but the urediniospores are smaller and
have equatorial germ pores.
Cyperaceae
Puccinia cf. scirpi DC. Fl. Fr. II: 223 (1805) on Ficinia
spp. from the Western Cape (Fig. 3)
Material examined: O n Ficinia spp. – South Africa,
Western Cape: Kleinmond, Kogelberg Nature Reserve, in
Oudebos River valley, on F. cf. trichodes, leg. E. Uhlmann
& R. Berndt no. RSA 426, 4 Nov 2005 (PREM no. to be
assigned, ZT Myc 330 4. II, III). Cederberge, road 303 from
Citrusdal to Ceres, Middelberg Pass, 32°37’18”S19°
07’58”E, on F. cf. levynsiae Arnold & Gordon-Gray, leg.
E. Uhlmann & R. Berndt no. RSA 44, 22 Oct 2004.
(PREM, ZT Myc 3305. II).
On Scirp us lacustris L. – France, Departement de l’Ain:
Pond near Dompierre-en-Dombes, leg. E. Mayor s. n., 30
Aug 1937 (ZT Myc 2789. II, III). Germany, Bavaria:
Ingolstadt, leg. E. Eichhorn s. n., 20 Sep 1932 (ZT Myc
2788. II, III). Hungary: locality illegible, leg. F. Greinich,
16 Sep 1926 (ZT Myc 2787. II, III).
Puccinia austrina McKenzie. – New Zealand: Campbell
Island, Mt. Honey, on Isolepis habra (Edgar) Soják, leg
EHC McKenzie no. EHC 116, 7 Mar 2000 (PDD 87917,
holotype. II).
Ficinia is a genus of about 60 species that are, with a
few exceptions, confined to the Cape Floristic Region
(Goldblatt and Manning 2000
). It is closely related to
Isolep
is and
both are currently included in tribe Cypereae of
Cyperaceae (Muasya et al. 2009). Two Puccinia spp are
known on members of Ficinia: Puccinia austrina on F.
nodosa (Rottb.) Goetgh. (= Scirpus nodosus Rottb.) in
Fig. 2 Phakopsora combretorum (isotype), subepidermal telium
composed of pale, superimposed teliospores. Bar 20 μm
486 Mycol Progress (2012) 11:483–497

Australia (McAlpine 1906,subUredo scirpi-nodosi
McAlp.) and P. conclusa Thüm. on F. ‘capitaltum’ Nees
(probably F. capitellum Nees is meant) in South Africa
(Jørstad 1956). A Puccinia collected on Ficinia in the
Fynbos vegetation of Western Cape Province could not be
assigned to the latter rust species, differing in non-locular,
early naked telia, teliospores with a thinner apex and
possibly by bald patches on the surface of the uredinio-
spores (e.g., Gäumann 1959; Mayor and Terrier 1957). It
should be mentioned that Barreto and Evans (1995)
synonymized P. conclusa with P. canaliculata Schw. ex
Lagerh. and presented a SEM micrograph of a uredinio-
spore of the type of P. canaliculata that may show an
indistinct tonsure proximal to the germ pore. It is uncertain
whether the single shown spore is representative for the
species. The present rust was very similar to P. austrina in
urediniospore characters (Fig. 3a, b) but had longer, more
darkly pigmented teliospores (Fig. 3c). P. isolepidis
Gjærum from Ethiopia and Uganda is the only other known
species on Isolepis (Gjærum 1990a). The specimens on
Ficinia are also similar to that species but have a thicker
urediniospore wall and a less thickened teliospore apex.
They also differed from U. scirpi-corymbosi Doidge and U.
scirpi-maritimi Doidge described from South Africa on
Scirpus spp. (Doidge 1939, 1941) and were preliminarily
assigned to P. scirpi, which has very similar uredi nio- and
teliospores. We noti ced only small differences in the
urediniospores which have a thicker spore wall (Fig. 3a)
and a smooth patch located proximal to the germ pores on
Ficinia (Fig. 3b). In P. scirpi, bald patches are present as
well—though sometimes indistinct—but they are placed
over and around the germ pores more or less symmetrically.
The presence of smooth patches in urediniospores of P.
scirpi has not been reported hitherto (e.g., Klebahn 1914;
Gäumann
1959; Zhuang 1988). P
. scirpi is heteroecious and
infects members of Menyanthaceae in its haplontic stage
(Aecidium nymphoidis DC.). Nymphoides indica (L.)
Kuntze, a host of A. nymphoidis DC. in Uganda (Gjærum
et al. 2003), also occurs in the Cape region (Goldblatt and
Manning 2000 ) and might serve as the alternate host. P. scirpi
would be a new member of the South African rust fungi.
The following description is based on the investigated
specimens on Ficinia: Spermogonia and aecia unknown.
Uredinia on abaxial side of leaves on straw-colored to brown
leaf spots, subepidermal, shortly elliptic or longitudinally
elongated, surrounded by the torn epidermis, ferrugineous
and pulverulent. Urediniospores obovoid, subglobose to
transversely broadly ellipsoid, generally slightly compressed
along axis between the 2 opposite, equatorial to super-
equatorial germ pores, 21–31×17.5–24.5 μm(25.3×
21.0 μm), wall ca. 1.5–2 μm thick, golden to orange brown,
finely and rather distantly echinulate except for two
conspicuous bald patches proximal to the germ pores. Telia
as uredinia, but blackish brown to black, subcompact and
early naked. Teliospores subclavate to subpyriform, more
rarely ellipsoid, sometimes slightly deformed, at the septum
hardly or slightly constricted, occasionally pronouncedly
constricted, distal cell±round, rounded at the apex, about as
Fig. 3 Puccinia aff. scirpi (ZT Myc 3304). a Urediniospores, optical
section. b Urediniospores, focus on spore surface. Note large bald
patch proximal to germ pores. c Teliospores. Bars 20 μm
Mycol Progress (2012) 11:483–497 487

long as the conical proximal cell which is tapering towards the
subhyaline to light brown and slightly thick-walled pedicel,
33–65×15.5–23.5 μm(47.3×19.3μm), wall smooth, light to
dark chestnut brown and 1–2 μm thick in distal cell with apex
to 2–5(6) μm thick, in proximal cell pallid brown to light
chestnut and about 1 μm thick; germ pores indistinct, apical in
distal cell, probably close to the septum in the proximal cell;
one-celled spores few to numerous.
Puccinia subcoronata Henn. Hedwigia 34:94 (1895), a new
record for South Africa and the Old World on a new host
plant (Fig. 4)
Syn.: Puccinia antioquiensis Mayor. Mém. Soc. Neuch.
Sci. Nat. 5: 473 (1913) (fide Jackson 1926 ).
Material examined: South Africa , KwaZulu-Natal:
Durban, Westville, Roosfontein Nature Reserve, on Cype-
rus albostriatus Schrad., leg. AR Wood no. 705, 27 Dec
2006 (ZT Myc 3302. II, III) and no. 753, 15 Jun 2009
(PREM 59915, ZT Myc 2786. II, III). Brazil : Goyaz, near
Goyaz, on Cyperus diffusus Vahl, leg. E. Ule no. 1985, Feb
1893 (HBG, holotype. II, [III]).
Puccinia collected on Cyperus in Natal could not be
identified as any of the Puccinia species reported to date on
Cyperus in Africa [P. canaliculata, P. conclusa, P. cyperi
Arth., P. cyperi-cristati Gjærum, P. cyperi-fastigiati Doidge,
P. cyperi-tegetiformis Kern, P. flavo-virens Jacks., P.
morganae Doidge, P. morotoensis Gjærum, P. pegleriana
Doidge, P. philippinensis Syd. & P. Syd., P. pottsii Doidge,
P. romag noliana Maire & Sacc., P. transvaalensis Doidge]
(e.g. Doidge 1927; Gjærum 1984). It had some resemblance
to the widespread P. canaliculata though it differed in
shorter teliospores (28–56 vs. 39–64
μm) with a less
thicke
ned spore apex (2.5–5(6) vs. 3–10 μm) and a more
weakly pigmented spore wall (Barreto and Evans 1995;
Kern 1919; Zhuang 1988). It should be noted that Barreto
and Evans (1995) synonymize P. canaliculata with P.
conclusa, P. philippinensis and P. romagnoliana.
The species which it resembled most was P. subcoronata.
This rust is characterized by thin-walled uredinio- and
teliospores and an only slightly thickened teliospore apex
(Mayor 1914;Kern1919). The South African specimen
tallied well with all characters of P. subcoronata. However,
we observed the presence of a few one-celled mesospores
and that telia were immersed in a chestnut brown stromatic
hyphal mass (Fig. 3), characters that have not been described
for P. subcoronata. In the type of P. subcoronata,onlytwo
tiny telia were discovered of which one was sectioned. It was
bounded by a palisade-like sterile hyphal layer and contained
a single one-celled spore among two-celled teliospores. The
present specimen is therefore assigned to P. subcoronata.
This is the first report of this rust fungus outside of the New
World on a new host species.
Uromyces cypericola Gjærum. Lidia 3: 22 (1990), new
for South Africa, with a description of the urediniospores
(Fig. 5)
Material examined: South Africa, KwaZulu-Natal: Sc ott-
borough, Roberts Nature Reserve, on Cyperus albostriatus
Schrad., leg. AR Wood no. 706, 5 Jun 2006 (PREM 59913.
ZT Myc 3303. (II), III).
ArustfunguscollectedonCyperus in KwaZulu-Natal was
identified as U. cypericola hitherto only known from the
type specimen from Kenya on Cyperus cyperoides Kük. ssp.
pseudoflavus (Kük.) Lye (Gjærum 1990b). It had shorter
teliospores than the type of U. cypericola [22–31(35)×13–
20 μm(25.9×16.5μm) vs. 27–40×15–19 μm] and bore
urediniospores (Fig. 5
) so far unknown in U. cy
pericola:
Fig. 5 Uromyces cypericola (ZT Myc 3303), urediniospores. Bar 20 μm
Fig. 4 Puccinia subcoronata (ZT Myc 2786), section through telium
showing subepidermal position (ed epidermis) and the presence of a
thin hyphal peridium (arrow) attached to the epidermis. Bar 100 μm
488 Mycol Progress (2012) 11:483–497

Uredinia not observed; urediniospores scattered on the
surface of telia, obovoid to subglobose, wall light orange
brown, ca. 1.5(−2) μm thick, more or less evenly echinulate
with spines ca. 2–3 μmapartand0.5–1 μm long, germ pores
2, more or less opposite, equatorial or super-equatorial, with
or without(?) a bald patch proximal to the pores.
Other Uromyces spp. known on Cyperus, namely U.
bermudianus Cummins, U. cyperi Henn., U. cyperinus
Speg. and U. notabilis Wakef., did not tally with the
characters of the present specimen. This is the second
collection of U. cypericola, a new report for South Africa
and a new host species.
Euphorbiaceae
Crossopsora antidesmae-dioicae (Syd. & P. Syd.) Arth.
& Cummins. Philipp. J. Sci., C, Bot. 61: 474 (1937), first
report from South Africa
Material examined: South Africa, KwaZulu-Natal:
Uvongo, Skyline Nature Reserve, on Antidesma venosum
E. Mey. ex Tul., leg. AR Wood no. 720, 29 May 2008
(PREM 60114, ZT Myc 579. II).
Crossopsora anti desmae-dioicae has been reported on
the same host species as the South African specimen from
Ivory Coast (Viennot-Bourgin 1958) and Uganda (Cum-
mins 1945). It is otherwise known from China, Indonesia,
Japan, New Guinea, the Philippines and Vietnam. The
known localities indicate that C. antidesmae-dioicae is
widely distributed in warm regions of the Old World.
Phakopsora stratosa (Cooke) Arth. Bull. Torrey Bot. Club
44: 508 (1917), first report of spermogonia and aecia
and observations on the morphology of the uredinia
and urediniospores (Fig. 6)
Material examined: South Africa, KwaZulu-Natal: Durban,
Yellowwood Park, Kenneth Stainbank Nature Reserve, on
Croton sylvaticus Hochst., leg. AR Wood no. 698, 14 Dec
2006 (PREM 60109, ZT Myc 581. 0, I, II, III); Inanda, on
C. sylvaticus, leg. JM Wood no. 466, 12 Jan 1881 (PREM
10363, Isotype. II); Winkle Spruit, on C. sylvaticus, leg. IB
PoleEvans s. n., 14 Feb 1913 (PREM 5639, II); Eshow e, on
C. sylvaticus, leg. APD McClean s. n., Apr 1941 (PREM
33093, II, III).
Phakopsora stratosa has been recorded previously from
South Africa on C. sylvaticus (Sydow and Sydow 1915;
Wood 2006) and from Ethiopia on C. macrostachys
(Castellani and Ciferri 1950). An investigation of recently
collected material revealed the presence of spermogonia
and
aecia
that have not been observed hitherto and some
morphological characters that supplement existing descrip-
tions (Doidge 1927; Sydow and Sydow 1915):
Spermogonia subcuticul ar with flat hymenium, type 7,
120–200(−250) μ m diam., amphigenous on leaves, in
dense groups on large, round, brown spots that become
necrotic wit h age, or on sm all gall-like swellings of
branchlets and leaf petioles; groups of spermogonia
surrounded by old and mostly empty aecia. Aecia Uredo-
like. The few aeciospores present were pedicellate and
identical with the urediniospores. Uredinia amphigenous on
leaves, scattered or in small groups, pustulate with a
peridium adhering to the overarching host epidermis
(Milesia-type), pulveru lent, cinnamon brown; uredinio-
spores (Fig. 6) pedicellate, narrowly obovoid to subclavate,
slightly asymmetrical with a ventral, slightly flattened side
and a dorsal, more convex side, spore wall pale golden
brown, indistinctly bi-layered, the inner layer being more
deeply pigmented, ca. 3–4 μm thick at sides a round the
two, more or less equatorial and opposite germ pores, up to
10 μm thic k at apex, echinulate by rather coarse spines
which are ca. 4–6 μm apart – except for a smooth patch on
the ventral side of the spore.
Fig. 6 Phakopsora stratosa (ZT Myc 581), urediniospores. a Optical
section. b F ocus on spore surface. Note lateral patch l acking
echinulation. Bar 20 μm
Mycol Progress (2012) 11:483–497 489

Both Sydow and Sydow (1915) and Doidge (1927)
described urediniospores with four germ pores and did not
mention the asymmetry of the spores and the presence of a
smooth ventral patch. However, after staining, only two
pores were observed in all specimens studied by us
including an isotype (PREM 10363). The occurrence of
spermogonia and aecia together with uredinia and telia
prove that P. stratosa is a macrocyclic autoecious rust
species. Unfortunately, the aecia were old and mostly
parasitizedbyaTuberculina species, so that a full
description cannot be given.
Fabaceae
Sphaerophragmium dalbergiae Dietel. Hedwigia 32: 30
(1893), a revised description of the urediniospores
Material examined: South Africa, KwaZulu-Natal: Umko-
maas, Empisini Nature Reserve, on Dalbergia armata E.
Mey., leg. AR Wood no. 709, 6 Jun 2007 (PREM 59914.
II); Durban, on D. armata, leg. IB PoleEvans s. n., 13 Apr
1911 (PREM 1371, II); Durban, Umgeni beach, on D.
armata, leg. EM Doidge s. n., 7 Jul 1912 (PREM 2524. II).
Lohsomboon et al. (1994) stated that the number of germ
pores of the urediniospores of S. dalbergiae could hardly be
discerned and gave it as 1–2 in accordance with the
descriptions provided by Monoson (1974, one pore) and
Dietel ( 1893, two pores). Viennot-Bourgin (1958 ) counted
four pores in an equatorial position. After staining the
spores, germ pores were more easily visible and 4–7 pores
were observed most often in an approx. equatorial position,
more rarely tending to be scattered. Our urediniospore
measurements of 23–34×13–15.5 μm (27.0×14.2 μm, for
22 spores, PREM 59914) tallied quit e well with the
measurements published by Lohsomboon et al. (1994 )in
a specimen on D. armata from Natal [21–34×11–22 μm,
(27×16.5 μm)], while Viennot-Bourgin (1958) measured
20–31×15–25 μm(26×24.5μm) in a specimen on
Dalbergia sp. from the Ivory Coast. The second mean
value may be an error. The uredinial paraphyses observed
in our specimen agreed well with the description and
illustration published by Viennot-Bourgin (1958).
Malvaceae
Endophyllum
mpenjatiense A. R. Wood & R. Berndt, spec.
nov. (Fig. 7)
Etymology: after the locus classicus, the Mpenjati Nature
Reserve.
Differ t ab Aecidiis spp. Hibiscorum spp. micromorphologia
et germinatione sporarum basidiis.
In foliis Hibisci sp. verosimiliter.
Telia Aecidium-like, mainly on abaxial side of leaves, in
dense groups on bleached leaf spots, telia cup-shaped, 150–
290 μm diam., short, with peripheral peridium whose
margin is frayed or lacerating into a few lobes, peridial cells
only weakly adherent and easily separating, irregularly
shaped but generally more or less rectangular, outer
periclinal wall ca. 6 μm thick, fine ly and distantl y
verrucose, inner wall ca. 2–3 μm thick, more coarsely and
very variably verrucose with irregular and rather distant
warts; spore mass yellow when fresh, ebony or whitish in
herbarium, subpulverulent, cottony-fibrous u pon germina-
tion of basidia, teliospores subglobose, ellipsoid, broadly
ellipsoid, often slightly deformed or subangular, 17–21.5×
12–16 (18.8×14.1 μm), spore wall ≤ 1 μm, very finely and
densely verruculose with some interspersed bigger warts,
becoming almost smooth or smooth at sp ore base,
germinating in situ with basidia upon maturity, basidia
50–85×6–10 μm(n=8), 1 or 2 septate, producing two thin-
walled basidiospores on sterigmata; basidiospores ca.
10 μm in diameter.
On leaves of cf. Hi biscus sp.
South Africa, KwaZulu-Natal: Mpenjati Nature Reserve
between Ramsgate and Port Shepstone, on non-fl owering
Malvaceae plant, probably Hibiscus sp., leg. AR Wood no.
719, 28 May 2008 (holotype PREM 60113, isotype ZT
Myc 583. III).
Additional material examined: Aecidium garckeanum
Henn. Ethiopia (Abessinien, Coloni a Eritrea): Unter Geleb,
alt. 1,700 m, auf Hibiscus micrant hus, leg. Schweinfurth s.
n., 11 Apr 1891 (B 700014687 : Lectotype, designated here.
0, I). Am Sabor Geleb, alt. 2,200 m, auf H. crassinervius,
leg. Schweinfurth s. n., 20 Apr 1891 (B 700014688:
Syntype. 0, I).
Aecidium garckeanum Henn. and A. hibisci-surattensis
Mayor & Vienn.-Bourg. are known on Hibiscus spp. in
Africa, the former also from South Africa (Doidge 1927 ).
T
ype
material of A. garckeanum was compared with E.
mpenjatiense and found to be different in characters of
spores, peridium and life cycle. As Hennings (1891) did not
designate a type specimen for A. garckeanum, one of the
syntypes kept in Berlin, B 700014687, is selected here as
lectotype. It showed spermogonia of group V / type 4
(classification acc. to Hiratsuka and Cummins 1963) that
had not b een described before. Marchal and Steyaert (1929)
proposed Puccinia garckiana Marchal & Steyaert to which
they assigned A. garckeanum as the aecial anamorph.
Sydow and Sydow (1924) suggested that A. garckeanum
may be the same as A. erythrobasis Berk. & Broome
described from Sri Lanka but this species has larger spores
(Gjærum 1985) and is also different from the present
species. A. hibisci-surattensis Mayor & Vienn.-Bourg. is
very similar to E. mpenjatiense but has slightly larger
spores and a peridium smooth on the outside; the spore
490 Mycol Progress (2012) 11:483–497

germination has not been recorded (Mayor and Viennot-
Bourgin 1951).
Poaceae
The rust fungi of Ehrharta (Ehrharteae)
Material examined: Uromyces ehrhartae-giganteae: South
Africa, Western Cape: Cape Town, Mowbray, on E.
thunbergii Gibbs Russell (= E. gigantea Thunb.), leg. CP
Van der Merwe, 10 Feb 1924 (PREM 7392, holotype. III.
PUR F14562, isotype. III).
U. quaggafonteinus: South Africa, Northern Cape:
Kamiesberg M ountains, alon g gravel r oad between
Kamieskroon and Leliefontein, on E. calycina J.E. Sm., leg.
E. Uhlmann & R. Berndt, 15 Oct 2004 (RSA 21. II, III).
Kamiesberg Mountains, along gravel road from Kamiesk-
roon to Soebatsfontein, on E. calycina, leg. E. Uhlmann &
R. Berndt, 15 Oct 2004 (RSA 23. II, III). Western Cape:
Vanrhynsdorp, gravel road to Gifberg, on E. cf. calycina,leg.
Rössel, Uhlmann & Berndt, 12 Oct 2005 (RSA 186. II, III.
Specimen probably represents a mixed infection as two
kinds of urediniospores occur). Cederberge, Blinkbergpass,
on E. calycina, leg. Rössel, Uhlmann & Berndt, 17 Oct 2005
(RSA 240. (II), III. Specimen may represent a mixed
infection as two kinds of urediniospores occur).
Uromyc es indet.: South Africa, Northern Cape:
Kamiesberg Mountains, on a granite koppie near Leliefontein,
Fig. 7 Endophyllum mpenja-
tiense. a Teliospores germinat-
ing with two-celled basidia
(holotype, bars 10 μm). b
Teliospores with finely verrucu-
lose surface. Arrows indicate
spores where bigger warts are
interspersed among minute ones.
c Peridial cells: delicately ver-
rucose outer periclinal wall (ar-
row). d Peridial cells: irregularly
verrucose inner periclinal wall
(b–d, isotype, bars 20 μm)
Mycol Progress (2012) 11:483–497 491

on Ehrharta barbinodis Nees ex Trin., leg. E. Uhlmann &
R. Berndt, 15 O ct 2004 (RSA 17. II, III). Along gravel
road from Kamieskroon to Soebatsfontein, on E. cf.
barbinodis, leg. E. Uhlmann & R. Berndt, 17 Oct 2004
(RSA 24. I I, III). Barrydale, on E. erecta Lam. var. erecta,
leg. E. Uhlmann & R. Berndt, 23 Oct 2004 (RSA 51. (II),
III). Western Cape: Road from Clanwilli am to Pakhuis
Pass, on Ehrharta sp., leg. Röss el , Uhlmann & Berndt, 13
Oct 2 005 (RSA 193. (II ), III). Ceder berge, at Pakhuis
Pass, on E. cf. thunbergii,leg.Rössel,Uhlmann&Berndt,
13 Oct 2005 (RSA 201. III). Cederberge, Uitkyk Pass, on
E. cf. thunbergii,leg.Rössel&Berndt,15Oct2005(RSA
235. (II), III).
Three ru st f un gi have been des cr i bed on m em b ers o f
the genus Ehrharta in South Africa: Uromyces ehrhartae-
giganteae Doidge, U. quaggafontein us Mennicken &
Oberw., and Uredo ehrhartae-calycinae Doidge. Uromy-
ces ehrhartae-giganteae is only known in the telial state
whereas telio- and urediniospores were described for U.
quaggafontei nus. Mennicken and Oberwinkler (2004)
assigned Uredo ehrhartae-calycinae to U. ehrhartae-
giganteae as its uredinial anamorph. This assumption can
neither be verified nor falsified on grounds of morpholo-
gy, as the type specimens of U. ehrhartae-giganteae kept
in PREM and PUR do not contain any uredinia or
urediniospores.
We collected two species of Uromyces bearing telio- and
urediniospores on Ehrharta in the northern and western Cape
Provinces. One of them was identified as U. quaggafonteinus,
the other one could not be determined. It differed from U.
quaggafonteinus in larger urediniospores and from U.
ehrhartae-giganteae in the presence of a uredinial state. It is
very probable that the uredinial state is identical with Uredo
ehrhartae-calycinae as the urediniospores are most similar
(Table 1). Uromyces quaggafonteinus and the undetermined
species cannot be safely separated from U. ehrharta e-gigante ae
by teliospore characters. As urediniospores are unknown in U.
ehrhartae-giganteae both species may be distinct from it or
each of them may be the same as U. ehrhartae-g iganteae.
The available collections indicate that U. quaggafonteinus
may be restricted to E. calycina
while the undetermined
spec
ies appears to have a wider host range comprising E.
barbinodis, E. calycina, E. erecta. var. erecta, E. cf .
thunbergii and E. villosa. Uromyces ehrhartae-giganteae is
presently only known from E. thunbergii (= E. gigantea).
Rhamnaceae
Rust fungi of Ziziphus in southern Africa
Two rust fungi, Crossopsora ziziphi (S yd. & Butl.) Syd. &
P. Syd. and Phakopsora ziziphi-vulgaris Dietel, have been
recorded from species of Ziziphus in Africa. A specimen
recently collected on Z. mucronata Willd. was different
from both of them and is described as new.
Uredo sekhukhunensis R. Berndt & A. R. Wood, spec. nov.
(Fig. 8) Syn. Cronartium ziziphi auct. Doidge EM 1948.
Bothalia 4: 898. Non C. ziziphi Syd. & Butl. 1912.
Etymology: denominating the Sekhukhunene region
where the type specimen was collected.
Uredinia in gregibus parvis pagina abaxiali foliorum
insidentia sparsa nonnihil confluentia subepidermalia pallide
brunnea, annuliformiter corona ferruginea paraphysium cir-
cumdata, mature exposita pulverulentiaque; urediniosporae
ellipsoideae subclavatae subpyriformes ad oblongae, 26–42×
14.5–20.5 μm(32.3×17.2μm), pariete 1–2 μm crasso leniter
tenuiore hilum versus stramineo vel pallide– ad aureo-
brunneo, echinulato spinis sate delicatis hilum versus decres-
centibus, deinde levi, poris germinationis verosimiliter 4–6
approx. aequatorialibus papillis carentibus; paraphyses
periphericae basaliter junctae (uredinia typi Physopellae)sine
septis vel simplices septatae cylindricae ad subclavatae rectae
vel leniter incurvatae saepe irregulariter distortae, pariete
brunneo apicaliter et dorsaliter valde incrassato.
In foliis Ziziphi mucronatae Willd.
Uredinia scattered, but generally in small groups and
slightly confluent on abaxial side of leaves, pale brown,
with darker, ferrugineous border of paraphyses, subepider-
mal, early exposed and pulverulent; urediniospores ellip-
soid, subclavate, subpyriform to oblong, 26–42×14.5–
20.5 μm (32.3×17.2 μm), spore wall 1–2 μmthick,
slightly thinner towards hilum, straw-colored to pale or
orange brown, echinulate by sharp and rather delicate
spines which become finer towards the spore base and
vanish on the proximal third of the spores, germ pores
difficult
to
discern, probably 4–6, approx. equatorial to
almost scattered, without caps; paraphyses peripheral and
only basally united (uredinia of Physopella-type), without
or with a single septum, cylindrical to subclavate, sometimes
straight or slightly bent inwardly, generally distorted or
gnarled or geniculate, cell wall brown, irregularly thickened,
dorsally and apically often strongly so.
On leaves of Ziziphus mucronata Willd.
South Africa, Mpumalanga: S of Penge, track to Serowe
gorge, 24°26′S30°26′E, on Z. mucronata,leg.ARWoodno.
733, 27 Jul 2008 (PREM 60080, holotype. ZT Myc 952,
isotype. II).
Additional material investigated: on Z. mucr onata. – South
Africa, Limpopo: Between Acornhoek and Polokwane (Tza-
neen), leg. LCC Liebenberg no. 3608, Jun 1939 (PREM
32724. II). D’Nyala Nature Reserve, Ellisras, leg. M. van
Reenen no. 2/1, 28 Sep 1989 (PREM 50573. II). Mpumalanga:
Burgersfort, leg. LCC Liebenberg no. 6215, May 1959 (PREM
41983. II). Lowveld Botanical Garden, Nelspruit, leg. M. van
Reenen no. 5/11 , 26 Apr 1990 (PREM 50576. II). Berg-en-
492 Mycol Progress (2012) 11:483–497

Table 1 Rust fungi of Ehrharta; host plants and morphological characters of uredinio- and/or teliospores (n.a. not assessed); sources of data not assessed by the authors are added in parentheses
Rust taxon, investigated specimen,
host plant
Urediniospores Mean Teliospores Mean
Uromyces ehrhartae-giganteae
Doidge (type); on E. gigantea
Unknown 23–37×16–24 μm (Doidge 1927); (23)26–36
(38)×(16)19–24(26) μm (Cummins 1971);
26–42(45)×17–26 μm, tending to be
dimorphic, wall about 2–2.5 μm at sides and
5–8(12) μm at apex
32.1×21.7 μm
Uromyces indet.
RSA 193; on Ehrharta sp. Very few present, similar to following specimens
but not measured
n.a. tending to be dimorphic, chestnut brown ones
25–33×21–26.6(28) μm, more lightly
pigmented ones 27–38×14–19.5 μm
29.2×23.8 μm, 32.2×17.8 μm
general: 30.7×20.8 μm
RSA 51; on Ehrharta sp. 23–28×18–22 μm, germ pores 7–9, scattered
(inconspicuous)
25.0×20.4 μm26–35×19–28 μm 29.4×24.0 μm
RSA 17; on E. barbinodis 23.5–28×18–21.5 μm, germ pores numerous,
scattered (difficult to see, number?)
25.1×19.7 μm24–32×19–25.5 μm 28.8×22.1 μm
RSA 24; on E. cf. barbinodis 22–27×17–23 μm, germ pores 5–7, scattered
(inconspicuous)
24.3×20.3 μm25–33(39)×18–26 μm 27.7×22.5 μm
RSA 186; on E. cf. calycina 21–25×20–22 μm, germ pores 7–10, scattered 22.5×20.9 μm not assessed as collection appeared to represent
a mixed infection
n.a.
RSA 201; on Ehrharta cf. thunbergii Not present 27–39×19–29 μm 32.5×24.2 μm
RSA
235;
on E. cf. thunbergii 20–25×20–24 μm, germ pores 8–12, scattered 22.4×21.6 μm 24.5–37.5×19–26.5 μm 29.7×23.6 μm
Mennicken no. 190, 192, 20; on E.
calycina and E. villosa sub U.
ehrhartae-giganteae
20–31×19–26 μm, germ pores 6–8, scattered
(Mennicken and Oberwinkler 2004)
n.a. 26–43(46)×19–32 μm, tending to be dimorphic
(Mennicken and Oberwinkler 2004)
n.a.
Uredo ehrhartae-calycinae Doidge;
on E. calycina
(20)22–27(29)×(17)19–21 μm, germ pores 5–8,
scattered, wall 3–4 μm thick (Cummins 1971)
n.a. Unknown
Uromyces quaggafonteinus
Mennicken no. 139 (type); on E. calycina 26–36×24–32 μm, germ pores 5–9, scattered, wall
2.5–4 μm thick (Mennicken and Oberwinkler 2004)
n.a. 28–39×24–32 μm (Mennicken and
Oberwinkler 2004)
n.a.
RSA 23;
on E. calycina 27–34(37)×24–28.5 μm, germ pores 6–8, scattered 30.7×26.1 μm26–34×22.5–27 μm 29.9×24.5 μm
RSA 21; on E. calycina 25–33×22–27 μm, germ pores 6–8, scattered 29.3×24.7 μm27–36(39)×21.5–27(28) μm 31.0×23.5 μm
RSA 240; on E. calycina 26–33(35)×23–28 μm, germ pores 6–8, ± scattered,
wall ca. 3–4 μmthick
29.6×25.7 μm abundantly present but not assessed as collection
may represent a mixed infection
n.a.
RSA 186; on E. cf. calycina 25–32×23–25 μm, germ pores 7–10(11), scattered 27.7×24.3 μm not assessed as collection appeared to represent
a mixed infection
n.a.
Mycol Progress (2012) 11:483–497 493

Dal, Kruger National Park, leg. M. van Reenen, 15 Jul 1991
(PREM 50696. II). Skukuza research camp, Kruger National
Park, leg. AR Wood no. 646, 6 Apr 2006 (PREM 59108. II).
Crossopsora ziziphi (Syd. & Butl.) Syd. & P. Syd.
Zambia: Gwembe Valley, on Z. abyssinica Hochst. ex A.
Rich., leg. A. Angus, 17 Jun 1961 (PREM 43807, IMI
90070. II). Vietnam (as ‘Tonkin’): Cho Gonh, on Ziziphus
sp., leg. A. Petelot, Nov 1922 (Syd. Fungi exot. exsicc. no.
972, ZT Myc 1272, type. II, III). India, Goa State: between
Honavar and Jog Falls, ca. 22 km E of Honavar, on
Ziziphus sp., leg. R. Berndt, 29 Oct 1995 (HeRB B-210. II).
Karnataka State: on coastal road ca. 28 km N of Mangalore,
on Ziziphus sp., leg. ND Sharma & R. Berndt, 30 Oct 1995
(HeRB B-218. II, III).
Uredo sekhukh unensis (Fig. 9) differs from Crossopsora
ziziphi that has shorter, slightly asymmetrical urediniospores
with three more or less equatorial germ pores and uredinia with
straight thin-walled paraphyses. Phakopsora ziziphi-vulgaris
differs in smaller urediniospores and uredinia of Milesia-type;
the uredinial stage of the Indian Kuehneola ziziphi (Ramakr.
& Subram.) Thirum. in thick-walled and smaller uredinio-
spores with scattered germ pores according to the description
given by Ragunathan and Ramakrishnan (1973).
Fig. 8 Uredo sekhukhunensis (ZT Myc 952, isotype). a Uredinio-
spores. b Uredinial paraphyses. Bars 20 μm
Fig. 9 Dietelia cardiospermi (ZT Myc 949). a Section through young
telium still covered by host epidermis (ed) and cellular peridium (pd).
bar 50 μm. b Section through telium showing teliospore chains with
teliospores and intercalary cells (arrows). c Teliospores and basidia
scraped from mature telium, partly hidden by debris. Bar 20 μm
494 Mycol Progress (2012) 11:483–497

In Africa, Crossopsora ziziphi has been recorded from
Eritrea (sub Uredo zizyphi Pat., Castellani and Ciferri
1937), Zimbabwe (Whiteside 1966), and South Africa
(Doidge 1948, sub Cronartium ziziphi). The description
and illustration presented by Doidge (1948) indicate that
not C. ziziphi but another rust species is involved in South
Africa. Therefore, all South African specimens of C. ziziphi
kept in PREM were examined and all those on Z.
mucronata were found to represent the new species, U.
sekhukhunensis. C. ziziphi has to be deleted from the lists of
South African and Zimbabwean rust fungi.
Phakopsora ziziphi-vulgaris Dietel. Ann. Mycol. 8: 469
(1910), new for South Africa
Material examined: South Africa, Limpopo: Bela Bela
(Warmbaths), Klein Kariba Resort, on Z. mucronata, leg.
AR Wood no. 666, 23 May 2006 (PREM 60107, ZT Myc
580. II, III). India, Maharashtra State: Pune, Pune
University campus, near the guesthouse, on Z. cf. jujuba
Mill., leg. R. Berndt, 18 Nov 1995 (HeRB 4008. II/III).
Bihar State: Pusa, on Z. jujuba, leg. EJ Butler, 5 Feb 1913
(Sydow, Fungi exot. exsic c. no. 165, ZT Myc 953 and Z s.
n. II, III). Israel: Lidda, on Z. spina-christi Willd., leg. T.
Rayss, 14 Nov 1951 (ZT Myc 962. II).
Phakopsora ziziphi-vulgaris is widely distributed in
warm regions of the Old and New World. In Africa, it has
only been recorded so far in Eritrea (Castellani and Ciferri
1937), in Uganda (Gjærum 1998) and from the Cape Verde
Islands (Gjærum 1974). It was recent ly collected for the
first time in South Africa on Z. mucronata wi th both
uredinia and telia present. The South African specimen
differed from other specimens by very finely echinulate or
almost smooth urediniospores with slightly thicker spore
walls; the remaining characters agreed well, however.
Rosaceae
Pucciniastrum agrimoniae (Dietel) Tranzschel. Scripta Bot.
Horti Univ. Imper. Petrop. 4: 301 (1895), new host genera
and species from South Africa
Doidge (1950) listed P. agrimoniae for South Africa on the
indigenous Agrimonia bracteata E. Mey ex C.A. Mey. (sub
A. eupatoria L. var. capensis Harv.). It is also recorded
herein on Leucosidea sericea Eckl. & Zeyh. and Cliffortia
odorata L. To our knowledge, these are new host species
and genera and the first records of P. agrimo niae
on genera
other than Agrim
onia. All three genera belong to the same
tribe, Sanguisorbeae of Rosoideae (Potter et al. 2007). P.
agrimoniae shows a disjunct distri bution between the
northern and southern hemisphere, similar to some fern
rusts of the Pucciniastraceae recorded from South Africa
(Berndt 2008a). Teliospores have not been observed in
South Africa to our knowledge and the rust fungus may be
able to persist in the uredinial state.
Material examined: on Leucosidea sericea. – South
Africa, KwaZulu-Natal: Pietermaritzburg, Natal Botanical
Garden, leg. AR Wood no. 135, 21 Jan 1999 (PREM
59039. II). Pietermaritzburg, Natal Botanical Garden, leg.
AR Wood no. 561, 14 May 2004 (PREM 59040. II).
Mpumalanga: Long Tom Nature Reserve, Long Tom Pass,
NE of Lydenburg, leg. AR Wood no. 148, 12 May 1998
(PREM 59038. II). On Cliffortia odorata. – South Africa,
Western Cape: Knysna, Knysna Forest, Lelievlei Nature
Reserve, “Terblans Walk”, leg. E. Uhlmann & R. Berndt,
25 Oct 2004 (RSA 66. II). George, Outeniqua Nature
Reserve, leg. AR Wood no. 379, 4 June 2002 (PREM
59042. II). Cape Town, Skeleton Gorge in Kirstenbosch
Botanical Garden, leg. AR Wood no. 513, 17 March 2003
(PREM 59041. II).
Sapindaceae
Dietelia cardiospermi (Cooke) R. Berndt & A. R. Wood,
comb. nov. (Fig. 9)
Basionym: Aecidium cardiospermi Cooke. Grevillea 10:
125 (1882). Type on Cardiospermum microcarpum Kunth,
Cooke no. 537
Material examined: South Africa , KwaZulu-Natal:
Ngwavuma River about 1 km S of Ndumo village, on C.
halicacabum L. var. halicacabum, leg. R. & A. Urb an, 15
Jul 2008 (PREM 60085, ZT Myc 949. III).
“Aecidium serjaniae” Henn. Brazil: São Paulo, Campinas,
on leaves of Serjania sp., leg. F. Noack, 7 Nov 1896 (Sydow,
Uredineen no. 2197, ZT Myc 2992. I). Venezuela,State
Yaracui: between S. Pedro and Campo Elias (Panamericana
highway between Barquisimeto and San Felipe), on cf.
Serjania sp., leg. R. Berndt, 25 Nov 1993 (HeRB A-179,
located at ZT. I).
Aecidium cardiospermi was described on the type host
C. halicacabum L. var. microcarpum (Kunth.) Blume from
KwaZulu-Natal and found on the same plant species in
Zanzibar, Kenya (Sydow and Sydow 1903). We observed
that the ‘aeciospores’ germinated in situ producing basidia
and regard them as teliospores accordingly. Because of the
firmly adherent teliospores (Fig. 9a, b) we transfer A.
cardiospermi to Dietelia
(Buriticá 19
91
; B uriticá and
Hennen 1980)asD. cardiospermi. The formation and
exact morphology of the basidia was difficult to assess as
the peridial cells, basidia and compressed, empty telio-
spores formed a tangled mass on the mature sori. As far as
observable, the basidia were two-celled and formed
basidiospores on stout sterigmata (Fig. 9c). According to
our knowledge D. cardiospermi has not been reported from
Mycol Progress (2012) 11:483–497 495

outside of Africa. The host genus Cardiospermum is most
speciose in the New World, but C. halicacabum is regarded
as occurring naturally in South Africa by Arnold and de
Wet (1993). A number of Aecidium species were described
on Cardiospermum in the New World: A. serjaniae Henn.
1896, A. cardiospermophilum Speg. 1909 (= A. serjaniae
Speg. 1925) and A. reichei Dietel 1914. Aecidium ure-
dinoides Henn. may not grow on a sapindaceous host but a
member of Annonaceae (Sydow and Sydow 1924). Aeci-
dium serjaniae Henn. was excluded from rust fungi by
Sydow and Sydow (1924). The Aecidium issued in ‘Sydow,
Uredinee n ’ no. 2197 under that name is of uncertain
identity, therefore, but it has been found to be different
from the present speci es. Aecidium reichei and A. cardio-
spermophilum are different from the present rust as well
according to the descriptions given by Sydow and Sydow
(1924) and Lindquist (1982).
Acknowledgements We thank Anja Rössel and Elisabeth Uhlmann
for companionship and assistance in the field, the German Ministry of
Education and Research (BMBF) for funding the field stays of R.B.,
and the Western and Northern Cape Nature Conservation Boards in
South Africa for issuing collecting and export permits. Loans of
specimens granted by the curators and directors of B, HBG, PDD,
PREM, PUR and VPRI are gratefully acknowledged. Roger Peterson
and an unknown reviewer are thanked for helpful corrections and
comments on the manuscript.
References
Arnold TH, De Wet BC (eds) (1993) Plants of southern Africa: Names
and distribution. Memoirs of the Botanical Survey of South
Africa no. 62. The National Botanical Institute, Pretoria, South
Africa
Barreto RW, Evans HC (1995) Mycobiota of the weed Cyperus
rotundus in the state of Rio de Janeiro, with an elucidation of its
associated Puccinia complex. Mycol Res 99:407–419
Berndt R (2008a) The rust mycobiota of southern Africa: species
richness, composition and affinities. Mycol Res 112:463–471
Berndt R (2008b) The rust fungi (Uredinales) on ferns in South
Africa. Mycol Prog 7:7–19
Berndt R (2010) The Puccinia species of Berkheya (Asteraceae) with
description of four new species. Mycologia 102:1437–1449
Berndt R, Uhlmann E (2006) New species, reports, observations and
taxonomical changes of southern African rust fungi. Mycol Prog
5:154–177
Buriticá P (1991) Familias del orden Uredinales con ciclo de vida
completamente reducido. Rev Acad Colomb Cienc Exactas
18:131–148
Buriticá P, Hennen JF (1980) Pucciniosireae (Uredinales, Puccinia-
ceae) Flora Neotropica monograph no. 24. The New York
Botanical Garden, New York, USA
Castellani E, Ciferri R (1937) Prodromus mycoflorae africae orientalis
italicae. Biblioteca Agraria Coloniale, Istituto Agricolo Coloniale
Italiano, Firenze, Italy
Castellani E, Ciferri R (1950) Mycoflora erythraea, somala et
aethiopica – Suppl. I. Atti dell’ Istituto Botanico della Universitá
di Pavia, Laboratorio Crittogamico (suppl.), Serie 5, vol. H:
1–51
Crous PW, Rong IH, Wood A, Lee S, Glen H, Botha W, Slippers B, de
Beer WZ, Wingfield MJ, Hawksworth DL (2006) How many
species of fungi are there at the tip of Africa? Stud Mycol 55:13–33
Cummins GB (1945) Description of tropical rusts – VII. Bull Torrey
Bot Club 72:205–222
Cummins GB (1952) Uredinales from various regions. Bull Torrey
Bot Club 79:212–234
Cummins GB (1971) The rust fungi of cereals, grasses and bamboos.
Springer, New York
Dianese JC, Medeiros RB, Santos LTP, Furlanetto C, Sanchez M,
Dianese AC (1993) Batistopsora gen. nov. and new Phakopsora,
Ravenelia, Cerotelium, and Skierka species from the Brazilian
cerrado. Fitopatol Bras 18:436–450
Dietel P (1893) Sphaerophragmium dalbergiae n. sp. Hedwigia
32:30–32
Doidge EM (1927) A preliminary study of the South African rust
fungi. Bothalia 2:1–228
Doidge EM (1939) South African rust fungi, III. Bothalia 3:487–
512
Doidge EM (1941) South African rus t f ungi, IV. Bothalia 4:229–
236
Doidge EM (1948) South African rust f ungi, V. Bothalia 4:8 95
–
918
Doidge
EM
(1950) The South African fungi and lichens to the end of
1945. Bothalia 5:1–1094
Gäumann E (1959) Die Rostpilze Mitteleuropas. Büchler, Bern
Gjærum HB (1974) Uromyces anthyllidis and Phakopsora zizyphi-vulgaris
new to the Cape Verde Islands. V ieraea 3:100–102
Gjærum HB (1984) East African rusts (Uredinales), mainly from
Uganda. 2. On Cyperaceae. Mycotaxon 20:53–63
Gjærum HB (1985) East African rusts (Uredinales), mainly from
Uganda. 4. On families belonging to Apetalae and Polypetalae.
Mycotaxon 24:237–273
Gjærum HB (1990a) Puccinia isolepidis, a new cyperaceous rust
species. Lidia 3:19 –21
Gjærum HB (1990b) Uromyces cypericola, a new rust species in the
Cyperaceae. Lidia 3:22–24
Gjærum HB (1998) Rust fungi from various countries. II. Agarica
15:149–153
Gjærum HB, Namaganda M, Lye KA (2003) Additions to the rust
flora of Uganda. Lidia 6:33–61
Goldblatt P, Manning J (2000) Cape Plants. A conspectus of the Cape
flora of South Africa. Strelitzia 9:1–743
Hennings P (1891) Fungi africani. Bot Jahrb Syst 14:337–373
Hiratsuka Y (1973) The nuclear cycle and the terminology of spore
states in Uredinales. Mycologia 65:432–443
Hiratsuka Y, Cummins GB (1963) Morphology of the spermogonia of
the rust fungi. Mycologia 55:487–507
Holmgren PK, Holmgren NH, Barnett LC (1990) Index herbariorum,
Part I, 8th edn. The New York Botanical Garden, New York
Jackson HS (1926) The rusts of South America based on the Holway
collections – I. Mycologia 18:139–162
Jennings DM, Ford-Lloyd BV, Butler GM (1989) An aniline blue
squash technique for observation of urediniospore germ pores.
Mycol Res 92:230–231
Jørstad I (1956) Reliquiae Lagerheimianae – African Uredinales. Ark
Bot 3(ser. 2):563–598
Kern FD (1919) North American rusts on Cyperus and Eleocharis.
Mycologia 11:134– 147
Klebahn H (1914) Pilze III: Uredineen. In: Kryptogamenflora der
Mark Brandenburg. Vol. Va. Gebr. Bornträger, Leipzig
Lindquist JC (1982 ) Royas de la República Argentina y zonas
limítrofes. Instituto Nacional de Tecn ología Agropecuaria,
Colección Científica vol. 20. Buenos Aires, Argentina
Lohsomboon P, Kakishima M, Ono Y (1994) A monograph of
Sphaerophragmium. Mycol Res 98:907–919
496 Mycol Progress (2012) 11:483–497

Marchal É, Steyaert R-L (1929) Contribution à l’étude des champignons
parasites des plantes au Congo Belge. Bull Soc R Bot Belgique 61
(2. ser. vol. 11):160–169
Mayor E (1914) Contribution à l’étude des Urédinées de Colombie.
Mém Soc Sci Nat Neuchâtel 5:442–599
Mayor E, Terrier C (1957) Excursions mycologiques en Corse. Rev
Mycol Paris 22:113–129, + 1 plate
Mayor E, Viennot-Bourgin G (1951) Contribution à la connaissance des
micromycètes de la Côte-d’Ivoire. Bull Soc Mycol Fr 67:113–139
McAlpine D (1906) The rusts of Australia. The Government Printer,
Melbourne, Australia
Mennicken M, Oberwinkler F (2004) A contribution to the rust flora
(Uredinales) of southern Africa, with an emphasis on South
Africa. Mycotaxon 90:1–28
Mennicken M, Maier W, Oberwinkler F (2005) A contribution to the
rust flora (Uredinales) of southern Africa, with an emphasis on
Namibia. Mycol Prog 4:55–75
Monoson HL (1974) The genus Sphaerophragmium. Mycologia
66:791–802
Muasya AM, 6 coauthors (2009) Phylogeny of Cyperaceae based on
DNA sequence data: Current progress and future prospects. Bot
Rev 75:2–21
Potter D, 10 coauthors (2007) Phylogeny and classification of
Rosaceae. Plant Syst Evol 266:5–43
Ragunathan AN, Ramakrishnan K (1973) Rust fungi of Madras state.
VII. Kuehn eola, Hamaspora, Phragmotelium, Phragmidi um,
Diorchidium, Ravenelia, Hapalophragm ium and Nyssopsora.
Mysore J Agric Sci 7:73–86
Sydow H, Sydow P (1903) Neue und kritische Uredineen. Ann Mycol
1:324–334
Sydow P, Sydow H (1915) Monographia Uredinearum. Vol. III.
Pucciniaceae (excl. Puccinia et Uromyces). Gebr. Bornträger,
Leipzig, Germany
Sydow P, Sydow H (1924) Monographia Uredinearum. Vol. IV.
Uredineae imperfectae. Gebr. Bornträger, Leipzig, Germany
Viennot-Bourgin G (1958) Urédinales d’Afrique (5
e
note). Urédinales
de Côte-d’Ivoire (4
e
note). Uredineana 5:137–248
Viennot-Bourgin G (1959) Étude de micromycètes parasites récoltés
en Guinée. Ann Inst Natl Agron 45:1–91
Whiteside JO (1966) A revised list of plant diseases in Rhodesia.
Kirkia 5:87–196
Wood AR (2006) New and interesting records of southern African rust
fungi (Uredinales). S Afr J Bot 72:534–543
Wood AR (2007) Rust fungi (Uredinales) on Grewia species
(Tiliaceae) in South Africa, with Uredopeltis atrides comb.
nov. the new name for Ravenelia atrides. Mycol Prog 6:93–
99
Wood AR, Crous PW (2005) Morphological and molecular characterization
of Endophyllum species on perennial asteraceous plants in South
Africa. Mycol Res 109:387–400
Zhuang J-Y (1988) Species of Puccinia on the Cyperaceae in China.
Mycosystema 1:115–148
Mycol Progress (2012) 11:483–497 497