Mission impossible

444 Review
Spore Liberation and Dispersal in Smut Fungi*
M. Piepenbring'. G. Hagedorn 2, and F. Oberwinkler'
* Part 143 in the series "Studies in Heterobasidiomycetes· from the Botanical Institute, University of Tiibingen
1 Lehrstuhl Spezielle BotanikfMykologie, Botanisches Institut, Universitat Tiibingen, Germany
21nstitut fOr Mikrobiologie, Federal Biological Research Institute (BBA), Berlin, Germany
Received: May 6; Accepted: August 4,1998
Abstract: Teliospores are the most important diaspores of smut
fungi, albeit not the only ones, The role of basidia. basidio-
spores. secondary spores. yeast cells. and infected parts of the
host for dispersal has often been neglected. Many smut species
have soral structures like galls. perldla, and elaters, which cause
teliospores to be liberated over prolonged periods. This in-
creases the chance that at least some spores are released under
favourable wet climatic conditions and while host plants are
susceptible.
In this review. the diversity of dispersal units as well as vec-
tors of smut fungi are presented. The importance of timing of
diaspore liberation, flexibility in dispersal strategies, and the
genetic and evolutionary implications of dispersal strategies of
smut fungi are discussed.
The general considerations are complemented by examples
based on original field and laboratory observations: peridia of
Farysia corniculata and certain species of Sporisorium expose
the spore mass by hygroscopic movement under wet condi-
tions (hygrochasy) favourable for teliospore germination and
infection of a host plant. Basidia with firmly attached basidio-
spores liberated from spore balls of Doassansiopsis deformans.
branched basidiospores of Rhamphospora nymphaeae. needle-
shaped basidiospores of species of Entyloma. folded basidio-
spores of Mycosyrinx cissi, and stellate groups of yeast cells of
Trichocintractia utricu/icola show enlarged surfaces. which are
advantageous for dispersal in water. Galls filled with spore balls
of Doassansiopsis Iimnocharidis and witches' brooms formed by
spikelets infected by Cintractia standleyana separate from the
host and fall into water where they are dispersed.
Key words: Ustilaginales. Tilletiales. Microbotryum. dlaspores,
teliospores, deposition, vectors. evolution.
1. Introduction
The smut fungi (Basidiomycetes; Ustilaginales, Tilletiales, and
related groups as well as Microbotryales; see Bauer et al.,
1997) are an important ecological group of obligate parasites
of herbs. They are characterized by usually thick-walled
teliospores produced in sori in various organs of their host
plants, including flowers, leaves, stems, and roots (Fig.I). The
spore mass can be enclosed by peripheral host tissue when
Bot.Acta 111 (1998) 444 - 460
© Georg Thieme VerlagStuttgart· New York
young (sori with exposed spore mass in Figs.ta, b), it can be
located in galls formed by host tissue (Figs.Ic, d), or covered
at least when young by peridia of sterile fungal cells (Figs.Ie,
f). Usually, the tissue enclosing the spore mass ruptures and
liberates the teliospores (Figs.la-e, e-f) or the spores
remain inside the tissue until they are released by its decay
(Fig.ld). Teliospores rarely develop exposed on the host tissue
surface. Teliospores are the most important diaspores of smut
fungi. However, their germination products (i.e. basidia,
basidiospores, secondary spores, yeast cells, or hyphae) may
also spread the fungus. Sometimes the smut fungus is
dispersed together with host organs.
Structures of smut fungi related to spore liberation or
dispersal are often described morphologically (e.g. Vanky,
1987,1994), but their adaptive significance is rarely discussed.
Some information on liberation and dispersal processes in
smut fungi can be found in the works of Fischer and Holton
(1957) and Ingold (1971). However, smut fungi are largely
ignored in most discussions of dispersal mechanisms, even in
otherwise comprehensive reviews specific to fungi (e.g.
Malloch and Blackwell,1992; Lacey, 1996).
The present article tries to fill this gap for those interested in
smut fungi. The different dispersal units, the liberation of
teliospores, and dispersal agents (vectors) utilized by smut
Fig.l Host plants with sori of smut fungi. •
a) Ustilago schroeteriana Henn. on Paspalum virgaWm L. The
spikelets and the peripheral tissue of two inflorescences are
destroyed by the development of the teliospore mass (Costa Rica,
MP 11). b) Microbotryum violaceum (Pers.: Pers.) Oeml ~~ Oberw. on
Saponaria officinalis L. The teliospore mass fills the anthers and is
scattered on the petals (Germany, without number). c) Til/etia ayresii
Berk. on Panicum maximum jacq. In this rather old inflorescence,
only spikelets with sac-shaped sori are left while healthy spikelets
fell off. Note the beetle feeding on the teliospores (CD5ta Rica, MP
617), d) Juncus bufonius L. with a gall (arrow) caused by Entorrhiza
casparyana (Magnus) Lagerh. Teliospores are located inside the cells
of the root gall (Costa Rica, MP 742). e) Cintractia scleriae (DC) Ling
on Rhynchospora corymbosa (L.) Britt. The sori envelop peduncles of
the sterile inflorescence. Note remains of white peridia adjacent to
the black telispore mass (Costa Rica, MP 93; photo R. Berndt). f)
Sphacelotheca d. hydropiperis (Schum.) de Baryon Polygonum
punctatum Elliot.The teliospore mass is enclosed by sori with central
columellae and peridiallobes replacing flowers (Costa Rica, MP142;
photo R. Berndt).

Dispersal in Smut Fungi Bot. Acta 111 (1998) 445
Figs.la-f

446 Bot. Acta 111 (1998)
fungi are discussed. Dispersal units and some structures
related to diaspore liberation and dispersal are illustrated
with smut fungi observed mostly in Central America and
Germany. Some aspects of the dispersal of plant pathogenic
fungi in general are discussed at the end. We hope that this
presentation encourages explicit studies of the many phe-
nomena which we currently can only speculate about.
2. Material and Methods
Smut fungi were observed in their natural environment in
Central America, Mexico, and Cuba (several journeys from
1992-1996; for Costa Rica see Piepenbring, 1996) and
Germany. Simple experiments of the effect of water on soral
structures and flotation on dispersal units in amphibious
habitats were carried out with fresh material. The hygroscopic
movement of peridial lobes was observed in the field and on
herbarium material. Hand sections for anatomical observa-
tions were mounted in water. Germination ofteliospores were
obtained on 1% agar (WA) or 0.2% malt extract agar (MEA);
cultures were grown on malt yeast peptone agar (MYP). All
media were supplemented with the antibiotic, tetracycline
(ca. 500 ppm).
Collection or herbarium numbers of illustrated specimens are
cited in the legends of the figures. Most of the material is
deposited in the private herbarium of the first author
(MP= collection numbers, H.U.P. = herbarium numbers), du-
plicates in the Botanische Staatssammlung Mtinchen (M).
3. Diaspores
3.1 Teliospores
Most smuts are frequently spread by variously ornamented
teliospores, which may be single, form pairs, groups, or balls
(Figs.2 and 5). They are adapted to long-term survival and
long-distance dispersal. A thick, usually pigmented, wall
prevents desiccation and protects against harmful UV radia-
tion. Some smuts produce teliospores permanently embedded
in host tissue (e.g. from species of Enry/oma and Rhampho-
spora, Fig.2e), which are usually smooth and hyaline when
seen using the light microscope. Most features of teliospore
dispersal are discussed together with the various dispersal
vectors in section "5. Dispersal vectors".
3.2 Basidiaasdispersalunits
Smuts exhibit a remarkable flexibility in using other devel-
opmental stages than teliospores as dispersal units. In some
smut species entire basidia (e.g, Doassansiopsis deformans,
Fig.3e, camp. Setchell, 1892: PI. I, Figs. 51-58; Cintractia
scieriae (DC) Ling, cornp, Piepenbring, 1996) separate from
the teliospore and become dispersal units. In other species
the cells of the phragmobasidium separate and can be
dispersed. e.g. the upper three cells of the phragmobasidium
in Microbotryum violaceum (Pers.) G. Deml & Oberw. (Dernl,
1986), the upper two cells in Ustilago maydis (DC) Corda
(Fig.3a; Ramberg and Mclaughlin, 1980), and even single
basidial cells in Ustilago /ongissima (Ingold,1983).
M. Piepenbring, G. Hagedorn, and F. Oberwinkler
3.3 Basidiospores and secondaryspores
The importance of basidiospores as secondary dispersal units
is highly variable among smut fungi. The number of basidio-
spores is usually limited in the holobasidial smuts, while
most phragmobasidial smuts produce a potentially indeter-
minate number of basidiospores by a repetitive process
(Fig.3a). Basidiospores usually separate from the basidium
(Figs.3a, b, d), but those of Doassansiopsis deformans (Fig.3e)
and most species of Enry/oma (Figs. 3e, f) and Urocystis remain
firmly attached to it. Folded basidiospores of Mycosyrinx cissi
(Fig.3h; cornp, Piepenbring and Bauer,1995) develop directly
on the teliospore and are liberated without a typical basidium
being formed.
The basidiospores of most phragmobasidial smut species may
grow into a haploid yeast stage by budding for a prolonged
time (Fig.3a). The yeast cells can grow on sugar exudates on
the leaf surface, leafaxils, or in flowers, where they can be
dispersed by splash dispersal, insects, or wind. In some
phragmobasidial species the haplophase is reduced by early
mating, either among basidiospores or among the basidial
cells themselves by clamp connections (Fig.3b). The like-
lihood of this event depends upon environmental conditions
and varies strongly from species to species. It has been
observed in many species of Microbotryum and Usti/ago.
Mating of the phragmobasidial cells themselves is common,
e.g, in some species of Cintractia and Ustilago (Fig.3b). In the
case of Ustilago avenae (Pers.) Rostrup and the embryo
infecting loose smuts of barley and wheat (Usti/ag nuda
(jensen) Kellermann and Ustilago tritid (Pers.) Rostrup)
frequently no basidiospores are produced at all. Forsuccessful
infection the teliospores of these three species must germi-
nate on the stigmata of grass flowers, where the conditions
are probably not favourable for the budding of yeast cells. In
the latter two species, germination and the mating of mono-
karyotic cells must occur rapidly so that the infective
dikaryotic mycelium reaches the embryo in time (comp,
"3.4 Dispersal together with host organs").
In many species of the holobasidial smuts, the basidiospores
mate while still attached to the basidium (Figs.3e, e). The
fertilised pairs sometimes detach from the basidium and are
therefore potential dispersal units. These mating units or the
resulting dikaryotic mycelium can produce secondary spores
as further dispersal units. In species of. e.g., Til/etia and
Enry/oma, secondary spores can be actively discharged into
the air, i.e.they are ballistospores (Figs.3e, g).
Fig.2 Teliospores andelatersofsmut fungi. ~
a) Ustilago affinis Ellis & Ev. Warty teliospores (Costa Rica, MP
1177). Bar =4 Jim. b) Ustilago reticulata liro. Reticulate teliospores
with warts in the meshes (Germany, MP 2352). Bar = 4,um. c)
Rhamphospora nymphaeae Cunn. One teliospore with adjacent
hyphae and host tissue. Warty ornamentation of the teliospores
covered here by the sheath can be seen in the light microscope
(Costa Rica, MP 675). Bar =4Jim. d) Farysia corniculata K. Vanky.
Teliospores and part of an elater made of sterilefungal cells (Costa
Rica, MP 736). Bar = 4Jim. e) Trichocintractia utriculicola (Henn.) M.
Piepenbr. Filamentous thick-walled fungal cells, one isbroken (Costa
Rica, MP 94). Bar =2Jim. f) Melanopsichium pennsylvanicum Hirschh.
Partsof two teliospores (one broken) embedded in a matrix which
is hyaline when seen in the light microscope (Costa Rica. MP 801).
Bar=4Jim.

Figs.2a-f

448 Bot. Acta 111 (1998)
While most smuts species employ teliospores as their princi-
pal dispersal units, the teliospores of species of Enty/omaand
some other small genera are usually reduced to mere resting
structures which remain embedded in the host tissue (mainly
in blades of leaves). The teliospores germinate within the host
tissue and secondary spores develop on hyphae protruding
through the stomata (Fig.3 g), assuming the role of primary
dispersal units (Kaiser,1936).
3.4 Dispersal together with host organs
Many smut fungi are dispersed together with diaspores of
their host plants. This may be a sporadic event in some
species, but can be an integral part of the dispersal strategy of
other species. Interestingly, several smut fungi with sori in
ovaries do not utilize the fruit dispersal strategies of their
hosts, but on the contrary, prevent the abscission of these
organs (Fig.te). Such a strategy may prolong the period of
wind dispersal, cornp. "4. Teliospore liberation".
In many cases the smut spores simply adhere to the outside of
the seed or fruit. Usually the teliospores remain dormant
together with the host organ and infect the germinating
seedling. In some seed-borne smut species, the smut develop-
ment starts before the germination of the seed. In Til/etia
caries(DC) L.-R. & C. Tul. the optimum germination temper-
ature of the teliospores is slightly below the germination
temperature of wheat, giving the fungus a developmental
advantage in the spring (see review of literature in Schuh-
mann, 1962: 471). Other examples are Usti/ago avenae and U.
hordei(Pers.) Lagerh.which germinate and mate between hull
and caryopsis. Parasitic hyphae may penetrate the pericarp in
these species, but the embryo remains unaffected and the
seedling is infected after overwintering (Tapke,1940).
The only known cases where the embryo is infected before
dispersal and overwintering of the seeds are found in the
loose smuts of wheat and barley (Usti/ago tritid and U. nuda,
respectively). Dusting seeds or seedlings of wheat or barley
with smut spores of these species results in almost no
infection. Instead, teliospores deposited on the stigmata of
the flowers germinate, copulate, and the dikaryotic mycelium
grows down to the developing caryopsis. It invades the
embryo, where it enters into a resting phase until the grain
germinates (Brefeld and Falk, 1905, the literature pertaining
to seedling and embryo infection was reviewedlast by Fischer
and Holton, 1957: 146ff).
Since the anther smuts of Caryophyllaceae (Microbotryum
vio/aceumand related species; Fig.tb) can infect flowers, it is
tempting to assume that the fungus is dispersed together
with the seed. Brefeld indeed conjectured this and published
an account apparently proving it (Brefeld and Falk, 1905).
Several generations of workers tried to repeat these results
and failed. Baker (1947) reviewed the literature and con-
ducted conclusive experiments showing that seed-borne
infection is in fact a rare exception, even when seeds are
artificially inoculated with high spore concentrations. He
showed that the teliospores of M. vio/aceum have no dor-
mancy and probably germinate immediately in the soil. The
resulting haploid yeast phase either fails to overwinter, or
becomes so diluted with time that successful mating and
establishment of the infective dikaryon is very rare. Further-
more, while the dikaryotic mycelium of most cereal smuts
(e.g. species of Ustilago) can grow for some time saprophyti-
cally, the development of fertilised yeast cells of M.vio/aceum
is arrested unless host chemicals exuded by the host plants
(e.g. tocopherol, Castle and Day,1984) are present. Instead, M.
vio/aceum infects new host plants in the same season and
overwinters in organs of their perennial host plants (see
Thrall et aI.,1993a).
Other host organs than seeds may be used for dispersal as
well. Vegetative propagules may be infected, or systemically
growing smuts may spread together with the clonal growth of
the host, e.g. through runners or rhizomes. For example
Usti/ago hypodytes (Schlecht.) Fries is dispersed with the
rhizomes of its hosts (Bond, 1940). Transmission may be
incomplete. Wennstrom and Ericson (1990) found that as
much as two thirds of the winter buds (vegetative propagules)
of Trientalis plants infected with the systemic smut fungus
Urocystis trientalis (Berk. & Br.) B. Lindeb. grew into healthy
plants.
While some smut fungi exploit natural dispersal strategies of
their hosts (e.g. Usti/ago spinificis Ludwig in the spikelets of
the tumbleweed Spinifex, Poaceae), other smuts modify the
morphology of their hosts for their dispersal. Galls falling out
of leaves of Limnocharis f1ava infected by Doassansiopsis
limnocharidis and witches' brooms of Cintractia stand/eyana,
which fall from the host plant (Figs.6a-e; comp, 3.3.2) are
produced only under the influence of the smut fungus.
4. Teliospore liberation
Few smut species (e.g. Ustilago hypodytes, species of To/ypo-
sporium and some species of Cintractia) develop teliospores
on the surface of host organs. Usually, young teliospores are
embedded in a host organ, which ruptures during teliospore
liberation. The sorus is frequently enlarged by hypertrophic
growth of host tissue (e.g. species of Entorrhiza Fig.td,
Fig.3 Germination products of smut fungi. •
a) Ustilago maydis. Germinated teliospore with basidial cells
separated into two pairs. The basidial cells develop basidiospores on
sterigmata (Costa Rica, MP 858). b) Ustilago tenuispora Ciferri.
Germinated teliospore with basidial cells conjugated by clamp
connections. Before copulation the basidial cells produced basidio-
spores which are now conjugated by bridges. Dikaryotic cells grow
with hyphae (Costa Rica, MP 143). c) Ooassansiopsis deformans
(Setch.) Dietel. Basidia with firmly attached, conjugated basidio-
spores and some secondary spores (Honduras, MP 2066). d)
Rhamphospora nymphaeae. Three germinated teliospores. From left
to right: a young holobasidium with three apical basidiospores;
older basidium with detached basidiospores; older basidium with
ramified basidiospores developing secondary spores (Germany,
H.U.P. 112). e) Entyloma bidentis Henn. Basidium with sessile
conjugated basidiospores and sickle-shaped secondary spores
(Costa Rica, MP 826). f) Entyloma calendulae (Oudem.) de Bary.
Basidium with sessile basidiospores (conjugation not visible) and
needle-shaped secondary spores (Germany. H.U.P. 550). g) Entyloma
ranunculi-repentis Sternon. Stoma of the abaxial epidermis of an
infected leaf with hyphae producing needle- and sickle-shaped
secondary spores (Germany, MP 2357). h) Mycosyrinx cissi (DC) G.
Beck. One pair of teliospores and two single teliospores with folded
basidiospores. The basidium probably corresponds to the teliospore
(Costa Rica, MP 1017). i) Trichocintractia utriculicola. Stellate groups
of yeast cells after 10 days on MYP (Honduras. MP2075).

Dispersal in Smut Fungi
Figs.3a-i
J-I
5}Jm
"/'-':'. c;;;JP.>/ 0",' :...... C.:.-:-.. : .
"..... ..
Bot. Acta Tll (1998) 449

450 Bot. Acta 111 (1998) M. Piepenbring. G. Hagedorn. and F. Oberwinkler
--------------------Melanopsichium, Mycosyrinx, Sporisorium, Thecaphora. Tilletia
Fig.Lc,Urocystis. and many Ustilago species). Young sari may
be concealed by host organs such as leaf sheaths (e.g. Ustilago
hypodytes, U. schroeteriana Henn. Fig.la). perigynia (e.g.
species of Anthracoidea and Farysia Fig.4a). or scales of
spikelets (e.g. species of Tolyposporium and some species of
Cintractia). Immature sari of species of Anthracoidea, Cintrac-
tia (Fig.Je), Farysia (Fig.4a). Sphacelotheca (Fig.H). Sporiso-
rium (Figs.4e-f). and other small genera are covered by
peridia of sterile fungal cells.
The soral envelope probably ruptures due to the pressure
exerted by the increasing volume of enclosed teliospores. In
many smuts the soral envelope continues to protect the spore
mass from being liberated too fast, e.g. being completely
washed away in a torrent rain. At the same time. peridial
membranes or host organs like glumes can act as a bellows
mechanism: water drops which hit the sorus depress the
covering and emit a puff of air mixed with spores. Examples
for smuts where a bellows mechanism is likely are Farysia
comiculata (Fig.4a). Trichocintractia utticulicola, and species
of Anthracoidea.
In species of Farysia. groups of long. sterile multicellular
fungal threads (Fig.2d) and in Trichocintractia utriculicola
(Henn.) M. Piepenbr. groups of thick-walled elongated cells
(Fig.2e) are interspersed with the teliospores. Their function
may be comparable with that of the capillitium of myxomy-
cetes or gasteromycetes, or the elaters of bryophytes and
pteridophytes (comp. Pacini and Franchi, 1993). They loosen
the spore mass and cause spores to be liberated gradually
over a prolonged period. In some species of Sporisorium (e.g.
S.reilianum(Kilhn) Langdon & Fullerton, S.cenchri (Lagerh.) K.
Vanky) the host axis is split by teliospores, produced within
the axis. into numerous columellae. corresponding to vascular
bundles with some attached parenchymatous tissue (Langdon
and Fullerton. 1975 and pers. observation). These brush-like
structures probably have a similar function.
Hygrochasy. the exposure of teliospores by opening peridial
lobes when these are wetted and closure upon drying,
actively influencing teliospore liberation is reported and
illustrated here for the first time for some smut fungi (Fig.4:
cornp, Piepenbring and Hagedorn 1998). The teliospore mass
of Farysia comiculata develops in a ruptured ovary of Carex
sp. and is covered by a peridial sac (Fig.4a). The peridium is
formed by hydrophilous. longitudinally-waved. elongated,
non-septate fungal cells (Fig.4b). which swell considerably
when brought into contact with water. Partial wetting results
in writhing movements. which cause the peridium to rupture
and the spore mass to be exposed. The peridia of sari of
Sporisorium species (Figs.4e-i) often consist of several irreg-
ular layers of hydrophilous. sterile. fungal cells which form
chains more or less parallel to the longitudinal axis of the
sorus (Fig.4g). The fungal peridium can further be covered by
remains of host tissue. The peridium usually ruptures longi-
tudinally, forming several peridial lobes. In the case of
Sporisorium ovarium (Fig.4d) and S. cruentum (Figs.4e, f). the
peridial lobes enclose the spore mass when dry and expose it
after a few minutes when wetted. In contrast. the peridium of
S. sorghi (Fig.4e) does not move. apart from some swelling.
when brought into contact with water. In transverse sections
of the peridia of S. sorghi and S. cruentum (Figs.4 hand i,
respectively), the anatomical structures responsible for the
movement are present only in the peridial lobes of the latter
species. The thick. persistent peridium of S. sorghi (Fig.4h)
consists of relatively small cells and is clearly delimited
towards the inner surface. The thin peridium of S. cruentum
(Fig.4i), however. shows an irregular texture with small.
densely arranged cells towards the remains of the host issue
on the outside. and a mixture of small and large cells with
large intercellular spaces towards the spore mass on the inner
side of the sorus. The inner part of the peridium absorbs
water more readily and the intercellular spaces allow the
tissue to swell more strongly than the tissue in the outer part,
resulting in the backward movement of the peridiallobes and
the exposure of the teliospore mass.
The teliospore masses of the majority of smuts, e.g. species of
Ustilago, Farysia, Spbacelotheca, and Sporisorium. are powdery
because hyaline gelatinous material around young teliospores
disappears when the spores mature. In contrast, most of the
spore mass of species of Anthracoidea and Cintractia is
agglutinated by remains of the hyaline matrix in which the
spores developed. Only the surface of the sorus becomes
gradually powdery. This leads to a gradual liberation of
teliospores when the peridium ruptures. The teliospore mass
of species of Melanopsichium is embedded in a gelatinous
matrix even when mature (Fig.2f). The matrix is fluid during
teliospore liberation in wet environmental conditions.
Teliospores of species of Entorrhiza, Entyloma, Melanotaenium,
Doassansia, Doassansiopsis, Burtillia, Rhamphospora. and
Nannfeldtiomyces usually remain embedded in the host tissue
until its decay. In a very few species of these genera, the sorus
may rupture and expose the non-powdery spore mass, e.g, in
Entyloma microsporum, Doassansiopsis deformans, and some
species of Melanotaenium.
Powdery masses of teliospores are often exposed in high
positions on the host plant. Ingold (1971: 219ff) notes that
barley plants infected with Ustilago nuda are usually a little
taller than healthy ones. Uninfected heads bend over. while
Fig.4 Sari of smut fungi with peridia opening upon wetting ~
(hygrochasy).
a-b) Farysia corniculata on Carex /emanniana Boott (Costa Rica. MP
736). a) A locally ruptured sorus showinq the teliospore mass and
elaters enclosed by a peridium (brown), with scale. remains of the
perigynium, and the ruptured ovary (not visible) at the base. b)
Surface view of undulated, longitudinally-orientated, non-septate,
agglutinated fungal cells of the peridium. c-f) Sari of species of
Sporisorium. c) S. sorghi Ehrenb. ex Link on Sorghum bicolor (L.)
Moench. Cylindrical sari and some healthy spikelets of the host
(Nicaragua, MP 2036a). d) S. ovarium (Griffiths). K. Vanky on
Urochloa fasciculata (Sw.) R. Webster. Three dry sari with more or
less ruptured peridia in spikelets, the middle one redrawn after
contact with water for 2 min (Panama. MP 2214). e-f) S. cruentum
(Kuhn) K. Vanky on Sorghum halepense (L.)Pers.(Mexico. MP 1935).
e) Dry sari in spikelets of the host. f) Isolated sorus dry and after
contact with water for 4 min. g-i) Anatomy of the peridia shown in
Figs.4c and 4e-f (hand sections in water). g-h) S. sorghi. g) Surface
view of the peridium. From left to right: direction of the longitudinal
axis of the sorus. h) Transversesection of the peridium. From top to
bottom: remains of host tissue, sterile fungal cells (a 5011m thick
part has not been drawn). inner surface of the peridium with a
single teliospore. i) S. cruentum. Transverse section of the peridium
with two adjacent teliospores. For further description see Fig.4 h.

1mm I
Figs.4a-i
Bot A tc a 111 (1998) 451

452 Bot. Acta 111 (1998)
-~--,------
infected inflorescences remain upright, thus improving the
exposure of smutted heads to the wind. By preventing the
abscission of infected floral organs, several smut species
prolong the time of teliospore exposure to the wind. This has
been observed by the authors in Cintractia jlmblistylicola
Pavgi & Mundkur on Pimbristyiis spadicea (L) Vahl, Spoliso-
lium veracruzianum (Zundel & Dunlap) M. Piepenbr. on
Panicum viscidel/um Scribn., Til/etia ayresii on Panicum max-
imum (Fig.te), Ustilago ixopholi Duran on lxophorus unisetus
(Presl) Schlecht., and for the insect dispersed anther smut
Microbotryum violaceum on species of Silene.
Direct liberation of spores by wind is probably rare, except for
cases noted above where the infected host organs stand above
the surrounding vegetation and are directly exposed to strong
winds (e.g. Fig.ta). Wind speed within the vegetation cover is
usually insufficient for direct liberation (comp. Lacey, 1996).
Instead, wind often causes indirect effects like turbulence
within the vegetation, stem vibrations, and fluttering of
leaves. The resulting acceleration forces, as well as friction
among leaves, lead to the liberation of spores into the air
(comp., e.g. Mills, 1967; Bainbridge and Legg,1976).
Liberation and dispersal of spores by rain-wash (ombrohy-
drochory) or rain-splash occurs in many smuts with spore
masses exposed to the rain. The teliospores of most wind or
rain dispersed smut species are hydrophobic due to physical
or chemical aspects of the spore surface. Even under wet
conditions, these spores remain powdery and are easily
carried away by the wind.
In general, water drops not only liberate spores from hydro-
philic slimy spore masses considered typical for this libera-
tion mode (Fitt et aI.,1989), but also play an important role in
the liberation of dry, hydrophobic spores. Three mechanisms
can be distinguished: First, the bellows mechanism involving
the soral envelope, already mentioned above. Second, the
impact of raindrops on a dry surface creates a fast moving
radial air pressure front which directly liberates dry spores
(puff mechanism, Hirst and Stedman, 1963). Third, delicate
host organs hit by drops may vibrate sufficiently to release
spores (tap mechanism, Hirst and Stedman, 1963). The
generally observable increase of spore concentration in the air
at the onset of rain (see review in Malloch and Blackwell,
1992: 152) is probably due to these effects as well as to the
strong and gusty winds and resulting turbulence which often
accompanies rain. The spores liberated by water drops are
further dispersed by wind.
The difference between wet and dry splash dispersal is
relevant in terms of the resulting dispersal gradient. Wet
splash dispersal occurs in the smut fungi during the dispersal
of the generally hydrophilic, secondary dispersal units (basi-
dia, basidiospores, secondary spores, and yeast cells) dis-
cussed in the previous sections. Most spores caught in water
droplets have a very limited dispersal range. Only a small
fraction of wet-dispersed spores are caught in droplets
sufficiently small to be carried over long distances. The spore
concentration gradient is much steeper for splash-dispersed
than for wind-dispersed spores. Dry splash dispersal is typical
for many smut teliospores and is more similar to wind
dispersal, the steeper dispersal gradient being due to wash-
out effects.
Despite their hydrophobic surface, smut spores are efficiently
collected by rain drops (Hirst, 1959) and washed out from the
atmosphere. Upon deposition, the hydrophobic surface also
favours fixation of the spore to a hydrophobic target, such as
the waxy cuticle of a leaf surface (Clement et ai, 1994). The
importance of direct (wash-out, splash liberation) and indi-
rect (bellows, tap, and puff liberation, associated strong
winds) effects of rain has been shown by Sreeramulu (1962)
and Sreeramulu and Vittal (1972), who found that frequent
rain and strong winds reduce the length of the dissemination
period because the supply of spores is depleted earlier.
Many host species of smuts may grow under an overgrowth
of trees or shrubs, sheltered from the direct action of rain.
This is no obstacle to splash liberation and dispersal. In fact,
leaf drip splash is often more effective than direct rain splash
because the water drops are larger (Gregory, 1973).
The open sori of, e.g. Sphacelotheca, form cup-like structures
(Fig.H). Whether these cups operate as typical "splash-cups",
the spores being dispersed by water drops hitting the cup, or
whether they operate as "wind cups", the spores being picked
up by turbulences of the wind at the edge of the cup, needs to
be investigated.
5. Dispersalvectors
5.1 Wind dispersal
Wind dispersal is the most common dispersal mode of smut
teliospores. It can be very far reaching: smut spores have been
found over Arctic regions and at high altitudes (comp. Fischer
and Holton, 1957: 183; Gregory, 1973: 272). Simmonds (1994)
gives a plausible argument that Ustilago scitaminea H. Sydow
succeeded in transatlantic wind dispersal and successfully
established itself as a major pest of surgar cane in the
Caribbean. Wind is also important for short-range dispersal
and local disease epidemics.
Dispersal gradients of wind-dispersed diaspores are steep:
the spore concentration and infection rate decreases approx-
imateley negative exponentially with distance (reviewed by
Fitt and McCartney, 1986; Fitt et aI., 1987). Such a steep
gradient seems to make long-distance dispersal a most
unlikely event. Yet, as Gregory (1982) points out, a part of the
diffusion cloud reaches sufficient heights to become subject
to different aerodynamic processes. This "escape fraction" is
transported much longer distances and its dispersal gradient
is much less steep. Dispersal gradient data specific to smut
fungi are found in Oort, 1940 (U. tlitici), Sreeramulu, 1962 (U.
Fig.5 Teliospores carrying appendages or forming more or less ~
persistent spore balls.
a) Cintractia nova-guineae Zundel. Teliospore with appendage
(appendage hyaline when seen with the light microscope; New
Guinea, BPI 172057). Bar= 5Jlm. b) Moesziomyces bullatus (Schrtiter)
K. Vanky, Part of a broken spore ball showing scattered teliospores
surrounded by thin-walled sterile cells (Cuba, MP 2151). Bar= 10Jlrn.
c-d) "Sorosporium" (=Sporisorium) everhartii Ellis & Gall. c) One
complete spore ball (Cuba, MP 2270). Bar= 20 urn. d) Part of a
broken spore ball showing smooth sterile internal cells. Bar= 5Jlm.
e) Thecaphora pustuJata Clinton. Permanent spore ball (Costa Rica,
H.U.V.13624). Bar~ 10urn, f) Sporisorium cenchri (Lagerh.) K.Vanky,
Very loose spore ball (Nicaragua, MP 1952). Bar= 10Jlm.

Figs.5a-f

454 Bot. Acta 111 (1998)
nuda), Sreeramulu and Vittal, 1972 (u. sdtaminea), Misra et
aI.,1989 (Ustilago scitaminea), Stepanov, 1935 (Til/etia caries),
and Stepanov, 1962 (T. tritid) (the latter two in Russian,
quoted after Fitt et aI.,1987).
In general, most wind-dispersed (anemopilous) diaspores are
characterized by a relatively small size and the frequent
presence of structures enlarging their surfaces. The surface of
many wind-dispersed smut teliospores is enlarged by differ-
ent ornamentations (Figs.2 and 5; comp, Vanky, 1991), by
appendages (e.g. some species of Cintractia, Fig.5a) or a
locally ruptured exosporium (Ustilago williamsii (Griffiths)
Lavrov). The adaptive value of different spore ornamentations
(small or large warts, reticulations) remains unclear. The
diversity of this characteristic seems to imply that it has some
significance, yet many species with smooth spores seem to be
well adapted to wind dispersal as well (e.g. Ustilago hordei
(Pers.) Lagerheim).
Some wind-dispersed teliospores form spore balls. While the
spore balls of Moesziomyces (Fig. 5b), many species of Urocys-
tis, and some species of Sporisorium (Figs. 5c, d) consist partly
of empty sterile cells, thereby increasing the surface without
proportional weight increase, the spore balls of other Urocys-
tis species (e.g. U. primulae (Rostrup) K. Vanky), Glomospo-
rium, Thecaphora (Fig.5e), Tolyposporium, and Sorosporium, as
well as the more or less persistent spore balls of many species
of Sporisorium (Fig.5f) consist predominantly of fertile cells.
The inner walls of these spore balls make them heavier than a
single cell of the same size.
One explanation for this unintuitive construction could be
that a shorter dispersal range is advantageous per se. Long-
distance dispersal initially increases the fitness of the patho-
gen, because it is more likely that a host plant encountered is
not yet infected. On the other hand, the frequency of host
plants is likely to decrease with distance, because many
plants have either a gregarious life form or are distributed in
patches. Also, long-distance dispersal means a longer expo-
sure to harmful UV radiation and desiccation. Evenwhen the
spores are protected against these influences, the viability of
the diaspores will decrease with the time of exposure. All
these factors will result in an optimal dispersal distance.
Another potential explanation is that the morphology of
wind-dispersed spores is a compromise between being small
and light to remain in the air as long as possible, and being
heavy enough to have sufficient momentum to impact on the
host plant while passing it (Ingold, 1971: 193ff). If a spore is
too light, it will be carried together with the air stream around
the host plant organ. Thus, light spores are deposited mainly
through sedimentation under the influence of gravity (e.g.
after boundary layer exchange, cornp, Gregory, 1966) or are
washed from the atmosphere by rain. For wind-dispersed
pathogenic fungi, it might be advantageous to sacrifice
average dispersal distance, confounded with random deposi-
tion on plant and non-plant material, in favour of an increased
probability of deposition on a suitable host plant.
The optimal size of spores for impaction onto an organ of a
host plant strongly depends on wind velocity and the size of
the target. When wind velocity is high, heavier spores are
favoured over lighter ones. Since wind conditions are variable
and unpredictable from year to year, size variations may be
advantageous. Many smuts produce spore balls of variable
size or spore balls which easily fragment (Fig. 5f). Similarly,
although species of Ustilago and many other genera do not
have spore balls, their spores are often dispersed in clumps
(U. scitaminea, 44% of spores or 11.5 %of dispersal units were
in clumps of 5 to 8, Sreeramulu an Vittal, 1972; U. avenae,
clumps of20 to 200, Mills,1967).
The impaction process is also influenced by the relation
between spore size and the size of the impaction target. Small
spores might be carried around large host plant organs, but
they will successfully impact on host organs with a small
diameter. Because this relationship is strongly influenced by
wind velocity, it cannot be tested in absolute terms. Still, it
can be seen to some extent in the smut fungi. Although the
small spores of Ustilago nuda (9 um) would be too small for
efficient impaction on leaf blades, they reach an impaction
efficiency of 50 to 75%on the narrow glumes and stigmata of
the flowers (Gregory, 1966; Hirst and Stedman, 1971). At the
other extreme, it can be noted that many, albeit not all, smuts
with large, wind-dispersed spore balls (e.g. Thecaphora
species, Sorosporium saponariae Rudolphi) parasitize hosts
with larger leaves.
It is interesting to note that the diameter of smut spore balls is
well within the size range of rust aecio- and urediniospores or
wind-dispersed pollen grains. Selectionforces seem to result in
the convergent evolution of dispersal units of similar size.
5.2 Water dispersal
This section deals only with dispersal of smut diaspores
floating on the water surface or submerged in water. Splash
dispersal by water drops is so intimately connected with
splash liberation that it has already been discussed in the
section concerning spore liberation (4.).
Floating can be achieved by means of unwettable spores
taking advantage of the surface tension of water, or by
becoming buoyant through trapped air. Dry smut teliospore
masses are mostly unwettable on first contact with water,
later they are usually submerged. Savile (1954) argues that
the presence of air trapped between nail-headed warts enables
spores of certain species of Anthracoidea (section Echinospor-
ae; in Savile as species of Cintractia) to float. Kukkonen
(1964), however, observed that less prominently warty telio-
spores float just as well. Floating seems to be generally
preferable to submerged dispersal, since the probability of
hitting a target host plant is much higher for spores dispersed
in the two dimensions of the water surface than in the three
dimensions of the water volume, especially when the added
effect of Brownian molecular motion is considered (Cox,
1983).
Diaspores dispersed submerged in water may be larger than
those for wind dispersal, since the weight/surface ratio is not
as critical as in the air.The weight can be partly balanced by oil
drops, and surface enlargement without proportional increase
of weight reduces the settling rate (Webster, 1959; Webster
and Davey, 1975). Diaspores of the so-called aquatic hypho-
mycetes often possess tetraradiate or sigmoid shapes (Ingold,
1966 and 1979; Webster and Davey, 1984), which seem to be
favourable for transport of submerged diaspores by currents
(Pijl, 1969: 63) and for anchoring them on the host plant

Dispersal inSmutFungi
surface (Webster, 1959; Pijl,1969: 68). Random search theory
predicts that a large effective diameter, as reached by filiform
or branched diaspores, enhances the probability of hitting the
target host plant (Cox, 1983). Similar examples of surface
enlargement as an adaption to the aquatic environment can
be found among the phytoplankton and the pollen of aquatic
plants (Cox,1983).
All smut diaspores usually transported by wind may occa-
sionally be dispersed by water. However,smuts on aquatic or
amphibious host plants show enlarged diaspores which can
be interpreted as an adaptation to hydrochory. Species of
Bunillia, Doassansia, Doassansiopsis, Nannfeldtiomyces, Nara-
simhania, Pseudodoassansia, and Tracya, all parasites of plants
growing in water or swamps, form large (diameter usually
100-250jlm), non-powdery balls of numerous teliospores
and sterile fungal cells.
The following illustrated structures may also be interpreted
as adaptations for transport by water. Doassansiopsis limno-
charidis provokes hypertrophic growth of spots in leaves of
Limnocharis flava (Figs.6a-e) growing in swamps with free
water. The thickened spots are filled with numerous spore
balls and fall from the leaf by necrosis of adjacent tissue.
Isolated fresh galls usually first float on water and later move
freely in it. The spore balls are liberated after the decay of the
host tissue. On Rhynchospora marisculus, growing in swamps
or lakes, Cintractia standleyana inhibits the elongation of
peduncles and pedicelli in the normally loosely ramified
inflorescences, causing the formation of dense balls (witches'
brooms. Figs. 6d-e). Fully grown witches' brooms easily fall
from the host plant. If they are sufficiently dry, they float on
the water surface because air bubbles are trapped between
the bracts. Wet balls sink, but move freely in currents.
Teliospores concealed between the scales are continuously
liberated during the voyage of the spikelet ball.
The stellate basidia of Doassansiopsis defonnans (Fig. 3e) and
the ramified basidiospores of Rhamphospora nymphaeae
(Fig. 3d) remind one of the tetraradiate spores of aquatic
hyphomycetes. Filiform or folded diaspores, perhaps also
adapted to dispersal in water. are found e.g. in the needle-
shaped spores of species of Entyloma (Figs.3f. g) and the
basidiospores of Mycosyrinx dssi (Fig. 3 h). Ward (1887)
observed that in species of Entyloma, needle-shaped spores
develop mostly under wet weather conditions and sickle-
shaped ones (i.e. ballistospores) under dry conditions. Typical
yeast cells of smuts bud with young yeast cells emerging at a
point distal to the site of attachment to the mother cell (e.g.
Jacobs et aI., 1994) and separate easily. The yeast cells of
Trichodntractia utriculicola infecting several species of Rhyn-
chospora growing in swamps. however, bud close to the site of
attachment to the mother cell and adhere to it forming more
or less spherical yeast cell groups (Fig.3i). These are very
similar to Candida cell groups, which Ingold (1966) interprets
as being adapted to water dispersal.
However, tetradiately-shaped spores are not necessarily an
adaptation to aquatic environments. Tetraradiate spores also
possess an improved capacity to adhere, e.g, to insect hairs,
compare e.g. Candida reukaufii Diddens & Lodder with tetra-
radiate spores adapted to insect dispersal (Ingold.1971: 244).
Bot. Acta 111 (1998) 455
5.3 Insect dispersal
Dispersal by insects (entomochory) is potentially more effec-
tive than other dispersal modes because insects are likely to
deposit their spore load preferentially on particular plants.
Although these plants need not exclusively be hosts of the
smut fungus. and many spores may be lost in flight or while
grooming. less spores are subject to random deposition than
in e.g. wind dispersal (Roy, 1994). In the case of a heteroge-
neous distribution of the host plant with multiple local
populations at intermediate distances. dispersal gradients for
vector transmission will be more closely related to the
distribution of the host plants than to the distances them-
selves. Depending on the vector species, long-distance dis-
persal can be excellent. especially since small insects with
short flight ranges are frequently carried over long distances
by strong winds. When pollinators of the host plant also
disperse the pathogenic fungus. the disease behaves as a
venereal disease of the plant (Alexander and Antonovics,
1988, Thrall et al., 1993b). This makes it difficult for the host
plant to develop escape mechanisms to avoid infection
because a shorter period of exposure to the pathogen vector
also leads to reduced pollination by the same vector.
Insect dispersal is of major interest in the case of the anther
smuts of the genus Microbotryum (Fig.Ib; see Baker, 1947;
[ennersten, 1983 and 1988; Alexander and Antonovics, 1988
and 1995; Alexander,1990, Elmqvist et al., 1993; Roche et aI.,
1995, Altizer et al.,1998. and references in these articles). The
sporulation of these smut fungi occurs in the anthers, where
pollen is replaced by teliospores, but otherwise the flower
remains intact. They are still visited by pollinators. although
they may be less attractive (Maurizio, 1940; Shykoff and
Bucheli, 1995; Altizer et al., 1998). The fungus even induces
female plants of dioecious hosts (e.g. Silene latifolia Poiret) to
produce flowers with teliospore-filled anthers and reduced
ovaries.
Likewise, spores of anamorphic developments in flowers
infected by Thecaphora seminis-convolvuli (Desm.) Ito. Ustila-
go oxalidis Ellis& Tracy, Urocystis primulae, and Ur. primulicola
Magnus are most likely to be dispersed by pollinating insects.
There seem to be no ecological studies about this so far. Ur.
trientalis develops similar spores on the underside of the
leaves. probably relying on wind dispersal.
Dispersal agents of flower-inhabiting smuts may also be non-
pollinating insects (Fig.Ic). Flower-visiting and pollen-feed-
ing beetles, of e.g. the families Nitidulidae and Malachiidae,
are found abundantly in many flowers, even in flowers
primarily pollinated by bumble bees or moths. Certain
insects, especially Phalacridae beetles, feed on smut telio-
spores (Kellermann and Swingle, 1890: 286). Although they
may prefer infected plants, they probably visit healthy plants
as well. where spores can be deposited. Beetles of various
families have been observed repeatedly on smut-infected
plants and teliospores were found in their digestive system
(obs. by the authors).
No observations about myrmecochory have been made in the
smut fungi so far.

456 Bot. Acta 111 (1998) M. Piepenbring, G. Hagedorn, and F.Oberwinkler
------,--------------------------------
Fig.6 Dispersal units formed by host tis-
sue.
a-c) Doassansiopsis limnocharidis (Cif.) K.
Vanky on Limnocharis {lava (L) Buchenau.
Leaves with galls (Panama. MP 1438). a-b)
The galls often develop in two longitudinal
rows and may curl the blade (a). c) The
holes indicate where sori separated from
the lamina. d-e) Cintractia standleyana Zun-
del on Rhynchospora marisculus Nees (EI
Salvador, MP 1779). d) Inflorescence with
several spherical witches' brooms formed
by dwarfed spikelets filled with smut sori. e)
Part of an inflorescence showing one wit-
ches' broom with some exposed spore
masses and three dwarfed spikelets close to
the centre of the drawing. In the spikelet on
the left side, an ovary with stylopodium and
bristles which permit floating of the fruit is
shown.
12mm
@®
12cm

5.4 Dispersal through humans
In the bunts of cereal crops (e.g. Tilletia caries, I contraversa
Kuhn, I foetida (Wallroth) Liro), the compact sori remain
enclosed by a thin ovary wall and the glumes for a prolonged
period of time. Few teliospores are liberated in the field. The
majority of spores are set free during threshing, thus
efficiently inoculating the crop and the seed material for the
next sowing. This dispersal strategy could be interpreted as
an adaptation to the agricultural environment. Co-evolution-
ary forces are well studied in the case of weed seeds
becoming sufficiently similar in size and weight to crop seeds
to pass simple seed cleaning methods. Indeed, in many Tilletia
species on wild grasses (e.g. I ayresii Fig. te) the spores are
liberated into the wind while the sori are still on the plant.
Still, the question whether the dispersal strategy observed in
the cereal bunts is the result of a co-evolution of the parasite
by agricultural practices or the result of the domestication of
the host plant alone, must remain open.
5.5 Multiple vectors
The host plants of plant pathogens usually have a patchy,
heterogeneous distribution in the landscape. Any single
fungal population is in danger of extinction if the size of its
local plant population decreases. In such metapopulation
situations, different dispersal mechanisms may be dominant
within and among populations. For example, long-distance
dispersal of water-dispersed fungi occurs probably by means
of aquatic birds. Although in Microbotryum the main dispersal
occurs through insects, wind dispersal to neighbouring plants
and seedlings plays an important role for the local epidemiol-
ogy of the disease (Alexander,1990).The reverse could be true
for Ustilago maydis. The species is primarily dispersed by
wind, but insects probably are important for the ability of this
locally infecting smut to produce multiple sori on the same or
adjacent plants. Since primary infected seedlings produce sori
and die after a few weeks (Brefeld, 1912: 101f), all sori on
adult plants are the result of secondary epidemiological
cycles. Infection of mature plant organs by U. maydis is only
possible at wounds caused e.g. by hail storms or earwigs
(Eckstein, 1931), and stem boring insects like frit flies
(Oscinella frit, compare the review in Niemann, 1962: 441).
Most likely,these insects also disperse smut spores.
In general, there are probably two ranges of distances for
optimal dispersal, short distances to reach uninfected plants
within the local population and long distances corresponding
to distances between populations of the host plant. Effective
far-distance dispersal probably has to occur more sporadically
than short-distance dispersal for the long-term survival of the
parasite. Such a bimodal dispersal gradient may be achieved
by employing different vectors.
5.6 Unknown vectors
The sori of some species of Melanotaenium, Thecaphora, and
Urocystis occur on the stem as well as in the root of their
hosts, and spores of species of Entorrhiza develop exclusively
in the roots of their host plants (Fig.td). No vector dispersing
such underground spores has been identified yet. It is
conceivable that root parasitizing organisms like nematodes
or other soil organisms playa role in the dispersal of these
smuts. This would be especially relevant for Entorrhiza where
Bot. Acta 111
the teliospores are buried deep in wet, often heavy clay soils.
Brefeld (1912: 82) assumed that Entorrhiza teliospores are
only resting structures and other spore forms would be
responsible for long-distance dispersal. He suggested that the
fungus could spread by long-reaching infective mycelia to
neighbouring plants. Entorrhiza does produce secondary
spores on the mycelium resulting from teliospore germination
(Weber, 1884; Fineran, 1982), but these spores are not
ballistospores, making it more difficult to be released into the
air. Currently, no mechanism for long-distance dispersal is
known, and the distribution pattern of Entorrhiza (Europe,
Canada, and Latin America - where ustilaginologists looked
for it!), remains puzzling.
6. GeneralAspects
6.1 Thetiming ofteliospore liberation and dispersal
Important factors for the successful completion of the life cycle
of a smut fungus are the liberation of diaspores under favour-
able climatic conditions, the presence of a vector for dispersal,
the deposition of the diaspores on a susceptible host plant, and
adequate temperature and moisture conditions for germination
and infection. The fungus usually cannot influence the spatial
component of these factors, but it can influence the temporal
component. Efficient dispersal is highly dependent upon the
timing of diaspore liberation. The correct time may depend
upon the time of day or year, the climatological conditions, or
the phenology of susceptible stages of the host plant.
Several flower-infecting smuts induce their hosts to present
their infected flowers slightly earlier than healthy plants, thus
providing optimal inoculum concentrations during anthesis
of healthy plants. Examples are Microbotryum violaceum
(Baker, 1947), Ustilago nuda (Freeman and Johnson, 1909;
Sreeramulu, 1962), and U. avenae (Mills, 1967). The whip-
shaped sori of Ustilago scitamineaon Saccharum cultivars also
develop well before the anthesis of the healthy plants (pers.
observation). Lee (1981) found that male plants of Silene
dioica (L.) Clairv. infected by M. violaceum flower stronger at
the onset of the flowering period, and infected female plants
have a prolonged flowering period towards the end. Infected
flowers are thus disproportionally present during times when
fewer healthy flowers compete for pollinator visitation, maxi-
mizing the potential for spread of the disease.
The developing teliospores of most smut fungi are protected
by some host or fungal tissue against desiccation, UV
radiation, and insects feeding on young fungal cells. After the
initial rupture of the sorus, these soral envelopes influence
teliospore liberation. They ensure that teliospores are liber-
ated over a prolonged period of time, so that at least some
spores are released under climatic conditions suitable for
further growth and at a time when host plants are susceptible
to infection. Agglutination of spores and elaters embedded in
the spore mass can add to this effect. This interpretation also
applies to the capillitium in myxomycetes and secondary
pollen presentation in Asteridae (Yeo, 1993).
The hygroscopic movement of the peridia of certain species of
Sporisorium and Farysia comiculata (Fig.4) promotes telio-
spore liberation under wet conditions (hygrochasy). High
humidity conditions may be relevant to all stages of dispersal.
First, high humidity conditions or showers are frequently
457

458 Bot. Acta 111 (1998)
----------------,
accompanied by strong and gusty winds, which liberate
spores either directly or through host plant agitation and
favour a long-distance dispersal of the spores. Second, most
smuts can only infect young meristematic tissues, like seed-
lings or growth points in leafaxils or inflorescences. Rain can
wash teliospores or germination products beneath the glumes
in grass inflorescences, or from the leaf blade of leaf axil buds,
places which are otherwise difficult to reach. Malik and Batts
(1960) found an increase of up to 50% in infection with
Ustilago nuda when water drops were sprinkled over the ears
which were previously dusted with smut spores. Finally, the
conditions for spore germination are improved.
Figure 7 illustrates the interrelationship between abiotic
factors and the stages of diaspore dispersal. Since the
occurrence of strong wind, rain, and moisture is often highly
correlated, a smut fungus can utilize different atmospheric
factors for different stages of dispersal without undue risk of
only one factor being present. For example, a certain smut
could be predominantly liberated by host plant agitation,
dispersed by wind, and deposited by wash out. Many smut
fungi seem to be flexible in their choice, utilizing different
strategies depending on the weather conditions.
6.2 Thegenetic lifecycle
Most basidiomycetes use basidiospores as predominant dia-
spores. In the smut fungi this role has been assumed by a
probasidium, the teliospore. The difference between the
dispersal of teliospores (dikaryotic or diploid) and the dis-
persal of haploid basidio- or secondary spores is genetically
relevant. Mating and dikaryotization are obligatory for the
infection process because haploid stages of smut fungi cannot
infect a host. In contrast to basidiospores, teliospores carry
the complementary mating types in a single dispersal unit,
maximizing the chances of dikaryotization. The same argu-
ment applies to the dispersal of entire basidia and the
dikaryotic spores produced after the mating of basidiospores
in Tilletia and Entyloma.
Adisadvantage of this strategy is obviously the high degree of
inbreeding resulting from it. In some species the inbreeding is
further increased by early copulation of basidiospores or
phragmobasidial cells themselves. Although spores are often
deposited together, these are likely to come from the same
sorus, thus being kin to each other. Only in cases of very
heavy infection pressure will spores from different sources be
deposited close enough to allow outcrossing. The chances of
outcrossing are greatly increased by the short-distance dis-
persal of haploid secondary dispersal units, e.g. yeast cells
from various deposition places being washed into a leaf axil
prior to mating and infection. Thus secondary units may have
a genetic importance far beyond their importance for dis-
persal.
A similar argument can be made for the rust fungi. Although
both their haplophase and dikaryophase are parasitic, both
mating types must be present for the successful completion
of the life cycle. Most dispersal units (aecio-, uredinio- and
teliospores) are dikaryotic. They function as a container,
guaranteeing that any founded population consist of both
mating types.
The correlation between the use of a premeiotic diaspore and
a fixed life cycle with at least one obligately phytopathogenic
stage is striking. Pathogens will generally profit from the
ability to disperse solopathogenic diaspores. The constant
need to found small isolated populations, sometimes after a
long-distance dispersal, makes a life cycle involving obligate
outcrossing disadvantageous. Baker (1955) observed a general
correlation between self-compatibility and long-distance dis-
persal in higher plants and animals. Therefore, although the
common presence of a teliospore in smut and rust fungi is
interesting, it should not be regarded as sufficient grounds to
assume a sister taxon relationship between them.
7. Acknowledgements
The authors thank C. T. Ingold, M. Scholler, S. Altizer, and an
unknown reviewer for critically revising earlier or later drafts
of the manuscript, R. Bauer and K. Vanky for interesting
discussions, V. Uhle-Schneider for improving and mounting the
illustrations, as well as R. Berndt, H. Schoppmann, and F.
Albrecht for photographic work.
8. References
Alexander, H. M. - Epidemiology of anther-smut infection of Silene
alba causedbyUstilago vialacea: Patternsofsporedeposition and
diseaseincidence.]. Eco!. 78 (1990), 166-179.
Alexander, H. M. and Antonovics,]. - Disease spread and population
dynamics of anther-smut infection of Silene alba caused by
Ustilago violacea.]. Eco!. 76(1988), 91-104.
Spore Agitation 'lap release"
liberation of host plant: wet or dry
friction. accele- (puff, bellows)
ration forces strong wind, splash liberation strong
turbulence water currents
"
flowing
Transport Wind ~ Rain I-- Water
splash dispersal
Deposition Gravity impaction wash out
impaction,
sedimentation
anchoring
t t
Germination
Moisture, water films
J& Infection improved germination conditions
Fig.7 The interrelationship between at-
mospheric factors (Wind, rain, flowing wa-
ter, moisture), gravity, and differentstages
of the diaspore pathway (liberation, trans-
port, deposition, and infection). The occur-
rence of wind, rain, and water is naturally
correlated.

Alexander, H. M. and Antonovics, j. - Spread of anther-smut disease
(Usti/ago vio/acea) and character correlations in a genetically
variable experimental population of Silene alba. j. Ecol. 83 (1995),
783-794.
Altizer. S. M., Thrall. P. H., and Antonovics, j. - Vector behavior and
the transmission of anther-smut infection in Si/ene alba. Am.
Midland Nat. 139 (1998), 147 -163.
Baindridge, A. and Legg, B. J. - Release of barley-mildew conidia
from shaken leaves. Trans. Br.Mycol. Soc. 66 (1976),495 - 498.
Baker, H. G. - Infection of species of Me/andrium by Ustilago vio/acea
(Pers.) Fuckel and the transmission of the resultant disease. Ann.
Bot. N. 5.11 (1947).333-348.
Baker. H. G. - Self-compatibility and establishment after "long-
distance" dispersal. Evolution 9 (1955), 347.
Bauer, R., Oberwinkler. F., and Vanky, K. - Ultrastructural markers
and systematics in smut fungi and allied taxa. Can. j. Bot. 75
(1997),1273 -1314.
Bond. T. E. T. - Observations on the disease of sea lyme-grass
(E/ymus arenarius L.) caused by Usti/ago hypodytes (Schlecht.)
Fries. Ann. Appl.BioI. 27 (1940), 330-337.
Brefeld, O. - Untersuchungen aus dem Gesammtgebiete der Mykolo-
gie. XV. Die Brandpilze und die Brandkrankheiten. V. Commis-
sions Verlag H. Schoningh, MOnster, 1912.
Brefeld, O. and Falk, R. - Untersuchungen aus dem Gesammtgebiete
der Mykologie. XIII.Die Bhiteninfektion bei den Brandpilzen und
die naturliche Verbreitung der Brandkrankheiten. I. Commissions
Verlag H. Schoningh, MOnster, 1905.
Castle, A. J. and Day, A. W. - Isolation and identification of alpha-
tocopherol as an inducer of the parasitic phase of Ustilago
vio/acea. Phytopathology 74 (1984),1194-1200.
Clement, J. A., Porter, R., Butt, T. M., and Beckett, A. - The role of
hydrophobicity in attachment of urediniospores and sporelings of
Uromyces viviae-fabae. Mycol. Res. 98 (1994),1217 -1228.
Cox, P. A. - Search theory, random motion, and the convergent
evolution of pollen and spore morphology in aquatic plants. Am.
Nat. 121 (1983),9-31.
Deml, G. - Keimung phragmobasidialer Brandpilze. Ber. Deutsch.
Bot. Ges. 99 (1986), 83-88.
Eckstein, F. - Uber die Bedeutung des Ohrwurms (Forficu/a auticu-
/aria) fur den Mais. Nachrichtenblatt des Deutschen Pflanzen-
schutzdienstes 11 (1931),2 - 3.
Elmqvist, T., Liu, D.. Carlsson, U., and Giles. B. E. - Anther-smut
infection in Silene tiiocia: variation in floral morphology and
patterns of spore deposition. Oikos 68 (1993), 207 - 216.
Fineran, J. M. - Teliospore germination in Entorrhiza casparyana
(Ustilaginales). Can.j. Bot. 60 (1982), 2903 -2913.
Fischer, G. W. and Holton, C. S.- Biology and control of the smut
fungi. Ronald Press, New York, 1957.
Fitt, B. D. L. and McCartney, H. A. - Spore dispersal in relation to
epidemic models. In: K. j. Leonard and W. E. Fry, eds., Plant
disease epidemiology. Vol. 1: Population dynamics and manage-
ment. pp 311 - 345. Macmillan, New York, 1986.
Fitt, B. D. L., Gregory, P. H., Todd, A. D., McCartney, H. A., and
Macdonald, O. C. - Spore dispersal and plant disease gradients; a
comparison between two empirical models. j. Phytopathol. 118
(1987),227 -270.
Fitt, B. D. L.,McCartney, H. A., and Walklate, P.j. - Role or rain in the
dispersal of pathogen inoculum. Annu. Rev. Phytopathol. 27
(1989),241-270.
Freeman, E. M. and Johnson, E. C. - The loose smuts of barley and
wheat. Bull. U.S. Dept. Agric. Bur. plant Industr. No. 152 (1909).
Gregory, P. H. - Dispersal. In: G. C. Ainsworth and A. S. Sussman,
eds., The fungi, an advanced treatise. Vol. II: The fungal organism.
pp, 709- 732. Academic Press, New York, 1966.
Gregory. P. H. - The Microbiology of the Atmosphere. 2nd edn.
Leonard Hill, London, 1973.
Bot. Acta 111 (1998) 459
Gregory, P. H. - Disease gradients of wind-borne plant pathogens:
Interpretation and misinterpretation. In: K. S. Bilgrami, R. S.
Misra, and P. C.Misra, eds., Advancing Frontiers of Mycology and
Plant Pathology. pp. 107 -117. Today and Tomorrow's Publishers,
New Delhi, 1982.
Hirst.], M. - Spore liberation and dispersal. In: C.S. Holton, ed., Plant
Pathology Problems and Progress, 1908 -1958. pp. 529-538. Uni-
versity of Wisconsin Press, Madison, Wisconsin, 1959.
Hirst, J. M. and Stedman, O. J. - Dry liberation of fungus spores by
raindrops. j. Gen. Microbiol. 33 (1963), 335 - 344.
Hirst, j. M. and Stedman, O. j. - Patterns of spore dispersal in crops.
In: T. F. Preece and C. H. Dickinson, eds., Ecology of leaf surface
microorganisms. pp. 229 - 237. Academic Press, London. 1971.
Ingold, C. T. - The tetraradiate aquatic fungal spore. Mycologia 58
(1966),43-56.
Ingold, C. T. - Fungal spores. Their liberation and dispersal.
Clarendon, Oxford, 1971.
Ingold, C.T. - Advances in the study of so-called aquatic hyphomy-
cetes. Am.J. Bot. 66 (1979), 218-226.
Ingold, C. 1. - The basidium in Ustilago. Trans. Br. Mycol. Soc. 81
(1983),573-584.
Jacobs, c. w., Mattichak, S. J.. and Knowles, j. F. - Budding patterns
during the cell cycle of the maize smut pathogen Ustilago maydis.
Can.j. Bot. 72 (1994), 1675-1680.
jennersten, O. - Butterfly visitors as vectors of Usti/ago vio/acea
spores between caryophyllaceous plants. Oikos 40 (1983), 125-
130.
jennersten, O. - Insect dispersal of fungal disease: effects of Ustilago
infection on pollinator attraction in Viscaria vulgaris. Oikos 51
(1988),162 -170.
Kaiser, W. - Zur Biologie und Entwicklungsgeschichte einiger
Enty/oma-Arten. Angew. Bot. 18 (1936),81-131.
Kellermann, W. A. and Swingle, W. T. - Report of the Botanical
Department. Report on the loose smuts of cereals. Kansas Agric.
Exp. Sta. Annu. Rep. 2 (1890), 213 - 288.
Kukkonen, I. - Taxonomic studies on the species of the section
Echinosporae of Anthracoidea. Ann. Bot. Fenn. 1 (1964),161-177.
Lacey,j. - Spore dispersal-its role in ecology and disease: the British
contribution to fungal aeriobiology (centenary review). Mycol.
Res. 100 (1996), 641-660.
Langdon, R. F. N. and Fullerton, R. A.- Sorus ontogeny and sporo-
genesis in some smut fungi. Aust.], Bot. 23 (1975). 915-930.
Lee, j. A. - Variation in the infection of Silene dioica (L.) Clairv. by
Ustilago vio/acea (Pers.) Fuckel in North West England. New
Phytol. 87 (1981).81-89.
Malik, M. M. S. and Batts. C.C.V.- The determination of the reaction
of barley varieties to loose smut. Ann. Appl. BioI. 48 (1960), 39-
50.
Malloch, D. and Blackwell. M. - Dispersal of fungal diaspores. In: G.
C. Carroll and D. 1. Wicklow, eds.. The fungal community: its
organization and role in the ecosystem. pp 147 -171. Marcel
Dekker. New York,1992.
Maurizio, A. - Pollenanalytische Beobachtungen 9. Die Nektarpro-
duktion in gesunden und von Ustilago vio/acea befallenen Bluten
von Melandrium a/bum (Miller) Garcke. Ber. Schweiz. Bot. Ges. 51
(1940): 77-95.
Mills, J. T. - Spore dispersal and natural infection in the oat loose
smut (Ustilago avenae). Trans. Br.Mycol. Soc. 50 (1967), 403 - 412.
Misra. S. R., Ram. R. S.. and Singh. K. - Effect of intercrops on
dispersal of smut (Ustilago scitaminea) spores in sugarcane field.
Indian]. Agric. Sci. 59 (1989). 114-117.
Niemann, E. - Ustilaginales, In: H. Richter, ed., Handbuch der
Pflanzenkrankheiten, begr. Von P. Sorauer, Bd.3. 4. Lieferung:
Basidiomyceten 6. Edn. pp. 276-316. Paul Parey, Berlin 1962.
Oort, A. J. P. - De verspreiding van de sporen van tarwestuifbrand
(Ustilago tritici)door de lucht. (The dissemination of the spores of
loose smut (Ustilago tritici) through the air.). Tijdschr. PI. Ziekt. 46
(1940),1-18.

460 Bot. Acta 111 (1998)
Pacini, E.and Franchi, G.G.- Roleof the tapetum in pollen and spore
dispersal. Plant Syst. Evol.(Suppl.) 7 (1993), 1-11.
Piepenbring, M. - Smut fungi (Ustilaginales and Tilletiales) in Costa
Rica.Nova Hedwigia Beih.113,1996.
Piepenbring, M. and Bauer, R. - Noteworthy germinations of some
Costa Rican Ustilaginales. Mycol.Res.99 (1995), 853-858.
Piepenbring, M. and Hagedorn, G. - Beobachtungen zur Freisetzung
und Verbreitung von Diasporen neotropischer Brandpilze. Biele-
felder Okologische Beitrage 12 (1998), 136.
Pijl, L. van del' - Principles of dispersal in higher plants. Springer,
Berlin, Heidelberg, New York, 1969.
Ramberg, j. E. and McLaughlin, D. j. - Ultrastructural study of
promycelial development and basidiospore initiation in Ustilago
maydis.Can.j. Bot.58 (1980), 1548-1561.
Roche, B. M., Alexander, H. M. and Maltby, A. D. - Dispersal and
disease gradients of anther-smut infection of Silene alba at
different life stages. Ecology76 (1995),1863 - 1871.
Roy, B.A. - The use and abuse of pollinators by fungi. Trends Ecol,
Evol. 9 (1994), 335-339.
Savile, D. B. O. - Cellular mechanics, taxonomy and evolution in the
Uredinales and Ustilaginales. Mycologia 46 (1954), 736-761.
Schuhmann, G. - Tilletiaceae. In: H. Richter, ed., Handbuch del'
Pflanzenkrankheiten, begr. Von P. Sorauer, Bd.3, 4. l.ieferung:
Basidiomyceten, 6. Edn. pp. 458-525. Paul Parey, Berlin,(1962).
Setchell, W. A. - An examination of the species of the genus
Doassansia Cornu. Ann. Bot.(London) 6 (1892), 1-48 + 2 Pis.
Shykoff,J. A. and Bucheli, E. - Pollinator visitation patterns, floral
rewards, and the probability of transmission of Microbotryum
violaceum, a veneral disease of plants. j. Ecol. 83 (1995): 189-
198.
Simmonds, N.W. - Some speculative calculations of the dispersal of
sugarcane smut disease. Sugar Cane 1 (1994),2 - 5.
Sreeramulu, T.- Aerial disseminations of barley loose smut (Ustilago
nuda).Trans. Br.Myco!.Soc. 45 (1962), 373 -384.
Sreeramulu, T. and Vittal, B. P. R. - Spore dispersal of the sugarcane
smut (Ustilago scitaminea). Trans. Br. Mycol. Soc. 58 (1972), 301-
312.
Stepanov, K. M. - Dissemination of infective diseases of plants by air
currents. (In Russian.) Bull. PI. Prot. Leningrad, Ser. 2. Phytopa-
thology 8 (1935), 1-68.
Stepanov, K.M.- Gribnye epifitotii. SeI'khozizdat, Moscow, 1962.
Tapke, V. F.- Studies on the natural inoculation of seed barley with
covered smut (Ustilago hordei).J.Agric.Res.60 (1940), 787 - 810.
Thrall, P. H. Biere, A., and Antonovics, J. - Plant life-history and
disease susceptibility - the occurrence of Ustilago violacea on
different species within the Caryophyllaceae. j. Ecol. 81 (1993a),
489-498.
Thrall, P. H., Antonovics, j., and Hall, D. W. - Host and pathogen
coexistence in sexually transmitted and vector-borne diseases
characterized by frequency-dependent disease transmission. Am.
Nat. 142 (1993b), 543 - 552.
Vanky, K. - Illustrated genera of smut fungi. - Gustav Fischer Verlag,
Stuttgart, New York, 1987.
Vanky, 1<. - Spore morphology in the taxonomy of Ustilaginales.
Trans. Mycol.Soc.japan 32 (1991),381 -400.
Vanky, 1<. - European Smut Fungi. Gustav Fischer Verlag, Stuttgart,
[ena, New York, 1994.
Ward, H. M. - On the structure and life-history of Entyloma
Ranunculi (Bonorden). Phil. Trans. R. Soc. London, Ser. B, 178
(1887),173 -185 + PIs.10-13.
Weber, C. - Ober den Pilz del' Wurzelanschwellungen von ]uncus
bujonius. Bot.Zeit. (Berlin) 42 (1884), 369 - 379.
Webster, J. - Experiments with spores of aquatic hyphomycetes. I.
Sedimentation, and impaction on smooth surfaces. Ann. Bot. 23
(1959), 595 - 611.
Webster. j. and Davey, R. A. - Sedimentation rate and trapping
efficiency of cells of Candida aquatica. Trans. Br. Mycol. Soc. 64
(1975),437 -440.
M. Piepenbring, G. Hagedorn. and F. Oberwinkler
Webster, J. and Davey. R. A. - Sigmoid conidial shape in aquatic
fungi. Trans. Br.Mycol. Soc.83 (1984),43 - 52.
Wennstrtim, A. and Ericson, L. - The interaction between the clonal
herb Trientalis europaea and the host specific smut fungus
Urocystis trientaiis. Oecologia 85 (1990), 238-240.
Yeo, P. F. - Secondary pollen presentation. Form. function and
evolution. Plant Syst. Evol. Suppl. 6. Springer-Verlag, New York,
1993.
M. Piepenbring, F. Oberwinlder
Lehrstuhl Spezielle BotanikfMykologie
Botanisches Institut
Universitat Tiibingen
Aufdel' Morgenstelle 1
D-72 076 Tiibingen
Germany
Section Editor: G.Cottsberger

Mission impossible

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Mission impossible

Similar to Mission impossible (20)

Recently uploaded

Recently uploaded (20)

Mission impossible