12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
10.1146/annurev.phyto.40.021202.110417
Annu. Rev. Phytopathol. 2002. 40:13?43
doi: 10.1146/annurev.phyto.40.021202.110417
Copyright c? 2002 by Annual Reviews. All rights reserved
EVOLUTIONARY ECOLOGY OF PLANT DISEASES IN
NATURAL ECOSYSTEMS
Gregory S. Gilbert
University of California, Environmental Studies Department, Santa Cruz,
California 95064; e-mail: ggilbert@cats.ucsc.edu
Key Words plant community dynamic, plant diversity, density dependence,
Janzen-Connell hypothesis, local adaptation
n Abstract Plant pathogens cause mortality and reduce fecundity of individual
plants, drive host population dynamics, and affect the structure and composition of
natural plant communities. Pathogens are responsible for both numerical changes in
host populations and evolutionary changes through selection for resistant genotypes.
Linking such ecological and evolutionary dynamics has been the focus of a growing
body of literature on the effects of plant diseases in natural ecosystems. A guiding
principle is the importance of understanding the spatial and temporal scales at which
plants and pathogens interact. This review summarizes the effects of diseases on popu-
lations of wild plants, focusing in particular on the mediation of plant competition and
succession, the maintenance of plant species diversity, as well as the process of rapid
evolutionary changes in host-pathogen symbioses.
INTRODUCTION
Plant pathogens play Jekyll and Hyde roles in the structure, dynamics, and evo-
lution of natural plant communities. As destructive agents, plant pathogens can
cause mortality, reduced fitness of individual plants, rapid declines of populations
of particular host species, or dramatic shifts in the structure or composition of
plant communities. At the same time, pathogens can help maintain plant species
diversity, facilitate successional processes, and enhance the genetic diversity and
structure of host populations. Understanding the impacts of diseases in natural
plant communities requires integrating numerical dynamics, rapid evolutionary
changes, and spatial structure of both host and pathogen populations, and an ap-
preciation for how destructive actions at one scale can be the foundation for positive
outcomes at another.
There have been a number of important reviews of plant diseases in natural
ecosystems. Early contributions included the original summons to plant ecologists
by Harper in 1977 (120), the first significant review of the field by Dinoor & Eshed
in 1984 (84), and Burdon?s seminal Diseases and Plant Population Biology in
0066-4286/02/0901-0013$14.00 13
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
14 GILBERT
1987 (46). The following decade saw a number of important reviews on various
aspects of plant diseases in natural communities (10, 14, 29, 48, 52, 82, 140). Here
I expand upon and update what has already been said.
In this review I address three broad areas of the role of plant diseases in natural
plant ecosystems. First, I look briefly at how pathogens with different life histories
affect wild plants at different stages in their life cycles, focusing on effects of re-
duced survivorship, growth, and fecundity that could influence numerical dynamics
of the host. Second, I look at diseases that affect host population dynamics or that
affect plant community diversity, structure, and dynamics by mediating plant com-
petition or successional processes. Finally, I explore some of the rapid evolutionary
changes associated with plant diseases in natural ecosystems, particularly looking
at patterns and consequences of local adaptation. In examining these themes, I
try to integrate existing theoretical frameworks with empirical examples from the
literature. Whenever possible, examples come from studies in natural ecosystems,
but in some cases, I refer to experimental studies in agriculture or model systems
to indicate what we might expect to find in natural systems. Finally, for each broad
question, I point out key areas in which critical data are lacking. Perhaps because
of early development of theory for the role of plant diseases in high diversity sys-
tems, there is an unusual depth to the literature of diseases in natural communities
in the tropics, and that is reflected in this review.
HOST MORTALITY, GROWTH, AND REPRODUCTION
Pathogens can affect host population dynamics through direct effects on the sur-
vival, growth, and fecundity of individual plants. Examples in agricultural sys-
tems pervade the literature, and although documentation of direct effects outside
of crop plants is rarer, a wide range of pathogen types clearly affect wild host
plants in many of the same ways. Plants killed by disease before reproducing
do not contribute to the next generation. Additionally, because fecundity is usu-
ally correlated with plant growth and size, diseases that affect growth are likely
to reduce reproductive output, as do pathogens that directly attack flowers and
developing fruits. Because the impacts of diseases depend on the life stage of
the plant that is attacked and the life history of the pathogen, I have categorized
examples of wild plant diseases as (a) seed decay, (b) seedling diseases, (c) fo-
liage diseases, (d ) systemic infections, (e) parasitic plants, ( f ) cankers, wilts, and
diebacks, (g) root and butt rots, and (h) floral diseases. Similarly, I categorized
the effects of the diseases on (A) host survival, (B) growth, or (C ) fecundity. In
Table 1 I use this cross-categorization to summarize a collection of representa-
tive studies of the effects of diseases on wild plants. The discussion is organized
to follow the life cycle of a plant, and I highlight some important themes and
emphasize disease systems seldom included in discussion of diseases in natural
ecosystems.
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 15
TABLE 1 Representative studies of the direct effects of plant diseases on individual plants in
natural systems
Type of effects on host
Plant stage Survival Growth Fecundity
Seed decay (73?75, 146, 156, 160) N.A. N.A.
Seedling diseases (16, 25, 27, 28, 30, 92, ? N.A.
142, 170, 222)
Foliar diseases (13, 76, 77, 164, 165, (85, 88, 144, 154, 216) (13)
182, 214, 231)
Viruses, viroids, (11, 13, 67, 70, 140, (70, 94, 112, 116, 140, (9, 58, 65, 94, 140,
and phytoplasmas 188, 204, 221, 224) 188, 190?193, 205, 224) 185, 186, 188)
Parasitic plants (19, 34, 161, 218) (161, 202, 218) (194)
Canker, wilt, and (1, 20, 21, 43, 77, 96, (99, 102, 131) ?
dieback diseases 100, 102, 114, 131,
141, 147, 159, 178,
180, 195, 196)
Root diseases and (82, 83, 117, 125, (18, 41) (167)
butt rots 181, 189, 223)
Floral infections N.A. N.A. (8, 9, 12, 15, 17, 58,
140, 207, 217, 227)
N.A. D not applicable, ? D no examples found, but effects are probable.
Seed Decay
The highest rates of disease-related mortality of plants in natural systems are
usually due to seed and seedling diseases. Fresh seeds are susceptible to fungal
infection, but rates of attack are often affected by handling by frugivorous ani-
mals. Seeds of the common forest tree Strychnos mitis (Strychnaceae) in Uganda
suffered 88% mortality, primarily from fungal attack, although when fruits were
first handled by frugivorous monkeys fungal attack was nearly eliminated (146).
In contrast, fungal attack of seeds of fresh blueberries (Vaccinium angustifolium,
Ericaceae) in Nova Scotia increased dramatically when fruits were first consumed
and passed by birds (74).
Loss of soil seed bank to fungal attack ranged from 10% in the invasive shrub
Mimosa pigra (Mimosaceae) in tropical Australia (156) to 47% and 39% annual
mortality of the seed bank of two tropical pioneer tree species in a lowland moist
tropical forest in Panama (75), and from negligible to more than 90% mortality for
seeds of five plant species in the soil of Wyoming shrub-steppe (73). In general,
impacts of seed pathogens may be greatest for plant species that rely heavily
on periodic disturbances for regeneration from long-lived seed banks, including
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
16 GILBERT
forest-gap-specialist pioneer species and fire-released species from prairie and
chaparral habitats. However, a major lacuna in research remains the effects of
pathogens on seeds in the soil.
Seedling Diseases
Damping-off diseases of seedlings are probably the most widely studied lethal dis-
eases of plants in natural communities. Augspurger and coworkers (25, 27, 28, 142)
first brought damping-off to the attention of plant ecologists with a series of stud-
ies of seedling mortality in the tropical forest of Barro Colorado Island (BCI) in
Panama. Damping-off affected 80% of tested species (25) and was the primary
cause of death for seedlings of six of nine tree species, killing up to 74% of a par-
ent tree?s annual reproductive output (28). Most studies of damping-off in natural
ecosystems have been done in tropical forests, but Packer & Clay (170) found that
black cherry seedlings (Prunus serotina, Rosaceae) in Indiana suffered up to 65%
mortality from Pythium.
Environmental heterogeneity can have strong effects on the development of
damping-off caused by Oomycetes like Pythium and Phytophthora. For seedlings
of several tropical trees, damping-off was more severe in soils with blocked
drainage (92) or in areas of deep shade (25, 28, 222) than in drier or sunnier
areas. Such effects of environmental heterogeneity are common for damping-
off caused by Oomycetes, which rely on water-filled soil pores for zoospore
motility.
Foliar Diseases
Foliar pathogens, including the fungi, bacteria, and viruses that cause spots, necro-
sis, chlorosis, shot-hole, early senescence, and leaf abscission, all reduce leaf area
and associated photosynthetic activity that would otherwise contribute to plant
growth and reproduction. For seedlings in particular, foliar diseases may signifi-
cantly increase mortality, since the probability of seedling survival is strongly af-
fected by seedling size (104). Defoliation of Larix (Pinaceae) by the needle-cast
fungus Mycosphaerella laricinia reduced seedling radial growth, root biomass,
and new shoot production in the following year (144). Although foliar diseases
may sometimes kill plants directly, reduced plant size from foliar or other diseases
also places affected seedlings at a competitive disadvantage to larger, healthier
neighbors (220). This disadvantage may be most important in light-limited envi-
ronments, such as forest understories or dense grasslands.
Foliar diseases can also affect growth and reproduction of larger plants, although
they probably rarely kill them. Fungal pathogens accounted for 34% of leaf damage
on the leaves of 10 tree species in a lowland tropical forest in Panama, and were the
leading cause of damage on 4 of the host species (35). Two thirds of host species and
43% of leaves surveyed in a Mexican tropical rain forest were damaged by fungi,
but leaf area damaged was on average less than 1%, and always less than 20% (98).
Similarly, foliar infection of seedlings in forest fragments in the central Amazon
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 17
was less than 2% (38). The petiolar pathogen Phylloporia chrysita caused 52%
reduction in growth rates on infected individuals of the tropical tree Erythrochiton
gymnanthus (Rutaceae) in Costa Rica (88).
Effects of foliar plant pathogens are not restricted to terrestrial plant commu-
nities. Marine slime molds in the genus Labyrinthula cause wasting disease in a
wide range of seagrasses worldwide (214). Labyrinthula infection of turtlegrass
(Thalassia testudinum) causes necrotic lesions on the leaves and has dramatic ef-
fects on photosynthetic efficiency (85). Wasting disease decimated populations of
eelgrass (Zostera marina) in the early 1930s, and again in the 1980s and 1990s
on both the eelgrass (164, 165) and turtlegrass (182, 231). Temperature, light, and
sediment nutrients all strongly influence production of defensive phenolic com-
pounds in seagrasses (45, 213, 215), and some authors suspect that outbreaks may
be related to changes in environmental conditions (213). Recently, Jackson postu-
lated that the ultimate cause of wasting outbreaks may have been the much earlier
ecological demise of green sea turtles, which were dominant grazers on seagrasses
(133). Loss of the dominant herbivores may have released turtlegrass populations
from top-down population regulation, resulting in dense host populations and al-
tered nutrient cycling, making Thalassia vulnerable to an outbreak of the endemic
pathogen Labyrinthula.
Systemic Infections
In 1990, Harper made a plea for increased work on the role of viruses in natural
plant communities (120). However, despite the recognition that viruses, viroids,
and phytoplasmas cause systemic and persistent infections in a wide range of host
plants, their importance in natural communities has still received much less atten-
tion than other pathogens. There are numerous reports of wild plants as reservoirs
for phytoplasmas, viruses, and viroids that affect economically important crops
(60, 90, 127, 137, 145, 148, 157), but the effects of these infections on the wild
hosts are rarely studied (112, 116, 191, 193). With the increased availability and
portability of PCR-based and immunologically based assays for phytoplasmas and
viruses, our understanding of their role in plant population dynamics outside of
agriculture should increase dramatically.
Some fungi and Oomycetes cause systemic infections with severe effects. Sys-
temic infection of seedlings of Shepherds Purse, Capsella bursa-pastoris (Brassi-
caceae), by Albugo candida or Peronospora parasitica caused up to 88% mortal-
ity (13), and the systemic smut Urocystis trientalis reduced survival of Trientalis
europaea (Primulaceae) by 50% (221).
In contrast, systemic infections of grasses by fungal endophytes are often viewed
as important mutualisms because some endophytes protect the host plant from her-
bivory and increase the plant?s competitive ability (67, 70, 224). Such antiherbivory
benefits to the host plant may not be universal, however (205), and in many cases
host growth, competitive ability, and reproduction suffer directly from endophyte
infection in the absence of herbivory [see reviews in (66, 140, 188)].
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
18 GILBERT
Parasitic Plants
Dwarf mistletoe (Arceuthobium spp., Viscaceae) is often a major pathogen of
coniferous trees, causing reduced host growth (161, 202), increased mortality
(19, 34, 161), and reduced cone and seed production (194). Dwarf mistletoe can
cause up to 65% reduction in growth in severely infected Douglas-fir (Pseudot-
suga menziesii) individuals, and mortality in severely infected stands was three- to
fourfold greater than in healthy stands (161). Watson has provided a recent review
of the ecology of mistletoes (218).
Cankers, Wilts, and Diebacks
In the past century, North American forests have suffered devastating effects from
introduced diseases including chestnut blight (20), Dutch elm disease (100, 147),
pitch canker of Monterey pine (114), beech bark disease (131), white pine blister
rust (195), and recently, sudden oak death (96). These epidemics were particu-
larly dramatic due to the widespread, rapid death of mature host trees when the
pathogens girdle them or otherwise interfere with vascular transport. Sudden oak
death, caused by Phytophthora ramorum, has killed tens of thousands of true
oaks (Quercus spp.) and tanoaks (Lithocarpus densiflorus) (Fagaceae) in Califor-
nia since its appearance in 1994 (96, 180). The known taxonomic host range of
P. ramorum is growing rapidly, with infections detected in at least five host genera
outside of the Fagaceae (180). Sudden oak death threatens to transform oak wood-
lands of the Pacific coast of the United States, as well as oak-dominated systems
elsewhere. Important canker and vascular diseases are not necessarily caused by
introduced pathogens. The endemic Hawaiian tree koa (Acacia koa), a keystone
species in upper-elevation forests, suffers from a dieback disease caused by the
systemic wilt pathogen Fusarium oxysporum f. sp. koae (21). Epidemics of dra-
matic consequence may have prehistoric precedent. Fossil evidence from England
indicates there was a large neolithic decline of elms, thousands of years before the
Dutch elm disease epidemic of recent years (110, 177). However, such dramatic
impacts of introduced diseases may become increasingly frequent phenomena with
growth of international trade, precipitating the need for strengthened policies on
the movement of biological materials (113).
Root and Butt Rots
Root and butt rots of trees can cause significant mortality of mature forest trees.
The introduced pathogen Phytophthora lateralis has caused severe mortality on
the California-Oregon endemic Port-Orford Cedar (Chamaecyparis lawsoniana,
Cupressaceae) (117), and Phytophthora cinnamomi introduced to the Jarrah
(Eucalyptus marginata, Myrtaceae) forests of Western Australia caused widespread
destruction of Jarrah and dozens of other susceptible host species (82, 189, 223).
The native fungi Phellinus weirii and Heterobasidion annosum cause high mortal-
ity in conifer forests in western North America (83, 125, 181). The mortality caused
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 19
by all these pathogens is particularly notable because it affects large, established
trees, and death of dominant species can lead to dramatic shifts in forest structure.
Such effects are not new?there is some evidence for epidemic tree mortality in
the Late Triassic. A relatively thin, but widespread stratum in the Petrified Forest
National Park in Arizona shows that certain trees were damaged by decay similar
to present-day damage caused by H. annosum (72).
Root rots are not always lethal, but may reduce host growth or reproduction. In
British Columbia, Armillaria ostoyae reduced radial growth of Douglas-fir by up
to 60% relative to healthy trees (41). Similarly, H. annosum reduced mean annual
radial growth of loblolly pine (Pinus taeda) by 36% (18).
Floral Diseases
The most direct effect of disease on plant fecundity, short of killing the host, is
the attack of flowers or developing fruits, thereby preventing fruit production. Few
floral diseases have been studied in natural systems, but available data suggest
that floral diseases may have large effects on host fecundity. In the Appalachian
Mountains of the eastern United States, flower galls of flame azalea (Rhododen-
dron calendulaceum, Ericaceae) caused by Exobasidium vaccinii led to a 50%
reduction in flower number, as well as reduced fruit production (227). Similarly,
attack of flowers and ovaries of the tropical understory tree Faramea occiden-
talis (Rubiaceae) by the rust Aecidium farameae reduced fruit set by 75% (207).
The herb Plantago lanceolata (Plantaginaceae) infected with the floral pathogen
Fusarium moniliforme var. subglutinans were more likely to be male-sterile, and
had reduced seed production compared to healthy plants (12).
By far, however, the best known of floral diseases in wild plants is the sex-
ually transmitted, pollinator-vectored anther smut (Microbotryum violaceum) of
Silene spp. (Caryophyllaceae). The cost of infection to host fitness is great, since
the fungus replaces stamens and staminoids with spore-bearing structures (8). In-
fected populations produced lower densities of seedlings (58), but infection caused
significant host mortality only in years of mild winters (204). Differential suscepti-
bility to anther smut across host genotypes, and strong effects on host reproduction
combine to link numerical effects of disease on the population with evolutionary
responses through selection for resistant genotypes (17). [See excellent reviews in
the evolutionary ecology of this disease system by Jarosz, Alexander, and cowork-
ers (17, 140)].
In a bizarre interaction, systemic infection of Arabis spp. (Brassicaceae) by
the rust Puccinia monoica inhibits flowering and causes a radical morphological
change in the host plant, with clusters of infected leaves creating pseudoflowers of
similar appearance to unrelated co-occurring flowers (185). Pollinators attracted
to the pseudoflowers fertilize the rust. The rust blocks sexual reproduction in the
host plants and can also affect reproduction in nonhost neighbors. The increased
density of showy pseudoflowers can attract additional pollinator visitation locally,
but may reduce pollinator efficiency when pollinators spend time ?pollinating?
pseudoflowers (186).
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
20 GILBERT
Recent years have witnessed significant growth in the number of studies of
plant diseases in natural ecosystems. Nevertheless, the number of empty or nearly
empty cells in Table 1 indicates the need for studies on a diversity of pathogen
types and their effects on various stages in the host life cycle.
POPULATION DYNAMICS, SPECIES
INTERACTIONS, AND DIVERSITY
The past two decades have seen an expansion in our appreciation of the ecologi-
cal dynamics of plant diseases in natural communities. Three interrelated themes
have emerged as leading areas of ecological inquiry: (a) the importance of plant
diseases in density-dependent population dynamics, (b) the importance of com-
pensatory and cross-generational effects, and (c) the direct effects of pathogens on
shifting competitive interactions among plant species, effects on plant community
succession, and the maintenance of plant species diversity.
Plant Diseases and Density-Dependent Population Dynamics
Density dependence is important to all aspects of plant demography from seed
production and germination to plant growth and mortality. The mechanisms un-
derlying density-dependent responses include plant competition, seed-germination
inhibition, seed predation, competition for pollinators, herbivory, and of course,
diseases. Host density can in turn affect disease incidence and severity either
through direct effects on host-pathogen encounter rates or through indirect effects
on disease development by changing environmental factors (49). These effects
on diseases are created through (a) changes in the number of hosts available for
infection through space or time, (b) the distance between susceptible hosts and the
probability of transmission of pathogen propagules among plants, (c) effects of
competition on host vigor, (d ) frequency-dependent effects on vector behavior or
herbivore damage, (e) effects on the density and composition of associated plant
species in the community, and ( f ) the physical environment in which the pathogen
and host interact.
In their classic 1982 review, Burdon & Chilvers (49) evaluated studies on 46
host-pathogen combinations for patterns of density-dependent disease incidence.
Fungal diseases overwhelmingly showed positive density dependence. The few
cases of negative density dependence were associated either with short-term effects
where primary inoculum was limiting at high host densities (as for some soilborne
pathogens), or with heteroecious rusts where lower densities of one host allowed
the development of higher densities of the obligate alternate host plant. On the
other hand, vector-borne viral diseases usually showed negative density-dependent
incidence, either because infective vectors were limiting at high host densities or
because vector behavior changed with respect to crop density. Although most of
the examples reviewed were from managed settings, it is clear both that density-
dependent disease development can be an important factor in disease ecology
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 21
anywhere, and that natural history traits of the pathogen are key to the outcome.
Since 1982 there have been numerous publications on density-dependent disease
development in natural systems congruent with the patterns described in the review
(56, 58, 97, 121, 154) and few contrary examples (7, 62).
Counterweights to Numerical Effects
Disease can cause significant plant mortality, but it is only important to population
dynamics if the reduced number of plants leads to decreased seed production at the
population level. For example, local density of seedlings may be reduced through
damping-off diseases, but if the remaining healthy plants respond to the reduced in-
traspecific competition by increased growth and reproduction, these compensatory
responses can offset numerical losses (14, 16). In a field experiment, Pythium re-
duced both the number and average size of seedlings of the annual Kummerowia
stipulacea (Papilionaceae), but at maturity plant size was largest in the plots with
the highest initial disease incidence (16). The compensatory response of the sur-
vivors to reduced competition eliminated differences in seed production between
diseased and nondiseased plots. Similarly, Portulaca oleraceae (Portulacaceae)
compensated for disease loss when as much as 50% of a stand was infected with
Cucumber mosaic virus (94). Even strong effects of diseases on fitness of individ-
ual plants may not affect numerical population dynamics. Research on the ability
of diseases to regulate host population dynamics must focus not just on numerical
responses in one generation, but should incorporate the compensatory responses
to reduced intraspecific competition on neighborhood fitness.
Cross-generational effects may modulate the numerical response of subsequent
host generations. Diseased maternal plants may produce inferior seed, reducing the
potential fitness of the next generation of plants (139). In contrast, it is possible that
defenses induced in maternal plants may be transmitted to the progeny, increasing
resistance to disease in the next generation. This, in effect, allows the inheritance
of increased disease resistance without selection for resistant genotypes. Although
not yet explored in disease systems, such maternally transmitted resistance has
been shown in response to herbivores (4).
In summary, the past two decades of research have shown that although plant
diseases can and do have significant impacts on plant populations in natural ecosys-
tems, predicting the effects of a particular disease requires far more than deter-
mining the effects on an individual plant. Compensatory responses in surviving
plants, multigenerational variation in numerical responses, cross-generational in-
duced resistance and effects on offspring vigor, and the dynamics of selection for
more resistant host genotypes and more virulent pathogens all contribute to the
effects of plant pathogens on the population biology of their hosts.
Competition, Succession, and Plant Diversity
Competitive interactions and host-pathogen dynamics may be strongly inter-
dependent. Stress from competition for scarce resources can alter susceptibility
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
22 GILBERT
to disease so that bottom-up competitive interactions lead to increased top-down
effects of pathogens. Similarly, differential susceptibility or tolerance to diseases
can lead to different competitive interactions in the presence or absence of disease.
Alexander & Holt (14) provide an excellent recent review on interactions between
diseases and plant competition.
DISEASE AND COMPETITION Competition may stress plants and make them less
tolerant to infection. The rust Puccinia recondita has a large impact on plant growth
of Impatiens capensis (Balsaminaceae) at high natural densities but not at thinned
densities (154). Thus, competition reduced host tolerance to rust infection.
Although differences in shade tolerance are usually attributed to physiological
differences, the presence of disease may create differences among individuals in
their abilities to compete for light. Survival of stump sprouts of Salix viminalis
(Salicaceae) is directly related to their height and thus ability to compete for light.
Infection by the rust Melampsora epitea greatly reduced shoot growth, and reduced
the plant?s competitive ability, ultimately killing infected stumps (216).
Specialist pathogens like rusts can affect the competitive interactions between
susceptible and nonsusceptible plant species or genotypes. Paul & Ayres found
that inoculation of groundsel (Senecio vulgaris, Asteraceae) with the rust Puccinia
lagenophorae significantly reduced groundsel growth, and reduced the competitive
ability of groundsel with both lettuce (Lactuca sativa) (175) and the co-occurring
weed petty spurge (Euphorbia peplus, Euphorbiaceae) (174). Rust infection di-
rectly affected the competitive ability of the host beyond the effects of infection
on plant growth. Similar effects of rust infection have been found for intraspecific
competition, including between infected and noninfected groundsel (175) and
between susceptible and resistant skeletonweed (Chondrilla juncea, Asteraceae)
(51). Research of this sort is sorely needed outside of agroecosystems.
APPARENT COMPETITION Apparent competition in plants describes the situation
where two species appear to be in competition for some limiting resource, but the
negative effect is actually due to an indirect interaction via an herbivore or pathogen
that attacks them both (126). Generalist pathogens may play an unrecognized role
in interspecific interactions in natural communities, mimicking or modifying com-
petitive interactions (14). An intriguing example of this exists between the annual
legume Chamaecrista fasciculata (Caesalpiniaceae) and the large, dominant peren-
nial grass Andropogon gerardii (Poaceae), which coexist in the Kansas tallgrass
prairies. Holah & Alexander (123) found that both species grew more poorly in
soil from the root zone of Chamaecrista than in Andropogon soil, and that the ef-
fect was associated with fungi found uniquely on Chamaecrista roots. Pathogenic
fungi ?cultured? on Chamaecrista roots shift the competitive outcome against
the dominant perennial grass, facilitating coexistence of the two plant species.
Understanding the host ranges of pathogens within a local plant assemblage and
the possible adaptation by plants to actively culture pathogens that increase their
competitive ability is a largely unexplored, but potentially fruitful field.
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 23
SOIL FEEDBACKS In other cases, host-specific microorganisms cultured in the rhi-
zosphere of a host may build up over time, with increasingly negative biological
effects on the host. Bever and coworkers (39, 40) developed the concept of feed-
back as a framework for the role of soil microbial communities in regulating plant
population dynamics and plant diversity. Negative feedbacks?the buildup of detri-
mental pathogens over time?are widely recognized in agricultural settings, where
crop rotation is a standard practice to escape the buildup of disease pressure. Crop
rotation, when integrated over time, effectively increases the plant species diversity
in a particular field. In natural communities, negative feedbacks could help main-
tain local plant diversity by preventing individual species from increasing to com-
plete dominance. Bever (39) looked at the effects of soil microbial communities in
soils ?cultured? by one of four old-field species: Krigia dandelion (Asteraceae) and
three grasses (Danthonia, Anthoxanthum, and Panicum). Krigia mortality doubled,
and grass growth rate decreased when grown in their own soil compared to being
grown in soil from any other species. For some species but not others, the effects
were associated with accumulation of Pythium in the soil, indicating host-specific
effects of the cultured pathogens (163). These studies suggest that the local devel-
opment of host-specific soilborne pathogens may establish a temporally dynamic
mosaic of deleterious ?home? and benign ?away? sites, and contribute to the main-
tenance of host diversity in grasslands. Such effects on plant diversity could be
quite strong. Based on spatial lottery models, as long as there is no intrinsic cor-
relation between host susceptibility and competitive ability, plant diversity should
increase monotonically with increased addition of host-specific enemies (169).
Negative feedback loops may be important in many kinds of natural systems,
but traits of pathogens and hosts that favor longer-term spatial structure are likely
to be key elements. In particular, pathogen dispersal distances need to be more
limited than host dispersal distances; thus many soilborne pathogens may be more
likely to be involved in negative feedbacks than would wind-dispersed pathogens.
Similarly, sexually reproducing plants with mobile seeds are more likely to escape
from the fouled ?home? sites than would primarily clonal hosts. Strongly dynamic
systems with high levels of disturbances may not maintain enough spatial structure
for feedbacks to develop.
DISEASES IN PLANT COMMUNITY SUCCESSION Diseases can cause changes in plant
community composition through the dynamic patchworks of Bever?s old-field
communities, or through a predictable succession of seres. In Europe, marram
grass (Ammophila arenaria, Poaceae) colonizes and thrives in the wind-blown
sands of coastal foredunes, but declines once the dunes are stabilized. A complex
of pathogenic soil fungi and nematodes develops in the Ammophila rhizosphere
through negative biotic feedback (79?81, 210, 212), which causes the degeneration
of the dominant Ammophila, and allows the resistant species, sand fescue (Festuca
rubra ssp. arenaria, Poaceae), to ultimately dominate the stabilized dunes (211). In
the northeastern United States, Ammophila breviligulata dominates the high beach
zone and is associated with high densities of growth-reducing parasitic nematodes
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
24 GILBERT
(151, 152), but greenhouse studies suggest the development of mycorrhizal net-
works may be more important than negative feedback for plant species succession
in this system (143, 151). Further, coordinated work on mycorrhizae, pathogenic
fungi, and pathogenic nematodes in these parallel systems would be valuable in
understanding variation in microbial control of plant successional processes.
Forest succession can also be driven by differential susceptibility to pathogens.
In particular, dramatic changes in species composition can be driven by epidemics
from introduced pathogens. Loss of American chestnut to chestnut blight in the
1910s to 1940s led to a peak in recruitment of Quercus rubra, Tsuga canadensis,
and a suite of other species that now dominate forests in the eastern United States
(2, 3, 5, 42, 93). Native pathogens can have similarly important effects on forest
succession. Phellinus weirii, the cause of laminated root rot, affects most conifer
species but with different degrees of severity. Phellinus spreads radially forming
?infection centers,? removing the overstory of highly susceptible Douglas-fir and
mountain hemlock (Tsuga mertensiana) in its wake (82, 125). The removal of the
overstory has major effects on the plant community, with local resistant species
then becoming dominant in the infection centers (124). Hansen & Goheen (118)
provide an excellent recent review of the importance of Phellinus in driving forest
structure and successional processes in western conifer forests.
These two systems, soilborne pathogens in beach grasses and Phellinus in
conifer forests, demonstrate the potential for native pathogens to drive succession in
low-diversity plant communities, and to affect the spatial distribution of susceptible
host species and their resistant competitors. In high-diversity plant communities,
such as tropical rain forests, the effects of disease on the spatial patterns and
coexistence of plant species may be more complex, and are brought together under
the framework of the Janzen-Connell hypothesis.
THE JANZEN-CONNELL HYPOTHESIS The puzzle of how so many plant species can
coexist in species-rich ecosystems like tropical rain forests has been the focus of
a tremendous body of literature [see reviews in (63, 111, 229)]. Given a suite of
species competing for a single resource, the competitive exclusion principle pre-
dicts that only one species, the competitive dominant, will persist. Processes that
increase negative interactions within a species relative to negative interactions
between species break the trajectory toward low diversity caused by interspe-
cific competition, and are essential to stabilize species coexistence (63). Specialist
natural enemies, including plant pathogens, herbivores, and seed predators, can
stabilize coexistence in tropical forests by limiting populations of each species in-
dependently. Janzen (136) and Connell (71) brought into the ecological limelight
a model first proposed by Gillett (109) in 1962, now known as the Janzen-Connell
hypothesis. There was considerable empirical support for the Janzen-Connell hy-
pothesis in tropical forests as early as 1984 (64), and it is still the leading conceptual
framework for understanding how plant diseases affect forest community structure
and diversity.
The Janzen-Connell hypothesis begins with the recognition that in natural
forests, most seeds do not disperse far, but instead form a ?seed shadow,? so that
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 25
the seedlings in a forest are highly clumped around their mother tree. A parent tree
could act as a reservoir for specialist pathogens or pests that are then transmitted
to nearby offspring, and high densities of seedlings near to the parent tree would
result in greater pressure from specialist natural enemies due to density-dependent
attack. With proportionally greater mortality of seedlings close to the parent tree
than at greater distances, intraspecific clumping should decrease through time,
so that the spatial distribution of mature trees should be less clumped than ex-
pected through random mortality throughout the original seedling distribution.
The Janzen-Connell hypothesis also predicts that distance- and density-
dependent mortality caused by specialist natural enemies should reduce interspe-
cific competition. Seeds of nonsusceptible plant species that dispersed into the seed
shadow of a susceptible species would then be at a competitive advantage over
the more numerous progeny of the susceptible species. In a diverse forest with
associated specialist natural enemies, individual trees would then have stronger
negative effects on survival of nearby conspecifics than on heterospecifics. Rare
species would be more likely than common species to find pathogen-free space,
and thus enjoy a competitive advantage. Intraspecific apparent competition would
then stabilize species coexistence and act to maintain tree diversity.
According to the mechanisms outlined in the Janzen-Connell hypothesis, for
plant diseases to favor higher host species diversity, (a) the pathogens, if not host
specialists, must have differential effects on different local host species; (b) there
must be a diversity of pathogens with different host preferences; and (c) both hosts
and pathogens must be dispersal limited (ubiquitous pathogens would not produce
distance-dependent disease gradients except through distance-dependent environ-
mental interactions). Specialist pathogens that are most likely to cause Janzen-
Connell?like effects are those that (a) infect both mature and juvenile plants, but
cause little damage to trees and have severe effects on seeds, seedlings, or saplings;
(b) have strong density-dependent infection or impacts on hosts; or (c) have long-
lived soilborne propagules that permit the local buildup of pathogen inoculum.
JANZEN-CONNELL EFFECTS IN TROPICAL FORESTS There is strong evidence that
plant diseases affect rain forest diversity, dynamics, and spatial structure, as pre-
dicted by the Janzen-Connell model. First, there is overwhelming support for the
importance of density-dependent mortality in shaping the diversity and dynamics
of tropical forests (e.g., 119, 219, 225). The first empirical tests of the role of plant
diseases for the Janzen-Connell hypothesis were by Augspurger and coworkers,
as described above under Seedling Diseases (25, 27, 28). In a series of studies on
BCI in Panama, they showed that damping-off diseases were important causes of
seedling mortality for a range of tree species, but that susceptibility varied widely
across host species (28). Seedlings were most likely to escape damping-off at
greater distances from the parent tree when close to the parent, and the proportion
of seedlings dying was greater in areas with higher seedling density (27, 142). For
Platypodium elegans (Papilionaceae), the median distance of offspring to the par-
ent tree increased from 15 to 31 meters from seed dispersal to surviving saplings,
greatly reducing species clumping as compared to what would be expected from
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
26 GILBERT
random mortality (26). Thus, damping-off of seedlings in the lowland moist trop-
ical forest of BCI shows all the predicted elements of the Janzen-Connell mecha-
nism for the maintenance of diversity among tree species: strong distance-to-adult
and density-dependent mortality with differential effects across host species, caus-
ing a temporal reduction in clumping of individual species. Direct effects on local
maintenance of species diversity were not investigated. Recently, however, Harms
et al. (119) followed seed rain and seedling recruitment at 200 sites on BCI, and
found strong negative density-dependent effects on 53 focal species, which led to a
significant increase in local species diversity in the transition from seeds to seedling
recruits over four years. The studies by Augspurger and coworkers strongly sug-
gest that a large proportion of this seedling mortality could have been caused by
damping-off pathogens.
Other diseases are also important for Janzen-Connell?like mortality patterns in
tropical forests. Also on BCI, the canker-fungus Botryosphaeria dothidea reduced
growth rate and increased mortality of seedlings and small saplings of Tetragas-
tris panamensis (Burseraceae), and canker formation was significantly greater on
seedlings under the canopy of parent trees than at further distances (102). Another
canker disease of various species of Ocotea and Nectandra (Lauraceae) caused
by Phytophthora sp. (105, 106) showed that Janzen-Connell effects may continue
to be important determinants of the spatial distribution of mortality of susceptible
hosts decades old. Mortality of juvenile Ocotea whitei (1?30 cm dbh) and reduc-
tion in clumping of juveniles over an 8-year period was consistent with spatial
patterns of the canker disease. In a study of 13,000 seedlings in two cohorts of
O. whitei at the same site, mortality was similarly density dependent and caused a
significant shift away from parent trees over time (104). Particularly for forest trees
like O. whitei that reproduce on a supra-annual basis so that seedling densities vary
greatly among years (104), disease-driven reductions in host density may lead to
reductions in disease pressures, setting up coupled density-dependent cycles for
both the host and pathogen populations.
The Janzen-Connell theory predicts that adult individuals should be less clumped
than expected from random mortality of the original seed distribution, but not
necessarily randomly or regularly distributed in space. Although in some cases
distance-dependent mortality can cause a shift from an initially aggregated dis-
tribution of seedlings to a random distribution of adults (36, 37), many tropical
species retain a significantly clumped distribution at maturity (92, 105, 132). This
may be due either to the overwhelming effect of seed shadows or to the effects
of environmental heterogeneity on plant survival. Regardless, this pattern is not
contrary to the predictions of the Janzen-Connell hypothesis. True tests of the
Janzen-Connell hypothesis do not rely on static analyses of the spatial distribution
of entire species or even juveniles with respect to distance from conspecific adults,
but rather look at the changes in degree of clumping over time compared to the
initial distribution of seeds and seedlings.
JANZEN-CONNELL EFFECTS IN TEMPERATE FORESTS The Janzen-Connell hypoth-
esis was developed as an explanation for maintenance of high diversity in tropical
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 27
forests and most tests of the model have been in those ecosystems, but it is now
clear that this phenomenon can also be important in temperate forests. Packer &
Clay (170) provide the first complete test of the Janzen-Connell hypothesis in a
temperate forest, as well as the first direct evidence for the local maintenance of
species diversity. In Indiana, black cherry seedlings suffered 100% mortality from
damping-off caused by Pythium spp. in soil from beneath parent trees, whereas
mortality was low for competing tree species in the same soil. Disease-induced
mortality decreased sharply with distance from the parent trees, causing a signifi-
cant shift in the spatial distribution of seedlings away from the mother tree. This
study clearly shows that the negative feedback of increased soilborne pathogens
beneath the parent canopy puts less susceptible seedlings of locally rare species
at a competitive advantage over the parent?s much more common offspring, thus
maintaining local plant diversity in a temperate forest.
Although the Janzen-Connell effect clearly can be important in temperate
forests, several studies suggest it may be less common there than in tropical forests
(122, 128?130). At least three factors could contribute to the difference between
forest types. First, lower diversity and corresponding higher density of individual
species in temperate forests may lead to a strong overlap in seed and seedling
shadows from different mothers of the same species, negating the advantage of
dispersal (128). Second, the Janzen-Connell effect depends on differential suscep-
tibility of hosts to a particular pathogen. In some temperate forests, particularly
in conifer forests where most individuals may belong to just a few genera in one
or two families, many pathogens may be capable of infecting a majority of in-
dividuals in the forest. Nevertheless, the expanding foci of infection, differential
effects on local tree species, and greater survival of less susceptible conifer species
in the wake of diseases caused by Phellinus wierii (118, 124, 125) and the Het-
erobasidion annosum complex (95, 181) could be considered special cases of the
Janzen-Connell effect. Finally, for Janzen-Connell effects to play an important
role in tree diversity and distributions, disease effects cannot be overshadowed by
large catastrophic disturbances such as fire or windstorms that can cause catas-
trophic stand replacement (208). Fire disturbance is much more frequent in forests
of western North America (44, 197, 198) than in the wet tropics (209), but even in
tropical systems diseases may be less likely to play a key role in those areas that
are subject to frequent hurricanes, landslides, or other disturbances (24).
HOST DENSITY AND PATHOGEN SPECIFICITY The Janzen-Connell effect requires
a degree of host specificity, but for a specialist fungus to remain a viable part of a
forest ecosystem a suitable host must not only be present, but present at sufficiently
high density to ensure that the fungus can colonize new host individuals. Dispersal-
limited, host-specialized pathogens would not likely dominate in a high-diversity
forest where most host species will be at low density; such forests should be
dominated by fungi with broad local host ranges. In contrast, a low-diversity
forest, with relatively high densities of each of the component species, should
favor the development and maintenance of a fungal community dominated by
host specialists, as long as the available hosts are not all too closely related to
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
28 GILBERT
facilitate host specialization. The effects of host community diversity and structure
on fungal diversity and host specificity are largely unexplored, except for recent
studies of the diversity and host specificity of wood-decay polypore fungi in Central
American tropical forests. In the high-diversity moist tropical forests of BCI in
Panama (103) none of 43 species showed significant host preferences, and in the
medium-diversity dry tropical forest of Costa Rica (150) only 3 of 32 species
showed host specificity. The fungal assemblages in these forests included many
rare species, but the common fungi were generalists. In contrast, in a naturally low-
diversity mangrove forest in Panama (only 3 host tree species present, each from a
different family), the polypore fungal community is strongly dominated by just a
few common species, each highly host specific (107). Considerable work has been
done in agricultural settings on how the structure of the plant assemblage affects the
development of disease (166). Additional research on the effects of host density,
diversity, and spatial structure on the diversity and specificity of pathogenic fungi
is clearly needed to understand the dynamic effects of pathogens on competitive
interactions in natural plant communities.
RAPID EVOLUTION AND LOCAL ADAPTATION
The Red Queen Hypothesis
The minority advantage afforded rare, resistant species in the Janzen-Connell
model, or more generally by disease-driven apparent competition, is of course
not static. As a rare host becomes more common, its own pathogens may increase
in abundance or virulence through density-dependent feedback and frequency-
dependent selection. Additionally, pathogens from one host may evolve to acquire
new hosts (96, 180, 187). Natural selection for increased virulence on hosts that re-
cently became common occurs each time a resistant agronomic cultivar is defeated
by a new race of pathogen. Wild hosts can also evolve resistance to pathogens, lead-
ing to the coevolutionary arms race known as the Red Queen hypothesis (69, 86).
Rapid evolution of interactions between hosts and microorganisms are common
in natural systems as well (199). In wild plant-pathogen interactions, hosts should
evolve toward increased levels of resistance, and pathogens should evolve toward
optimum levels of virulence to maximize their own fitness (89, 149). Rates of
evolutionary change may not be equal between hosts and pathogens, since differ-
ences between partners in rates of molecular change in coevolutionary processes
are correlated with generation time (115). Short-lived plants may have generation
times similar to their pathogens, and both numerical and evolutionary responses
will be important in the dynamics of host-pathogen interactions. For instance, the
introduction of the rust Puccinia chondillina for biological control of the inva-
sive skeleton weed Chondrilla juncea (Asteraceae) in Australia was followed by a
dramatic shift within the host population from dominance by the susceptible host
biotype to resistant biotypes, due to a reduction in the otherwise superior compet-
itive ability of the susceptible biotype (50, 51). In forest systems, the generation
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 29
time of most pathogens will be much faster than that of the host species, and at
least in the short term (a few tree generations), pathogens may pass through many
generations under strong selection for increased virulence before resistant hosts
reach reproductive maturity. Understanding the interplay between numerical and
evolutionary dynamics in natural plant-pathogen systems is a key challenge for
the coming decade (17, 22).
The ability of hosts to respond evolutionarily to selection pressures from patho-
gens will depend not only on generation time, but also on host genetic variability
and rates of outcrossing. Burdon et al. (55) compared two distinct wild metapop-
ulations of Linum marginale infected with the rust Melampsora lini; one host
metapopulation was strongly inbred and the other was strongly outcrossed. They
found that the outcrossed populations showed consistently higher overall levels
of resistance, numbers of resistance phenotypes, and levels of polymorphism for
specific virulence factors in the pathogen. Many systemic fungal parasites castrate
their host plants, allowing only clonal reproduction or autogamy and avoiding Red
Queen challenges of sexual recombination in the host (68). Both intrinsic vari-
ability in host outcrossing rates and pathogen manipulation of the host may affect
the response to selection in evolutionary arms races. Additionally, metapopulation
structure, including effects of genetic drift and gene flow, may be equally important
in determining the dynamics of reciprocal selection (200).
Local Adaptation and the Scale of Dispersal
The spatial scale of dispersal is central to the ecological and evolutionary dynam-
ics of plant-pathogen interactions (203). The dispersal of pathogens to susceptible
hosts, the dispersal of pollen for outcrossing in plants, and the dispersal of seeds
away from areas of high disease pressure, all determine the spatial scale of eco-
logical interactions including disease persistence, host distribution, and epidemic
development, as well as the relative isolation and genetic differentiation of locally
adapting pathogens or plants. Below I briefly examine recent research on the role
of dispersal in ecological and evolutionary dynamics of host-pathogen systems.
PATHOGEN DISPERSAL GRADIENTS Although there is a long tradition of study of
the spatial spread of plant diseases in agronomic and forestry settings, our detailed
knowledge of the spatial scale of pathogen dispersal is limited to a fairly narrow
range of plant pathogens, and an even smaller set of pathogens important in natu-
ral systems. There have been several excellent reviews of the extensive literature
on pathogen dispersal through the air currents, by rain splash, through the soil or
subsurface water, and by vectors (31, 32, 57, 91), with nearly all examples coming
from agronomic settings. Power law and exponential models often provide rea-
sonably good fits to observed dispersal gradients. Fitt et al. (91) fit 325 data sets
of spore or pollen deposition or disease gradients to the two models and found
that both models worked well, with exponential models working somewhat bet-
ter for splash-dispersed systems and power law slightly better for wind-dispersed
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
30 GILBERT
systems. They note that the exponential model has the advantage of providing a
?half distance? measure where the observed deposition decreases by half with a
constant distance increment, and show that half distances are greater for airborne
than for splash-dispersed or soilborne diseases. However, these two models and
all current empirical methods do not adequately measure or describe the tail of
the dispersal curve; improvement both of measurement abilities and mathematical
descriptors for dispersal tails is an important current area in spatial ecology.
Empirical measurement of dispersal gradients for a broad range of pathogen
types in natural ecosystems, and appropriate mathematical description of disper-
sal gradients, would provide important insights into the scales of host-pathogen
interactions and the potential range of types of interactions in natural systems.
Although there are few direct studies of pathogen dispersal in natural ecosys-
tems, the available literature suggests that some fungal spores may travel hundreds
of kilometers (176, 201). However, damage from UV irradiation (173, 183, 184)
and susceptibility to desiccation (206) limit regular long-distance dispersal of
most pathogens, with dispersal usually contained within tens to hundreds of me-
ters for airborne and less than one meter for splash-dispersed pathogens. Some
pathogens may move longer distances through ?stepping stones? of infection
and reproduction, achieving further dispersal in a series of shorter steps (33,
158, 176).
SPATIAL PATTERNS Our understanding of the role of the spatial scale of dispersal
for structuring natural microbial communities and populations is growing rapidly.
In a wet tropical forest, Lodge & Cantrell (155) found no significant resemblance
between litter basidiomycete and microfungal communities in small patches sepa-
rated by more than 100 m. Arnold et al. (23) found significant differences in fungal
endophyte communities within a host species across sites separated by a few hun-
dred meters in a moist tropical forest. Both studies suggest that distance has a
strong effect on determining fungal community composition, most likely through
the independent effects of dispersal limitation and environmental heterogeneity
on component species. Traditional phytosociological studies of the spatial scale of
turnover in fungal communities must be complemented with studies of the spatial
distribution of individual species, populations, and genotypes.
Individual fungal species often show strong spatial patterns, reflecting both
environmental heterogeneity and dispersal limitation. Fungal infection is often
aggregrated within individual hosts. Infection by the common endophyte Discula
quercina was spatially aggregated among leaves within a individual host Quercus
garryana (Fagaceae) (226). Similarly, the tropical leaf-colonizing fungus Scole-
copeltidium mayteni on Trichilia spp. (Burseraceae) was highly correlated among
leaflets within a leaf, but showed high variation among leaves (108). Carroll (59)
reviewed endophyte distributions and suggests that within trees the height within
the crown is the most important determinant of spatial structure, whereas within
individual branches the age of the leaf is most important. Although these stud-
ies suggest that propagule dispersal may limit where fungi can infect host plants,
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 31
environmental conditions often ultimately control whether infection takes place
or disease develops (61, 108, 138, 230). Unfortunately, however, except for impor-
tant pathogens of widespread crops, our understanding of the effects of dispersal
and environment on the large-scale biogeographic distribution of pathogens re-
mains poor. Collaborative work between pathologists working in separate regions
or habitats could help further our understanding in the area.
Within a fungal species, genotypes are also often spatially aggregated. Isolates
of the chestnut blight fungus Cryphonectria parasitica showed strong genetic
structure, with isolates of the same genotype closer together than expected at
random (162). Genets of Armillaria spp. in nine forest sites in New York were
dominated by very small genets (one sample point) with a low frequency of larger
genets up to 44 m long (228). A number of studies show significant genetic struc-
turing of fungi in natural ecosystems over large geographic distances (e.g., 134,
135, 168). Understanding how different life histories of fungi affect the scale of
genetic structure will provide much insight into the pattern and likelihood of local
adaptation in plant-pathogen interactions.
METAPOPULATIONS Dispersal limitation will also strongly control the evolution-
ary and ecological dynamics of host-pathogen metapopulations by coupling the
host and pathogen populations of connected demes and decoupling the dynam-
ics of more distant ones. Ericson et al. (87) found strong variation among years
and among islands in the dynamics of infection of Valeriana salina by Uromyces
valerianae in 30 discrete populations in an archipelago in Sweden. Pathogen pop-
ulations commonly went extinct and recolonized healthy host populations, and
disease was more likely to spread between neighboring populations. Similarly,
wild flax (Linum marginale) infected with the rust Melampsora lini showed asyn-
chronous disease dynamics and large differences in diversity and frequency of
virulence and resistance phenotypes among demes in a metapopulation, but demes
closer together had more similar resistance phenotypes than more distant demes
(54). Burdon & Thrall (53) have provided a good review of the role of spatial
structure for ecological and evolutionary dynamics of plant-pathogen interactions,
with particular reference to its effects on coevolutionary dynamics both within sin-
gle populations and across a metapopulation. Plant-pathogen systems are among
the few where both numerical and genetic dynamics have been studied in a spa-
tial context, and where the genes under selection are unambiguous. Research on
the spatial context of the evolutionary ecology of plant-pathogen systems is im-
portant not only to disease ecologists, but to the field of evolutionary ecology in
general.
LOCAL ADAPTATION If the evolutionary dynamics of plant-pathogen interactions
are spatially heterogeneous (47, 199), then the dynamics described for interspecific
interactions in the Janzen-Connell hypothesis and other forms of disease-driven
host dynamics can be extended to ecological interactions between genotypes of
the same host species. Rare genotypes should be at an advantage if microbial
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
32 GILBERT
feedback processes are driven by the locally dominant host genotypes (153). A
pathogen that is reliably ?cultured? for generations on the leaves or roots of a
single mother tree may adapt to that host genotype and increase in virulence over
time; consequently, it may be more virulent on offspring of that tree than on rare
seedlings of a more distant mother of a different genotype (101, 179). Isolated host
patches with independent local coevolutionary trajectories should lead to different
types of disease interactions when rare seeds disperse from one patch to another.
There are numerous examples of local adaptation in agricultural systems (e.g., 6),
but few published examples from natural systems. Two genetically isolated, but
nearby (50 m) populations of the annual, selfing legume Amphicarpaea bracteata
(Papilionaceae) differed in susceptibility to the specialist Synchytrium decipiens
(171), with one population completely resistant to the pathogen and the other
showing almost no resistance (172). On the other hand, Davelos et al. (78) did
reciprocal transplants of the clonal prairie grass Spartina pectinata among patches
with low and high incidence of species of Puccinia rusts. There was no effect of
rare-genotype advantage in transplants at a 4.5-km scale, although at 120 km there
was an indication of lower disease on transplanted hosts. We still have a limited
understanding of the scale of local adaptation of pathogens in natural ecosystems.
Reciprocal transplant and common garden experiments will provide significant
advances in our understanding of the scale of local adaptation in plant-pathogen
systems.
CONCLUSIONS
We must bring a solid understanding of the life histories of both pathogens and
plant hosts to our efforts to appreciate the effects that plant disease can have in
natural ecosystems. Further empirical work should focus on improving our under-
standing of pathogen dispersal and host specificity in natural communities, and
on the dynamics and spatial scale of rapid evolutionary changes in plant-pathogen
symbioses. Just a few decades ago, documenting that diseases were parts of nat-
ural systems was exciting news, and our expectations of their effects in natural
plant communities were simple extensions of our understanding of agroecosys-
tems. Now we are quickly developing a library of studies on wild plants set within
an integrated framework of evolutionary ecology. This empirical work both fuels
and builds on an increasingly robust body of theory, underscoring the fundamental
place of diseases in ecology and evolution.
ACKNOWLEDGMENTS
I thank Barbara Ayala, Katrina Dlugosch, Neil Gilbert, Cynthia Hays, Matt
Kauffman, Ingrid Parker, Doug Plante, and Yuri Springer for helpful discussions
and comments on the manuscript. Research related to this review was supported
by a grant from the National Science Foundation (DEB-9806517).
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 33
The Annual Review of Phytopathology is online at http://phyto.annualreviews.org
LITERATURE CITED
1. Abrams MD, McCay DM. 1996. Vege-
tation-site relationships of witness trees
(1780?1856) in the presettlement forests
of eastern West Virginia. Can. J. For. Res.
26:217?24
2. Abrams MD, Orwig DA, Dockry MJ.
1997. Dendroecology and successional
status of two contrasting old-growth oak
forests in the Blue Ridge Mountains,
U.S.A. Can. J. For. Res. 27:994?1002
3. Agrawal A. 1995. Use of dendrochrono-
logical methods to estimate an ecological
impact date of the chestnut blight. Va. J.
Sci. 46:41?47
4. Agrawal A, Laforsch C, Tollrian R. 1999.
Transgenerational induction of defences
in animals and plants. Nature 401:61?63
5. Agrawal A, Stephenson SL. 1995. Recent
successional changes in a former chest-
nut-dominated forest in Southwestern
Virginia. Castanea 60:107?13
6. Ahmed HU, Mundt CC, Coakley SM.
1995. Host-pathogen relationship of ge-
ographically diverse isolates of Septoria
tritici and wheat cultivars. Plant Pathol.
44:838?47
7. Alexander HM. 1984. Spatial patterns of
disease induced by Fusarium moniliforme
var. subglutinans in a population of Plan-
tago lanceotata. Oecologia 62:141?43
8. Alexander HM. 1987. Pollination limita-
tion in a population of Silene alba infected
by the anther-smut fungus, Ustilago vio-
lacea. J. Ecol. 75:771?80
9. Alexander HM. 1989. An experimental
field study of anther-smut disease of Si-
lene alba caused by Ustilago violacea:
genotypic variation and disease incidence.
Evolution 43:835?47
10. Alexander HM. 1992. Fungal pathogens
and the structure of plant populations and
communities. See Ref. 58a, pp. 481?97
11. Alexander HM, Antonovics J. 1995.
Spread of anther-smut disease (Ustilago
violacea) and character correlations in a
genetically variable experimental popu-
lation of Silene alba. J. Ecol. 83:783?
94
12. Alexander HM, Antonovics J, Rausher
MD. 1985. Relationship of phenotypic
and genetic variation in Plantago lanceo-
lata to disease caused by Fusarium moni-
liforme var. subglutinans. Oecologia 65:
89?93
13. Alexander HM, Burdon JJ. 1984. The ef-
fect of disease induced by Albugo can-
dida (white rust) and Peronospora para-
sitica (downy mildew) on the survival and
reproduction of Capsella bursa-pastoris
(shepherds purse). Oecologia 64:314?
18
14. Alexander HM, Holt RD. 1998. The in-
teraction between plant competition and
disease. Perspect. Plant Ecol. Evol. Syst.
1:206?20
15. Alexander HM, Maltby A. 1990. Anther-
smut infection of Silene alba caused by
Ustilago violacea: factors determining
fungal reproduction. Oecologia 84:249?
53
16. Alexander HM, Mihail JD. 2000. Seed-
ling disease in an annual legume: conse-
quences for seedling mortality, plant size,
and population seed production. Oecolo-
gia 122:346?53
17. Alexander HM, Thrall PH, Antonovics J,
Jarosz AM, Oudemans PV. 1996. Popula-
tions dynamics and genetics of plant dis-
ease: a case study of anther-smut disease.
Ecology 77:990?96
18. Alexander SA, Skelly JM, Webb RS.
1981. Effects of Heterobasidion annosum
on radial growth in southern pine beetle-
infested loblolly pine. Phytopathology 71:
479?81
19. Alfaro RI, Bloomberg WJ, Smith RB,
Thomson AJ. 1985. Epidemiology of
dwarf mistletoe [Arceuthobium tsugense]
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
34 GILBERT
in Western hemlock [Tsuga heterophylla]
stands in south coastal British Columbia.
Can. J. For. Res. 15:909?13
20. Anagnostakis SL. 1987. Chestnut blight
the classical problem of an introduced
pathogen. Mycologia 79:23?37
21. Anderson RC, Gardner DE, Daehler CC,
Meinzer FC. 2001. Dieback of Acacia
koa in Hawaii: ecological and patholog-
ical characteristics of affected stands. For.
Ecol. Manag. 5576:1?14
22. Antonovics J. 1994. The interplay of nu-
merical and gene-frequency dynamics in
host-pathogen systems. In Ecological Ge-
netics, ed. L Real, pp. 129?45. Princeton,
NJ: Princeton Univ. Press
23. Arnold AE, Maynard Z, Gilbert GS, Coley
PD, Kursar TA. 2000. Are tropical fungal
endophytes hyperdiverse? Ecol. Lett. 3:
267?74
24. Attiwill PM. 1994. The disturbance of
forest ecosystems: the ecological basis
for conservation management. For. Ecol.
Manag. 63:247?300
25. Augspurger CK, Kelly CK. 1984. Path-
ogen mortality of tropical tree seedlings:
experimental studies of the effects of
dispersal distance, seedling density, and
light conditions. Oecologia 61:211?17
26. Augspurger CK. 1983. Offspring recruit-
ment around tropical trees?changes in
cohort distance with time. Oikos 40:189?
96
27. Augspurger CK. 1983. Seed dispersal of
the tropical tree, Platypodium elegans,
and the escape of its seedlings from fungal
pathogens. J. Ecol. 71:759?71
28. Augspurger CK. 1984. Seedling survival
of tropical tree species?interactions of
dispersal distance, light-gaps, and patho-
gens. Ecology 65:1705?12
29. Augspurger CK. 1988. Impact of patho-
gens on natural plant populations. In Plant
Population Ecology, ed. AJ Davy, MJ
Hutchings, AR Watkinson, pp. 413?33.
Oxford: Blackwell Sci. Publ.
30. Augspurger CK. 1990. Spatial patterns of
damping-off disease during seedling re-
cruitment in tropical forests. See Ref. 52,
pp. 131?44
31. Aylor DE. 1986. A framework for examin-
ing inter-regional aerial transport of fun-
gal spores. Agric. For. Meteor. 38:263?
88
32. Aylor DE. 1990. The role of intermittent
wind in the dispersal of fungal pathogens.
Annu. Rev. Phytopathol. 28:73?92
33. Aylor DE, Taylor GS, Raynor GS. 1982.
Long-range transport of Tobacco Blue
Mold spores. Agric. Meteor. 27:217?32
34. Baker FA, French DW. 1991. Radial en-
largement of mortality centers caused
by Arceuthobium pusillum Peck in black
spruce stands. For. Sci. 37:364?67
35. Barone JA. 1998. Host-specificity of foli-
vorous insects in a moist tropical forest. J.
Anim. Ecol. 67:400?9
36. Barot S, Gignoux J, Menaut JC. 1999. De-
mography of a savanna palm tree: predic-
tions from comprehensive spatial pattern
analyses. Ecology 80:1987?2005
37. Barot S, Gignoux J, Menaut JC. 1999.
Seed shadows, survival and recruitment:
how simple mechanisms lead to dynamics
of population recruitment curves. Oikos
86:320?30
38. Benitez-Malvido J, Garc??a-Guzman G,
Kossmann-Ferraz ID. 1999. Leaf-fungal
incidence and herbivory on tree seedlings
in tropical rainforest fragments: an exper-
imental study. Biol. Cons. 91:143?50
39. Bever JD. 1994. Feedback between plants
and their soil communities in an old field
community. Ecology 75:1965?77
40. Bever JD, Westover KM, Antonovics J.
1997. Incorporating the soil community
into plant population dynamics: the util-
ity of the feedback approach. J. Ecol. 85:
561?73
41. Bloomberg WJ, Morrison DJ. 1989. Rela-
tionship of growth reduction in Douglas-
fir to infection by Armillaria root dis-
ease in southeastern British Columbia
[Canada]. Phytopathology 79:482?87
42. Bradshaw RHW, Miller NG. 1988. Re-
cent successional processes investigated
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 35
by pollen analysis of closed-canopy for-
est sites. Vegetatio 76:45?54
43. Brooks FE, Ferrin DM. 1994. Branch die-
back of southern California chaparral veg-
etation caused by Botryosphaeria doth-
idea. Phytopathology 84:78?83
44. Brown PM, Swetnam TW. 1994. A cross-
dated fire history from coast redwood near
Redwood National Park, California. Can.
J. For. Res. 24:21?31
45. Buchsbaum RN, Short FT, Cheney DP.
1990. Phenolic-nitrogen interactions in
eelgrass, Zostera marina L: possible im-
plications for disease resistance. Aquat.
Bot. 37:291?97
46. Burdon JJ. 1987. Diseases and Plant Pop-
ulation Biology. Cambridge: Cambridge
Univ. Press. 208 pp.
47. Burdon JJ. 1989. Pattern and patchiness
in plant-pathogen interactions: causes and
consequences. Annu. Rev. Ecol. Syst. 20:
119?36
48. Burdon JJ. 1991. Fungal pathogens as
selective forces in plant populations and
communities. Austr. J. Ecol. 16:423?32
49. Burdon JJ, Chilvers GA. 1982. Host den-
sity as a factor in plant disease ecology.
Annu. Rev. Phytopathol. 20:143?66
50. Burdon JJ, Groves RH, Cullen JM. 1981.
The impact of biological control on the
distribution and abundance of Chondrilla
juncea in south-eastern Australia. J. Appl.
Ecol. 18:957?66
51. Burdon JJ, Groves RH, Kaye PE, Speer
SS. 1984. Competition in mixtures of sus-
ceptible and resistant genotypes of Chon-
drilla juncea differentially infected with
rust. Oecologia 64:199?203
52. Burdon JJ, Leather SR, eds. 1990. Pests,
Pathogens and Plant Communities. Ox-
ford: Blackwell Sci. Publ. 333 pp.
53. Burdon JJ, Thrall PH. 1999. Spatial and
temporal patterns in coevolving plant and
pathogen associations. Am. Nat. 153:S15?
S33
54. Burdon JJ, Thrall PH. 2000. Coevolution
at multiple spatial scales: Linum margi-
nale-Melampsora lini?from the indivi-
dual to the species. Evol. Ecol. 14:261?
81
55. Burdon JJ, Thrall PH, Brown AHD. 1999.
Resistance and virulence structure in two
Linum marginale-Melampsora lini host-
pathogen metapopulations with differ-
ent mating systems. Evolution 53:704?
16
56. Burdon JJ, Wennstrom A, Ericson L,
Muller WJ, Morton R. 1992. Density-
dependent mortality in Pinus sylvestris
caused by the snow blight pathogen Pha-
cidium infestans. Oecologia 90:74?79
57. Campbell CL, Noe JP. 1985. The spatial
analysis of soil-borne pathogens and root
diseases. Annu. Rev. Phytopathol. 23:
129?48
58. Carlsson U, Elmqvist T. 1992. Epidemio-
logy of anther-smut disease (Microbo-
tryum violaceum) and numeric regulation
of populations of Silene dioica. Oecologia
90:509?17
58a. Carroll GC, Wicklow DT, eds. 1992. The
Fungal Community. Its Organization and
Role in the Ecosystem. New York: Marcel
Dekker
59. Carroll G. 1995. Forest endophytes: pat-
tern and process. Can. J. Bot. 73:S1316?
24
60. Chavez LB, Varon de AF, Morales F, Cas-
tano M, Arroyave J, Galvez G. 1999. Re-
conocimiento, transmisio?n y hospederas
de pato?genos virales del maracuya? (Passi-
flora edulis Sims) en Colombia. Fitopatol.
Colomb. 23:24?31
61. Chellemi DO, Britton KO. 1992. Influ-
ence of canopy microclimate on incidence
and severity of dogwood anthracnose.
Can. J. Bot. 70:1093?96
62. Chellemi DO, Britton KO, Swank WT.
1992. Influence of site factors on dogwood
anthracnose in the Nantahala Mountain
range of western North Carolina. Plant
Dis. 76:915?18
63. Chesson P. 2000. Mechanisms of main-
tenance of species diversity. Annu. Rev.
Ecol. Syst. 31:343?66
64. Clark DA, Clark DB. 1984. Spacing
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
36 GILBERT
dynamics of a tropical rain forest tree:
evaluation of the Janzen-Connell model.
Am. Nat. 124:769?88
65. Clay K. 1984. The effect of the fungus
Atkinsonella hypoxylon (Clavicipitaceae)
on the reproductive system and demogra-
phy of the grass Danthonia spicata. New
Phytol. 98:165?76
66. Clay K. 1990. Insects, endophytic fungi
and plants. See Ref. 52, pp. 111?30
67. Clay K, Holah J. 1999. Fungal endophyte
symbiosis and plant diversity in succes-
sional fields. Science 285:1742?44
68. Clay K, Kover P. 1996. Evolution and
statis in plant-pathogen associations.
Ecology 77:997?1003
69. Clay K, Kover PX. 1996. The Red Queen
Hypothesis and plant/pathogen interac-
tions. Annu. Rev. Phytopathol. 34:29?
50
70. Clay K, Marks S, Cheplick GP. 1993. Ef-
fects of insect herbivory and fungal en-
dophyte infection on competitive interac-
tions among grasses. Ecology 74:1767?
77
71. Connell JH. 1971. On the role of natural
enemies in preventing competitive exclu-
sion in some marine animals and in rain
forest trees. In Dynamics of Numbers in
Populations (Proc. Adv. Stud. Inst., Oster-
beek 1970), ed. PJ Boer, GR Graadwell,
pp. 298?312. Wageningen: Cent. Agric.
Publ. Document.
72. Creber GT, Ash SR. 1990. Evidence of
widespread fungal attack on Upper Trias-
sic trees in the southwestern USA. Rev.
Palaeobot. Palynol. 63:189?96
73. Crist TO, Friese CF. 1993. The impact of
fungi on soil seeds: implications for plants
and granivores in semiarid shrub-steppe.
Ecology 74:2231?39
74. Crossland DR, Kloet SPV. 1996. Berry
consumption by the American Robin, Tur-
dus migratorius, and the subsequent ef-
fect on seed germination, plant vigour,
and dispersal of the Lowbush Blueberry,
Vaccinium angustifolium. Can. Field-Nat.
110:303?9
75. Dalling JW, Swaine MD, Garwood NC.
1998. Dispersal patterns and seed bank
dynamics of pioneer trees in moist tropi-
cal forest. Ecology 79:564?78
76. Daughtrey ML, Hibben CR. 1994. Dog-
wood anthracnose: a new disease threat-
ens two native Cornus species. Annu. Rev.
Phytopathol. 32:61?73
77. Daughtrey ML, Hibben CR, Britton KO,
Windham MT, Redlin SC. 1996. Dog-
wood anthracnose: understanding a dis-
ease new to North America. Plant Dis. 80:
349?58
78. Davelos AL, Alexander HM, Slade NA.
1996. Ecological genetic interactions
between a clonal host plant (Spartina
pectinata) and associated rust fungi (Puc-
cinia seymouriana and Puccinia spargan-
ioides). Oecologia 105:205?13
79. De Rooij-Van Der Goes PCEM. 1995. The
role of plant-parasitic nematodes and soil-
borne fungi in the decline of Ammophila
arenaria (L.) Link. New Phytol. 129:661?
69
80. De Rooij-Van Der Goes PCEM, Van
Der Putten WH, Peters BAM. 1995. Ef-
fects of sand deposition on the interac-
tion between Ammophila arenaria, plant-
parasitic nematodes, and pathogenic
fungi. Can. J. Bot. 73:1141?50
81. De Rooij-Van Der Goes PCEM, Van Der
Putten WH, Van Dijk C. 1995. Analysis
of nematodes and soil-borne fungi from
Ammophila arenaria (Marram grass) in
Dutch coastal foredunes by multivariate
techniques. Eur. J. Plant Pathol. 101:149?
62
82. Dickman A. 1992. Plant pathogens and
long-term ecosystem changes. See Ref.
58a, pp. 499?520
83. Dickman AW, Cook SA. 1989. Fire and
fungus in a mountain hemlock forest. Can.
J. Bot. 67:2005?16
84. Dinoor A, Eshed N. 1984. The role and
importance of pathogens in natural plant
communities. Annu. Rev. Phytopathol. 22:
443?66
85. Durako MJ, Kuss KM. 1994. Effects of
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 37
Labyrinthula infection on the photosyn-
thetic capacity of Thalassia testudinum.
Bull. Mar. Sci. 54:727?32
86. Ebert D, Hamilton WD. 1996. Sex against
virulence: the coevolution of parasitic dis-
eases. Trends Ecol. Evol. 11:79?82
87. Ericson L, Burdon JJ, Muller WJ. 1999.
Spatial and temporal dynamics of epi-
demics of the rust fungus Uromyces va-
lerianae on populations of its host Valeri-
ana salina. J. Ecol. 87:649?58
88. Esquivel RE, Carranza J. 1996. Patho-
genicity of Phylloporia chrysita (Aphyl-
lophorales: Hymenochaetaceae) on Ery-
throchiton gymnanthus (Rutaceae). Rev.
Biol. Trop. 44:137?45
89. Ewald PW. 1995. The evolution of viru-
lence: a unifying link between parasitol-
ogy and ecology. J. Parasitol. 81:659?69
90. Ferris MA, Castello JD, Sinclair WA.
1989. Effects of virus and mycoplasma-
like organism infection on green and white
ash. Phytopathology 79:579?83
91. Fitt BDL, Gregory PH, Todd AD, McCart-
ney HA, Macdonald OC. 1987. Spore dis-
persal and plant disease gradients; a com-
parison between two empirical models. J.
Phytopathol. 118:227?42
92. Forget PM. 1994. Recruitment pattern
of Vouacapoua americana (Caesalpini-
aceae), a rodent-dispersed tree species in
French Guiana. Biotropica 26:408?19
93. Foster DR, Zebryk T, Schoonmaker P,
Lezberg A. 1992. Post-settlement history
of human land-use and vegetation dy-
namics of a Tsuga canadensis hemlock
woodlot in central New England. J. Ecol.
80:773?86
94. Friess N, Maillet J. 1996. Influence of cu-
cumber mosaic virus infection on the in-
traspecific competitive ability and fitness
of purslane (Portulaca oleracea). New
Phytol. 132:103?11
95. Garbelotto M, Otrosina WJ, Cobb FW,
Bruns TD. 1998. The European S and
F intersterility groups of Heterobasidion
annosum may represent sympatric proto-
species. Can. J. Bot. 76:397?409
96. Garbelotto M, Svihra P, Rizzo DM. 2001.
Sudden oak death syndrome fells 3 oak
species. Calif. Agric. 55:9?19
97. Garcia-Guzman G, Burdon JJ, Ash JE,
Cunningham RB. 1996. Regional and lo-
cal patterns in the spatial distribution of
the flower-infecting smut fungus Sporiso-
rium amphilophis in natural populations
of its host Bothriochloa macra. New Phy-
tol. 132:459?69
98. Garc??a-Guzma?n G, Dirzo R. 2001. Pat-
terns of leaf-pathogen infection in the
understory of a Mexican rain forest: inci-
dence, spatiotemporal variation, and
mechanisms of infection. Am. J. Bot. 88:
634?45
99. Gavin DG, Peart DR. 1993. Effects of
beech bark disease on the growth of Amer-
ican beech (Fagus grandifolia). Can. J.
For. Res. 23:1566?75
100. Gibbs JN, Wainhouse D. 1986. Spread of
forest pests and pathogens in the northern
hemisphere. Forestry 59:141?54
101. Gilbert GS. 1995. Rain forest plant dis-
eases: the canopy-understory connection.
Selbyana 16:75?77
102. Gilbert GS, De Steven D. 1996. A canker
disease of seedlings and saplings of Tetra-
gastris panamensis (Burseraceae) caused
by Botryosphaeria dothidea in a low-
land tropical forest. Plant Dis. 80:684?
87
103. Gilbert GS, Ferrer A, Carranza J. 2002.
Polypore fungal diversity and host density
in a moist tropical forest. Biodiv. Conserv.
In press
104. Gilbert GS, Harms KE, Hamill DN,
Hubbell SP. 2001. Effects of seedling size,
El Nin?o drought, seedling density, and dis-
tance to nearest conspecific adult on 6-
year survival of Ocotea whitei seedlings
in Panama. Oecologia 127:509?16
105. Gilbert GS, Hubbell SP, Foster RB. 1994.
Density and distance-to-adult effects of a
canker disease of trees in a moist tropical
forest. Oecologia 98:100?8
106. Gilbert GS, Hubbell SP, Foster RB. 1994.
Spatial distribution of a canker disease of
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
38 GILBERT
Lauraceae in a lowland moist tropical for-
est. Phytopathology 84:868
107. Gilbert GS, Sousa WP. 2002. Host spe-
cialization among wood-decay polypore
fungi in a Caribbean mangrove forest.
Biotropica. In press
108. Gilbert GS, Talaro N, Howell CA,
Symstad A. 1997. Multiple-scale spatial
distribution of the fungal epiphyll Scole-
copeltidium on Trichilia spp. in two low-
land moist tropical forests. Can. J. Bot.
75:2158?64
109. Gillett JB. 1962. Pest pressure, an under-
estimated factor in evolution. Syst. Assoc.
Publ. Number 4:37?46
110. Girling MA, Greig J. 1985. A first fossil
record of Scolytus scolytus (elm bark bee-
tle): its occurrence in elm decline deposits
from London (England, UK) and the im-
plications for neolithic elm disease. J. Ar-
chaeol. Sci. 12:347?52
111. Givnish TJ. 1999. On the causes of gra-
dients in tropical tree diversity. J. Ecol.
87:193?210
112. Gleason ML, Parker SK, Engle TE, Flynn
PH, Griffiths HM, et al. 1997. Ash
yellow occurrence and association with
slow growth of green ash in Iowa and
Wisconsin cities. J. Arboricult. 23:77?
82
113. Goodell K, Parker IM, Gilbert GS. 2000.
Biological impacts of species invasions:
implications for policy makers. In Incor-
porating Science, Economics, and Sociol-
ogy in Developing Sanitary and Phytosan-
itary Standards in International Trade, ed.
J Caswell, pp. 87?117. Washington, DC:
Natl. Acad. Press
114. Gordon TR, Storer AJ, Wood DL. 2001.
The pitch canker epidemic in California.
Plant Dis. 85:1128?39
115. Hafner MS, Sudman PD, Villablanca FX,
Spradling TA, Demastes JW, Nadler SA.
1994. Disparate rates of molecular evolu-
tion in cospeciating hosts and parasites.
Science 265:1087?90
116. Han Y, Castello JD, Leopold DJ. 1991.
Ash yellows, drought, and decline in ra-
dial growth of white ash. Plant Dis. 75:
18?23
117. Hansen EM, Goheen DJ, Jules ES, Ullian
B. 2000. Managing Port-Orford-cedar and
the introduced pathogen Phytophthora
lateralis. Plant Dis. 84:4?14
118. Hansen EM, Goheen EM. 2000. Phellinus
weirii and other native root pathogens as
determinants of forest structure and pro-
cess in western North America. Annu. Rev.
Phytopathol. 38:515?39
119. Harms KE, Wright SJ, Calderon O, Her-
nandez A, Herre EA. 2000. Pervasive
density-dependent recruitment enhances
seedling diversity in a tropical forest. Na-
ture 404:493?95
120. Harper JL. 1977. The Population Biology
of Plants. London: Academic. 892 pp.
121. Helander ML, Sieber TN, Petrini O, Neu-
vonen S. 1994. Endophytic fungi in Scots
pine needles: spatial variation and conse-
quences of simulated acid rain. Can. J.
Bot. 72:1108?13
122. Hiura T, Fujiwara K. 1999. Density-
dependence and co-existence of conifer
and broad-leaved trees in a Japanese
northern mixed forest. J. Veg. Sci. 10:843?
50
123. Holah JC, Alexander HM. 1999. Soil
pathogenic fungi have the potential to
affect the co-existence of two tallgrass
prairie species. J. Ecol. 87:598?608
124. Holah JC, Wilson MV, Hansen EM. 1993.
Effects of a native forest pathogen, Phelli-
nus weirii, on Douglas-fir forest composi-
tion in western Oregon. Can. J. For. Res.
23:2473?80
125. Holah JC, Wilson MV, Hansen EM.
1997. Impacts of a native root-rotting
pathogen on successional development of
old-growth Douglas fir forests. Oecologia
111:429?33
126. Holt RD. 1977. Predation, apparent com-
petition, and the structure of prey commu-
nities. Theor. Pop. Biol. 12:197?229
127. Hopkins DL, Adlerz WC. 1988. Natu-
ral hosts of Xylella fastidiosa in Florida
[USA]. Plant Dis. 72:429?31
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 39
128. Houle G. 1992. Spatial relationship bet-
ween seed and seedling abundance and
mortality in a deciduous forest of north-
eastern North America. J. Ecol. 80:99?
108
129. Houle G. 1995. Seed dispersal and seed-
ling recruitment?the missing link(s).
Ecoscience 2:238?44
130. Houle G, McKenna MF, Lapointe L.
2001. Spatiotemporal dynamics of Floer-
kea proserpinacoides (Limnanthaceae),
an annual plant of the deciduous forest
of eastern North America. Am. J. Bot. 88:
594?607
131. Houston DR. 1994. Major new tree dis-
ease epidemics: beech bark disease. Annu.
Rev. Phytopathol. 32:75?87
132. Hubbell SP. 1979. Tree dispersion, abun-
dance, and diversity in a tropical dry for-
est. Science 203:1299?309
133. Jackson JBC. 2001. What was natural in
the coastal oceans? Proc. Natl. Acad. Sci.
USA 98:5411?18
134. James TY, Moncalvo J-M, Li S, Vilgalys
R. 2001. Polymorphism at the ribosomal
DNA spacers and its relation to breed-
ing structure of the widespread mush-
room Schizophyllum commune. Genetics
157:149?61
135. James TY, Vilgalys R. 2001. Abundance
and diversity of Schizophyllum commune
spore clouds in the Caribbean detected
by selective sampling. Mol. Ecol. 10:471?
79
136. Janzen DH. 1970. Herbivores and the
number of tree species in tropical forests.
Am. Nat. 104:501?27
137. Jarausch W, Jarausch-Wehrheim B, Danet
JL, Broquaire JM, Dosba F, et al. 2001.
Detection and identification of European
stone fruit yellows and other phytoplas-
mas in wild plants in the surroundings
of apricot chlorotic leaf roll-affected or-
chards in southern France. Eur. J. Plant
Pathol. 107:209?17
138. Jarosz AM, Burdon JJ. 1988. The effect
of small-scale environmental changes on
disease incidence and severity in a natu-
ral plant-pathogen interaction. Oecologia
75:278?81
139. Jarosz AM, Burdon JJ, Muller WJ. 1989.
Long-term effects of disease epidemics. J.
Appl. Ecol. 26:725?33
140. Jarosz AM, Davelos AL. 1995. Tansley
Review No. 81: effects of disease in wild
plant populations and the evolution of
pathogen aggressiveness. New Phytol.
129:371?87
141. Keane RE, Arno SF, Brown JK, Tomback
DF. 1990. Modelling stand dynamics in
whitebark pine (Pinus albicaulis) forests.
Ecol. Model. 51:73?95
142. Kitajima K, Augspurger CK. 1989. Seed
and seedling ecology of a monocarpic
tropical tree, Tachigalia versicolor. Ecol-
ogy 70:1102?14
143. Koske RE, Gemma JN. 1997. Mycorrhi-
zae and succession in plantings of beach-
grass in sand dunes. Am. J. Bot. 84:118?30
144. Krause SC, Raffa KF. 1992. Comparison
of insect, fungal, and mechanically in-
duced defoliation of larch: effects on plant
productivity and subsequent host suscep-
tibility. Oecologia 90:411?16
145. Kyriakopoulou PE, Guinchedi L, Hadidi
A. 2001. Peach latent mosaic and pome
fruit viroids in naturally infected culti-
vated pear Pyrus communis and wild pear
P. amygdaliformis: implications on possi-
ble origin of these viroids in the Mediter-
ranean region. J. Plant Pathol. 83:51?62
146. Lambert JE. 2001. Red-tailed guenons
(Cercopithecus ascanius) and Strychnos
mitis: evidence for plant benefits beyond
seed dispersal. Int. J. Primatol. 22:189?
201
147. Lanier GN, Schubert DC, Manion PD.
1988. Dutch elm disease and elm yellows
in Central New York USA. Out of the fry-
ing pan into the fire. Plant Dis. 72:189?94
148. Larsen KJ, Whalon ME. 1988. Field mon-
itoring of X-disease leafhopper vectors
(Homoptera: Cicadellidae) and infected
chokecherry in Michigan [USA] peach
and cherry orchards. Great Lakes Ento-
mol. 21:61?68
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
40 GILBERT
149. Lenski RE, May RM. 1994. The evolution
of virulence in parasites and pathogens:
reconciliation between two competing hy-
potheses. J. Theor. Biol. 169:253?65
150. Lindblad I. 2000. Host specificity of some
wood-inhabiting fungi in a tropical forest.
Mycologia 92:399?405
151. Little LR, Maun MA. 1996. The ?Am-
mophila problem? revisited: a role for my-
corrhizal fungi. J. Ecol. 84:1?7
152. Little LR, Maun MA. 1997. Relationships
among plant-parasitic nematodes, mycor-
rhizal fungi and the dominant vegetation
of a sand dune system. Ecoscience 4:67?
74
153. Lively CM, Dybdahl MF. 2000. Parasite
adaptation to locally common host geno-
types. Nature 405:679?81
154. Lively CM, Johnson SG, Delph LF, Clay
K. 1995. Thinning reduces the effect of
rust infection on jewelweed (Impatiens
capensis). Ecology 76:1859?62
155. Lodge DJ, Cantrell S. 1995. Fungal com-
munities in wet tropical forests: variation
in time and space. Can. J. Bot. 73:S1391?
S8
156. Lonsdale WM. 1993. Losses from the
seed bank of Mimosa pigra: soil micro-
organisms vs. temperature fluctuations. J.
Appl. Ecol. 30:654?60
157. MacClement WD, Richards MG. 1956.
Viruses in wild plants. Can. J. Bot. 34:
793?99
158. Madden LV. 1997. Effects of rain on
splash dispersal of fungal pathogens. Can.
J. Plant Pathol. 19:225?30
159. Madrin?a?n S, Schultes RE. 1995. Colom-
bia?s national tree: the wax palm Cerox-
ylon quindiuense and its relatives. Elaeis
7:35?56
160. Masaki T, Tanaka H, Shibata M, Naka-
shizuka T. 1998. The seed bank dynam-
ics of Cornus controversa and their role
in regeneration. Seed Sci. Res. 8:53?
63
161. Mathiasen RL, Hawksworth FG, Edmin-
ster CB. 1990. Effects of dwarf mistletoe
on growth and mortality of Douglas-Fir
in the Southwest [USA]. Great Basin Nat.
50:173?80
162. Milgroom MG, Lipari SE. 1995. Spatial
analysis of nuclear and mitochondrial
RFLP genotypes in populations of the
chestnut blight fungus, Cryphonectria
parasitica. Mol. Ecol. 4:633?42
163. Mills KE, Bever JD. 1998. Maintenance
of diversity within plant communities: soil
pathogens as agents of negative feedback.
Ecology 79:1595?601
164. Muehlstein LK. 1992. The host-pathogen
interaction in the wasting disease of eel-
grass Zostera marina. Can. J. Bot. 70:
2081?88
165. Muehlstein LK, Porter D, Short FT. 1988.
Labyrinthula sp., a marine slime mold
producing the symptoms of wasting dis-
ease in eelgrass, Zostera marina. Mar.
Biol. 99:465?72
166. Mundt C. 2002. Influence of plant-geno-
type mixtures on disease management.
Annu. Rev. Phytopathol. 40:341?70
167. Newsham KK, Fitter AH, Watkinson AR.
1994. Root pathogenic and arbuscular
mycorrhizal fungi determine fecundity of
asymptomatic plants in the field. J. Ecol.
82:805?14
168. Paavolainen L, Kurkela T, Suhonen J,
Hantula J. 2001. The genetic population
structure of Pyrenopeziza betulicola, the
causative agent of birch leaf spot disease.
Mycologia 93:258?64
169. Pacala SW, Crawley MJ. 1992. Herbi-
vores and plant diversity. Am. Nat. 140:
243?60
170. Packer A, Clay K. 2000. Soil pathogens
and spatial patterns of seedling mortality
in a temperate tree. Nature 404:278?81
171. Parker MA. 1991. Local genetic differen-
tiation for disease resistance in a selfing
annual. Biol. J. Linnean Soc. 42:337?50
172. Parker MA. 1993. Constraints on the evo-
lution of disease resistance in an annual
legume. Heredity 71:290?94
173. Parnell M, Burt PJA, Wilson K. 1998. The
influence of exposure to ultraviolet radi-
ation in simulated sunlight on ascospores
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 41
causing Black Sigatoka disease of banana
and plantain. Int. J. Biomet. 42:22?27
174. Paul ND. 1989. The effects of Puccinia
lagenophorae on Senecio vulgaris in com-
petition with Euphorbia peplus. J. Ecol.
77:552?64
175. Paul ND, Ayres PG. 1987. Water
stress modifies intraspecific interference
between rust (Puccinia lagenophorae
Cooke)-infected and healthy groundsel
(Senecio vulgaris L.). New Phytol. 106:
555?66
176. Pedgley DE. 1986. Long distance trans-
port of spores. In Plant Disease Epidemi-
ology, ed. KJ Leonard, WE Fry, pp. 346?
65. New York: McGraw Hill
177. Perry I, Moore PD. 1987. Dutch elm dis-
ease as an analogue of Neolithic elm de-
cline. Nature 326:72?73
178. Peterken GF, Mountford EP. 1998. Long-
term change in an unmanaged population
of wych elm subjected to Dutch elm dis-
ease. J. Ecol. 86:205?18
179. Rice WR. 1983. Parent-offspring patho-
gen transmission: a selective agent pro-
moting sexual reproduction. Am. Nat.
121:187?203
180. Rizzo DM, Garbelotto M, Davidson JM,
Slaughter GW, Koike ST. 2002. Phytoph-
thora ramorum as the cause of extensive
mortality of Quercus spp. and Lithocar-
pus densiflorus in California. Plant Dis.
86:205?14
181. Rizzo DM, Slaughter GW, Parmeter JR.
2000. Enlargement of canopy gaps asso-
ciated with a fungal pathogen in Yosemite
Valley, California. Can. J. For. Res. 30:
1501?10
182. Robblee MB, Barber TR, Carlson PR
Jr., Durako MJ, Fourquerean JW, et al.
1991. Mass mortality of the tropical sea-
grass Thalassia testudinum in Florida Bay
(USA). Mar. Ecol. Prog. Ser. 71:297?99
183. Rotem J, Aust HJ. 1991. The effect of
UV and solar radiation and temperature
on survival of fungal propagules. J. Phy-
topathol. 133:76?84
184. Rotem J, Wooding B, Aylor DE. 1985.
The role of solar radiation, especially UV,
in the mortality of fungal spores. Phy-
topathology 75:510?14
185. Roy BA. 1993. Floral mimicry by a plant
pathogen. Nature 362:56?58
186. Roy BA. 1996. A plant pathogen influ-
ences pollinator behavior and may influ-
ence reproduction of nonhosts. Ecology
77:2445?57
187. Roy BA. 2001. Patterns of association be-
tween crucifers and their flower-mimic
pathogens: Host jumps are more common
than coevolution or cospeciation. Evolu-
tion 55:41?53
188. Saikkonen K, Faeth SH, Helander M, Sul-
livan TJ. 1998. Fungal endophytes: a con-
tinuum of interactions with host plants.
Annu. Rev. Ecol. Syst. 29:319?43
189. Shearer BL, Tippett JT. 1989. Jarrah
Dieback: The Dynamics and Management
of Phytophthora cinnamomi in the Jarrah
(Eucalyptus marginata) Forest of South-
Western Australia. Como, Aust.: Dep.
Conserv. Land Manag.
190. Sinclair WA, Griffiths HM. 2000. Varia-
tion in aggressiveness of ash yellows phy-
toplasmas. Plant Dis. 84:282?88
191. Sinclair WA, Griffiths HM, Davis RE.
1996. Ash yellows and lilac witches-
broom: phytoplasmal diseases of concern
in forestry and horticulture. Plant Dis.
80:468?75
192. Sinclair WA, Griffiths HM, Lee I-M.
1994. Mycoplasmalike organisms as
causes of slow growth and decline of trees
and shrubs. J. Arboricult. 20:176?89
193. Sinclair WA, Griffiths HM, Treshow M.
1993. Impact of ash yellows mycoplasma-
like organisms on radial growth of natu-
rally infected white, green, and velvet ash.
Can. J. For. Res. 23:2467?72
194. Singh P, Carew GC. 1989. Impact of east-
ern dwarf mistletoe in black spruce forests
of Newfoundland [Canada]. Eur. J. For.
Pathol. 19:305?22
195. Smith RSJ. 1996. Spread an intensifica-
tion of blister rust in the range of sugar
pine. In Sugar Pine Status, Values, and
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
42 GILBERT
Roles in Ecosystems, Publ. 3362, ed. BBJ
Kinloch, M Marosy, ME Huddleston, pp.
112?18. Oakland, CA: Univ. Calif., Div.
Agric. Nat. Resourc.
196. Tattar TA, Klekowski EJ, Stern AI. 1994.
Dieback and mortality in red mangrove,
Rhizophora mangle L., in southwest
Puerto Rico. Arboricult. J. 18:419?29
197. Taylor AH. 2000. Fire regimes and forest
changes in mid and upper montane forests
of the southern Cascades, Lassen Volcanic
National Park, California, U.S.A. J. Bio-
geogr. 27:87?104
198. Taylor AH, Skinner CN. 1998. Fire his-
tory and landscape dynamics in a late-
successional reserve, Klamath Moun-
tains, California, USA. For. Ecol. Manag.
111:285?301
199. Thompson JN. 1982. The Coevolutionary
Process. Chicago: Univ. Chicago Press.
376 pp.
200. Thompson JN, Burdon JJ. 1992. Gene-
for-gene coevolution between plants and
parasites. Nature 360:121?25
201. Thomsen IM, Koch J. 1999. Somatic com-
patibility in Amylostereum areolatum and
A. chailletii as a consequence of symbio-
sis with siricid woodwasps. Mycol. Res.
103:817?23
202. Thomson AJ, Alfaro RI, Bloomberg WJ,
Smith RB. 1985. Impact of dwarf mistle-
toe [Arceuthobium tsugense] on the
growth of western hemlock [Tsuga het-
erophylla] trees having different patterns
of suppression and release. Can. J. For.
Res. 15:665?68
203. Thrall PH, Burdon JJ. 1997. Host-patho-
gen dynamics in a metapopulation con-
text: the ecological and evolutionary con-
sequences of being spatial. J. Ecol. 85:
743?53
204. Thrall PH, Jarosz AM. 1994. Host-patho-
gen dynamics in experimental popula-
tions of Silene alba and Ustilago violacea:
I. Ecological and genetic determinants of
disease spread. J. Ecol. 82:549?59
205. Tibbets TM, Faeth SH. 1999. Neotypho-
dium endophytes in grasses: deterrents
or promoters of herbivory by leaf-cutting
ants? Oecologia 118:297?305
206. Tisserat NA, Kuntz JE. 1983. Longevity
of conidia of Sirococcus clavigignenti-
juglandacearum in a simulated airborne
state. Phytopathology 73:1628?31
207. Travers SE, Gilbert GS, Perry EF. 1998.
The effect of rust infection on reproduc-
tion in a tropical tree (Faramea occiden-
talis). Biotropica 30:438?43
208. Trummer LM, Hennon PE, Hansen EM,
Muir PS. 1998. Modeling the incidence
and severity of hemlock dwarf mistletoe
in a 110-year-old wind-disturbed forests
in southeast Alaska. Can. J. For. Res. 28:
1501?8
209. Turcq B, Siffedine A, Martin L, Absy ML,
Soubies F, et al. 1998. Amazonia rainfor-
est fires: a lacustrine record of 7000 years.
Ambio 27:139?42
210. Van der Putten WH, Maas PWT, Van Gu-
lik WJM, Brinkman H. 1990. Character-
ization of soil organisms involved in the
degeneration of Ammophila arenaria. Soil
Biol. Biochem. 22:845?52
211. Van Der Putten WH, Peters BAM.
1997. How soil-borne pathogens may af-
fect plant competition. Ecology 78:1785?
95
212. Van der Putten WH, Van Dijk C, Troelstra
SR. 1988. Biotic soil factors affecting the
growth and development of Ammophila
arenaria. Oecologia 76:313?20
213. Vergeer LHT, Aarts TL, De-Groot JD.
1995. The ?wasting disease? and the ef-
fect of abiotic factors (light intensity,
temperature, salinity) and infection with
Labyrinthula zosterae on the phenolic
content of Zostera marina shoots. Aquat.
Bot. 52:35?44
214. Vergeer LHT, Den Hartog C. 1994. Omni-
presence of Labyrinthulaceae in sea-
grasses. Aquat. Bot. 48:1?20
215. Vergeer LHT, Develi A. 1997. Phenolic
acids in healthy and infected leaves of
Zostera marina and their growth-limiting
properties towards Labyrinthula zosterae.
Aquat. Bot. 58:65?72
12 Jul 2002 9:19 AR AR165-PY40-02.tex AR165-PY40-02.SGM LaTeX2e(2002/01/18) P1: IKH
DISEASES IN NATURAL ECOSYSTEMS 43
216. Verwijst T. 1993. Influence of the patho-
gen Melampsora epitea on intraspecific
competition in a mixture of Salix vimi-
nalis clones. J. Veg. Sci. 4:717?22
217. Washitani I, Okayama Y, Sato K, Taka-
hashi H, Ohgushi T. 1996. Spatial varia-
tion in female fertility related to inter-
actions with flower consumers and
pathogens in a forest metapopulation of
Primula sieboldii. Res. Pop. Ecol. 38:
249?56
218. Watson DM. 2001. Mistletoe?a keystone
resource in forests and woodlands world-
wide. Annu. Rev. Ecol. Syst. 32:219?49
219. Webb CO, Peart DR. 1999. Seedling den-
sity dependence promotes coexistence of
Bornean rain forest trees. Ecology 80:
2006?17
220. Weiner J. 1990. Asymmetric competition
in plant populations. Trends Ecol. Evol.
5:360?64
221. Wennstrom A, Ericson L. 1990. The in-
teraction between the clonal herb Tri-
entalis europaea and the host specific
smut fungus Urocystis trientalis. Oecolo-
gia 85:238?40
222. Wenny DG. 2000. Seed dispersal, seed
predation, and seedling recruitment of a
neotropical montane tree. Ecol. Monog.
70:331?51
223. Weste G, Marks GC. 1987. The biology of
Phytophthora cinnamomi in Australasian
forests. Annu. Rev. Phytopathol. 25:207?
29
224. Wilkinson HH, Siegel MR, Blankenship
JD, Mallory AC, Bush LP, Schardl CL.
2000. Contribution of fungal loline al-
kaloids to protection from aphids in a
grass-endophyte mutualism. Mol. Plant-
Microbe Interact. 13:1027?33
225. Wills C, Condit R. 1999. Similar non-ran-
dom processes maintain diversity in two
tropical rainforests. Proc. R. Soc. London,
Ser. B 266:1445?52
226. Wilson D, Carroll GC. 1994. Infection
studies of Discula quercina, an endophyte
of Quercus garryana. Mycologia 86:635?
47
227. Wolfe LM, Rissler LJ. 1999. Reproduc-
tive consequences of a gall-inducing
fungal pathogen (Exobasidium vaccinii)
on Rhododendron calendulaceum (Eri-
caceae). Can. J. Bot. 77:1454?59
228. Worrall JJ. 1994. Population structure of
Armillaria species in several forest types.
Mycologia 86:401?7
229. Wright SJ. 2001. Plant diversity in trop-
ical forests: a review of mechanisms of
species coexistence. Oecologia 130:1?
14
230. Zak JC, Sinsabaugh R, Mackay WP.
1995. Windows of opportunity in desert
ecosystems: Their implications to fun-
gal community development. Can. J. Bot.
73:S1407?S14
231. Zieman JC, Fourqurean JW, Frankovich
TA. 1999. Seagrass die-off in Florida Bay:
long-term trends in abundance and growth
of turtle rass, Thalassia testudinum. Estu-
aries 22:460?70