Host-Dependent Sporulation and Species Diversity of Arbuscular Mycorrhizal Fungi in a

Mown Grassland
Author(s): James D. Bever, Joseph B. Morton, Janis Antonovics and Peggy A. Schultz
Source: Journal of Ecology , Feb., 1996, Vol. 84, No. 1 (Feb., 1996), pp. 71-82
Published by: British Ecological Society

Stable URL: https://www.jstor.org/stable/2261701

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

British Ecological Society is collaborating with JSTOR to digitize, preserve and extend access to
Journal of Ecology

This content downloaded from

170.150.25.187 on Sat, 25 Jun 2022 19:20:35 UTC
All use subject to https://about.jstor.org/terms
Journal of
Host-dependent sporulation and species diversity of
Ecology 1996,
84, 71-82 arbuscular mycorrhizal fungi in a mown grassland

JAMES D. BEVER,*t JOSEPH B. MORTON,t JANIS ANTONOVICS*

and PEGGY A. SCHULTZ*
*Department of Botany, Duke University, Durham, NC 27708 and tDivision of Plant and Soil Sciences, 401
Brooks Hall, PO Box 6057, West Virginia University, Morgantown, WV 26505-6057 USA

Summary

1 In laboratory microcosm experiments, co-occurring plant species were found to

support very different rates of sporulation of arbuscular mycorrhizal (AM) fungi.
These differences were not affected by the time of harvest, suggesting that they reflect
host-dependent differences in fungal growth rates, rather than host-dependent timing
of sporulation.
2 Spore counts in field soil and estimates from sorghum trap cultures showed that the
association of AM fungi with particular host plants in the field was positively cor-
related with the sporulation rates observed on those hosts in the microcosm experi-
ments.
3 The AM fungal species richness observed at the field site was high relative to
estimates made in previous studies. 23 distinct species of AM fungi were found, seven
of which have not been previously described.
4 The host-dependence of the relative growth rates of fungal populations may play
an important role in the maintenance of fungal species diversity.

Keywords: arbuscular mycorrhizae, community ecology, glomalean fungi, microcosm,

population growth rates, species diversity

Journal of Ecology (1996) 84, 71-82

explanations for this pattern have been discussed

Introduction
(Morton 1990; Bever 1992). On a local scale, however,
The majority of vascular plants associate with ar- the determinants of AM fungal species diversity are
buscular mycorrhizal (AM) fungi (frequently called ves- poorly understood (Morton et al. 1995).
icular arbuscular endomycorrhizal or VAM fungi or While most AM fungi can associate with a wide
glomalean fungi). Despite the importance of AM array of hosts, a growing body of work suggests that
fungi in the physiology and nutrition of plants, as their performance relative to each other depends upon
well as in plant community and ecosystem processes the host species involved. The sporulation rates of
(Miller 1987; Allen & Allen 1990; Allen 1991; Brun- AM fungi have been found to be host-dependent in
drett 1991; Hartnett et al. 1993; Miller & Jastrow laboratory systems (Daft & Hogarth 1983; Hetrick &
1994), factors affecting diversity of AM fungi are Bloom 1986; Koomen et al. 1987; Hung & Sylvia
poorly understood. By current estimates (Law & 1988), and such differences in sporulation rates may
Lewis 1983) it appears that, on a global scale, there explain changes in the composition of the AM fungal
are vastly fewer species of AM fungi relative to the community in response to the changes in the plant
number of species of mycorrhizal dependent plant community (Schenck & Kinloch 1980; Johnson et al.
hosts. It has been suggested that this low diversity 1991a, 1992; Sanders & Fitter 1992; Hendrix et al.
results from the low host-specificity of AM fungi rela- 1995). These host-dependent sporulation rates also
tive to other fungi (Law & Lewis 1983), though other may reflect host-dependence in the rate of production
of infective propagules (i.e. spores and hyphae), which
is a major component of the growth rates of fungal
* Correspondence: James D. Bever, Department of
populations.
Botany, Duke University, Durham, NC 27708, USA. Tel.:
?) 1996 British 9196842598. Fax: 919 6845412. E-mail: bever@sunl. Host-dependence of fungal population growth
Ecological Society botany.duke.edu. rates may be a mechanism for the maintenance of

This content downloaded from

170.150.25.187 on Sat, 25 Jun 2022 19:20:35 UTC
All use subject to https://about.jstor.org/terms
72 fungal species diversity. Enhancement of fungal spec- 35 years, and consists predominantly of grasses. More
Host-dependence ies diversity is likely when the host-dependence than 25 plant species occur frequently, without any
of AMfungi involves changes in the ranking of fungal population attaining clear cut dominance. This plant community
growth rates on different plant hosts. With such host- is divided temporally into summer and winter assem-
dependence, a diverse plant community would rep- blages (Fowler & Antonovics 1981). The plants,
resent a heterogeneous selective environment (sensu pathogens and soil of this site have been intensively
Brandon 1990) to the AM fungi. Such heterogeneous studied over the last 20 years (e.g. Fowler & Antono-
environments have long been considered to be a pri- vics 1981; Clay 1982; Alexander 1984; Ellstrand &
mary mechanism for the maintenance of species diver- Antonovics 1985; Schmitt & Antonovics 1986;
sity (MacArthur 1972; Levin 1976; Levins 1979), but Moloney 1988; Kelley 1989; Broaddus 1991; Ron-
the importance of this process in AM fungal com- sheim 1992; Bever 1994). The soil is a sandy loam of
munities is poorly understood. Inference from pat- the White Store series with a pH of 5.0 and low fertility
terns of distribution of AM fungi in the field suggest with extractable phosphorus concentrations of 1.5-
host species, as well as abiotic factors and competitive 6,ug g-' (Fowler & Antonovics 1981).
interactions, may be important in maintaining fungal We studied four perennial plant species that are
species diversity (Koske 1981; Koske 1987; Gemma abundant at the study site: Allium vineale L., Anthox-
et al. 1989), but only a few studies have used exper- anthum odoratum L., Panicum sphaerocarpon Ell., and
imental manipulation to evaluate the roles of these Plantago lanceolata L. Dynamics of these populations
different factors (Sanders & Fitter 1992; Johnson et have been investigated previously (Allium: Ronsheim
al. 1992). 1992; Bever 1992; Anthoxanthum: Antonovics &
Furthermore, because host-dependent population Ellstrand 1984; Schmitt & Antonovics 1986; Kelley
growth rates of AM fungi may result in the devel- 1989; Bever 1994; Panicum: Bever 1994; Plantago:
opment of distinct fungal communities on plant hosts, Antonovics & Primack 1982; Alexander 1984), includ-
these fungal communities may, in turn, affect the per- ing the role of the soil community in the dynamics of
formance of plant species relative to each other. Fun- Anthoxanthum and Panicum (Bever 1994). For each
gal species have indeed been observed to have differ- of these species, individual plants were collected from
ential effects on the growth of plant species (Nemec a 5-m x 15-m region of the field and cloned to pro-
1978; Powell et al. 1982; Ollivier et al. 1983; Wilson duce the replicates in our study. In this paper, clones
1988; Ravnskov & Jakobsen 1995). Given such mut- derived from the same field-collected individual will
ual interdependence of plant and fungal relative be identified as being of the same 'genotype'.
growth rates, coexistence of both plants and fungi
would be determined by whether their interaction
Methods
leads to negative or positive feedback (Bever 1992).
Negative feedback occurs if a particular plant species
GENERAL PROTOCOLS
promotes the growth of one AM fungal species more
than that of others, but this fungal species has a less We tested for host-dependent differences in AM fun-
positive effect on growth of that plant species (but a gal population growth rates using experimental
more positive effect on other plant species) than do microcosms composed of components which
other AM fungi. Negative feedback would be coexisted in the field. Specifically, we started with
expected to promote coexistence of multiple sets of replicates of a single laboratory mycorrhizal
interactants, whereas positive feedback would lead to community, grew them with Allium, Anthoxanthum,
the predominance of one particular interactant pair Panicum, and Plantago, and then tested for differ-
(Bever 1992). entiation of the mycorrhizal fungal communities by
In the present study, we test for host-dependence monitoring sporulation over successive harvests. We
of the growth rate of AM fungi by growing a single examined the spatial correlations between the dis-
fungal community in association with different plant tribution of AM fungi and the distribution of the same
species and monitoring sporulation. We also examine four plant species in the field using three sampling
the distribution of fungi in the field with respect to methods: (1) Sorghum trap cultures. Sorghum plants
these same plant species and ask whether the plant- serve as hosts for fungi collected from underneath
fungal associations in the field are similar to the pat- individuals of each of the four plant species; (2) Trans-
terns of host-dependence observed experimentally in plant trap cultures. Field collected individuals of
the glasshouse. Anthoxanthum, Panicum, and Plantago were trans-
planted into soil free of AM fungi; and (3) Field-
collected spore counts. Direct counts were made of
STUDY SYSTEM
field-collected spores from underneath individuals of
The plants and AM fungi used in these experiments each of the four plant species. We then compared the
? 1996 British
were obtained from a mown field on the campus of patterns observed in the field to the patterns of host-
Ecological Society,
Journal of Ecology, Duke University, Durham, North Carolina. The field dependent differentiation of the AM fungal com-
84. 71-82 has been maintained by annual mowing for at least munities observed in the microcosm.

This content downloaded from

170.150.25.187 on Sat, 25 Jun 2022 19:20:35 UTC
All use subject to https://about.jstor.org/terms
73 In September, 1992, we identified six locations were planted into 24 6-cm x 25-cm pots (30-50 seeds
J.D. Bever et al. within a 5-m x 15-m study area of the field in which per pot) containing the soil mixtures and grown in the
an individual of each of the four plant species was glasshouse in Durham, NC, under natural sunlight
located within 0.5 m of each other (see Fig. 1). These and cool conditions for 5.5 months. They were wat-
individuals and the soil and roots from underneath ered and fertilized as described for the microcosm
them (c. 500 mL) were used for the studies described experiment.
below.

Transplant trap cultures

EXPERIMENTAL MICROCOSMS
Plants were removed from the field as described pre-
Portions of the soil and roots from underneath the 6 viously, their roots washed free of soil, and planted
individuals of each of the four plant species were into an autoclaved mixture of equal parts soil and
pooled, diced, and thoroughly mixed with equal parts sand. Allium was not used in this experiment because,
of sand which had been autoclaved for two hours. as a winter perennial, it did not have roots at the time
Two mycorrhizae-free individuals of each of seven of collection. Because the root systems of individual
genotypes of each of the four plant species (a total of plants were only partially sampled and the root archi-
56 pots) were planted into 6-cm x 25-cm pots con- tecture differed between plant species, the quantity
taining this mixture. These genotypes were derived and nature of roots associated with these transplants
from individuals collected at random from this same could not be held constant between species. The plants
site in the previous year. Individuals of Anthoxanthum were grown for 5 months in the glasshouse in North
and Panicum were cloned by dividing plants into Carolina and watered and fertilized as described for
the microcosm experiment.
tillers, cutting off the roots, and surface sterilizing the
tillers in 10% Clorox for 10 min. (Bever 1994). Allium
was cloned from asexually produced aerial bulbils
Field-collected spore counts
(Ronsheim 1992; Bever 1992). Genotypes of Plantago
were cloned by rooting surface-sterilized leaf cuttings In April 1993, three adjacent soil cores (1.7cm in
(Wu & Antonovics 1975; Teramura et al. 1981). diameter and 6 cm deep) were taken from underneath
The plants were grown in a glasshouse in Durham, a further seven sets of individuals of each of the four
NC, under natural sunlight and cool conditions (4- species. All sets had the same clumped arrangement
21 ?C) with one replicate being harvested after 4.5 as described above (see Fig. 1) and were taken from
months and the second replicate being harvested after the original 5-m x 1 5-m area. These three cores were
6 months. The plants were watered as needed and pooled so that the soil volume totalled % 50 cm3.
fertilized during the second and fourth month with a
100 mL of 1/4 strength Hoaglands solution modified
IDENTICATION AND ENUMERATION OF
to contain a reduced concentration of phosphorus
SPORES
(Millner & Kitt 1992).
All samples were stored in sealed plastic bags at 4?C
for up to 1 month until spores could be counted.
DISTRIBUTION OF AM FUNGI IN THE FIELD
Spores were extracted from the soil by blending a
Sorghum trap cultures solution of a known quantity of soil (either 50 or
100 cm3) in water for two 5-s bursts in a Waring
Soil and roots from underneath each of the six sam-
blender, collecting the spores by pouring the slurry
pled plants of the four species (Fig. 1) were mixed
through a 45-,um sieve, and separating the spores from
separately with equal parts autoclaved sand. Sorghum
sand, soil, and organic debris using sucrose gradient
vulgare was used as a host plant because it grows well
centrifugation (Daniels & Skipper 1982). Spores were
in the glasshouse and is a suitable host for a wide
examined under a binocular stereomicroscope and
variety of fungi (Morton et al. 1993). Sorghum seeds
Nomarski interference microscope. Species identi-
fication was based on spore colour, size, surface orna-
4------ 15 m
mentation, and wall structure using reference cultures
from the International Collection of Vesicular and
t 0 a: I
Arbuscular Mycorrhizal Fungi, INVAM (Morton et
I o o Allium vineale
al. 1993), and species descriptions (Schenck & Perez
E a a Anthoxanthum odoratum
0 Panicum sphaerocarpon I 1990). Permanent slide vouchers were made of all
0 v Plantago lanceolata
0 ~~~~~~~~~~0 fungi, and the majority were subsequently established
0 ~~~~~0a
0 a v in single species cultures and deposited in INVAM.
A taxonomic treatment of AM fungi at this site will
?) 1996 British Fig. 1 Sampling design of 5-rn x
be published separately. Only spores which appeared
Ecological Society, field in North Carolina. Individuals of each of the four
Journal of Ecology, species were collected from six locations with individuals at to be viable (based on colour, shape, surface
84, 71-82 a given location being within 0.5 m of each other. conditions, and examination of spore contents) were

This content downloaded from

170.150.25.187 on Sat, 25 Jun 2022 19:20:35 UTC
All use subject to https://about.jstor.org/terms
74 counted. Spores of all species were counted at 30 x species were tested individually using univariate
Host-dependence under a binocular stereomicroscope except the small- analysis of variance. The similarity of patterns of
of AM fungi spored species of Acaulospora (A. trappei, A. D2, A. differentiation observed in the community microcosm
D4) and Glomus (Gl. fasciculatum and Gl. D4). For experiments to those observed in the field was tested
these species, spores were counted at 42 x within 30 by regressing the mean ranks of spore numbers of
optical squares located randomly across the plate. different fungal species on different plant hosts
The average of counts from two separate extractions between the sorghum traps and the direct field counts
were used in the analysis of the microcosm experi- against those observed in the community microcosm
ment. experiment.

STATISTICAL ANALYSIS Results

Species in which only a few spores were found or

DIFFERENTIATION OF AM FUNGAL
where counting techniques appeared unreliable
COMMUNITIES ON PLANT HOSTS IN THE
(especially the rarer small spored species) were not
MICROCOSMS
included in statistical analyses. The spore counts were
not normally distributed and the variance of the At the first harvest (4.5 months) the abundance of AM
counts was not homogeneous across treatments, as fungal spores present in the pots differed significantly
has been found in other studies of AM fungi (St. John between host species (Table 1, Fig. 2), indicating the
& Koske 1988). Conversion of the spore counts into original fungal community had strongly differentiated
ranks of counts greatly improved the normality and in response to being grown with the different plant
homogeneity of variances; therefore, analyses were species. The sporulation of nine of the 14 most com-
performed on the ranks. These ranks were analysed mon fungal species depended significantly on host
with a two-way multivariate analysis of variance with species (Fig. 2, Table 2). Furthermore, the ranks of
profile contrasts (Morrison 1976) using the general the spore numbers of the different fungal species were
linear models procedure of SAS (SAS 1986). The over- significantly altered when grown with different plant
all effects and the interaction of the effects with the species, as tested by the interaction of rank profiles
profile were tested using Wilk's Lambda criterion with the plant species effect (Table 1, Fig. 2). That is,
because it is derived from a likelihood-ratio approach fungal species sporulated differentially on the different
(SAS 1986); however, multivariate tests using Pillai's species of plant hosts. However, there was no sig-
Trace and Hotelling-Lawley Trace gave similar nificant effect of plant genotype within species on the
results. The interaction of effects with the profiles (e.g. composition of the AM fungal species, but the power
the dependence of the shape of the profiles on plant of the analysis to detect such effects was low because
species) is of particular interest because it specifically there were only two replicates and the analysis was
tests whether the relative ranks of spore numbers of based on ranks.
the fungal species (i.e. relative abundance) varied with The total number of fungal spores and the relative
treatments. When the multivariate tests were sig- sporulation of the fungal species differed between the
nificant, the ranks of spore numbers of AM fungal two harvests (Table 1). For example, Gi. gigantea

Table 1 Multivariate analysis of variance of effects of plant species, genotype within plant species, and harvest on spore counts
of AM fungi in the microcosm experiment. Spore counts were made for each pot and these counts were ranked within each
fungal species; the analysis was then carried out on the ranks. The heading 'overall differences' presents tests for effects of
treatments on total sporulation (considered across all fungal species). The heading 'interaction with the profile' presents tests
for the specific hypothesis that the relative rank of sporulation of the different fungal species depends upon the designated
treatment effects. The plant species effect was tested over the variation among genotypes within species. Other effects were
tested over the residual error

Num. Den. Wilk's

Effect d.f. d.f. lambda F P

Overall differences
plant 48 28 8.0x 104 5.8 0.0001
genotype within plant 384 176 5.0 x 10'7 0.9 NS
harvest 16 9 8.0 x 10-2 5.9 0.005
plant x harvest 48 28 2.0 x 10-2 1.5 NS

Interaction with profile

plant 45 31 3.0x 10-3 4.1 0.0001
?) 1996 British genotype within plant 360 178 1.0 x 10b 0.9 NS
Ecological Society, harvest 15 10 1.0 x 10-' 5.8 0.004
Journal of Ecology, plant x harvest 45 31 2.5 x 10-2 1.7 0.07
84, 71-82

This content downloaded from

170.150.25.187 on Sat, 25 Jun 2022 19:20:35 UTC
All use subject to https://about.jstor.org/terms
 A. DI A. trappel Gl. gigantea
75 70 a 2000 5

J.D. Bever et al.

50 50a4
40 3a

30

21 2
I 0
2 . bS0 '4opr . 4G.D
10 b ~~~~bc c0
0l 2 1 ,,LO M 0
Allium Anthox. Panicum Plantago Allium Anthox. Panicum Plantago Allium Anthox. Panicum Plantago

S. calospora A. D4 GI. D3
40 250 3a

200~~
30

20 b ab 1.5

10 * ~~~~~~~~~~~50 0.5
Allium Anthox. Panicum Plantago Allium Anthox. Panicum Plantago Allium Anthox. Panicum Plantago

GI. DI A. 3 S. pellucida
3 1.2 a 0.35a

2.5 10.
0.8 E~~~~~~~~~~~~~02
'E 2 E 0.8 'E~~~~~~~~~AD .2
8 ab 0.2
1.5 ab 06ab
1 E K E mum 04 ~~~~~~~~~~~~~~~~~~~~0.15

0.5 0.2 0.05

Allium Anthox. Panicum Plantago Allium Anthox. Panicum Plantago Allium Anthox. Panicum Plantago

Fig. 2 Sporulation of AM fungi in association with the four plant species in the experimental microcosm. F
significantly different sporulation between hosts are presented, and are arranged in decreasing order of t
volume. Although all statistical tests were performed on the ranks of the spore counts, mean sporulation
of interpretation. Significant differences in sporulation are indicated by different letters.

produced a greater number of spores at the first har- nificantly influenced by the species of plant under
vest than the second, while Acaulospora mellea which the soil was obtained (Table 4). However, both
showed the opposite trend. However, the effect of the total number of spores, and the relative abundance
the host species on total sporulation did not depend of spores of different species did depend on the
significantly on the harvest (Table 1). location in the field from which the inocula were col-
lected (Table 4), indicating that AM fungal species
DISTRIBUTION OF AM FUNGI IN THE FIELD were not uniformly distributed even in this small
area.
Twenty-three distinct AM fungal species were ident-
ified from this site, seven of which have not been
described previously (Table 3). No single sampling Transplant trap cultures
procedure revealed all of the fungal species, and while
The overall abundance of spores observed was sig-
there was qualitative agreement in the abundances of
nificantly influenced by both the species of plant
the different species recorded by the different sampling
which was transplanted as well as by the location of
methods, there were also some marked discrepancies.
that plant in the field (Table 4). The relative abun-
For example, Scutellospora calospora spores were very
dance of spore types in the transplant traps was sig-
abundant in the Sorghum trap cultures, but were much
nificantly affected by plant species but not by location
less so using the other methods. Also, the transplant
(Table 4), indicating that the composition of the fun-
trap cultures revealed the lowest diversity and total
gal communities differed between plant species.
abundance of fungal spores. The results of the indi-
vidual sampling methods are described below. The
spore abundances in the experimental microcosms Field-collected spore counts
were qualitatively similar to those obtained by direct
In the direct counts of field soil, neither the overall
field sampling.
abundance of spores nor the relative abundance of
spores of different species depended on the plant spec-
?) 1996 British Sorghum trap cultures
ies under which the soil was sampled, although the
Ecological Society,
Journal of Ecology, Neither overall abundance of spores nor the relative latter effect approached significance (P < 0.08, Table
84, 71-82 abundance of spores of different species were sig- 4). Location of the sample in the field influenced both

This content downloaded from

170.150.25.187 on Sat, 25 Jun 2022 19:20:35 UTC
All use subject to https://about.jstor.org/terms
76 Table 2 Results from univariate analyses of variance of effects of plant species, genotype within plant species, and harvest on
spore counts of AM fungi in microcosm experiment, considering each fungal species separately. Table shows the sums of
Host-dependence
squares and significance levels of effects for each fungal species. Spore counts were made for each pot and these counts were
of AMfungi
ranked within each fungal species; the analysis was then carried out on the ranks. The plant species effect was tested over the
variation among genotypes within species. Other effects were tested over the residual error

Genotype
within Plant x
Plant plant harvest Harvest Error

d.f. 3 24 1 3 24
A. DI 8628**** 2138 944** 354 2513
A. trappei 4116*** 4326 1170** 1037 3969
Gi. gigantea 2392* 5124 462 2019* 4425
S. calospora 2295** 3882 3045*** 237 5145
Gl. D2 1115 4458 5904**** 206 2936
A. D4 1981**** 1248 325t 374 2052
S. reticulata 1093 5487 45 286 5532
A. D2 1232 5258 952* 625 4008
Gl. D3 1770*** 1512 18 159 2511
Gl. D2 1514* 3802 138 1806 5440
A. mellea 322 2497 3180**** 655 3029
A. D3 1221* 3278 208 514 2779
S. pellucida 764** 1306 5 537 2215
Gi. rosea 599 3043 833* 103 4282
S. heterogama 70 3325 71 281 3287
Gl. clarum 144 1922 9 159 2585

tP< 01, *P < .05 **P < .1, ***P < 001, ****P < 0.01.

the overall abundance of spores and the relative ranks teristic of the site (Morton et al. 1995). Moreover,
of spore abundance of the different fungal species given that no single sampling procedure revealed all
(Table 4). of the fungal species (Table 3), it is unlikely that all
of the fungal species present at this site were dis-
covered in this study. In fact, further sampling at
Correlation offield distribution with differentiation in
this site has revealed several additional species (J. D.
the microcosms
Bever, unpublished data).
The composition of the AM fungal community associ- The glasshouse microcosm experiment provides
ated with the four plant species in the field, as esti- strong evidence of host-dependent differentiation of

mated from the sorghum trap cultures and from direct AM fungal communities. Not only did overall sporu-

counts of field soil, was significantly similar to the lation of the AM fungi vary significantly on different

composition of the AM fungal communities observed plant hosts, but, most importantly, fungal species had

in the glasshouse microcosms (P < 0.04, R2 = 0.08 distinct patterns of differential sporulation such that

and P < 0.05, R2 = 0.08, for sorghum traps and field for a given pair of plant species, one was a better

soil, respectively). When only the nine fungal species host for one AM fungal species, but a worse host for

which had significantly different sporulation on plant another. For example, A. DI and A. D4 sporulated

hosts in the microcosm experiment (Table 3) were most prolifically on Allium while on this same host A.

included in the analysis, the strength of the regressionstrappei and Gi. gigantea had their lowest sporulation
(Fig. 2, Table 2). The mechanism of this host-depen-
increased (P < 0.03, R2 = 0.17 and P < 0.04,
dent sporulation was not investigated, but it could
R' = 0.12 for sorghum traps and field soil, respec-
result directly from host-dependent growth of the
tively) (Fig. 3). There was no significant correlation
fungi, indirectly through host-dependent changes in
between the field distribution of the fungi as estimated
the mineral soil, or indirectly through host-mediated
from the transplant trap cultures, and their host
interactions of individual AM fungal species with
related abundances in the glasshouse microcosms.
each other or with other components of the soil com-
munity. Regardless of the mechanism generating the
Discussion host-dependence, the strength of the host effect sug-
gests that host-dependent processes are important in
The 23 species of AM fungi identified from the 75_m2
this system.
area of the field site in this study probably represents Direct evidence of host-dependence in nature was
? 1996 British
Ecological Society, the highest species richness ever recorded for AM much weaker: of the three methods used to examine
Journal of Ecology, fungi within such a small area, and may reflect the association of fungal distribution with plant species,
84, 71-82 intensity of sampling rather than any unique charac- a significant plant species effect was only observed in

This content downloaded from

170.150.25.187 on Sat, 25 Jun 2022 19:20:35 UTC
All use subject to https://about.jstor.org/terms
77 Table 3 Abundance of AM fungal species present in a 5 m x 15 m region of a mown grassland in North Carolina, as detecte
J.D. Bever et al. using four methods (see text for detailed description). Data are presented in terms of total spore volume, to give a bet
indication of overall abundance. For each species and each method, the total spore volume per 100 cm3 soil is given (in uni
of 106 Pm3). Spore volume was calculated by multiplying the average number of spores observed by the average volume of
individual spores, as calculated from their average diameter and the equation of a sphere. The average diameter used in the
calculations are reported. Species present but whose counts were unreliable (see text) are indicated by a 'P'

Total spore volume detected by different methods

Sorghum Transplant Experimental Spore diam. INVAM

AM fungia cultures cultures Field soil microcosm (Pm) number

Acaulospora
bireticulata 0 0 19.7* 0 180
mellea 71.4 0 7.9 2.4 101 NC 148
trappei 48.0 45.6 13.7 198.2 65 NC 112A
sp. D1' 216.2 95.6 217.1 211.6 280 NC 168
sp. D22 2.9 9.2 46.0 5.9 69
sp. D33 0 0 13.4 2.2 69
sp. D44 0 0 9.1 10.3 201 NC 174
Gigaspora
gigantea 102.2 34.2 61.1 64.3 352 NC 120A
rosea 23.7 3.7 P 1.6 260 NC 121A
ramisporophora 0 . 0 0 P 320 NC 175
Glomus
clarum P 0 P 0.1 104 NC 112B
constrictum P 0 P P 170
fasciculatum 0 3.0 0 0 55
leptotichum P 0 0 0 160 NC 171
mosseae 3.8 0 P P 143
sp. Dl1 7.1 2.2 0.6 0.9 106 NC 172
sp. D26 16.4 0.9 31.6 12.7 83
sp. D37 77.5 0 7.1 3.0 160
sp. D48 p 0 p p 53
Scutellospora
calospora 223.4 4.4 16.3 40.2 154 NC 146
heterogama 1.0 P 1.3 1.0 165 NC 141
pellucida 23.5 0 21.1 3.0 330 NC 155
reticulata 1.6 0 2.4 7.9 237 NC 202

aVouchers available upon request to first author.

*Not included in statistical analysis because of missing data.
'Spores are red to orange in colour and 220-360 jim in diameter, with a wall structure resembling that of A. laevis
2Spores are lightly straw coloured and 50-70 ,um in diameter, with a spore wall and inner wall structure resembling that of A.
delicata.
3Spores are copper coloured and 130-230 ,um in diameter, with a textured surface and laminae of the spore wall which shatters
when squashed.
4Spores are straw coloured and 55-80 ,um in diameter, with spore wall structure resembling that of A. scrobiculata.
5Spores are white and 60-120 jim in diameter, with a spore wall structure which most closely resembles that of Gl. diaphanum
6Spores are often formed in sporocarps, red to red-orange in colour, and 60-90 jm in diameter, with a wall structure which
most closely resembles that of GI. etunicatum.
7Spores are tan in colour and 130-180 ,m in diameter, having an ephemeral outer wall covering a thick, robust laminate
wall.
8Spores are formed in sporocarps, yellow to red-brown in colour, and 40-60 ,um in diameter. This category is likely composed
of several separate sporocarpic species and for that reason was not included in analyses.

the transplant traps. Moreover, this method has the geners; bulk soil samples taken from under each plant
weakness that the quantity of the inoculum is not included the roots of other plant species.
controlled and is likely to depend on the plant species. The strongest evidence that the host-specific effects
The absence of a significant effect of plant species on seen in the microcosm were relevant to the field situ-
the distribution of spores in field soil and sorghum ation was the correlation between the patterns of host-
traps may not be surprising given our sampling tech- dependent differentiation observed in the microcosms
nique. Real differences in the fungal communities in and those found in the field (Fig. 3). While this cor-
the rhizospheres of the four plant species may have relation is weak, it does suggest that the process of
? 1996 British
Ecological Society, been obscured by the fact that these plants are grow- differential sporulation observed in the laboratory
Journal of Ecology, ing in close proximity and their root systems certainly may be active under natural conditions, though this
84, 71-82 overlap with each other as well as with other con- pattern was difficult to detect with our sampling tech-

This content downloaded from

170.150.25.187 on Sat, 25 Jun 2022 19:20:35 UTC
All use subject to https://about.jstor.org/terms
78 Table 4 Multivariate analysis of effect of plant species and sample location on spore numbers of AM fungi in field soil sam
Host-dependence as assessed using three methods. Spore counts were made for each sample and these counts were ranked within each fung
species; the analysis was then carried out on the ranks. The heading 'overall differences' presents tests for effects of treatment
of AMfungi
on total sporulation (considered across all fungal species). The heading 'interaction with the profile' presents tests for th
specific hypothesis that the relative rank of sporulation of the different fungal species depends upon the designated treatm
effects

Num. Den. Wilk's

Effect d.f. d.f. lambda F P

Overall differences
Sorghum trap cultures
plant species 42 16 1.0 x 10-' 1.4 ns
location 70 28 2.5 x 10-5 3.3 0.0004
Transplant trap cultures
plant species 18 2 1.4 x 108 92.5 0.001
location 45 8 1.6x 10-9 15.5 0.0003
Direct counts of field soil
plant species 45 10 9.0 x 10 2.1 ns
location 90 24 1.0 x 106 2.7 0.004
Interaction with profile
Sorghum trap cultures
plant species 39 18 1.5 x 10-3 1.5 ns
location 65 32 8.0x 10-5 3.1 0.0003
Transplant trap cultures
plant species 16 4 1.0 x 10-3 7.6 0.03
location 40 11 2.3 x 10' 1.8 ns
Direct counts of field soil
plant species 42 13 2.0 x 10' 2.1 0.08
location 84 29 2.5 x 106 3.1 0.0005

niques in a heterogeneous field. This host-dependence relation within a fungal species and generality of this
was seen across a wide range of plant and fungal correlation across fungal and host species is not
taxa, further suggesting that such host-dependence of known. Differences in sporulation, as observed in this
sporulation rates are of general importance among study, may therefore reflect differences in the growth
co-occurring plants and AM fungi. rates of fungal populations or differences in the allo-
To interpret the ecological consequences of the cation of resources to sporulation relative to hyphal
differential sporulation rates of the AM fungi extension. In this study, we tested sporulation at two
observed in this study, it is important to know if these
times of harvest, and while total spore number was
differences result in corresponding differential rates significantly different between the two harvests, the
of population growth. However, measuring the rate strong patterns of differentiation with respect to plant
of growth of a fungal population is problematical host were not significantly affected by time of harvest.
because fungal individuals are usually difficult to dis- This suggests that the differential sporulation rates
tinguish and observe directly (for a discussion of the are not simply due to timing of allocation of resources
difficulties for estimating growth rate of fungal geno- to sporulation, and supports the general expectation
types, see Antonovics & Alexander 1989). Most wor- that these sporulation rates are likely to be correlated
kers in epidemiology use the rate of production of with fungal population growth rates.
new infections/lesions per infected individual/lesion Host-dependence of fungal population growth
as a phenomenological measure of fungal spread; and rates would play an important role in maintaining
this measure itself is a function of the rate of pro- AM fungal species diversity whenever there is a rever-
duction of infectious propagules, the time to disease sal of the ranks of population growth rates on distinct
expression, and the duration of the infectious period. hosts (MacArthur 1972; Levin 1976; Levins 1979), as
However, in AM fungi, the rate of production of observed in our microcosm experiment. Furthermore,
infective propagules has been an elusive measurement if the relative rate of growth of a plant species is also
because new infections can result either from the ger- dependent upon the species of AM fungi with which
mination of spores (which are identifiable to species) it associates, as has frequently been observed (Nemec
or from the extension of hyphae from previously 1978; Powell et al. 1982; Ollivier et al. 1983; Wilson
infected roots (which cannot currently be dis- 1988; Ravnskov & Jakobsen 1995), then the resulting
tinguished taxonomically). While sporulation has dynamics may be complex, with alternate mechanisms
? 1996 British
Ecological Society, been shown to be correlated with fungal biomass at for the coexistence of fungal species (Bever 1992). The
Journal of Ecology, some level (Gazey et al. 1992; Jasper et al. 1993; power of these mechanisms for maintaining diversity
84, 71-82 Franke & Morton 1994), the linearity of this cor- is enhanced by the plant and fungal life histories.

This content downloaded from

170.150.25.187 on Sat, 25 Jun 2022 19:20:35 UTC
All use subject to https://about.jstor.org/terms
 20-
79 correlation by using systematic and consistent sam-
J.D. Bever et al. : 18 pling methods over a wide range of plant com-
munities.
g 16- . ..-
Host-dependent fungal sporulation may also
E 14- depend upon the season. The plant community at this

12 -
site is temporally divided into distinct warm and cool
season associations (Fowler & Antonovics 1981), with
10-
U
the four plant species used in this study being active
0 8 during the cool season. These experiments were per-
formed in a glasshouse under the seasonally appro-
<
6 6- I I l l I
priate cool-weather conditions, perhaps facilitating
the detection of host-dependent responses. Identi-
fication of similar responses in the field soil as in
22- the glasshouse experiments may have been further
facilitated by the sampling of the field soil in spring,
20-
at which time fungi active during the cool season may
,, 18 - be expected to sporulate. Seasonal dependence has
been observed in fungal sporulation (Gemma et al.
r 16 *
1989) and in the interaction of plants and AM fungi
a 14 - in another grassland site, a prairie in Kansas, with

12-
warm and cool season grasses showing their greatest
response to fungi at their respective optimal tem-
0 - peratures (Bentivenga & Hetrick 1991; Hetrick et al.

8 - 1994).
15 20 25 30 35 40 45 50 Host-dependent responses in other organisms have
Abundance in Experimental Microcosms
been found to vary between (Rank)
geographical isolates and
ecological circumstance
Fig. 3 Regression of ranks of number of spores (Fox & of
Morrow
AM 1981; Bur-
fungi
don 1987). from
in association with four plant hosts Geographical
(a)isolates of single species
sorghum trapsof
and (b) field samples against the corresponding ranks from AM fungi have been found to differ in their responses
an experimental microcosm. Each point represents the mean
to edaphic factors (Gilden & Tinker 1981; Stahl &
rank of spore numbers of a fungal species in association with
a plant species. The lines were derived by regression with
Smith 1984; Louis & Lim 1988; Boerner 1990; Stahl
P < 0.03 and P < 0.04 for the sorghum traps and field et al. 1990; Stahl & Christensen 1991), but ecotypic
specialization of AM fungi on host species has not
samples, respectively. Here, we only include the fungal spe-
cies which were found to have sporulated differentially been
on investigated. It is interesting that in contrast to
host-plant species in the microcosm experiment, although
this study, Sanders & Fitter (1992) did not find high
the results are similar if all fungal species are included.
sporulation rates of Scutellospora calospora on Plan-
tago lanceolata, in a seminatural grassland com-
munity in England. Distinct patterns in geo-
Because of the perennial habit of the plants (quite graphically distinct populations and differing
long relative to the potential rate of turnover of the ecological circumstances suggests that host-depen-
fungi), the fungi have a high probability of infecting dence may not be a species-level characteristic in AM
their optimal host plant in successive generations, fungi.
thereby expanding the range of conditions in which The sorghum traps, transplant traps, and direct
diversity may be maintained. Moreover, with long counts of spores in the field all showed evidence of
lived plants and limited fungal dispersal, host-depen- spatial structure in the AM fungal community at the
dence may maintain AM fungal diversity on a local relatively small scale of a few metres. Similar small-
scale. scale differentiation was observed in a detailed study
An expectation of the hypothesis that host-depen- of a sand dune by Friese & Koske (1991). Other inves-
dence of fungal population growth rates maintains tigations had previously documented differentiation
AM fungal species diversity is that species diversity of of the AM fungal community across broader scales
fungi will correlate with diversity of VA mycorrhizal (Walker et al. 1982; Hetrick & Bloom 1983; Sylvia
plant species. It may not be a coincidence that this 1986; Koske & Tews 1987; Koske 1987; Johnson et
study site not only has a high AM fungal species al. 1991 b). Such small scale differentiation may be
diversity, but also has a stable but diverse plant com-the result of local variation in edaphic factors or in
munity (Fowler & Antonovics 1981); many of the vegetation composition; such effects were not assessed
previous studies of AM fungal diversity have largely directly in the present study but previous inves-
? 1996 British
Ecological Society, focused on samples taken from single plant species or tigations in the same field had shown small scale vari-
Journal of Ecology, from agricultural settings where plant species diver- ation in both these features (Fowler & Antonovics
84, 71-82 sity is low. It would be interesting to test such a 1981; Antonovics et al. 1988; Moloney 1988).

This content downloaded from

170.150.25.187 on Sat, 25 Jun 2022 19:20:35 UTC
All use subject to https://about.jstor.org/terms
80 The combined results of the four experiments in temperature effects on mycorrhizal activity and depen-
Host-dependence this study indicate that caution must be exercised in dence of cool- and warm-season tallgrass prairie grasses.
Canadian Journal of Botany, 70, 1596-1602.
of AMfungi characterizing AM fungal species diversity. As has
Bever, J.D. (1992) Ecological and evolutionary dynamics
been noted previously (Anderson et al. 1983; Tews & between plants and their soil communities. PhD thesis,
Koske 1986), spores of fungal species have clumped Duke University.
distributions even within a very small area; therefore Bever, J.D. (1994) Feedback between plants and their soil

repeated sampling from apparently similar locations communities in an old field community. Ecology, 75,
1965-1977.
may yield different compositions of fungal species.
Boerner, R.E. (1990) Role of mycorrhizal fungus origin in
Furthermore, the host-dependent sporulation rates growth and nutrient uptake by Geranium robertianum.
indicate that the AM fungal diversity detected using American Journal of Botany, 77, 483-489.
trap cultures may depend upon the species of host Brandon, R. (1990) Adaptation and Environment. Princeton
University Press, Princeton.
plant used in the trap. Finally, a single trapping pro-
Broaddus, L.E. (1991) Natural selection on gynodioecy in
cedure is unlikely to capture all AM fungal species
Plantago lanceolata L. PhD thesis, Duke University.
present in the sample, and a diversity of approaches Brundrett, M. (1991) Mycorrhizas in natural ecosystems.
is necessary for more complete descriptions of AM Advances in Ecological Research, 21, 171-313.
fungal species diversity at a given location. Burdon, J.J. (1987) Diseases and Plant Population Biology.
Cambridge University Press, Cambridge.
Clay, K. (1982) Environmental and genetic determinants of
cleistogamy in a natural population of the grass Dan-
thonia spicata. Evolution, 36, 734-741.
Acknowledgements
Daft, M.J. & Hogarth, B.G. (1983) Competitive interactions
amongst four species of Glomus on maize and onion.
We thank S. P. Bentivenga, E. Macklin, S. Sturmer,
Transactions of the British Mycological Society, 80, 339-
K. Heldreth, A. Fitter, and two anonymous reviewers 345.
for helpful comments on versions of this manuscript. Daniels, B.A. & Skipper, H.D. (1982) Methods for the recov-
This work was made possible by USDA grant 92- ery and quantitative estimation of propagules from soil.
37101-7461 to J.D.B. and NSF grant DIR-9015519 Methods and Principles of Mycorrhizal Research (ed. N.
C. Schenck), pp. 29-35. American Phytopathological
to J.B.M.
Society, St. Paul.
Ellstrand, N.C. & Antonovics, J. (1985) Experimental stud-
ies of the evolutionary significance of sexual repro-
duction II. A test of the density-dependent selection
References hypothesis. Evolution, 39, 657-666.
Fowler, N.L. & Antonovics, J. (1981) Competition and
Alexander, H.M. (1984) Spatial patterns of disease induced coexistence in a North Carolina grassland. I. Patterns in
by Fusarium monofiliforme var. subglutinans in a popu- undisturbed vegetation. Journal of Ecology, 69, 825-841.
lation of Plantago lanceolata. Oecologia, 62, 141-143. Fox, L.R. & Morrow, P.A. (1981) Specialization: species
Allen, E.B. & Allen, M.F. (1990) The mediation of com- property or local phenomenon? Science, 211, 887-893.
petition by mycorrhizae in successional and patchy Franke, M. & Morton, J.B. (1994) Ontogenetic comparisons
environments. Perspectives on Plant Competition (eds J. of arbuscular mycorrhizal fungi Scutellospora het-
B. Grace & D. Tilman), pp. 367-390. Academic Press, erogama and Scutellospora pellucida: revision of taxo-
San Diego. nomic character concepts, species descriptions, and
Allen, M.F. (1991) The Ecology of Mycorrhizae. Cambridge phylogenetic hypothesis. Canadian Journal of Botany,
University Press, New York. 72, 122-134.
Anderson, R.C., Liberta, A.E., Dickman, L.A. & Katz, A.J. Friese, C.F. & Koske, R.E. (1991) The spatial dispersion of
(1983) Spatial variation in vesicular-arbuscular spores of vesicular-arbuscular mycorrhizal fungi in a
mycorrhiza spore density. Bulletin of the Torrey Bot- sand dune: microscale patterns associated with the root
anical Club, 110, 519-525. architecture of American beachgrass. Mycological
Antonovics, J. & Alexander, H.M. (1989) The concept of Research, 95, 952-957.
fitness in plant-fungal pathogen systems. Plant Disease Gazey, C., Abbott, L.K. & Robson, A.D. (1992) The rate of
Epidemiology, Vol. 2. Genetics, Resistance and Man- development of mycorrhizas affects the onset of sporu-
agement (eds K. J. Leonard & W. E. Fry), pp. 185-214. lation and production of external hyphae by two species
McGraw-Hill, New York. of Acaulospora. Mycological Research, 96, 643-650.
Antonovics, J. & Ellstrand, N.C. (1986) The fate of dispersedGemma, J.N., Koske, R.E. & Carreiro, M. (1989) Seasonal
progeny: experimental studies with Anthoxanthum. Gen- dynamics of selected species of V-A mycorrhizal fungi
etic Differentiation andDispersal in Plants (eds Jacquard, in a sand dune. Mycological Research, 92, 317-321.
P, Heim, G. & Antonovics, J.), pp. 369-381. Springer- Gilden, A. & Tinker, P.B. (1981) A heavy-metal-tolerant
Verlag, Berlin. strain of a mycorrhizal fungus. Transactions of the Brit-
Antonovics, J. & Primack, R. (1982) Experimental ecological ish Mycological Society, 77, 648-649.
genetics in Plantago. VI. The demography of seedling Hartnett, D.C., Hetrick, B.A.D., Wilson, G.W.T. & Gibson,
transplants of P. lanceolata. Journal of Ecology, 70, 55- D.J. (1993) Mycorrhizal influence on intra- and inter-
75. specific neighbour interactions among co-occurring
Antonovics, J., Ellstrand, N.C. & Brandon, R.N. (1988) prairie grasses. Journal of Ecology, 81, 787-795.
Genetic variation and environmental variation: expec- Hendrix, J.W., Guo, B.Z. & An, S.Q. (1995) Divergence of
? 1996 British tations and experiments. Plant Evolutionary Biology (eds mycorrhizal communities in crop production systems.
Ecological Society, Gottlieb, L. D. & Jain, S. K.), pp. 275-303. Chapman Plant and Soil, 170, 131-140.
Journal of Ecology, and Hall, London. Hetrick, B.A.D. & Bloom, J. (1983) Vesicular-arbuscular
84, 71-82 Bentivenga, S.P. & Hetrick, B.A.D. (1991) Seasonal and mycorrhizal fungi associated with native tall grass

This content downloaded from

170.150.25.187 on Sat, 25 Jun 2022 19:20:35 UTC
All use subject to https://about.jstor.org/terms
81 prairie and cultivated winter wheat. Canadian Journal soilless production of vesicular-arbuscular mycorrhizal
of Botany, 61, 2140-2146. fungi. Mycorrhiza, 2, 9-15.
J.D. Bever et al.
Hetrick, B.A.D. & Bloom, J. (1986) The influence of host Moloney, K.A. (1988) Fine-scale spatial and temporal vari-
plant production and colonization ability of vesicular- ation in the demography of a perennial bunchgrass.
arbuscular mycorrhizal spores. Mycologia, 78, 32-36. Ecology, 69, 1588-1598.
Hetrick, B.A.D., Wilson, G.W.T. & Schwab, A.P. (1994) Morrison, D.F. (1976) Multivariate Statistical Methods.
Mycorrhizal activity in warm- and cool-season grasses: McGraw-Hill Book Co., New York.
variation in nutrient-uptake strategies. Canadian Journal Morton, J.B. (1990) Species and clones of arbuscular
of Botany, 72, 1002-1008. mycorrhizal fungi (Glomales, Zygomycetes); their role
Hung, L.L. & Sylvia, D.M. (1988) Production of vesicular- in macro- and microevolutionary processes. Mycotaxon,
arbuscular mycorrhizal fungus inoculum in aeroponic 37, 493-515.
culture. Applied and Environmental Microbiology, 54, Morton, J.B., Bentivenga, S.P. & Bever, J.D. (1995)
353-357. Discovery, measurement, and interpretation of diversity
Jasper, D.A., Abbott, L.K. & Robson, A.D. (1993) The in symbiotic endomycorrhizal fungi (Glomales, Zyg-
survival of infective hyphae of vesicular-arbuscular omycetes). Canadian Journal of Botany, in press.
mycorrhizal fungi in dry soil: an interaction with sporu- Morton, J.B., Bentivenga, S.P. & Wheeler, W.W. (1993)
lation. New Phytologist, 124, 473-479. Germ plasm in the International Collection of Arbu-
Johnson, N.C., Pfleger, F.L., Crookston, R.K., Simmons, scular and Vesicular-Arbuscular Mycorrhizal Fungi
S.R. & Copeland, P.J. (199la) Vesicular-arbuscular (INVAM) and procedures for culture development,
mycorrhizas respond to corn and soybean cropping documentation, and storage. Mycotaxon, 48, 491-
history. New Phytologist, 117, 657-663. 528.

Johnson, N.C., Tilman, D. & Wedin, D. (1992) Plant and Nemec, S. (1978) Response of six citrus rootstocks to three
soil controls on mycorrhizal fungal communities. Ecol- species of Glomus, a mycorrhizal fungus. Proceedings of
ogy, 73, 2034-2042. the Florida State Horticulture Society, 91, 10-14.
Johnson, N.C., Zak, D.R., Tilman, D. & Pfleger, F.L. Ollivier, B., Bertheau, Y., Heim, H.G. & Gianinazzi-Pear-
(1991 b) Dynamics of vesicular-arbuscular mycorrhizae son, V. (1983) Influence de la variete de Vigna ungu-
during old field succession. Oecologia, 86, 349-358. iculata dans l'expression de trois associations endo-
Kelley, S.E. (1989) Experimental studies of the evolutionary mycorrhiziennes a vesicules et arbuscules. Canadian
significance of sexual reproduction. V. A field test of the Journal of Botany, 61, 354-358.

sib-competition lottery hypothesis. Evolution, 43, 1054- Powell, C. Ll., Clarke, G.E. & Verberne, N.J. (1982) Growth
1065. response of four onion cultivars to isolates of VA
Koomen, I., Grace, C. & Hayman, D.S. (1987) Effectiveness mycorrhizal fungi. New Zealand Journal of Agricultural
Research, 25, 465-470.
of single and multiple mycorrhizal inocula on growth of
clover and strawberry plants at two pH's. Soil Biology Ravnskov, S. & Jakobsen, I. (1995) Functional compatibility
and Biochemistry, 19, 539-544. in arbuscular mycorrhizas measured as hyphal P trans-

Koske, R.E. & Tews, L.L. (1987) Vesicular-arbuscular port to the plant. New Phytologist, 129, 611-618.
Ronsheim, M.L. (1992) The relative advantages of sexual and
mycorrhizal fungi of Wisconsin sandy soils. Mycologia,
79, 901-905. asexual reproduction in Allium vineale. PhD thesis, Duke
University.
Koske, R.E. (1981) A preliminary study of interactions
St. John, T.V. & Koske, R.E. (1988) Statistical treatment of
between species of vesicular-arbuscular fungi in a sand
endogonaceous spore counts. Transactions of the British
dune. Transactions of the British Mycological Society,
Mycological Society, 91, 117-121.
76, 411-416.
SAS (1986) SAS System for Linear Models. SAS Institute,
Koske, R.E. (1987) Distribution of VA mycorrhizal fungi
Cary, NC.
along a latitudinal temperature gradient. Mycologia, 79,
Sanders, I.R. & Fitter, A.H. (1992) Evidence for differential
55-68.
responses between host-fungus combinations of ves-
Law, R. & Lewis, D.H. (1983) Biotic environments and the
icular-arbuscular mycorrhizas from a grassland. Myco-
maintenance of sex - some evidence from mutualistic
logical Research, 96, 415-419.
symbioses. Biological Journal of the Linnean Society, 20,
Schenck, N.C. & Kinloch, R.A. (1980) Incidence of
249-276.
mycorrhizal fungi on six field crops in monoculture on
Levin, S.A. (1976) Population dynamic models in het-
a newly cleared woodland site. Mycologia, 72, 445-456.
erogeneous environments. Annual Review of Ecology and
Schenck, N.C. & Perez, Y. (1990) Manual for the Identi-
Systematics, 7, 287-3 10.
fication of VA Mycorrhizal Fungi. Synergistic Publi-
Levins, R. (1979) Coexistence in a variable environment.
cations, Gainesville.
American Naturalist, 114, 765-783.
Schmitt, J. & Antonovics, J. (1986) Experimental studies of
Louis, L. & Lim, G. (1988) Differential response in growth
the evolutionary significance of sexual reproduction. IV.
and mycorrhizal colonisation of soybean to inoculation
Effect of neighbor relatedness and aphid infestation on
with two isolates of Glomus clarum in soils of different
seedling performance. Evolution, 40, 830-836.
P availability. Plant and Soil, 112, 37-43.
Stahl, P.D. & Christensen, M. (1991) Population variation
MacArthur, R.H. (1972) Geographical Ecology. Harper and in the mycorrhizal fungus Glomus mosseae: breadth of
Row, New York.
environmental tolerance. Mycological Research, 95,
Miller, M.A. & Jastrow, J.D. (1994) Vesicular-arbuscular 300-307.
mycorrhizae and biogeochemical cycling. Mycorrhizae Stahl, P.D. & Smith, W.K. (1984) Effects of different geo-
in Plant Health (eds F. L. Pfleger & R. G. Linderman), graphic isolates of Glomus on the water relations of
pp. 189-212. American Phytopathological Society, St. Agropyron smithii. Mycologia, 76, 261-267.
Paul. Stahl, P.D., Christensen, M & S.E. Williams. (1990) Popu-
Miller, MA. (1987) The ecology of vesicular-arbuscular lation variation in the mycorrhizal fungus Glomus mos-
? 1996 British mycorrhizae in grass- and shrublands. Ecophysiology of seae: uniform garden experiments. Mycological
Ecological Society, VA Mycorrhizal Plants (eds G. R. Safir), pp. 135-170. Research, 94, 1070-1076.
Journal of Ecology, CRC Press, Boca Raton, FL. Sylvia, D.M. (1986) Spatial and temporal distribution of
84, 71-82 Millner, P.D. & Kitt, D.G. (1992) The Beltsville method for vesicular-arbuscular mycorrhizal fungi associated with

This content downloaded from

170.150.25.187 on Sat, 25 Jun 2022 19:20:35 UTC
All use subject to https://about.jstor.org/terms
82 Uniola paniculata in Florida foredunes. Mycologia, 78, Wilson, D.O. (1988) Differential plant responses to inocu-
728-734. lation with two VA mycorrhizal fungi isolated from a
Host-dependence
Teramura, A., Antonovics, J. & Strain, B.R. (1981) Exper- low-pH soil. Plant and Soil, 110, 69-75.
of AMfungi
imental ecological genetics in Plantago. IV. Effects of Wu, L. & Antonovics., J. (1975) Experimental ecological
temperature on growth rates and reproduction in three genetics in Plantago. I. Induction of roots and shoots on
populations of Plantago lanceolata, L. (Plantiginaceae). leaves for large scale vegetative propagation and metal
American Journal of Botany, 68, 425-434. tolerance testing in P. lanceolata. New Phytologist, 75,
Tews, L.L. & Koske, R.E. (1986) Toward a sampling strat- 277-282.
egy for vesicular-arbuscular mycorrhizas. Transactions
of the British Mycological Society, 87, 353-358.
Walker, C., Mize, C.W. & McNabb, H.S. (1982) Populations
of endogonaceous fungi at two locations in central Iowa. Received 9 November 1994
Canadian Journal of Botany, 60, 2518-2529. revised version accepted 31 July 1995

? 1996 British
Ecological Society,
Journal of Ecology,
84, 71-82

This content downloaded from

170.150.25.187 on Sat, 25 Jun 2022 19:20:35 UTC
All use subject to https://about.jstor.org/terms