1
Little evidence that farmers should consider abundance or diversity of
arbuscular mycorrhizal fungi when managing crops
Megan H. Ryan1, James H. Graham2
1School of Agriculture and Environment and Institute of Agriculture, The University of Western
Australia, 35 Stirling Highway, Crawley, WA 6009, Australia
2Department of Soil and Water Sciences, Citrus Research and Education Center, University of Florida,
Lake Alfred, FL 33850, USA
Corresponding author: Megan Ryan: [email protected]
2
Summary
Arbuscular mycorrhizal fungi (AMF) are ubiquitous in agroecosystems and often stated to be
critical for crop yield and agroecosystem sustainability. But, should farmers modify
management to enhance abundance and diversity of AMF? We address this question with a
focus on field experiments that manipulated colonisation by indigenous AMF and report crop
yield, or investigated community structure and diversity of AMF. We find the literature
presents an overly optimistic view of the importance of AMF in crop yield due, in part, to
flawed methodology in field experiments. A small body of rigorous research only sometimes
reports a positive impact of high colonisation on crop yield, even under phosphorus
limitation. We suggest that studies vary due to the interaction of environment and genotype
(crop and mycorrhizal fungal). We also find that the literature can be overly pessimistic about
the impact of some common agricultural practices on mycorrhizal fungal communities and
that interactions between AMF and soil microbes are complex and poorly understood. We
provide a template for future field experiments and a list of research priorities, including
phosphorus-efficient agroecosystems. However, we conclude that management of AMF by
farmers will not be warranted until benefits are demonstrated at the field scale under
prescribed agronomic management.
Key words: agronomy, crop sequences, experimental methodology, phosphorus-efficient
agroecosystems, trade-balance model, soil microbes, yield
3
Contents
Summary 2
I. Introduction 4
II. Investigating activity of AMF in agroecosystems 5
III. Crop benefit from AMF: agronomic and mycorrhizal literature differ 6
IV. Flawed methodology leads to benefits of mycorrhizas being overstated 7
V. Rigorous methodology suggests low colonisation by AMF can sometimes reduce crop yield 10
VI. Predicting when mycorrhizas matter for crop yield 12
VII. Crop genotype 14
VIII. Fungal genotype 16
IX. Complex interactions between the mycorrhizal fungal and soil microbial communities 22
X. Phosphorus-efficient agroecosystems 22
XI. Conclusions 24
Acknowledgements 24
Author contributions 25
References 25
4
I. Introduction
Arbuscular mycorrhizal fungi (AMF) provide benefits to host plants (Smith & Read 2008)
and show functional diversity assumed to have implications for agroecosystem management
(Verbruggen & Kiers, 2010). Their ability to enhance plant uptake of relatively immobile
nutrients, particularly phosphorus (P), and several micronutrients, is their most-recognised
benefit. However, AMF may also potentially improve soil health through their external
hyphae that sustain the soil food web (Newsham et al., 1995), aid soil structure maintenance
(Miller & Jastrow, 1990), and provide a large number of other benefits to host plants such as
pathogen and herbivory protection, alleviation of water stress, enhanced tolerance of salinity,
low pH and heavy metals, and biofortification of grain with micronutrients.
AMF have received increasing attention as part of a popular paradigm that considers an
abundant, active and diverse community of AMF as essential for agroecosystem
sustainability (see overview of this approach in Verbruggen et al., 2012). However, when we
previously reviewed the literature on the role of AMF in commercial agroecosystems (Ryan
& Graham, 2002) we concluded that AMF did not play a vital role in the nutrition and growth
of crops. Here, 16 years later, we review the subsequent literature. To ensure relevance to
commercial agroecosystems, we focus on field experiments that manipulate indigenous AMF
and report crop yield; the outcome of most immediate relevance to farmers. We also explore
the impact of variation in mycorrhizal fungal community structure and diversity in
agroecosystems. We rarely refer to the field inoculation literature as it was recently reviewed
(Hart et al., 2018) and inoculation is little used in commercial agroecosystems in spite of
some reports of yield benefits (e.g. Pellegrino & Bedini, 2014). We primarily consider
broadacre field crops in the high-yielding systems in Europe, North America, China and
Australia. However, many of our examples and conclusions will be relevant to other systems.
Many studies we review involve wheat (Triticum aestivum); this reflects its dominance in the
field-based literature on AMF and it being the only crop subject to meta-analysis of field data
(Pellegrino et al., 2015). Overall, we use this review to ask if there is sufficient evidence to
recommend farmers consider impact on AMF when making management decisions and to
suggest high priority areas for future research.
5
II. Investigating activity of AMF in agroecosystems
Quantifying abundance of AMF
Measuring the activity of AMF in agroecosystems, particularly impact on host plant yield, is
complicated by the difficulties with producing non-colonised control plants. Comparisons are
therefore mostly made amongst treatments differing in the abundance of AMF. This
approach cannot quantify the overall contribution of AMF to crop yield, but it does allow us
to ask whether farmers should consider AMF in their farm management. Abundance is most
often quantified as percentage of root length colonised and this is the measurement most
often used in meta-analyses of the impact of AMF on plant growth (e.g. Lekburg & Koide,
2005). However, differences among treatments or genotypes in percentage of root length
colonised may not reflect provision of nutrients by AMF to the host plant (e.g. Kaeppler et
al., 2000) due to variation in: root growth; the proportion of colonisation that consists of
arbuscules; the density of colonisation across the root; the density of external hyphae, and;
other variables that impact the activity of AMF such as microbial community. Other
measures could better reflect mycorrhizal fungal activity including percentage of root length
containing arbuscules, total length of colonised root and density of external hyphae (Kabir et
al., 1998; Sawers et al., 2017). However, the studies we review generally report only
percentage of root length colonised, except Ren et al. (2018) and Mai et al. (2018) who report
only external hyphal density. It is important that future studies report both, if feasible. Note
that measuring external hyphal density with confidence depends upon an ability to distinguish
living hyphae of AMF from dead hyphae and hyphae of other soil fungi (e.g. Gavito et al.
2003) and that the time required for this work may be prohibitive. More knowledge is also
likely required on the optimal sample size and sampling location (i.e., soil depth and
placement relative to crop rows) for collecting external hyphae in the field.
Can results of glasshouse experiments be extrapolated to the field?
In comparison with field experiments, it is far simpler to measure activity of AMF in
glasshouse experiments where non-colonised controls are easily established, total root length
quantifiable and management costs low. Many useful insights can be gained from glasshouse
experiments, particularly into the physiology of the symbiosis (Smith & Read, 2008) and
many aspects of physiology may be impossible, with current experimental techniques, to
investigate under field conditions. However, we believe that any conclusions about the
benefits of AMF for crop growth and yield, that is, the mycorrhizal growth response (MGR),
require verification under field conditions across a range of environments, as is routine in
6
other areas of agronomic research such as crop breeding (Dreccer et al., 2018) and crop
rotation (Angus et al., 2015). Note that MGR is defined here as the difference in growth
between colonised plants and non-colonised plants - or plants with high and low colonisation
- at a defined level of resource supply to the plant (Janos, 2007; Johnson et al., 2015). The
reasons why results from glasshouse experiments may not be applicable to the field involve:
environmental extremes and variation being minimised or absent (e.g. no very cold night
temperatures, no occasional periods of high or low water availability, homogenous sieved soil
and absence of a soil profile); the impact of soil sterilisation treatments on soil properties and
soil microbes; nutrients being manipulated to ensure only P is limiting plant growth; absence
of sward conditions (Jeffery et al., 2017) or unrealistic plant densities; the impact of
constraining roots within pots on rooting depth, root density and root system architecture; the
frequent watering required for plants in pots, and; the impact of pots on soil temperature
gradients (see Poorter et al., 2012). As we are interested in agronomic relevance, in this
review we focus whenever possible on field experiments.
III. Crop benefit from AMF: agronomic and mycorrhizal literature differ
Mycorrhizal literature suggests a need to manage AMF to increase crop yield
A crop yield benefit from high colonisation by AMF is a core claim of researchers of AMF.
Smith & Read (2008) state “there are many instances where crop productivity is influenced
by AM symbiosis”, Martín-Robles et al. (2018) note the “global importance of AM in
agriculture”, Bakhshandeh et al. (2017) say that AMF “may play a key role in crop rotation
systems decreasing the dependency on fertilizers” and Srivastava et al. (2017) claim “AMF
could reduce up to 50% usage of chemical fertilizers for optimal agriculture production”. An
optimistic view is nearly universally maintained, even when Thirkell et al. (2017) note that
colonisation does not necessarily translate to enhanced yield, their paper title asks “are
mycorrhizal fungi our sustainable saviours?”
Crop agronomy literature suggests no need to manage AMF
The crop agronomy literature rarely is focussed on AMF. However, this large literature can
provide insight into the need to consider AMF when making management decisions. For
instance, both short (< 12 months) and long (> 12 months) bare plant-free fallows and non-
mycorrhizal crops may often reduce the level of colonisation by AMF in following crops
(Thompson, 1987; Lekburg & Koide, 2005; Bowles et al., 2016). Thus, if high colonisation
provides a yield benefit, yield penalties should occur for crops following fallow or non-
7
mycorrhizal crops. This issue can be explored through the meta-analysis of Angus et al.
(2015) which was based on > 900 comparisons from around the world and quantified the
impact of non-cereal “break crops” on the yield of following wheat. They found no difference
between the yield benefit from mycorrhizal legumes - such as field peas (Pisum sativum) and
faba beans (Vicia faba) – and the yield benefit of non-mycorrhizal lupins (Lupinus
angustifolius). There was also no difference between non-mycorrhizal brassicas - such as
canola (Brasica napus) - and mycorrhizal flax (Linum usitatissimum). Angus et al. (2015)
suggest that reduced colonisation following non-mycorrhizal break crops may acually benefit
yield of following wheat. Similarly, in Western Australia, the results of 167 field experiments
showed the highest yield benefit for wheat after a break crop was for non-mycorrhizal lupins
(0.60 t ha-1), followed by field pea, canola, oats (Avena sativa) and long bare fallow (0.30 t
ha-1) (Seymour et al., 2012). Thus, these two meta-analyses contrast with the mycorrhizal
literature as they do not suggest that wheat routinely suffers a yield penalty if colonisation by
AMF is reduced. However, crop rotation effects involve complex changes in many factors
(soil nutrients, soil structure, root and shoot pathogens, soil microbes, etc.; Angus et al. 2015)
and the majority of studies included in these two meta-analyses were likely of crops with no
P limitation, and hence little chance of a large mycorrhizal benefit. We therefore present case
studies of wheat, and other crops, where these other factors were carefully controlled (or at
least considered) and where crop P status was known and, sometimes, deficient (Section V).
However, we first consider whether the mycorrhizal literature projects a realistic expectation
of mycorrhizal benefit.
IV. Flawed methodology leads to benefits of mycorrhizas being overstated
Meta-analyses
As meta-analyses identify broad trends they can strongly influence the literature, it is
imperative that they be rigorously conducted and reviewed. We critique two meta-analyses.
Lekberg & Koide (2005) compiled 290 field and glasshouse experiments that reported the
impact of agricultural practices, including inoculation, on plant growth. They reported that
increased colonisation increased average yields in the field by 23%. They stated that their aim
was to determine the effects of various agricultural practices on colonisation and yield. Yet,
only a small number of field studies were included (~24) and some of these were conducted
in soils of little general agricultural relevance, for instance, reclaimed mined soils in Australia
(Noyd et al., 1996) and eroded soil in China (Wu et al., 2002). Also, some studies involved
8
perennial plants and pre-inoculated transplants: treatments of little relevance to broadacre
agriculture.
Pellegrino et al. (2015) found a “strong significant” relationship between the level of
colonisation by AMF and wheat (Triticum aestivum) yields in the field and concluded that “a
high mycorrhizal infection potential of agricultural soil should be ensured” through
appropriate management. However, they considered only 16 studies where colonisation by
indigenous AMF was manipulated. Four of these did not provide grain yields and four
compared treatments that confounded the impact of changes in colonisation level with
changes in other factors likely to greatly impact yield including P fertiliser addition and
agroecosystems differing significantly in P availability (where increasing P is related to
increasing yield and decreasing colonisation) (Ryan et al., 2004; Covacevich et al., 2007;
Teng et al., 2013) and cultivar (Kirk et al., 2011). Of the remaining studies, one had
treatments that did not differ in colonisation (Germida & Walley, 1996). The remaining seven
studies involved a mixture of tillage and residue retention treatments (Duan et al., 2010) (no
correlation between AMF and yield), four farming system treatments (Hildermann et al.,
2010), crop rotation treatments (Karasawa & Takebe, 2012) (positive correlation), tillage and
crop rotation treatments (Gao et al., 2010) (no correlation), and three crop rotation studies in
southern Australia (Ryan et al., 2002; Ryan & Angus, 2003; Ryan et al., 2008) (negative or
no correlation). We note several concerns with this meta-analysis including: the paucity of
studies with valid methodology for determining the consequences of differing levels of
colonisation by AMF for crop yield; the failure to question whether the co-occurrence of high
colonisation and high yield always resulted from a strong empirical relationship, and; failure
to acknowledge that a significant proportion of studies showed no, or a negative, relationship
between colonisation level and yield.
Field experiments
Review papers and meta-analyses rely on the quality of source information. As noted by
Thirkell et al. (2017) and Lekberg & Helgason (2018), data from field experimentation on
AMF are notably lacking. Lekberg & Helgason (2018) attribute this to the difficulties with
inducing variation in abundance of AMF, without changing other variables, and the logistics
of field research. Furthermore, we suggest that field studies in agroecosystems have often
lacked sufficient rigour for the co-occurrence of high colonisation and increased yields to be
confidently assigned as causative. In part, this may reflect sampling of experiments not
9
designed to assess activity of AMF. However, it also reflects a lack of consideration for the
many other factors that can vary among treatments in agronomic field experiments, especially
crop sequence treatments.
These problems are illustrated by McGonigle et al. (2011) who concluded that “rotation of
flax after canola should be avoided” based on poor early season flax growth after canola,
compared with after wheat, and its association with a small decrease in colonisation (from
~45% to 35% of root length containing arbuscules) and reduced uptake of copper, P and zinc
(Zn) (Manitoba, Canada). They did not report yield. They also did not consider other factors
that could negatively affect flax growth following canola including nutrient removal by
canola from the nutrient-poor soil. Canola is nutrient-rich, and may have higher nutrient
removal in grain than wheat, even when yields are lower (Ryan & Kirkegaard, 2012).
Similar problems are highlighted by Bakhshandeh et al. (2017) who sampled wheat following
chickpea (Cicer arietinum) and canola (northern New South Wales – NSW, Australia). Soil
bicarbonate-extractable P was variable, but no lower than 28 mg kg-1; addition of P fertilizer
did not increase yield. Bakhshandeh et al. (2017) state an expectation of finding colonisation
by AMF positively related to wheat grain yield, but do not consider or cite any of the relevant
local literature (e.g. Ryan & Angus 2003 or Owen et al., 2010). They report high wheat yield
following chickpea and ascribe it to higher colonisation based on a weak positive correlation
between grain yield and the percentage of root length containing arbuscules (R2=0.18-0.20).
However, the study did not report the standard set of parameters that agronomists use to
determine the key variables influencing crop yield in rotation experiments. For instance, the
most likely reason for a yield benefit when wheat follows chickpea rather than canola is
greater soil mineral nitrogen (N) and water in the soil profile (Felton et al., 1997) due to
biological N-fixation by chickpea and its shallower roots and a shorter growing season.
Bakhshandeh et al. (2017) only report soil mineral N ~3 months after sowing the wheat and
did not identify that the extremely low values (< 4.2 mg kg–1, KCl-extracted) indicated most
mineral N had been taken up by the wheat and hence differences at sowing had been
translated into crop biomass and N content (not reported). They did not report soil water or
consider a major crop pathogen in the region, crown rot (Fusarium pseudograminearum)
(Felton et al., 1997; Kirkegaard et al., 2004); greater Zn removal by the canola was also not
considered. As Bakhshandeh et al. (2017) did not measure key explanatory variables or place
their research in the context of the local literature, the most plausible explanations for
10
enhanced yield of wheat following chickpea were not explored and it is likely that their
conclusions about the role of AMF in crop yields are incorrect.
A need to critically examine co-occurrence of low colonisation and poor growth/yield
When high colonisation by AMF coincides with high yield, it is generally assumed that the
high colonisiation was necessary for the high yield. However, there are examples where this
has been shown to not be the case. For instance, a cotton (Gossypium hirsutum) growth
disorder was characterised by stunted growth, low plant P content, symptoms of Zn
deficiency and low colonisation by AMF (Nehl et al., 1996) (northern NSW). However, the
disorder was also associated with root browning, and bioassays showed low levels of
mycorrhizal fungal inoculum were not the cause of the low colonisation (Nehl et al., 1998).
Thus, low colonisation was a consequence of the growth disorder: the cause was not
determined. Similarly, on a soil with “optimum” availability of P, maize (Zea mays) growth
was up to 3-fold greater following soybean (Glycine max) than following canola while
colonisation by AMF was around 65% of root length following canola and 85% following
soybean (Koide & Peoples, 2012) (Pennsylvania USA). However, the higher yields following
soybean were not ascribed to the higher colonisation due to no evidence of yield being driven
by crop P or N nutrition. For instance, for a given colonisation level, maize growth following
soybean varied nearly 4-fold suggesting other patchy growth-limiting factors. These could
not be identified, but may have included allelopathic effects from unusually abundant canola
residues due to pod shattering.
In summary, we suggest that some researchers have not been sufficiently critical in
interpreting field experiment results and this, along with failure to apply agronomic context
and methodology, has often resulted in overstating of the importance of high colonisation by
AMF for crop yield.
V. Rigorous methodology suggests low colonisation by AMF can sometimes
reduce crop yield
Grant et al. (2009) examined the impact of AMF on flax in 2-year cropping sequences,
established in three consecutive years, at two sites with low soil P (< 13 mg kg-1 bicarbonate
extractable P) (Manitoba, Canada). The sequences had canola and wheat in the first year and
flax in the second (and also included P fertiliser and tillage treatments). While P fertiliser had
little impact on flax growth or yield, flax establishment, early season growth, and yield were
11
all greater after wheat; by an average of 14% for yield. Colonisation by AMF, five weeks
after seeding, was greater following wheat by an average of 3.5% points, but this would have
been magnified in total colonised root length as flax root biomass to 30 cm soil depth was at
least 2-fold greater following wheat (Monreal et al., 2011). Grant et al. (2009) suggest lower
flax yield following canola could have resulted from decreased colonisation, allelopathy from
residues or early season competition from canola volunteers. Overall, in spite of this being an
extensive study with a comprehensive sampling regime, the role of AMF in crop yield
remains unclear. The absence of a P fertiliser response suggests P uptake by AMF was not
the mechanism of any mycorrhizal benefit. The reason for the large flax root systems
following wheat is unknown and alone could account for increased yield through greater soil
exploration.
An agronomic study specifically designed to investigate the role of AMF in wheat growth
was undertaken on a low P soil (11 mg kg-1 bicarbonate – Colwell – extractable P) by Ryan
& Angus (2003) (southern NSW). Five first year treatments manipulated the level of
mycorrhizal fungal inoculum (Figure 1a). In the second year, wheat, field peas and non-
mycorrhizal canola were grown with and without P fertiliser (Figure 1b): canola was a
control to identify crop sequence effects not due to AMF and field peas were assumed to have
high MGR as reported for other grain legumes (e.g. Pellegrino & Bedini 2014). The design
was intended to maximise the chances of showing a positive impact of high colonisation by
AMF on crop yield, but none was evident even under P limitation (Figure 1c). Other studies
in this region are consistent with this result (Kirkegaard & Ryan 2014) including studies
which show a positive impact of canola on following wheat (Kirkegaard et al., 1997).
Interestingly, roots at the site shown in Figure 1a were colonised by both AMF and the
arbuscule-forming symbiont ‘fine root endophyte’ (Ryan & Kirkegaard, 2012).
In contrast to the findings of Ryan & Angus (2003), a field experiment on a very low P soil
(7.1 mg kg-1 bicarbonate extractable P) by Owen et al. (2010) found a positive impact of high
colonisation on crop growth (southern Queensland, Australia). Wheat was sown following
long bare fallow and canola, as well as a range of mycorrhizal crops. There was no impact of
first year treatment on soil water, soil nitrate-N, P, or Zn (0-30 cm) or the pathogen
Pratylenchus thornei. Colonisation was higher following mycorrhizal crops (up to 80% root
length) than following fallow and most canola varieties (15–39%). Low colonisation was
associated with shoot biomass after ear emergence being 2–3-fold lower and consequently
12
there was a strong positive linear correlation between biomass and colonisation (R2=0.75,
n=15). Colonisation also correlated strongly with heads per m2 (R2=0.71), but less strongly
with yield (R2=0.40), likely due to water limiting yield. These findings are consistent with
reports in this region of crops following long bare fallows sometimes showing poor growth
and symptoms of P and Zn deficiency, as well as low colonisation by AMF (‘long fallow
disorder’) (Thompson, 1987).
The above case studies demonstrate the challenges in agronomic experiments with
determining the factors responsible for yield variation among treatments and with
apportioning any impact to the level of colonisation by AMF. Application of standard
agronomic methodology would ensure high-quality results, allow rigorous comparison among
studies and, ultimately, enable identification of the factors that cause the impact of
colonisation level on yield to vary among studies. We therefore provide a template for this
purpose (Table 1).
VI. Predicting when mycorrhizas matter for crop yield
Mycorrhizal literature: trade-balance model
In the mycorrhizal literature, based on studies largely undertaken in the glasshouse, it has
been proposed that the conditions under which AMF overcome resource limitations may be
key to understanding variations in MGR (Hoeksema et al., 2010; Johnson et al., 2015). In this
context, there may be a need to consider nutrients other than P as while enhanced P uptake is
the most reported plant nutritional benefit from AMF, uptake of other nutrients may also be
aided especially under varying field conditions (Lekberg & Koide, 2014; van der Heijden et
al., 2008). For instance, N uptake through external hyphae can sometimes be significant
(Johansen et al., 1992; Hodge & Fitter, 2010; Fellbaum et al., 2012); although whether it can
be sufficient to be of agronomic relevance remains to be proven. In addition, the availability
of plant carbon (C) to support the symbiosis and the strength of the fungal C drain will also
impact MGR (Graham and Abbott 2000). Thus, the relative supply of P, N, other nutrients
(perhaps), rates of C supply from photosynthesis, and diversion of C to AMF, and factors that
affect crop growth rate such as the levels of light and water availability likely together
determine MGR and, thereby, result in a range of “mycorrhizal phenotypes” from mutualism
to commensalism and parasitism (Johnson et al., 1997, 2015; Johnson & Graham, 2013). This
concept has been termed the “trade-balance model” (Johnson et al., 2010).
13
Johnson et al. (2010) identified the most significant benefits for MGR under P limitation with
N and light not limiting. However, competition for N between AMF and host plants may also
be important (see Hodge & Fitter, 2010; Hodge & Storer, 2015; Johnson et al., 2015). For
instance, Thirkell et al. (2016) demonstrated that organic N could elicit “strong mutualism”
whereby both plants and AMF benefit from the addition of N in a P-limited system.
Similarly, Püschel et al. (2016) tested the impact of inoculation with an indigenous
mycorrhizal fungal community and a laboratory culture of Rhizophagus irregularis on
growth, C allocation and nutrition of Andropogon gerardii under varying N and P supply.
The plants competed with AMF for N when supply was low, resulting in no or a negative
MGR and N-uptake response. The mycorrhizal fungal community inoculant only increased
plant P nutrition under N fertilisation, whereas R. irregularis slightly increased P uptake
without N supply. Both inoculants consistently increased nutritional and growth benefits to
the host with increasing N supply coincident with increasing root colonisation.
The trade-balance model is based on research under glasshouse conditions and has not been
agronomically tested; glasshouse experiments may not well mimic the resource availability
under field conditions for the reasons discussed in Section II. Applying the trade-balance
model to field crops will be complicated by temporal change as, for instance, warmer drier
weather dries the topsoil and reduces P availability or greater early crop growth due to high
colonisation reduces soil water availability later in the season and, if N is abundant, induces
“haying-off” (van Herwaarden et al., 1998; Kirkegaard & Ryan, 2014).
Agronomy literature: crop growth models
Crop growth models are becoming increasingly sophisticated to include crop genotype (G)
and crop management (M) (e.g. stubble retention, sowing date, plant density) in addition to
soil and climate information (E, environment) (Kirkegaard & Hunt, 2010; Zhang et al., 2012;
Teixeira et al., 2017). Whilst initially developed in countries with high input agricultural
systems, such as Australia and the USA, these models are now being used in many countries
and agricultural systems (e.g., Seyoum et al., 2018). A better understanding of the
mechanisms underlying the relationship between level of colonisation by AMF and crop yield
could enable colonisation level, or other measures of mycorrhizal activity to be included in
crop growth models using the G × E × M framework. Prediction of the impact of colonisation
level on crop yield could then be made without large, logistically-challenging, expensive,
field experiments.
14
A G × E × M approach was used by Ryan & Kirkegaard (2012) to investigate why wheat and
other crops have been reported to suffer a yield penalty when colonisation by AMF is low, as
may occur following long bare fallow or non-mycorrhizal crops, in the subtropical northern
wheatbelt of Australia (Thompson, 1987; Owen et al., 2010), but not in the southern
temperate wheatbelt (Figure 1;Ryan and Kirkegaard, 2012 and references therein).
Differences in soil type (E) and P availability (E) were discounted due to southern wheatbelt
experiments on vertisols typical of the northern wheatbelt yielding no relationship between
colonisation and yield for wheat and flax even under P limitation (Ryan & Kirkegaard, 2012).
However, they used the crop growth model APSIM to demonstrate that autumn-sown crops
in the southern wheatbelt have an initial slow growth rate due to winter conditions (E), and
hence low P demand and, perhaps, little need for AMF to supply P. They also speculated that
low winter soil temperatures (< 10°C at 0–10 cm depth) could reduce the spread of, and P
flow through, the external hyphae of AMF (Cooper & Tinker, 1981; Gavito et al. 2003). For
instance, when Gavito et al. (2003) grew field pea in growth chambers with soil maintained
at either 10°C or 15°C, they found no development of external hyphae at 10°C, in spite of an
average of 40% of root length being colonised, while external hyphae were present at 15°C
(71% of root length colonised). Interestingly, root mass ratio was 31% lower at 15°C than at
10°C, suggesting greater importance for roots in soil exploration for P at 10°C. Hence, Ryan
and Kirkegaard (2012) hypothesised that E may dictate crop response to variation in the
colonisation level of AMF both directly (crop P demand) and indirectly (mycorrhizal fungal
activity). Ryan & Kirkegaard (2012) also suggested that genotype (crop and mycorrhizal
fungal community) could contribute to a higher MGR in the northern wheatbelt.
VII. Crop genotype
Which crop genotypes benefit most from AMF?
Crop genotype is undoubtedly important for understanding variation in MGR (Francis &
Read, 1995; Graham et al., 1991; Hetrick et al., 1993; Pringle & Bever, 2008). Plants with
root systems poorly able to explore soil for nutrients have traditionally been considered to
likely have a high MGR with short root hair length proposed as a key trait. For instance, in a
glasshouse experiment, Schweiger et al. (1995) found annual pasture species with shorter,
less frequent root hairs responded most to inoculation (Figure 2) and Jakobsen et al. (2005)
found AMF substituted for lack of root hairs on mutant barley (Hordeum vulgare). However,
a meta-analysis by Maherali (2014) detected no impact on MGR of root system architecture,
15
including root hair length and density, and Martín-Robles et al. (2018) found no relationship
between MGR and fine root traits. Moreover, the meta-analysis of Hoeksema et al. (2010) of
306 greenhouse and growth chamber studies found that plant functional group, rather than
root morphology, determined MGR, with C4 grasses more positively responsive to
inoculation than C3 grasses; a finding consistent with a high P demand by C4 grasses due to
high growth rates with high temperatures and irradiance (Ludlow, 1985).
We suggest that the outcomes of Maherali (2014) and Hoeksema et al. (2010) may be
problematic as their data likely originated from studies that included only two levels of P. As
shown in Figure 2, MGR can vary greatly among P levels, and in a manner that differs among
genotypes, due to different shaped P response curves for colonised and non-colonised plants
(Schweiger et al. 1995). Differences among studies in resource availability and growth
conditions (section VI), and mycorrhizal fungal and soil microbial communities (sections
VIII, IX) also likely obscure the role of root morphology. In addition, the mechanisms of root
morphological adaptation to low P, which may influence MGR, may be variable and
complex. For instance, subterranean clover cultivars differ in external critical P requirement
(i.e. the P level sufficient to achieve 90% of maximum yield) due to variation in ability to
acclimate to low P conditions through maintenance of dry matter allocation to roots in soil
layers relatively high in P (e.g. topsoil) and high specific root lengths which together confer a
high root length density (Haling et al., 2018). To improve our ability to predict the MGR of
crop genotypes we suggest that investigations of MGR should generate a P response curve to
allow external critical P requirements of non-colonised and colonised plants to be compared
(Schweiger et al., 1995; Kelly et al., 2001; Figure 2): a thorough assessment of root
morphology traits and their response to low P conditions should also occur. Field experiments
are required.
Should crop breeders focus on crop MGR or crop phosphorus-use efficiency?
The potential of crop domestication to reduce MGR is often raised as a concern (e.g. Martín-
Robles et al., 2018). At the same time, improvement of P-acquisition efficiency and P-
utilization efficiency, i.e., P-use efficiency (PUE), is increasingly being suggested as a goal
for crop breeders (Lynch, 2011; Leiser et al., 2016; Mart et al., in press). So, should crop
breeders be considering MGR and PUE simultaneously?
16
Lehmann et al. (2012) conducted a meta-analysis of 39 publications on 320 crop plant
genotypes and the importance of year of release on MGR, colonisation and PUE. Cultivars
released after 1950 were less colonised, but more responsive, than ancestral genotypes;
although, recently, Martín-Robles et al. (2018) found that domesticated crops only benefited
from AMF under P-limitation. Lehmann et al. (2012) found colonisation was a significant
explanatory variable for the MGR trend with year of cultivar release, but PUE was not. This
suggests that breeding for optimised plant–mycorrhizal interactions may be as or more
important than selecting for PUE. Kaeppler et al. (2000) discussed variation in MGR
attributable to differences in plant–mycorrhizal interaction versus that attributable to the host
genotype PUE at a given soil P supply. They suggested, based on their glasshouse experiment
with in-bred lines of maize, that if the genetic potential to accumulate biomass at low P in the
absence of AMF is of greater consequence in determining MGR than genetic potential to
benefit from the symbiosis, then selection for PUE would become the more important crop
improvement target. Mycorrhizal interaction with host genotype and PUE was also studied by
Sawers et al. (2017) who evaluated the relative growth of 30 highly diverse maize lines with
and without inoculation and found that host genetic factors influenced fungal growth strategy
with P uptake by the most responsive line linked to higher abundance of external hyphae.
Overall, it seems that the interactions of host genotype, MGR and PUE are complex. Little
data are available from field studies and MGR data in the literature most often come from
studies with only two levels of P application. Moreover, whilst the level of colonisation by
AMF may vary greatly among crop genotypes (e.g. Leiser et al., 2016; Ryan et al., 2016),
breeding for high colonisation may be difficult due to low heritability (Leiser et al., 2016).
Furthermore, breeding for improved root soil exploration for P could result in selection for a
lower MGR and such plants may limit colonisation (Graham et al., 1991). In view of the
above, and the uncertainties over the importance of AMF for crop yield in the field (Section
VI), it currently is not possible to advise crop breeders to consider AMF when selecting for
PUE.
VIII. Fungal genotype
It is frequently stated that the mycorrhizal fungal community in agroecosystems functions
sub-optimally such that higher applications of P fertiliser and other inputs are required
(Srivastava et al., 2017). This view is based on the idea that industrial agricultural practices
select for genotypes less beneficial for crop hosts, for example, AMF with increased host C
17
cost because they rapidly colonise and sporulate. Verbruggen & Kiers (2010) suggest that
agricultural practices disfavour development of a more complex and functional (i.e. more
beneficial) mycorrhizal fungal community. Selection pressures most often cited and
researched that may reduce mycorrhizal fungal diversity and functionality in agroecosystems
are (i) tillage, (ii) high fertiliser input and (iii) crop rotations with non-hosts or low plant
diversity. Below, we consider whether these practices consistently reduce diversity and
functionality of the mycorrhizal fungal community, as well as whether higher diversity is
desirable and whether diversity can be usefully manipulated with inoculants.
Tillage alters the mycorrhizal fungal community, but may not reduce abundance and
diversity of AMF
Shifts in mycorrhizal fungal community composition and genetic diversity have repeatedly
been documented when high-intensity tillage (e.g. mouldboard or chisel-disk ploughing) and
low-intensity tillage systems are compared (Börstler et al., 2010; Miller et al., 1995;
Boddington & Dodd, 2000; Jansa et al., 2002, 2003; Alguacil et al., 2008; Verbruggen et al.,
2010). In a recent meta-analysis of 54 field studies, Bowles et al. (2016) reveal a large
(~30%) increase in colonisation level, but only a slight (11%) increase in species richness of
AMF in response to less intensive tillage. However, some studies find no impact of tillage
intensity on mycorrhizal fungal abundance and diversity. For instance, in Switzerland, Jansa
et al. (2002, 2003) found an increased incidence of certain AMF following long-term (13
years) tillage treatments of differing intensity (conventional, chisel plough, no-tillage).
However, while the mycorrhizal fungal community structure was greatly affected by tillage,
no difference in diversity was detected among tillage treatments.
A shift in the community structure of AMF under less intensive tillage may, however, affect
crop growth. For instance, Köhl et al. (2014) used microcosms to study whether tillage-
induced changes in the mycorrhizal fungal community altered plant productivity and nutrient
uptake. They found no-tillage communities increased P supply, but not plant productivity,
compared with conventional tillage communities, likely due to a two-fold increase in length
of external hyphae. Similarly, Miller et al. (1995) demonstrated, from multiple field studies
of maize evaluating effects of tillage on colonisation and external hyphal development in the
rhizosphere, that P uptake, not yield, is greater with no-tillage.
18
The decline in mycorrhizal fungal abundance with high intensity tillage may be relatively
short-term. Rasmann et al. (2009) evaluated the impact of 3-4 years of management system
(bahiagrass [Paspalum notatum] pasture, weed fallow, organic, disk fallow, and
conventional) on subsequent tomato (Solanum lycopersicum) cropping in a long-term
commercial tomato field (Florida, USA). Although colonisation of tomato was dramatically
decreased in soil actively undergoing intensive tillage, there was no long-term suppressive
effect on mycorrhizal fungal infectivity. Indeed in the disk fallow treatment following 14
years of intensive agricultural management - including four consecutive years of continuous
soil disturbance - inoculum density and infectivity of AMF was not significantly impacted
compared to the other farming practices investigated after tomato roots were present for a
single growing season. Thus, it seems the mycorrhizal fungal community may rapidly recover
from periods of intense disturbance and the absence of host tissue.
In a more extreme comparison, Lekberg et al. (2012) contrasted non-disturbed grassland soil
with intensely disturbed soil, finding 32 operational taxonomic units (OTUs) distributed
across five families of AMF (Zeeland, Denmark). Soil disturbance did not significantly alter
community composition and OTU-richness. OTUs from undisturbed soil were also common
after severe disturbance. Approximately 40% of all sequences within a sample were a single
OTU. They concluded that the assembly and abundance of the grassland mycorrhizal fungal
community indicated a high level of resilience and disturbance tolerance.
Fertilisation selects for tolerant AMF, but are they less functional?
Studies of long-term fertiliser trials report N fertiliser alters diversity of AMF, depending on
their sensitivity (Oehl et al., 2004), and may stimulate colonisation by certain species
(Toljander et al., 2008). Most commonly cited is the apparent tolerance of R. irregularis to
nutrient enrichment, as its abundance increased following eight years of N and P additions
compared to non-fertilised controls (Johnson, 1993). In maize fields in Central Italy, Boriello
et al. (2012) found that Glomeraceae were the main colonisers in fertilised plots, but they co-
occurred with Gigasporaceae and Paraglomus regardless of the management practices
applied. Members of Diversisporaceae and Entrophosporaceae were detected in N fertilised
and the untreated plots, respectively. In general, mycorrhizal fungal communities were
primarily influenced by N fertilisation and, to a lesser extent, by tillage. However, the
relationship between AMF selected under high fertility and their symbiotic functioning has
19
not been adequately evaluated; thus it remains unconfirmed if this vital agronomic practice
selects for a less functional mycorrhizal fungal community.
Crop rotation and other crop management practices promote diversity of AMF
Crop rotation has been demonstrated to promote mycorrhizal fungal communities (Oehl et
al., 2004; Hijri et al., 2006) that may closely resemble those from natural sites (Verbruggen
et al., 2010). Oehl et al. (2010) found that crop rotation, including a perennial grass–clover
mixture, increased diversity of AMF compared to continuous monoculture, and promoted
even more diversity than found in natural grasslands. Moreover, species of AMF not
recovered from continuous monoculture fields were detected after eight months in
microcosms constituted from these fields. Apparently, slower-sporulating genotypes
increased in abundance with host plants when propagated longer than an annual crop season.
Somewhat analogous to Oehl et al. (2010), Jansa et al. (2002, 2003) used trap cultures to
show that mycorrhizal fungal community composition is affected more by the species of trap
plant than by the tillage treatment of the field soil. Thus, agroecosystems with a diversity of
host species and host functional groups (annual/perennial, legume/grass etc) may promote
diversity of AMF.
Do more abundant or diverse or communities of AMF promote agroecosystem productivity
and sustainability?
Hoeksema et al. (2010) found that MGR was lower when plants were inoculated with one
species of AMF, compared to multiple species. However, colonisation by multiple species
can result in competitive, synergistic or antagonist interactions (Maherali & Klironomos,
2007) depending on the species of AMF, time of harvest, and environmental conditions (Daft
& Hogarth, 1983; Alkan et al., 2006; Jansa et al., 2008). For instance, Argüello et al. (2016)
found that diversity of AMF promotes cooperation between plants and AMF. However,
promoting diversity and inoculum density of AMF may trade-off their ability to stimulate
plant productivity in some instances. Janoušková et al. (2013) tested in potted field soil the
effects of isolates of Claroideoglomus claroideum, Glomus intraradices and G. mosseae,
each inoculated into a background synthetic mycorrhizal fungal community of the three
isolates, on colonisation and plant growth. They found a transient positive response to an
increase in root colonisation for the least competitive isolate. The two other more competitive
isolates responded negatively to intra- and interspecific inoculations, and in some cases,
reduced plant growth. Similarly, Thonar et al. (2014) found trade-offs that depended on the
20
species of AMF when they co-inoculated barrel medic (Medicago truncatula) with
Claroideoglomus and Rhizophagus. Even though the species competed for root colonisation
sites, co-inoculated plants were larger and higher in P than plants inoculated with each
species alone. In contrast, inoculation with Gigaspora and Rhizophagus, which facilitated
each other’s root colonisation, produced plants that were smaller than those with each species
inoculated separately. In the field, prior root occupancy may also impact the assembly and
functionality of AMF in soil microbial communities (Dumbrell et al., 2010).
For natural mycorrhizal fungal communities, high diversity is not always linked to greater
productivity. Verbruggen et al. (2012) inoculated maize with soil and roots from three
organic and three conventional maize fields and measured productivity, and nutrient loss
during leaching events induced by simulated rain (Netherlands). Maize growth responded
negatively to inoculation, especially for inoculum from organic fields which harboured the
greater diversity of AMF (Verbruggen et al., 2010). Productivity was inversely correlated
with colonisation level, suggesting C allocation to AMF was responsible for plant growth
reduction. Soil inoculation did reduce P leaching after simulated rain in response to increased
hyphal density related to the abundance of particular mycorrhizal fungal types. These results
demonstrate the potential for AMF to provide agroecosystem benefits other than crop yield,
but that they may not be optimised together.
Thus it seems there are complex interactions in diverse mycorrhizal fungal communities
which may induce significant changes in plant host growth and/or P acquisition and,
therefore, increasing diversity is no guarantee of increased MGR and crop productivity.
Augmentation with indigenous AMF may benefit crops
The view that low diversity of AMF causes a lack of mycorrhizal benefit in agroecosystems
has driven the development of commercial inoculants (Vosátka et al., 2012). The
performance of commercial products has been variable (e.g., Lojan et al., 2017) and they are
thus little used (Tarbell & Koske, 2007; Faye et al., 2013). More broadly, although
inoculation has the potential to increase growth and P uptake, particularly under low soil P
availability (e.g. Pellegrino & Bedini, 2014), numerous studies report little or no positive, and
even negative MGRs (Klironomos, 2003; Smith & Smith, 2011; Tawaraya, 2003).
Use of indigenous inoculant would mitigate bulk and expense, and avoid questions about the
desirability of exotic inoculants (Hart et al., 2018). Several field studies demonstrate that
21
indigenous AMF, produced on-farm with mycotrophic hosts, may be an economical and
sustainable practice. In Italy, Pellegrino et al. (2011) assessed shoot dry weight and nutrient
uptake of berseem clover (Trifolium alexandrinum) inoculated with four isolates of AMF
either alone or mixed, and compared to indigenous inoculum. Inoculation increased
productivity of berseem clover and a following maize crop, with indigenous inoculum as, or
more, effective than the introduced AMF. However, a risk with on-farm inoculum production
is propagation of soil-borne pathogens and parasitic nematodes, so measures should be taken
to ensure quality control of inoculum quality with soil bioassays using the target crop as the
host.
Use of an indigenous soil community of AMF may not always enhance plant growth when
the plant has not previously been grown in the soil. For instance, in a glasshouse experiment,
Řezáčová et al. (2017) tested the effect of soil microbial inocula with and without indigenous
mycorrhizal fungal communities, with and without P addition, on plant biomass and C fluxes
from plant to soil of a C3 and C4 Panicum grass (not native to the field soil used). When the
inoculum contained AMF, all plants had lower P and N status, and less biomass, due to
greater allocation of C belowground. Řezáčová et al. (2017) interpreted this negative impact
from AMF as perhaps being due to symbiotic incompatibility due to the fungal and plant
partners being of different geographic origins: see also Hart & Reader (2002) and Maherali &
Klironomos (2007). Řezáčová et al. (2017) also recognised that the negative impacts could
have resulted from slow development of colonisation or differences in microbial communities
between inoculants. However, there are also reports of plant growth enhancement when a
non-indigenous inoculant is added to soil containing indigenous AMF. For instance, Köhl et
al. (2016) inoculated microcosms of a grass–clover mixture (Lolium multiflorum and
Trifolium pratense) grown in a glasshouse with non-indigenous Rhizoglomus irregulare. The
microcosms contained eight nonsterilised field soils varying in soil type, chemical
characteristics and indigenous mycorrhizal fungal community. Inoculation increased the
abundance of R. irregulare in all soils, irrespective of soil P availability (which varied from
53-210 mg kg-1, ammonium acetate EDTA extraction), the initial abundance of R. irregulare
and the abundance of indigenous mycorrhizal fungal communities. Inoculation with AMF did
not affect the grass but significantly enhanced clover yield in five of eight field soils. The
results demonstrate that inoculation can be successful, even when the inoculant is not
indigenous, soil P is high, and indigenous communities are abundant. However, Köhl et al.
22
(2016) noted that the amount of inoculum applied was large (corresponding to 1.4 × 105 L ha–
1), expensive and possibly commercially unrealistic.
IX. Complex interactions between the mycorrhizal fungal and soil microbial
communities
In their meta-analysis, Hoeksema et al. (2010) reported plant response to inoculation with
AMF was greater with the use of whole soil inoculum presumed to contain multiple species
of AMF as well as other soil microbes, or if other soil microbes were added back to the
background growing media. These findings suggests that AMF are more beneficial when a
diverse soil microbial and mycorrhizal fungal community is also present, that is, in a realistic
biotic context.
Pizano et al. (2017) examined, in the glasshouse, the effects of AMF and soil filtrates on the
growth of 11 host species from three habitats in a tropical montane landscape in Colombia
using reciprocal habitat inoculation. Most plant species received a similar benefit from AMF
but differed in their response to soil filtrates from the three habitats. Soil filtrate from pastures
had a net negative effect, while filtrates from coffee (Coffea arabica) plantations and forests
had a positive effect, on plant growth. Pasture grass, coffee, and five pioneer tree species
performed better with the filtrate from soils in which they rarely occurred, while four shade-
tolerant tree species grew similarly with all filtrates. These results suggest that some habitats
accumulate species-specific soil pathogens, while others accumulate soil microbes that
benefit indigenous plants and crops. Moreover, non-mycorrhizal soil microbes acting as
mediators of plant–soil feedbacks may exert even greater habitat and host-specific effects on
plants than AMF (Bever et al., 1997).
In summary, interactions among species of AMF in diverse communities and the soil
microbiota are complex and poorly understood. Little research has been conducted under
field conditions. Additional complexity is added by the recent discovery that fine root
endophyte, which can be abundant in agroecosystems and co-inhabit roots with AMF, is
likely not phylogenetically aligned with the AMF (Orchard et al., 2017a, b).
X. Phosphorus-efficient agroecosystems
We propose that higher abundance and diversity of AMF may be best promoted not by the
methodologically-challenging task of quantifying their role in crop yields, but as a secondary
consequence of the development of P-efficient agroecosystems. Such agroecosystems are
23
desirable due to rock phosphate being a non-renewable resource (Cordell & White, 2015) and
the environmental damage caused worldwide by movement of P originating from excessive
application of P fertilisers off farm into waterways and estuaries (Sharpley et al., 2015).
It has been recently proposed that, for pastures, reducing the critical external P requirement of
the least P-efficient species through its replacement with a species better able to explore soil
for P allows the target soil extractable P level to be lowered without yield penalty (Simpson
et al., 2014, 2015; Haling et al., 2016). This then confers savings in P fertiliser use due to less
P leaching (see Khan et al., 2018) and, in pasture soils, less conversion of P into poorly plant-
available organic forms (Simpson et al., 2015). The MGR of a pasture or crop species with P
as the primary limitation is also reflected in a decrease in external critical P requirement due
to greater plant growth at moderate to low P availability when colonised (Figure 2). Hence,
high colonisation could aid development of P-efficient agroecosystems if the fungi enhance
crop P uptake.
Even if AMF are not required for optimal crop P nutrition, if the pasture or crop external
critical P requirement comes close to coinciding with maximum colonisation and/or length of
external hyphae, then yield and any benefits of AMF for soil health might be desirably
balanced. This concept was explored by Mai et al. (2018) in a field experiment where cotton
was grown at four P application rates over two years (Xinjiang region, northwest China). A
moderate P rate favoured density of external hyphae, soil exploration by roots and efficient
uptake of fertiliser P, and corresponded to around 90% of maximum yield. The authors
therefore recommended that the target bicarbonate-extractable soil P could be halved from
the 30 mg kg-1 currently recommended to 15-20 mg kg-1. For more diverse agroecosystems,
with crop rotation, intercropping or mixed-species pastures, identifying the target soil P level
will be more complicated. For instance, when Deng et al. (2017) grew a wheat-maize rotation
over two years at six P rates (Hebei Province, China), yield and colonisation by AMF of
wheat, but not maize, responded strongly to P addition (Figure 3). However, at the target
bicarbonate-extractable soil P concentration, determined by the authors to be 22 mg P kg-1
(Olsen), yields and colonisation levels over the rotation were reasonable well balanced. Note
that for both these studies, the crop MGR and hence the yield impact of reduced colonisation
at a given P application rate due to, for instance, a preceding non-mycorrhizal crop or intense
tillage, is unknown.
24
XI. Conclusions
Our review shows the yield benefits of maintaining high abundance and diversity of AMF in
agroecosystems are often overstated due to an underlying optimism about the need for high
colonisation and a diverse community to maximise crop yield and failure to apply rigorous
agronomic methodology. We also show that the mycorrhizal fungal community may be more
resilient to many agricultural practices than often assumed. However, there is a paucity of
field studies and our ability to manage AMF to favour crop yield is compromised by
rudimentary knowledge of many important aspects of the symbiosis. These include: the
determinants of MGR for crop genotypes; the impact of resource balance on MGR in the
field; and the complexity of interactions among species of AMF in diverse communities and
between AMF and soil microbes. This lack of knowledge means that mycorrhizal fungal
abundance and diversity cannot be currently included in crop growth models. To address this
situation, we recommend field experiments be conducted in collaboration with local
agronomists and use a standard rigorous agronomic framework (Table 1). Researchers with
an agricultural focus who operate experimentally in the glasshouse should take care to ensure
treatments and environmental conditions as relevant to the field as possible. We also suggest
eight research priorities (Table 2).
We conclude that adjusting farm management to favour abundance or diversity of AMF
cannot be widely recommended based on current knowledge and that gains in agricultural
yields and sustainability (e.g., efficient use of inputs, reduced nutrient losses) are likely more
easily achievable through rectifying soil nutrient deficiencies, diversification through crop
rotation and intercropping (especially with legumes), crop selection and breeding, and
improved understanding of G (crop) × E × M (Leiser et al., 2016; Wani et al., 2017;
Kirkegaard & Hunt, 2010; Kermah et al., 2017; Plaza-Bonilla et al., 2017). We suggest
research on AMF in agroecosystems be routinely placed into the context of the broader
literature on crop agronomy, breeding and agroecosystem sustainability.
Acknowledgements
Thank you to Tony Stewart and the UWA pasture science team for stimulating discussion and
to Larry Duncan and three anonymous reviewers for helpful comments. MHR was funded by
ARC Future Fellowship FT140100103.
25
Author contributions
MHR and JHG shared equally the research and writing of this manuscript.
References
Alguacil MM, Lumini E, Roldán A, Salinas-García JR, Bonfante P, Bianciotto V. 2008.
The impact of tillage practices on arbuscular mycorrhizal fungal diversity in subtropical
crops. Ecological Applications 18: 527–536.
Alkan N, Gadkar V, Yarden O, Kapulnik Y. 2006. Analysis of quantitative interactions
between two species of arbuscular mycorrhizal fungi, Glomus mosseae and G. intraradices,
by real-time PCR. Applied and Environmental Microbiology 72: 4192–4199.
Angus JF. 1981. Basic data set for interpretation of regional experimental programs. Plant
and Soil 58: 315–326.
Angus JF, Kirkegaard JA, Hunt JR, Ryan MH, Ohlander L, Peoples MB. 2015. Break
crops and rotations for wheat. Crop & Pasture Science 66: 523–552
Argüello A, O'Brien MJ, van der Heijden MG, Wiemken A, Schmid B, Niklaus PA and
Maherali H. 2016. Options of partners improve carbon for phosphorus trade in the
arbuscular mycorrhizal mutualism. Ecology Letters 19: 648–656.
Bakhshandeh S, Corneo PE, Mariotte P, Kertesz MA, Dijkstra FA. 2017. Effect of crop
rotation on mycorrhizal colonization and wheat yield under different fertilizer treatments.
Agriculture, Ecosystems & Environment 247: 130–136.
Bever JD, Westover K, Antonovics J. 1997. Incorporating the soil community into plant
population dynamics: the utility of the feedback approach. Journal of Ecology 85: 561–573.
Boddington CL, Dodd JC. 2000. The effect of agricultural practices on the development of
indigenous arbuscular mycorrhizal fungi. I. Field studies in an Indonesian ultisol. Plant and
Soil 218: 137–144.
Borriello R, Lumini E, Girlanda M, Bonfante P, Bianciotto V. 2012. Effects of different
management practices on arbuscular mycorrhizal fungal diversity in maize fields by a
molecular approach. Biology and Fertility of Soils 48: 911–922.
Börstler B, Thiery O, Sykorova Z, Berner A, Redecker D. 2010. Diversity of
mitochondrial large subunit rDNA haplotypes of Glomus intraradices in two agricultural
field experiments and two semi-natural grasslands. Molecular Ecology 19: 1497–1511.
Bowles TM, Jackson LE, Loeher M, Cavagnaro TR. 2016. Ecological intensification and
arbuscular mycorrhizas: a meta-analysis of tillage and cover crop effects. Journal of Applied
Ecology 54: 1785–1793.
26
Cooper KM, Tinker PB. 1981. Translocation and transfer of nutrients in vesicular-
arbuscular mycorrhizas IV. Effect of environmental variables on movement of phosphorus.
New Phytologist 88: 327–339.
Cordell D, White S. 2015. Tracking phosphorus security: indicators of phosphorus
vulnerability in the global food system. Food Security 7: 337–350.
Covacevich F, Echeverria HE, Aguirrezabal LAN. 2007. Soil available phosphorus status
determines indigenous mycorrhizal colonization of field and glasshouse-grown spring wheat
from Argentina. Applied Soil Ecology 35: 1–9.
Daft MJ, Hogarth BG. 1983. Competitive interactions amongst four species of Glomus on
maize and onion. Transactions of the British Mycological Society 80: 339–345.
Deng Y, Feng G, Chen X, Zou C. 2017. Arbuscular mycorrhizal fungal colonization is
considerable at optimal Olsen-P levels for maximized yields in an intensive wheat-maize
cropping system. Field Crops Research 209: 1–9.
Dreccer MF, Fainges J, Whish J, Ogbonnaya FC, Sadras VO. 2018. Comparison of
sensitive stages of wheat, barley, canola, chickpea and field pea to temperature and water
stress across Australia. Agricultural and Forest Meteorology 248: 275–294.
Duan T, Shen Y, Facelli E, Smith SE, Nan Z. 2010. New agricultural practices in the Loess
Plateau of China do not reduce colonisation by arbuscular mycorrhizal or root invading fungi
and do not carry a yield penalty. Plant and Soil 331: 265–275.
Dumbrell AJ, Nelson M, Helgason T, Dytham C, Fitter AH. 2010. Idiosyncrasy and
overdominance in the structure of natural communities of arbuscular mycorrhizal fungi: is
there a role for stochastic processes? Journal of Ecology 98: 419–428.
Faye, A., Dalpe Y, Ndung’u-Magiroi K, Jefwa J, Ndoye I, Diouf M, Lesueur D. 2013.
Evaluation of commercial arbuscular mycorrhizal inoculants. Canadian Journal of Plant
Science 93: 1201–1208.
Fellbaum CR, Gachomo EW, Beesetty Y, Choudhari S, Strahan GD, Pfeffer PE, Kiers
ET, Bücking H. 2012. Carbon availability triggers fungal nitrogen uptake and transport in
the arbuscular mycorrhizal symbiosis. Proceedings of the National Academy of Sciences,
USA 109: 2666–2671.
Felton WL, Marcellos H, Alston C, Martin RJ, Backhouse D, Burgess LW, Herridge
DF. 1997. Chickpea in wheat-based cropping systems of northern New South Wales. II.
Influence on biomass, grain yield, and crown rot in the following wheat crop. Australian
Journal of Agricultural Research 49: 401–408.
27
Francis R, Read DJ. 1995. Mutualism and antagonism in the mycorrhizal symbiosis, with
special reference to impacts on plant community structure. Canadian Journal of Botany 73:
1301–1309.
Gao XP, Akhter F, Tenuta M, Flaten DN, Gawalko EJ, Grant CA. 2010. Mycorrhizal
colonization and grain Cd concentration of field-grown durum wheat in response to tillage,
preceding crop and phosphorus fertilization. Journal of the Science of Food and Agriculture
90: 750–758.
Gavito ME, Schweiger P, Jakobsen I. 2003. P uptake by arbuscular mycorrhizal hyphae:
effect of soil temperature and atmospheric CO2 enrichment. Global Change Biology 9: 106–
116.
Germida JJ, Walley FL. 1996. Plant growth-promoting rhizobacteria alter rooting patterns
and arbuscular mycorrhizal fungi colonization of field-grown spring wheat. Biology and
Fertility of Soils 23: 113–120.
Graham JH, Abbott LK. 2000. Wheat responses to aggressive and nonaggressive
arbuscular mycorrhizal fungi. Plant and Soil 220: 207–218.
Graham JH, Eissenstat DM, Drouillard DL. 1991. On the relationship of a plant’s
mycorrhizal dependency and its rate of mycorrhizal colonization. Functional Ecology 5:773–
779.
Grant CA, Monreal MA, Irvine RB, Mohr RM, McLaren DL, Khakbazan M. 2009.
Crop response to current and previous season applications of phosphorus as affected by crop
sequence and tillage. Canadian Journal of Plant Science 89: 49–66.
Haling RE, Yang Z, Shadwell N, Culvenor RA, Stefanski A, Ryan MH, Sandral GA,
Kidd DR, Lambers H, Simpson RJ. 2016. Root morphological traits that determine
phosphorus-acquisition efficiency and critical external phosphorus requirement in pasture
species. Functional Plant Biology 43: 815–826.
Haling RE, Brown LK, Stefanski A, Kidd DR, Ryan MH, Sandral GA, George TS,
Lambers H, Simpson RJ. 2018. Differences in nutrient foraging among Trifolium
subterraneum cultivars deliver improved P-acquisition efficiency. Plant and Soil.
doi:10.1007/s11104-017-3511-7
Hart MM, Reader RJ. 2002. Taxonomic basis for variation in the colonization strategy of
arbuscular mycorrhizal fungi. New Phytologist 153: 335–344.
Hart MM, Antunes PM, Chaudhary VB, Abbott LK. 2018. Fungal inoculants in the field:
Is the reward greater than the risk? Functional Ecology 32: 126–135
28
Hetrick BAD, Wilson GWT, Cox TS. 1993 Mycorrhizal dependence of modern wheat
cultivars and ancestors: a synthesis. Canadian Journal of Botany 71: 512–518.
Hijri I, Sykorovà Z, Oehl F, Ineichen K, Mader P, Wiemken A, Redecker D. 2006.
Communities of arbuscular mycorrhizal fungi in arable soils are not necessarily low in
diversity. Molecular Ecology 15: 2277–2289.
Hildermann I, Messmer M, Dubois D, Boller T, Wiemken A, Mäder P. 2010. Nutrient
use efficiency and arbuscular mycorrhizal root colonisation of winter wheat cultivars in
different farming systems of the DOK long-term trial. Journal of the Science of Food and
Agriculture 90: 2027–2039.
Hodge A, Fitter AH. 2010. Substantial nitrogen acquisition by arbuscular mycorrhizal fungi
from organic material has implications for N cycling. Proceedings of the National Academy
of Sciences, USA 107: 13754–13759.
Hodge A, Storer K. 2015. Arbuscular mycorrhiza and nitrogen: Implications for individual
plants through to ecosystems. Plant and Soil 386: 1–19.
Hoeksema JD, Chaudhary VB, Gehring CA, Johnson NC, Karst J, Koide RT, Pringle
A, Zabinski C, Bever JD, Moore JC et al. 2010. A meta-analysis of context-dependency in
plant response to inoculation with mycorrhizal fungi. Ecology Letters 13: 394–407.
Jakobsen I, Chen BD, Munkvold L, Lundsgaard T, Zhu YG. 2005. Contrasting phosphate
acquisition of mycorrhizal fungi with that of root hairs using the root hairless barley mutant.
Plant Cell and Environment 28: 928–938.
Janos DP. 2007. Plant responsiveness to mycorrhizas differs from dependence upon
mycorrhizas. Mycorrhiza 17: 75–91.
Janoušková M, Kraka K, Wagg C, Štorchovác H, Caklováa P, Vosátka M. 2013. Effects
of inoculum additions in the presence of a preestablished arbuscular mycorrhizal fungal
community. Applied and Environmental Microbiology 79: 6507–6515.
Jansa J, Mozafar A, Anken T, Ruh R, Sanders IR, Frossard E. 2002. Diversity and
structure of AMF communities as affected by tillage in a temperate soil. Mycorrhiza 12: 225–
234.
Jansa J, Mozafar A, Kuhn G, Anken T, Ruh R, Sanders IR, Frossard E. 2003. Soil
tillage affects the community structure of mycorrhizal fungi in maize roots. Ecological
Applications 13: 1164–1176.
Jansa J, Smith FA, Smith SE. 2008. Are there benefits of simultaneous root colonization by
different arbuscular mycorrhizal fungi. New Phytologist 177: 779–789.
29
Jeffery RP, Simpson RJ, Lambers H, Kidd DR, Ryan MH. 2017. Plants in constrained
canopy micro-swards compensate for decreased root biomass and soil exploration with
increased amounts of rhizosphere carboxylates. Functional Plant Biology 44: 552–
562.Johansen A, Jakobsen I, Jensen ES. 1992. Hyphal transport of N-15-labeled nitrogen
by a vesicular-arbuscular mycorrhizal fungus and its effect on depletion of inorganic soil-N.
New Phytologist 122: 281–288.
Johnson NC. 1993. Can fertilization of soil select less mutualistic mycorrhizae? Ecological
Applications 3: 749–757.
Johnson NC, Graham JH. 2013. The continuum concept remains a useful framework for
studying mycorrhizal functioning. Plant and Soil 363: 411–419.
Johnson NC, Graham JH, Smith FA. 1997. Functioning of mycorrhizal associations along
the mutualism-parasitism continuum. New Phytologist 135: 575–585.
Johnson NC, Wilson GWT, Bowker MA, Wilson JA, Miller RM. 2010. Resource
limitation is a driver of local adaptation in mycorrhizal symbioses. Proceedings of the
National Academy of Sciences, USA 107: 2093–2098.
Johnson NC, Wilson GWT, Wilson JA, Miller RM, Bowker MA. 2015. Mycorrhizal
phenotypes and the law of the minimum. New Phytologist 205: 1473–1484.
Kabir Z, O'Halloran IP, Fyles JW, Hamel C. 1998. Dynamics of the mycorrhizal
symbiosis of corn (Zea mays L.): effects of host physiology, tillage practice and fertilization
on spatial distribution of extra-radical mycorrhizal hyphae in the field. Agriculture,
Ecosystems & Environment 68:151–163.
Kaeppler SM, Parke JL, Mueller SM, Senior L, Stuber C, Tracy WF. 2000. Variation
among maize inbred lines and detection of quantitative trait loci for growth at low
phosphorus and responsiveness to arbuscular mycorrhizal fungi. Crop Science 40: 358–364.
Karasawa T, Takebe M. 2012. Temporal or spatial arrangements of cover crops to promote
arbuscular mycorrhizal colonization and P uptake of upland crops grown after
nonmycorrhizal crops. Plant and Soil 353: 355–366.
Kelly R, Edwards D, Thompson J, Magarey R. 2001. Responses of sugarcane, maize, and
soybean to phosphorus and vesicular-arbuscular mycorrhizal fungi. Australian Journal of
Agricultural Research 52: 731–743.
Kermah M, Franke AC, Adjei-Nsiah S, Ahiabor BDK, Abaidoo RC, Giller KE. 2017.
Maize-grain legume intercropping for enhanced resource use efficiency and crop productivity
in the Guinea savanna of northern Ghana. Field Crops Research 213 (Supplement C): 38–50.
30
Khan A, Lu G, Ayaz M, Zhang H, Wang R, Lv F, Yang X, Sun B, Zhang S. 2018.
Phosphorus efficiency, soil phosphorus dynamics and critical phosphorus level under long-
term fertilization for single and double cropping systems. Agriculture, Ecosystems &
Environment 256. doi:https://doi.org/10.1016/j.agee.2018.01.006
Kirk AP, Entz MH, Fox SL, Tenuta M. 2011. Mycorrhizal colonization, P uptake and yield
of older and modern wheats under organic management. Canadian Journal of Plant Science
91: 663–667.
Kirkegaard JA, Hunt JR. 2010. Increasing productivity by matching farming system
management and genotype in water-limited environments. Journal of Experimental Botany
61: 4129–4143.
Kirkegaard JA, Ryan MH. 2014. Magnitude and mechanisms of persistent crop sequence
effects on wheat. Field Crops Research 164: 154–165.
Kirkegaard JA, Hocking PJ, Angus JF, Howe GN, Gardner PA. 1997. Comparison of
canola, Indian mustard and Linola in two contrasting environments. II. Break-crop and
nitrogen effects on subsequent wheat crops. Field Crops Research 52: 179–191.
Kirkegaard JA, Simpfendorfer S, Holland J, Bambach R, Moore KJ, Rebetzke GJ.
2004. Effect of previous crops on crown rot and yield of durum and bread wheat in northern
NSW. Australian Journal of Agricultural Research 55: 321–334.
Klironomos JN. 2003. Variation in plant response to native and exotic arbuscular
mycorrhizal fungi. Ecology 84: 2292–2301.
Köhl L, Lukasiewicz CE, van der Heijden MGA. 2016. Establishment and effectiveness of
inoculated arbuscular mycorrhizal fungi in agricultural soils. Plant, Cell and Environment 39:
136–146.
Köhl L, Oehl F, van der Heijden MGA. 2014. Agricultural practices indirectly influence
plant productivity and ecosystem services through effects on soil biota. Ecological
Applications 24: 1842–1853.
Koide R, Peoples M. 2012. On the nature of temporary yield loss in maize following canola.
Plant and Soil: 360: 259–269.
Lehmann A, Barto K, Powell JR, Rillig MC. 2012. Mycorrhizal responsiveness trends in
annual crop plants and their wild relatives—a meta-analysis on studies from 1981 to 2010.
Plant and Soil 355: 231–250.
Leiser WL, Olatoye MO, Rattunde HFW, Neumann G, Weltzien E, Haussmann BG.
2016. No need to breed for enhanced colonization by arbuscular mycorrhizal fungi to
improve low-P adaptation of West African sorghums. Plant and Soil 401: 51–64.
31
Lekberg Y, Helgason TC. 2018. In situ mycorrhizal function – knowledge gaps and future
directions. New Phytologist: doi:10.1111/nph.15064
Lekberg Y, Koide RT. 2005. Is plant performance limited by abundance of arbuscular
mycorrhizal fungi? A meta-analysis of studies published between 1988 and 2003. New
Phytologist 168: 189–204.
Lekberg Y, Koide RT. 2014. Integrating physiological community and evolutionary
perspectives on the arbuscular mycorrhizal symbiosis. Botany 92: 241–251.
Lekberg Y, Schnoor T, Kjøller R, Gibbons SM, Hansen LH, Al‐Soud WA, Sørensen SJ,
Rosendahl S. 2012. 454‐sequencing reveals stochastic local reassembly and high disturbance
tolerance within arbuscular mycorrhizal fungal communities. Journal of Ecology, 100: 151–
160.
Loján P, Senés-Guerrero C, Suárez JP, Kromann P, Schüßler A, Declerck S. 2017.
Potato field-inoculation in Ecuador with Rhizophagus irregularis: no impact on growth
performance and associated arbuscular mycorrhizal fungal communities. Symbiosis 73: 45–
56.
Ludlow MM. 1985. Photosynthesis and dry matter production in C3 and C4 pasture plants,
with special emphasis on tropical C3 legumes and C4 grasses. Functional Plant Biology 12:
557–572.
Lynch JP. 2011. Root phenes for enhanced soil exploration and phosphorus acquisition:
tools for future crops. Plant Physiology 156: 1041–1049.
Maherali H. 2014. Is there an association between root architecture and mycorrhizal growth
response? New Phytologist 204: 192–200.
Maherali H, Klironomos JN. 2007. Influence of phylogeny on fungal community assembly
and ecosystem functioning. Science 316: 1746–1748.
Mai W, Xue X, Feng G, Yang R, Tian C. 2018. Can optimization of phosphorus input lead
to high productivity and high phosphorus use efficiency of cotton through maximization of
root/mycorrhizal efficiency in phosphorus acquisition? Field Crops Research 216: 100–108.
Martín‐Robles N, Lehmann A, Seco E, Aroca R, Rillig MC, Milla R. 2018. Impacts of
domestication on the arbuscular mycorrhizal symbiosis of 27 crop species. New Phytologist
218: 322–334.
Mart BHR, Gerlinde BD, Gerwin FK, Oene O, Jan Willem G (in press) What root traits
determine grass resistance to phosphorus deficiency in production grassland? Journal of
Plant Nutrition and Soil Science. doi:10.1002/jpln.201700093
32
McGonigle TP, Hutton M, Greenley A, Karamanos R. 2011. Role of mycorrhiza in a
wheat–flax versus canola–flax rotation: A case study. Communications in Soil Science and
Plant Analysis 42: 2134–2142.
Miller RM, Jastrow JD. 1990. Hierarchy of roots and mycorrhizal fungal interactions with
soil aggregation. Soil Biology and Biochemistry 22: 579–584
Miller MH, McGonigle TP, Addy HD. 1995. Functional ecology of vesicular-arbuscular
mycorrhizas as influenced by phosphate fertilization and tillage in an agricultural ecosystem.
Critical Reviews in Biotechnology 15: 241–255.
Monreal MA, Grant CA, Irvine RB, Mohr RM, McLaren DL, Khakbazan M. 2011.
Crop management effect on arbuscular mycorrhizae and root growth of flax. Canadian
Journal of Plant Science 91: 315–324.
Nehl DB, Allen SJ, Brown JF. 1996. Mycorrhizal colonisation, root browning and soil
properties associated with a growth disorder of cotton in Australia. Plant and Soil 179: 171–
182.
Nehl DB, Allen SJ, Brown JF. 1998. Slow arbuscular mycorrhizal colonisation of field-
grown cotton caused by environmental conditions in the soil. Mycorrhiza 8: 159–167.
Newsham KK, Fitter AH, Watkinson AR. 1995. Multifunctionality and biodiversity in
arbuscular mycorrhizas. Trends in Ecology and Evolution 10: 407–411.
Noyd RK, Pfleger FL, Norland MR. 1996. Field responses to added organic matter,
arbuscular mycorrhizal fungi, and fertilizer in reclamation of taconite iron ore tailing. Plant
and Soil 179: 89–97.
Oehl F, Laczko E, Bogenrieder A, Stahr K, Bosch R, van der Heijden MGA, Sieverding
E. 2010. Soil type and land use intensity determine the composition of arbuscular
mycorrhizal fungal communities. Soil Biology and Biochemistry 42: 724–738.
Oehl F, Sieverding E, Mäder P, Dubois D, Ineichen K, Boller T, Wiemken A. 2004.
Impact of long-term conventional and organic farming on the diversity of arbuscular
mycorrhizal fungi. Oecologia 138: 574–583.
Orchard S, Hilton S, Bending GD, Dickie IA, Standish RJ, Gleeson DB, Jeffery RP,
Powell JR, Walker C, Bass D et al. 2017a. Fine endophytes (Glomus tenue) are related to
Mucoromycotina, not Glomeromycota. New Phytologist 213: 481–486.
Orchard S, Standish RJ, Dickie IA, Renton M, Walker C, Moot D, Ryan MH. 2017b.
Fine root endophytes under scrutiny: a review of the literature on arbuscule-producing fungi
recently suggested to belong to the Mucoromycotina. Mycorrhiza 27: 619–638.
33
Owen KJ, Clewett TG, Thompson JP. 2010. Pre-cropping with canola decreased
Pratylenchus thornei populations, arbuscular mycorrhizal fungi, and yield of wheat. Crop &
Pasture Science 61: 399–410.
Pellegrino E, Bedini S. 2014. Enhancing ecosystem services in sustainable agriculture:
Biofertilization and biofortification of chickpea (Cicer arietinum L.) by arbuscular
mycorrhizal fungi. Soil Biology and Biochemistry 68: 429–439.
Pellegrino E, Bedini S, Avio L, Bonari E, Giovannetti M. 2011. Field inoculation
effectiveness of native and exotic arbuscular mycorrhizal fungi in a Mediterranean
agricultural soil. Soil Biology and Biochemistry 43: 367–376.
Pellegrino E, Öpik M, Bonari E, Ercoli L. 2015. Responses of wheat to arbuscular
mycorrhizal fungi: a meta-analysis of field studies from 1975 to 2013. Soil Biology and
Biochemistry 84: 210–217.
Pizano C, Mangan SA, Graham JH, Kitajima K. 2017. Host-specific effects of soil
microbial filtrates prevail over those of arbuscular mycorrhizae in a fragmented landscape.
Ecological Applications 27:1946–1957.
Plaza-Bonilla D, Nolot JM, Raffaillac D, Justes E. 2017. Innovative cropping systems to
reduce N inputs and maintain wheat yields by inserting grain legumes and cover crops in
southwestern France. European Journal of Agronomy 82: 331–341.
Poorter H, Bühler J, van Dusschoten D, Climent J, Postma JA. 2012. Pot size matters: a
meta-analysis of the effects of rooting volume on plant growth. Functional Plant Biology 39:
839–850.
Pringle A, Bever JD. 2008. Analogous effects of arbuscular mycorrhizal fungi in the
laboratory and a North Carolina field. New Phytologist 180: 162–175.
Püschel D, Janoušková, M, Hujslová M, Slavíková R, Gryndlerová H, Jansa J. 2016.
Plant–fungus competition for nitrogen erases mycorrhizal growth benefits of Andropogon
gerardii under limited nitrogen supply. Ecology and Evolution 6: 4332–4346.
Rasmann C, Graham JH, Chellemi DO, Datnoff LE, Larsen J. 2009. Resilient
populations of root fungi occur within five tomato production systems in southeast Florida.
Applied Soil Ecology 43: 22–31.
Řezáčová V, Slavíková R, Konvalinková T, Hujslová M, Gryndlerová H, Gryndler M,
Püschel D, Jansa J. 2017. Imbalanced carbon-for-phosphorus exchange between European
arbuscular mycorrhizal fungi and non-native Panicum grasses—A case of dysfunctional
symbiosis. Pedobiologia 62: 48–55.
34
Ren H, Gui W, Bai Y, Stein C, Rodrigues JLM, Wilson GWT, Cobb AB, Zhang Y,
Yang G. 2018. Long-term effects of grazing and topography on extra-radical hyphae of
arbuscular mycorrhizal fungi in semi-arid grasslands. Mycorrhiza 28: 117–127.
Ryan MH, Angus JF. 2003. Arbuscular mycorrhizae in wheat and field pea crops on a low P
soil: increased Zn-uptake but no increase in P-uptake or yield. Plant and Soil 250: 225–239.
Ryan MH, Graham JH. 2002. Is there a role for arbuscular mycorrhizal fungi in production
agriculture? Plant and Soil 244: 263–271.
Ryan MH, Kirkegaard JA. 2012. The agronomic relevance of arbuscular mycorrhizas in the
fertility of Australian extensive cropping systems. Agriculture Ecosystems & Environment
163: 37–53.
Ryan MH, Derrick JW, Dann PR. 2004. Grain mineral concentrations and yield of wheat
grown under organic and conventional management. Journal of the Science of Food and
Agriculture 84: 207–216.
Ryan MH, McInerney JK, Record IR, Angus JF. 2008. Zinc bioavailability in wheat grain
in relation to phosphorus fertiliser, crop sequence and mycorrhizal fungi. Journal of the
Science of Food and Agriculture 88: 1208–1216.
Ryan MH, Norton RM, Kirkegaard JA, McCormick KM, Knights SE, Angus JF. 2002.
Increasing mycorrhizal colonisation does not improve growth and nutrition of wheat on
Vertosols in south-eastern Australia. Australian Journal of Agricultural Research 53: 1173–
1181.
Ryan MH, Kidd DR, Sandral GA, Yang Z, Lambers H, Culvenor RA, Stefanski A,
Nichols PGH, Haling RE, Simpson RJ. 2016. High variation in the percentage of root
length colonised by arbuscular mycorrhizal fungi among 139 lines representing the species
subterranean clover (Trifolium subterraneum). Applied Soil Ecology 98: 221–232
Sandral G, Simpson R, Yang Z, Culvenor R, Kidd D, Lambers H, Ryan M. 2015.
Phosphorus efficient pastures: response of alternative legumes to fertiliser application. Paper
presented at the “Building Productive, Diverse and Sustainable Landscapes”. Proceedings of
the 17th Australian Agronomy Conference 2015, 21 – 24 September 2015, Hobart, Tasmania.
Sawers RJH, Svane SF, Quan C, Grønlund M, Wozniak B, Gebreselassie M-N,
González-Muñoz E, Chávez Montes RA, Baxter I, Goudet J, Jakobsen I, Paszkowski U.
2017. Phosphorus acquisition efficiency in arbuscular mycorrhizal maize is correlated with
the abundance of root-external hyphae and the accumulation of transcripts encoding PHT1
phosphate transporters. New Phytologist 214: 632–643.
35
Schweiger PF, Robson AD, Barrow NJ. 1995. Root hair length determines beneficial effect
of a Glomus species on shoot growth of some pasture species. New Phytologist 131: 247–254.
Seymour M, Kirkegaard JA, Peoples MB, White PF, French RJ, van Burgel AJ. 2012.
Break-crop benefits to wheat in Western Australia – insights from over three decades of
research. Crop & Pasture Science 63: 1–16.
Seyoum S, Rachaputi R, Chauhan Y, Prasanna B, Fekybelu S. 2018. Application of the
APSIM model to exploit G × E × M interactions for maize improvement in Ethiopia. Field
Crops Research 217: 113–124.
Sharpley AN, Bergstrom L, Aronsson H, Bechmann M, Bolster CH, Borling K, Djodjic
F, Jarvie HP, Schoumans OF, Stamm C et al. 2015. Future agriculture with minimized
phosphorus losses to waters: Research needs and direction. Ambio 44: S163–S179.
Simpson RJ, Richardson AE, Nichols SN, Crush JR. 2014. Pasture plants and soil fertility
management to improve the efficiency of phosphorus fertiliser use in temperate grassland
systems. Crop & Pasture Science 65: 556–575.
Simpson RJ, Stefanski A, Marshall DJ, Moore AD, Richardson AE. 2015. Management
of soil phosphorus fertility determines the phosphorus budget of a temperate grazing system
and is the key to improving phosphorus efficiency. Agriculture, Ecosystems & Environment
212: 263–277.
Smith SE, Read DJ. 2008. Mycorrhizal Symbiosis, Third Edition. Academic Press, Oxford
Smith SE, Smith FA. 2011. Roles of arbuscular mycorrhizas in plant nutrition and growth:
new paradigms from cellular to ecosystem scales. Annual Review of Plant Biology 62: 227–
250.
Srivastava P, Saxena B, Giri B. 2017. Arbuscular Mycorrhizal Fungi: Green
Approach/Technology for Sustainable Agriculture and Environment. In: Varma A, Prasad R,
Tuteja N (eds) Mycorrhiza - Nutrient Uptake, Biocontrol, Ecorestoration. Springer
International Publishing, Cham, pp. 355–386.
Tarbell TJ, Koske RE. 2007. Evaluation of commercial arbuscular mycorrhizal inocula in a
sand/peat medium. Mycorrhiza 18: 51–56.
Tawaraya K. 2003. Arbuscular mycorrhizal dependency of different plant species and
cultivars. Soil Science and Plant Nutrition 49: 655–668.
Teng W, Deng Y, Chen XP, Xu XF, Chen RY, Lv Y, Zhao YY, Zhao XQ, He X, Li B et
al. 2013. Characterization of root response to phosphorus supply from morphology to gene
analysis in field-grown wheat. Journal of Experimental Botany 64: 1403–1411.
36
Teixeira EI, Zhao G, Ruiter JD, Brown H, Ausseil A-G, Meenken E, Ewert F. 2017. The
interactions between genotype, management and environment in regional crop modelling.
European Journal of Agronomy 88: 106–115.
Thirkell TJ, Cameron DD, Hodge A. 2016. Resolving the ‘nitrogen paradox’ of arbuscular
mycorrhizas: fertilization with organic matter brings considerable benefits for plant nutrition
and growth. Plant Cell & Environment 39: 1683–1690.
Thirkell TJ, Charters MD, Elliott AJ, Sait SM, Field KJ. 2017. Are mycorrhizal fungi our
sustainable saviours? Considerations for achieving food security. Journal of Ecology 105:
921–929.
Thompson JP. 1987. Decline of vesicular-arbuscular mycorrhizas in long fallow disorder of
field crops and its expression in phosphorus deficiency in sunflower. Australian Journal of
Agricultural Research 38: 847–867.
Thonar C, Frossard E, Smilauer P, Jansa J. 2014. Competition and facilitation in synthetic
communities of arbuscular mycorrhizal fungi. Molecular Ecology 23: 733–746.
Toljander JF, Santos-Gonzalez JC, Tehler A, Finlay RD. 2008. Community analysis of
arbuscular mycorrhizal fungi and bacteria in the maize mycorrhizosphere in a long-term
fertilization trial. FEMS Microbiology Ecology 65: 323–338.
van der Heijden MGA, Bardgett RD, van Straalen NM. 2008. The unseen majority: soil
microbes as drivers of plant diversity and productivity in terrestrial ecosystems. Ecology
Letters 11: 296–310.
van Herwaarden AF, Farquhar GD, Angus JF, Richards RA, Howe GN. 1998. Haying-
off', the negative grain yield response of dryland wheat to nitrogen fertiliser I. biomass, grain
yield and water use. Australian Journal of Agricicultural Research 49: 1067–1081.
Verbruggen E, Kiers ET. 2010. Evolutionary ecology of mycorrhizal functional diversity in
agricultural systems. Evolutionary Applications 3: 547–560.
Verbruggen E, Röling WFM, Gamper HA, Kowalchuk GA, Verhoef HA, van der
Heijden MGA. 2010. Positive effects of organic farming on below-ground mutualists:
largescale comparison of mycorrhizal fungal communities in agricultural soils. New
Phytologist 186: 968–979
Verbruggen E, Kiers ET, Bakelaar PNC, Röling WFM, van der Heijden MGA. 2012.
Provision of contrasting ecosystem services by soil communities from different agricultural
fields. Plant and Soil 350: 43–55.
Verbruggen E, Röling WFM, Gamper HA, Kowalchuk GA, Verhoef HA, van der
Heijden MGA. 2010. Positive effects of organic farming on below-ground mutualists: large-
37
scale comparison of mycorrhizal fungal communities in agricultural soils. New Phytologist
186: 968–979.
Vosátka M, Látr A, Gianinazzi S, Albrechtová J. 2012. Development of arbuscular
mycorrhizal biotechnology and industry: current achievements and bottlenecks. Symbiosis 58:
29–37.
Wani SP, Anantha KH, Garg KK. 2017. Soil properties, crop yield, and economics under
integrated crop management practices in Karnataka, Southern India. World Development 93:
43–61.
Wu T, Hao W, Lin X, Shi Y. 2002. Screening of arbuscular mycorrhizal fungi for the
revegetation of eroded red soils in subtropical China. Plant and Soil 239: 225–235.
Zhang Y, Feng L, Wang E, Wang J, Li B. 2012. Evaluation of the APSIM-Wheat model in
terms of different cultivars, management regimes and environmental conditions. Canadian
Journal of Plant Science 92: 937–949.
38
Fig. 1 An agronomic crop sequence experiment specifically designed to investigate if crop
growth and yield differed among treatments with either low or high levels of colonisation by
arbuscular mycorrhizal fungi in a low phosphorus (P) soil (11 mg kg-1 bicarbonate – Colwell
– extractable P) (Ryan & Angus, 2003). (a) Initially high inoculum levels were manipulated
in Year 1 with five contrasting treatments: fallow maintained with tillage (TF), canola
(Brassica napus) (C), fallow maintained with herbicides (CF), linola (flax; Linum
usitatissimum) (L) and subterranean clover (Trifolium subterraneum) pasture (Pa); there were
four replicates (Reps). (b) In Year 2, three crops were grown, both with (+P) and without (–P)
P-fertiliser: canola (non-host), wheat (Triticum aestivum, assumed to have a moderate
mycorrhizal growth response - MGR) and field pea (Pisum sativum, assumed to have a high
MGR). Crop growth was much increased by addition of P fertiliser. (c) Yield of the –P Year
C
L
CFPa
TF
Rep 1
Rep 2
Rep 3
Rep 4
12 m
(a) Year 1 treatments
(b) Year 2 crops12 m
Canola Wheat Peas
+P -P +P -P +P -P
15 m
+P -P +P -P
(c) Yield and mycorrhizal
colonisation of Year 2 crops
grown without P fertiliser
Yie
ld (
t h
a-1
)
Colo
nis
atio
n (%
)
Canola
Wheat
Peas
15 m
Figure 1
39
2 mycorrhizal crops did not vary with the level of colonisation despite being strongly limited
by P. Year 1 treatments had no significant impact on other measured factors with the
potential to greatly influence Year 2 crop growth (crop emergence, weeds, fungal root
pathogens, pre-season soil moisture and mineral nitrogen). Photograph in (a) courtesy of John
Angus.
Fig. 2 Shoot dry weight response, and
fitted curves, of three annual pasture
species – (a) subterranean clover
(Trifolium subterraneum), (b) yellow
serradella (Ornithopus compressus) and (c)
annual ryegrass (Lolium rigidum) – grown
in a glasshouse with 12 levels of
phosphorus (P) application, with and
without inoculation with an arbuscular
mycorrhizal fungus (AMF) (Glomus sp.)
(adapted from Schweiger et al., 1995; note
different scale of axes in (c)). Also shown
is the average root hair length at 60% of
maximum shoot growth for non-colonised
plants and the approximate critical P
application level for 90% of maximum
growth (stars). Critical P application levels
required for 90% of maximum growth in a
field experiment (Yass, New South Wales,
Australia), where roots were colonised by
indigenous AMF, were also higher for
subterranean clover (55-83 kg P ha-1) than
for yellow serradella (22-49 kg P ha-1)
(Sandral et al., 2015).
Figure 2
(a) Clover
(b) Serradella
(c) Ryegrass
P application rate (mg P kg-1 soil)
40
Fig. 3 Grain yield and the percentage of root length colonised by arbuscular mycorrhizal
fungi for irrigated (a) wheat (Triticum aestivum) and (b) maize (Zea mays) in a double-
cropped rotation system grown with six levels of phosphorus (P) application over two years
in Hebei Province, China, in a low P soil (7 mg P kg-1 of bicarbonate extractable – Olsen –
P) (adapted from Deng et al., 2017). The critical level of soil extractable P was determined to
be 22 mg kg-1 to optimise yields of wheat and maize, and it was calculated that this could be
maintained through application of 45-50 kg P ha-1 in each wheat-maize rotation (see stars on
graph), with application to the wheat only.
Co
lon
isatio
n (%
) C
olo
nis
atio
n (%
)
(a) Wheat
(b) Maize
2010 2011
Yie
ld (
t h
a-1
)
Y
ield
(t
ha
-1)
Figure 3
41
Table 1. Experimental design template for assessment of the role of arbuscular mycorrhizal fungi (AMF) in crop sequence experiments and to produce rigorous
results applicable to commercial agroecosystems and comparable with results from experiments from other crops and regions. Prepared for wheat (Triticum
aestivum) in a 2-year crop sequence in Australia: modify for other crops and regions or if testing inoculants.
Planning stage
Identify key mycorrhizal and agronomic literature for the crop(s) of interest for the region of interest and develop an hypothesis with regard to the role of AMF
Identify a suitable site and record its management history
Talk to local agronomists, farmers and others to determine the crop management practices most commonly used for commercial crops
Decide on the treatments to be used to manipulate colonisation levels (e.g. pre-crop, tillage, inoculation) avoiding those likely to have a strong independent effect on yield (e.g.
some crop species, watering regime, phosphorus (P) fertiliser rate). If possible, use multiple means to achieve contrasting levels of colonisation (Figure 1). Consider including a
non-agronomically relevant treatment to maximise the chances of finding an impact of AMF on crop growth (e.g. no P fertiliser and a crop considered to have a high mycorrhizal
growth response) and a control non-mycorrhizal crop in Year 2 (Figure 1b)
Identify the factors most likely to influence yield that may vary among treatments after year 1 and decide how to ameliorate (e.g. weeds, disease, soil water, soil nutrients)
Decide how to characterise mycorrhizal fungal abundance and the mycorrhizal fungal community
Determine on a sampling strategy that will ensure differences among Year 2 treatments in nutrition, growth and yield can be confidently ascribed to a cause (see below)
Measurements
Record all major operations on site (tillage, sowing, application of chemicals and fertilisers, grazing etc.)
Continuously record key weather data (e.g. daily air and soil temperature, rainfall and soil moisture) and note any key events (e.g. frosts, floods)
Note dates of key crop development stages (emergence, flowering, maturity)
Note any other factors that may affect yields (e.g. Year 1 crop stubble loads, Year 2 crop lodging, pest outbreaks, grain loss though shattering)
Pre-sowing Frequency Why required Notes
Soil type, macro- and
micronutrients, pH, organic matter
(top 10 cm)
Year 1 – for site To identify limitations to growth
Soil mineral N and soil water
(top 1.5 m)
Year 2 – for each
treatment
Ideally should not differ as can greatly
impact yield
Mineral N likely to be higher following legumes
Soil water likely to be higher following crops that senesced earlier
Soil water and N may be stored at depth below shallow-rooted
species
42
High soil water and N may reduce yield in a dry finish due to
“haying-off”
In-crop (noting crop phenological stage) for Year 1 and Year 2 (may only require a subset in Year 1)
Characterisation of mycorrhizal
fungal community
At least three times Assess impact of agronomic treatments
Differences among treatments may change
as the season progresses
Assess abundance of AMF using percentage of root length colonised
and, if possible, morphology (e.g. frequency of arbuscules). Check
whether fine root endophyte is present. If resources allow, measure
external hyphal density. If possible, assess mycorrhizal fungal
community structure using molecular techniques.
Plant density Seedling stage Seedling density may affect yield Crop stubble type and density may impact emergence and
establishment
Crop shoot biomass (and weeds and
volunteers of previous crop, if
present)
At least three times Shoot biomass indicates yield potential prior
to events that may differentially impact yield
of treatments
Three sampling points allows P uptake rates to be calculated
throughout the growing season
Measure at anthesis as large treatment effects can occur later due to,
for example, frost if slight differences in flowering time, or drought if
initial differences in profile soil water, or “haying-off”.
Measure root biomass at one time point if possible
Root and shoot pest and disease
rating
As appropriate May cause differences among treatments Disease may be lower if the pre-crop is a different type to the crop
(e.g. cereal, pulse, brassica)
Soil microbiome If possible Likely to change with treatments Can be assessed using molecular techniques, but, expensive and
likely to be difficult to interpret results in terms of relevance to crop
growth and yield.
Grain yield Final harvest Most agronomically relevant measure of
crop success
Look for shedding due to small grain or uneven ripening
Yield components (e.g. straw dry
weight, individual grain weight)
Final harvest Total crop biomass, harvest index and
individual grain weight all provide insight
into the factors that affected crop yield
Frost at anthesis may cause a low harvest index
Drought during grain filling may reduce grain size
Shoot, straw and grain nutrient
concentrations
As required
Key references: Angus (1981), Kirkegaard & Ryan (2014), Ryan & Angus (2003), Grant et al. (2009), Owen et al. (2010)
43
Table 2. Research priorities for arbuscular mycorrhizal fungi (AMF) in agroecosystems.
1) Determine if percentage of root length colonised is an adequate proxy for abundance of
AMF in field trials or if more time-consuming alternatives should be used (e.g.,
percentage of root length containing arbuscules, colonisation density, colonised root
length in topsoil, or density of external hyphae in topsoil).
2) Determine the plant genotype factors that influence the mycorrhizal growth response
under field conditions.
3) Test the relationship between abundance of AMF and crop yield across a range of crop
management systems and locations with rigorous standardised methodology (Table 1).
4) Determine if mycorrhizal fungal community composition or diversity, or interactions
with the soil microbial community including fine root endophyte, are limiting the
mycorrhizal growth response of crops in the field.
5) Test resource balance concepts in the field including whether the mycorrhizal growth
response is greatest when phosphorus is limiting and crop growth rate high due to high
light, optimal temperature, sufficient water and no other growth-limiting nutrients or
environmental factors.
6) Modify a crop growth model with results of 1) to 5) and determine if it can predict under
which circumstances (crops, regions, soil extractable phosphorus level, etc.) management
unfavourable to abundance of AMF, e.g. high-intensity tillage, long bare fallows or non-
mycorrhizal crops in the rotation, results in a yield penalty.
7) Investigate under field conditions whether impacts of AMF on agroecosystem
sustainability can be quantified and are significant in an agronomic context (e.g. soil
structural maintenance, reduction in phosphorus leaching etc).
8) Further test whether the external critical phosphorus requirement framework can be used
to identify a soil extractable phosphorus level at which the benefits of mycorrhizas to
crop yield and agroecosystem sustainability, and crop yields, are desirably balanced.
