Break crops and rotations for wheat

Abstract

Wheat crops usually yield more when grown after another species than when grown after wheat. Quantifying the yield increase and explaining the factors that affect the increase will assist farmers to decide on crop sequences. This review quantifies the yield increase, based on >900 comparisons of wheat growing after a break crop with wheat after wheat. The mean increase in wheat yield varied with species of break crop, ranging from 0.5tha-1 after oats to 1.2tha-1 after grain legumes. Based on overlapping experiments, the observed ranking of break-crop species in terms of mean yield response of the following wheat crop was: oats<canola ≈ mustard ≈ flax<field peas ≈ faba beans ≈ chickpeas ≈ lentils ≈ lupins. The mean additional wheat yield after oats or oilseed break crops was independent of the yield level of the following wheat crop. The wheat yield response to legume break crops was not clearly independent of yield level and was relatively greater at high yields. The yield of wheat after two successive break crops was 0.1-0.3tha-1 greater than after a single break crop. The additional yield of a second wheat crop after a single break crop ranged from 20% of the effect on a first wheat crop after canola, to 60% after legumes. The mean yield effect on a third wheat crop was negligible, except in persistently dry conditions. The variability of the break-crop effect on the yield of a second wheat crop was larger than of a first wheat crop, particularly following canola. We discuss the responses in relation to mechanisms by which break crops affect soil and following crops. By quantifying the magnitude and persistence of break-crop effects, we aim to provide a basis for the decision to grow continuous cereal crops, strategic rotations or tactically selected break crops. In many wheat-growing areas, the large potential yield increases due to break crops are not fully exploited. Research into quantifying the net benefits of break crops, determining the situations where the benefits are greatest, and improving the benefits of break crops promises to improve the efficiency of wheat-based cropping systems.

Full text

Break crops and rotations for wheat
J. F. Angus
A,B,E
, J. A. Kirkegaard
A,B
, J. R. Hunt
A,B
, M. H. Ryan
C
, L. Ohlander
D
,
and M. B. Peoples
A
A
CSIRO Agriculture Flagship, GPO Box 1600, Canberra, ACT 2601, Australia.
B
E.H. Graham Centre for Agricultural Innovation, Charles Sturt University, Locked Bag 588,
Wagga Wagga, NSW 2678, Australia.
C
School of Plant Biology and Institute of Agriculture, MO84, University of Western Australia,
35 Stirling Highway, Crawley, WA 6009, Australia.
D
Department of Crop Production Ecology, Swedish University of Agricultural Sciences,
SE-750 07, Uppsala, Sweden.
E
Corresponding author. Email: john.angus@csiro.au
Abstract. Wheat crops usually yield more when grown after another species than when grown after wheat. Quantifying
the yield increase and explaining the factors that affect the increase will assist farmers to decide on crop sequences.
This review quantifies the yield increase, based on >900 comparisons of wheat growing after a break crop with wheat
after wheat. The mean increase in wheat yield varied with species of break crop, ranging from 0.5 t ha
–1
after oats to 1.2 t ha
–1
after grain legumes. Based on overlapping experiments, the observed ranking of break-crop species in terms of mean yield
response of the following wheat crop was: oats < canola  mustard  flax < field peas  faba beans  chickpeas  lentils
 lupins. The mean additional wheat yield after oats or oilseed break crops was independent of the yield level of the following
wheat crop. The wheat yield response to legume break crops was not clearly independent of yield level and was relatively
greater at high yields. The yield of wheat after two successive break crops was 0.1–0.3 t ha
–1
greater than after a single break
crop. The additional yield of a second wheat crop after a single break crop ranged from 20% of the effect on a first wheat crop
after canola, to 60% after legumes. The mean yield effect on a third wheat crop was negligible, except in persistently dry
conditions. The variability of the break-crop effect on the yield of a second wheat crop was larger than of a first wheat crop,
particularly following canola. We discuss the responses in relation to mechanisms by which break crops affect soil and
following crops. By quantifying the magnitude and persistence of break-crop effects, we aim to provide a basis for the
decision to grow continuous cereal crops, strategic rotations or tactically selected break crops. In many wheat-growing areas,
the large potential yield increasesduetobreakcropsare not fully exploited. Research into quantifying the net benefits of break
crops, determining the situations where the benefits are greatest, and improving the bene fits of break crops promises to
improve the efficiency of wheat-based cropping systems.
Additional keywords: canola, crop sequence, grain legumes, meta-analysis, nitrogen, oilseeds, root disease, soil water.
Received 1 September 2014, accepted 8 March 2015, published online 29 May 2015
Introduction
Crops generally produce greater yields when grown after
unrelated species. This yield benefit has come to be known as
the rotation or ‘break-crop’ effect. Crop rotation is an ancient
concept although the use of the term break crop is relatively
recent, first appearing in UK papers about wheat-based cropping
sequences in the 1960s (Selman 1969). The term originally meant
breaking a sequence of continuous wheat crops with another
species, but now extends to crops other than wheat and has also
picked up the connotation of breaking the life cycle of pathogens.
The distinction between rotation and break crop is that rotation,
strictly defined, refers to a recurring sequence of crops, forages
and fallows. More loosely defined, rotation refers to a cropping
sequence that contains fallows, or crops and forages in addition
to the locally dominant species, which in this review is wheat.
A break crop generally refers to a single alternative crop followed
by wheat.
The value of rotations in increasing yield has been recognised
since at least the 10th century BCE in China (Karlen et al. 1994)
and the 4th century BCE in Greece when Theophrastus wrote
that ‘wheat exhausts the land more than any other crop’ and
‘beans...even seem to manure it’ (Hort 1926). Other examples of
early recommendations for crop rotations are discussed by Curl
(1963). Fallowing was widespread in ancient Roman times when
agronomists recommended crop rotation, although the extent of
adoption of crop rotations is not clear (White 1970). It seems
likely that legume rotations would have been eventually needed to
supply nitrogen (N). There was widespread adoption of rotations
Journal compilation  CSIRO 2015 www.publish.csiro.au/journals/cp
CSIRO PUBLISHING
Crop & Pasture Science , 2015, 66, 523–552
Review
http://dx.doi.org/10.1071/CP14252

in continental Europe by the end of the 18th century, one of which
was popularised in eastern England as the Norfolk four-course
rotation, consisting of a sequence of turnip, barley, clover and
wheat (Evans 1998).
The first experimental justification of rotations was when
Daubeny (1845) reported a 10-year experiment in which crops
grown in rotation outyielded monocultures in about three-
quarters of comparisons. A more structured experiment by
Lawes and Gilbert (1894), which started after Daubeny
reported his results, showed the value of a rotation from a
comparison of continuous crops and a four-course rotation at
Rothamsted (Table 1). Mean yields of the broadleaf crops, turnip
and faba beans were much greater when grown in rotation than
when grown continuously, barley yields were similar, and wheat
yield was 0.57 t ha
–1
or 41% higher when grown in rotation. In
this example, the productivity of the rotation was clearly greater
than that of any continuous crop.
Despite the long practical and experimental evidence for the
advantages of rotations and break crops, many farmers grow
wheat in continuous monocultures, or at least grow many wheat
crops in sequence (Cook and Weller 2004; Robertson et al. 2010).
Factors that favour the expansion of cereal monoculture are the
increased availability of inexpensive (relative to grain price) N
fertiliser and herbicides. Factors that have favoured expansion of
break crops are better adapted, lower risk and higher yielding
broadleaf species, particularly canola, the global area of which has
increased annually at an average rate of 4.3% between 1961 and
2013 while wheat area has remained constant (FAOSTAT 2014).
Over the same period the areas of other temperate break crops,
flax, oats and grain legumes, have decreased, with average annual
rates of 0.6% 0.5% and 0.2%, respectively. Exceptions to the
reduction in temperate grain legumes have been in Canada, with
an average annual increase of 6% from 1990 to 2013 (Miller et al.
2015), with similar increases in adjacent parts of the USA. In
Australia, the area of grain legumes increased annually at an
average rate of 11% from 1980 to 2000, followed by a reduction
of about half the previous increase after a series of droughts.
The relatively poor adoption of break crops has puzzled and
frustrated commentators who have emphasised the rotational
benefits and diversified income (Leighty 1938; Robertson
et al. 2010; Preissel et al. 2015). Economic reasons for poor
adoption are the delay in obtaining the additional returns from the
second crop and relatively low prices where governments
subsidise staples such as wheat but not alternative crops
(Fischer et al. 2014). Another economic reason is poor price
stability; there are few international markets for grain legumes
comparable to those for cereals or oilseeds. Furthermore,
monoculture provides farmers with advantages of simpler
management and machinery than more diverse systems. Some
soils are less suitable for break crops than for wheat, and in some
wheat-growing regions such as the US Central Great Plains, there
are currently no well-adapted alternative crops to cereals. Where
alternative crops can be grown, diversified cropping sequences
are not always more profitable than continuous wheat. For
example, the net returns from continuous wheat may be
greater and more reliable than from a diversified cropping
sequence in dry environments in Australia (Kirkegaard et al.
2001; Whitbread et al. 2015) but less reliable in the US Northern
Great Plains (Miller et al. 2015). The results of individual crop-
sequence experiments in variable climates may be influenced
by the prevailing weather, so they should be interpreted in relation
to long-term conditions.
Furthermore, a biologically optimum crop sequence may
produce the wrong proportion of products for a market.
A rotation such as that in Table 1, consisting of two cereals in
4 years, would, if applied on farms, lead to only half the cropped
area being sown to cereals. Individual farms in some areas may
profitably adopt such a diversified system but a whole industry
based on such a system would not meet the requirements of
grain markets, as can be seen from the following estimates of
crops grown in wheat-based systems. Globally, the area of wheat
grown in 2008–10 was 222 Mha and the areas of the temperate
break crops, canola and cool-season grain legumes, were 32 and
25 Mha, respectively (FAOSTAT 2014). In South and East Asia,
26 Mha of wheat may be double-cropped with rice (Timsina
and Connor 2001), and globally, ~6 Mha is double-cropped with
soybean, mostly in South America (Calviño and Monzon 2009;
J. Monzon, pers. comm. 2014). It is difficult to estimate accurately
the areas of other wheat crops that benefit from a rotation or
break crop. These include wheat alternating with fallows in
some temperate dryland environments, or as a winter crop
alternating with a summer fallow in subtropical environments,
and the declining area of wheat grown in phased rotation with
pastures, mainly annual-legume-based, in Australia (currently
about 7M ha; Fischer et al. 2014). Together, these are likely to
total ~50 Mha. Assuming that the known break crops and fallows
are followed by wheat, these data suggest that globally, ~40% of
the area sown to wheat may not be preceded by an effective break
crop, forage or fallow.
The present paper complements previous reviews of break-
crop effects (Kirkegaard et al. 2008; Peoples et al. 2009),
rotations (Bullock 1992; Karlen et al. 1994), and disease
aspects of rotations (Curl 1963). The aim is not to advocate
fixed rotations or the more frequent use of break crops. One aim
is to quantify the value of break crops to wheat production and
to identify the conditions that cause variation in the magnitude
and persistence of the break-crop bene fit. With reliable
information about the costs and benefits of break crops,
farmers will be in a better position to choose continuous
cereals or break crops, either as a long-term strategy or as a
tactical response to conditions of field, season and markets.
Other aims are to highlight gaps in understanding the
responses of wheat to break crops, to point out researchable
problems that limit the size and reliability of the break-crop effect,
and to draw attention to the potential for break crops in other
cropping systems.
Table 1. Comparison of dry matter yields (t ha
–1
) of grain (or roots
and shoots of turnips) for crops grown continuously or in a four-course
rotation over eight cycles at Rothamsted, UK, 1852–83 (Lawes and
Gilbert 1894)
Crop names are defined in Table 3
Grazed turnips Barley Faba beans Wheat
Continuous 2.10 2.57 0.65 1.39
Rotation 3.44 2.55 1.45 1.96
Percentage yield increase 63 –1 123 41
524 Crop & Pasture Science J. F. Angus et al.

Types of crop-sequence experiments
Crop-sequence experiments can be classified into four general
groups (Table 2). The oldest is rotation, as described in the
Introduction. Many of the earliest rotation experiments were
not replicated but provided useful information to farmers
because the yield responses were so large. Many recent rotation
experiments were managed in phased designs in which each crop
in a sequence was grown each year (Patterson 1964). Such designs
allow statistically valid comparisons but become complicated
and expensive when comparing phases of different durations.
Mobile insect pests are more difficult to manage with fully
phased rotation experiments than with simpler designs
(P. R. Miller, pers. comm. 2014).
In addition to the Rothamsted experiment in the UK (Lawes
and Gilbert 1894), there have been long-term experiments in
Canada (Ripley 1941; Campbell et al. 1990), Sweden (Ebbersten
1980) and Germany (Sieling and Christen 2015). Reports of
many others are incomplete or in the ‘grey’ literature, probably
because the management methods, cultivars or even crop species
became irrelevant to contemporary farmers or because the
designers retired or lost interest by the time usable data
became available.
When a rotation consists of more than two crops, it is not
possible to conclude whether the whole sequence is needed to
provide the yield benefits to the cereal in the sequence or whether
part of the sequence provides most, or all, of the benefits. Only
two-course rotation experiments can quantify the effect of a
particular break crop on cereal yield, and even then, it may not
be clear whether the increased wheat yield is due to a single break-
crop phase or repeated cycles. An example, reported by Harris
(1995) and Ryan et al. (2008), was a two-course rotation
experiment in Tel Hadya, Syria, where the average annual
rainfall was 330 mm. Over 13 years, this experiment showed
consistent rankings of wheat yield in response to the pre-crop
treatments, which were, in increasing order: wheat, chickpeas,
lentils and fallow.
The strength of long-term rotation experiments is that
they provide reliable information about changes in soil
properties over time (Powlson and Johnston 1994), which in
the case of soil organic matter or structural improvements may
be incremental. Their weakness, apart from attribution to specific
crops within the rotation, is that the delay between the design of
an experiment and the eventual arrival of statistical significance
among treatments can be so long that the system under study
may become obsolete and no longer relevant to contemporary
farmers.
The second class of crop-sequence experiments, which
constitutes most of this review, involves break crops followed
by wheat. These biennial experiments compare replicated plots of
one or more break-crop species and a wheat control in the
first year, followed by wheat superimposed on the previous
plots in the second year. Break-crop experiments are more
attractive to modern funding systems than rotation
experiments, and allow flexibility to move to fields with
particular starting conditions (high disease, low available soil
N, etc.) in order to clarify the likely impact under specific
conditions. Such experiments can become valuable when
repeated over sites and years as information about explanatory
factors accumulates. It is even possible to see that the mechanisms
of break-crop benefits change over time as a cropping system
evolves (Seymour et al. 2012). Some of these experiments
investigate the interaction of other factors affecting the
following wheat crop, such as tillage, fertiliser or variety
(Olofsson 1993). Surveys are another source of data about
break crops; farmers reported yields of wheat after wheat and
after break crops on matched fields (Bourgeois and Entz 1996;
Angus et al. 2011; Moodie 2012), and these showed similar
results to experiments.
The third class of experiment involves extension of
the biennial experiments described above to examine the
persistence of the break-crop effect by growing wheat for
2 years after the break crop (Hanley and Ridgman 1978;
Olofsson and Wallgren 1984; Gan et al. 2003; Seymour et al.
2012; Kirkegaard and Ryan 2014; McBeath et al. 2015). These
experiments are particularly important because they can help to
quantify the full benefits of break crops in regions where cereals
constitute most of the cropped landand extended phases of cereals
after a single break-crop year are commonplace.
The fourth class of experiment tests two or more break crops
grown in sequence before wheat. Testing a ‘double break’ can
explore the possibility of a larger or more long-lasting effect than
from a single-break crop. This has become increasingly important
for control of herbicide-resistant weeds, the seedbanks of which
may take several seasons to diminish. Even when such sequences
are not commercially relevant, they can help to explain how
simpler crop sequences affect yield and are also useful in defining
the maximum yield potential. Variations and combinations of
these types of experiments are possible, for example, two break
crops followed by successive wheat crops (Ridgman and Walters
1982; Kirkegaard and Ryan 2014).
This review combines data from experiments on single break
crops, break-crop persistence and double break crops in the
Table 2. Types of crop-sequence experiments, with examples
Year Rotation Single break crop Break-crop persistence Double break
Control Treatment Control Treatment Control Treatment Control Treatment 1 Treatment 2
1 Wheat Break crop 1 Wheat Break crop Wheat Break crop Wheat Wheat Break crop 1
2 Wheat Break crop 2 Wheat Wheat Wheat Wheat Wheat Break crop 2 Break crop 2
3 Wheat Wheat Wheat Wheat Wheat Wheat Wheat
4 Wheat Break crop 1
5 Wheat Break crop 2
6 Wheat Wheat
Example Harris 1995 Gregory 1998 Hanley and Ridgman
1978
O’Donovan et al. 2014
Break crops and rotations for wheat Crop & Pasture Science 525

expectation that a large dataset will provide insights that are not
available from single experiments.
Materials and methods
The review is based primarily on data collated from published
reports of the effects of break crops on wheat harvested for grain,
plus some data from unpublished reports, mostly in south-eastern
Australia. In most cases, the control treatment was wheat after
wheat and the treatments were wheat after alternative crops
(Table 2). We do not review the break-crop effects of forages,
pastures and cover crops, because their productivity and botanical
composition are not always well defined, and grazing by animals
introduces further complexity. We also avoided crop sequences
where high-input crops such as potatoes or sugar beet preceded
wheat, owing to a risk that residual nutrients following excessive
fertiliser usein these crops would dominate the responses. Species
of crops and pathogens are referred to by their common names as
identified in Table 3.
The papers reviewed here focus on wheat because wheat has
generated most of the farm income in many global farming
systems in the past. In some contemporary systems, other
crops such as soybeans or canola may be more important. In
principle, it should be possible to quantify the effect of each crop
species on other subsequent crops from matrix experiments and
use the relationships between them, and an understanding of
processes, to recommend strategies and tactics for cropping
sequences (Kirkegaard et al. 2001; Krupinsky et al. 2006;
Williams et al. 2014; Malik et al. 2015). Wheat grows mainly
between latitudes 258 and 608, generally during winter and spring
at the lower latitudes, and during spring and summer at the higher
latitudes. At the higher latitudes there is time for only one crop
each year, and break crops are grown at the same time as wheat.
At mid and low latitudes within its range, wheat also grows in
sequence with these species, with fallows in dry regions, and in
double-crop rotations with summer crops such as cotton, maize,
sorghum, soybeans, rice and sunflower. Wheat also grows at
high elevations in the humid tropics, generally during a cooler
and drier season. Over half of the global wheat production comes
from developed countries, and three-quarters is grown in rainfed
systems (Fischer et al. 2014).
Most of the data come from reports of field experiments
conducted in Australia, Northern and Western Europe, and
North America, with a small number from Asia and South
America. None of the reports mentioned that the selected
experimental sites were chosen in the anticipation of a large
response to break crops, so no specific bias is expected in the
results. Appendix 1 shows the origin of the data and the break crop
sequences reported in each paper. Supplementary materials, held
at the journal’s website, consists of a spreadsheet containing all
the data used in the analyses. All experiments compared the
yield of wheat following a break crop with a control sequence of
wheat following wheat, except for a few cases where the control
sequence was barley–wheat, presumably because wheat–wheat
was not considered commercially relevant. In a later section, we
show that the yield of wheat after barley is similar to wheat after
wheat, so barley–wheat is a valid control treatment.
Many experimental series report results from several seasons
or several locations and generally these were each included as
separate comparisons. Where there were several levels of N
fertiliser, yield from the middle level or, where reported, from
only the optimum N level was included in the dataset. In other
experiments where break crops were assessed in combination
with other factors such as tillage, each treatment was included
separately. Data for winter and spring wheat were combined
because their yield response was similar (Miller and Holmes
2005). Where there were relatively few examples of a crop
species, results were combined with a closely related species;
for example, white mustard was combined with canola, and white
lupin with narrow-leaf lupin. The Supplementary data refer to the
species tested. In some experiments, the soil was inoculated with
pathogens to reduce the variability of the results (Kollmorgen
et al. 1983; Kirkegaard et al. 2004; Cunfer et al. 2006). The data
from those experiments were not used in estimating the yield
benefit of break crops.
Table 3. Names of crops, diseases and pathogens mentioned in the text
Common and alternative names Latin names
Crop species
Wheat Triticum aestivum L.
Barley Hordeum vulgare L.
Oats Avena sativa L.
Triticale Triticosecale Wittm. ex A. Camus
Rye Secale cereale L.
Maize Zea mays L.
Sorghum Sorghum bicolour (L.) Moench. S.lat
Rice Oryza sativa L.
Canola, rape, oilseed rape Brassica napus L.
Mustard, Indian mustard,
juncea canola, brown mustard,
oriental mustard
Brassica juncea L.
Turnip, turnip rape Brassica rapa var. rapa
Field peas Pisum sativum L
Narrow-leaf lupins, blue lupins Lupinus angustifolius L.
White lupins, albus lupins Lupinus albus L.
Chickpeas Cicer arietinum L
Faba beans, tick beans Vicia faba L.
Lentils Lens culinaris Medicus
Soybeans Glycine max (L.) Merr.
Flax, linseed, linola Linum usitatissimum L.
Safflowers Carthamus tinctorius L.
Sunflowers Helianthus annuus L.
Potato Solanum tuberosum L.
Sugar beet Beta vulgaris L.
Cotton Gossypium hirsutum L.
Diseases and pathogens
Take-all Gaeumannomyces graminis (Sacc.)
Arx and Oliv. var. tritici Walker
False eye spot Wojnowicia graminis (McAlp.) Sacc.
and Sacc.
Common root rot Bipolaris sorokiniana (Sacc.) Shoem
Crown rot Fusarium pseudograminearum,
F. culmorum
Eyespot Cercosporella herpotrichoides
Fron
Rhizoctonia bare patch Rhizoctonia solani
Cereal cyst nematode Heterodera avenae Woll.
Root-lesion nematodes Pratylenchus thornei Sher and Allen,
P. neglectus (Rensch) Filipjev
and Schuumans
526 Crop & Pasture Science J. F. Angus et al.

The largest set of data was obtained from series of crop-
sequence experiments conducted in Sweden from the 1930s
until late in the 20th Century by the Agricultural College of
Sweden and its successor, the Swedish University of Agricultural
Sciences. (Nilsson and Wallgren 1981; Olofsson and Wallgren
1984; Nilsson 1985; Wallgren 1987; Lindénand Engström 2006).
Each experimental series consisted of similar experiments at
several locations (usually >20), and generally on farmers’
fields. The earlier experiments consisted of two replicates, but
the data are reported only as mean, with no statistical analysis of
individual experiments, except by Lindén and Engström (2006).
The purpose of the earlier experimental programs was to assess
the average productivity of cropping sequences so that
government agencies could set prices to promote national self-
sufficiency of each commodity.
Statistical analyses
The usual method of analysing several datasets is by formal
meta-analysis, a method that assumes that the outcome of an
imposed treatment can validly be expressed as a percentage
difference. As shown below, often the yield response to a
break crop is best not expressed as a percentage difference, but
rather as an offset. That is, the mean yield response to a break crop
is independent of the yield of wheat after wheat. Formal meta-
analysis is therefore unsuitable for analysing these data. Another
reason for not conducting a formal meta-analysis was the
difficulty of weighting different datasets, some of which were
means of many years of a rotation experiment, whereas others
were from a single year. For several datasets, there was no
estimate of error for crop yields, and for others the estimated
error was reported in forms such as multiple range tests that cannot
be converted to standard deviation, as required for formal meta-
analysis. Although about half of the data included estimates of
variation, we analysed all data using only means of treatments
within each experiment. This approach constitutes meta-analysis,
defined by Glass (1976) as a large collection of analysis results
from individual studies for the purpose of integrating the findings.
In this review, each data point on a graph and observation is
the mean of the replicated plots of a treatment within an
experiment. The yield after a break crop is related to the yield
of wheat–wheat using linear regression. The coefficient of the
intercept is significantly greater than 0 when the ratio of the
coefficient to its standard error is >2. The coefficient of the slope
is significantly greater than 1.0 when it differed from this value
by more than twice its standard error. In comparing the effects
of two or more species with wheat–wheat, we represented each
species as a factor in a general linear model (Minitab Release
12; Minitab Inc., State College, PA, USA). The F-ratio of the
ANOVA then indicates whether there are significant differences
among species and the coefficients indicate whether the species
differ in their intercepts or slopes. Claims of treatment differences
in this review mean that the differences were statistically
significant at P = 0.05.
Data review
This section contains data about experiments in which plots of
one or more break crops and a wheat control are grown in the
first year and wheat crops are grown on the same plots in the
following year (break crops in Table 2). For each species or group
of break crops, we present the data as graphs showing the mean
yield of wheat growing after a break crop plotted against the
mean yield of wheat growing after wheat for the range of
experiments included, along with linear regressions between
yields of break crop–wheat and wheat–wheat.
Barley and oats
The yield of wheat growing after barley was similar to the yield
of wheat growing after wheat in 60 experiments (Fig. 1a).
A linear regression fitted to the data shows that the yield of
wheat following barley was not significantly greater than after
wheat (Table 4). The absence of a break-crop benefit presumably
reflects the similar biology, root-disease spectrum and nutrition
of wheat and barley.
0246810
Yield of wheat after oats (t ha
–1
)
Yield of wheat after wheat (t ha
–1
)
0246810
Yield of wheat after barley (t ha
–1
)
0
2
4
6
8
10
(a)
(b)
Fig. 1. Yield of wheat grown (a) after barley compared with wheat after wheat in the same experiments,
and (b) after oats compared with wheat after wheat growing together in a different set of experiments.
Symbol colours refer to regions where the experiments were conducted: green, Australia; blue, Sweden;
pink, other Europe; red, North America. The dashed lines represent equal yields and the solid lines
fitted equations reported in Table 4.
Break crops and rotations for wheat Crop & Pasture Science 527

The yield of wheat after oats was significantly greater than
wheat after wheat when compared over 150 experiments (Fig. 1b,
Table 4). The yield difference appeared as an intercept that was
significantly greater than 0 (0.53  0.14 t ha
–1
), but a slope that
was not significantly different from 1.0. This result suggests
that the break-crop effect of oats does not provide an average
percentage yield increase but that the average yield increase is
constant over a wide range of yields.
Oats were widely grown in sequence with wheat before
well-adapted broadleaf crops became available and when there
was a larger market for horse feed. The popularity of oats, apart
from its value as a break crop, may have been due to its value in
suppressing weeds when cut for hay before the development of
selective herbicides; with the development of herbicide-resistant
weeds, oats may yet make a comeback internationally, as it
has in Western Australia. The reason for interest in oats in this
review is they do not host root pathogens of wheat such as take-all
(Russell 1934), common root rot (Ledingham 1961) or Pythium
damping off (Olsson and Kadir 1994). Furthermore, oat roots
contain avenacin, a glucoside that inhibits the take-all fungus
(Turner 1961).
Kirkegaard et al. (1996a) reported Australian experiments in
which oats were a relatively poor break crop for wheat (i.e. there
was no average difference between yields after wheat and after
oats), so it appears that those results were not representative of
the wider response. Subsequent experiments in the same region
found that oats, triticale and barley could all provide significant
break-crop benefits to wheat (Kirkegaard and Ryan 2014). Other
experiments testing rye and triticale are collated in the
Supplementary data, but data were insufficient to allow a
reliable assessment of the break-crop effects.
Brassicas
Cereals are known to yield well after Brassica and other
cruciferous crops (Ebbersten 1980). These include the species
grown for forage as well as those harvested for grain, which are
reviewed here. Most of the reviewed data are from canola and
mustard, but some data from Sweden are from the related species
white mustard and turnip rape.
Figure 2 shows a comparison of canola–wheat with
wheat–wheat in 180 comparisons and Table 4 presents the
regression equation. The intercept of 0.80  0.17 t ha
–1
and
slope of 1.00  0.04 show that the mean yield response was
significant and uniform across the full range of yields reviewed
and, as with the oat data, not proportional to yield. There is no
evidence of differences in response between the different regions
(data not shown).
Wheat grown after mustard, another Brassica grain species,
appeared to yield more than after canola in the first comparisons
of these species as break crops (Angus et al. 1991; Kirkegaard
et al. 1994). In those experiments, much of the break-crop benefit
appeared to be due to control of root disease, leading to the
hypothesis that the different effect of these Brassica species was
an active process, called biofumigation (Angus et al. 1994), rather
than absence of host as had been widely assumed. This is
discussed further under Control of root diseases.
This hypothesis was tested in 35 experiments in south-
eastern Australia comparing canola–wheat and mustard–wheat
sequences with wheat–wheat, as summarised in Fig. 3a. A linear
Table 4. Regression equations relating yield (t ha
–1
) of wheat after break crop to wheat after wheat
s.e., Standard error
Crop sequence Intercept ± s.e. Slope ± s.e. R
2
(%) No. of comparisons Figure
Barley–wheat 0.16 ± 0.18 1.00 ± 0.04 89 60 Fig. 1a
Oats–wheat 0.53 ± 0.14 0.99 ± 0.04 85 150 Fig. 1b
Canola–wheat 0.80 ± 0.17 1.00 ± 0.04 77 180 Fig. 2
Mustard–wheat, canola–wheat 0.57 ± 0.22 1.00 ± 0.06 79 35 Fig. 3a
Flax–wheat, canola–wheat 1.26 ± 0.29 0.90 ± 0.07 65 42 Fig. 3b
All legumes 0.92 ± 0.12 1.06 ± 0.04 69 300 Fig. 4
Lupins 1.61 ± 0.45 0.94 ± 0.19 25 75
Other legumes 0.72 ± 0.11 1.08 ± 0.04 80 225
Fallow and break crop Fig. 5
Fallow–wheat 1.12 ± 0.30 0.90 ± 0.16 51 32
Break crop–wheat 0.75 ± 0.28 1.13 ± 0.15 65 32
Yield of wheat after wheat (t ha
–1
)
02468
Yield of wheat after canola (t ha
–1
)
0
2
4
6
8
Fig. 2. Yield of wheat after canola compared with wheat after wheat
growing in the same experiments. Experimental locations: *, Australia;
&, Sweden; ~, other Europe; $, North America. The dashed lines represent
equal yields and the solid line the fitted equation reported in Table 4.
528 Crop & Pasture Science J. F. Angus et al.

model showed no significant difference in the relationships and
they were best described with a single regression equation
(Table 4). The coefficients of this equation indicated that the
yield benefits in these comparisons were not proportional to yield.
The intercept is less than for the canola-only data, probably
because the experiments were in lower yielding environments
and seasons.
Other research showed that mustard and canola varieties with
different concentrations of glucosinolates were equally effective
in reducing the incidence or severity of root disease in the
following wheat and led to no consistent differences in wheat
yield (Kirkegaard et al. 2000; Smith et al. 2004). These results
lead to the conclusion that both biofumigation and absence of
host may be mechanisms by which brassicas control root disease
of wheat and for annual crops, biofumigation confers no
additional yield benefit over depriving wheat pathogens of a
host. For crop cycles of <1 year, the effects of biofumigation
may provide greater pathogen control than host-deprivation
(Brown and Morra 1997).
Flax
Bedford (1899) reported a ‘comparative small (wheat) yield
after flax that agrees with the general experience of farmers in
Manitoba’ in what was probably the first quantitative assessment
of a break crop. The poor reputation of flax as a break crop
continued when Zentner and Campbell (1988) reported that wheat
after flax yielded less than wheat after wheat because of weeds
carried over from the flax crops and lack of residues to conserve
soil water. On the other hand, flax can be grown with herbicides
that are not suitable for other broadleaf crops, and other
experiments have shown that the yield of wheat after flax is
comparable to wheat after field peas (Miller and Holmes 2005).
Flax, as a broadleaf crop, is a non-host to cereal root pathogens
and it might be expected to provide an effective break for root
diseases of wheat. Flax roots contain linamerin, a cyanogenic
glycoside that hydrolyses and releases cyanide in the root-
zone (Rovira 1969) and could therefore have the potential for
biofumigation effects (Angus et al. 1994). Kirkegaard et al.
(2000) found more take-all inoculum after flax than after
brassicas, suggesting that flax exerts a weaker biofumigation
effect. Flax can be highly dependent on arbuscular mycorrhizal
fungi (AMF) for growth and nutrient uptake (Thompson et al.
2013). In this respect, it differs from Brassica crops, which are not
hosts of AMF (Thompson and Wildermuth 1989).
Despite these differences, the value of flax as a break crop was
statistically indistinguishable from canola when both were grown
in the same 42 experiments (Fig. 3b, Table 4). Most of these
comparisons were undertaken in Australia, following the release
of linola, a form of flax that produces grain low in linolenic acid
(Green 1986).
Grain legumes
The effect of growing cereals after grain legumes has probably
been studied more than the rotational benefits of other crops, and
there are 300 comparisons of legume–wheat and wheat–wheat
sequences in the dataset. Wheat following legumes bene
fits from
increased availability of soil N, a break in the life cycle of
soilborne pathogens, and other mechanisms discussed below.
Unlike oats and oilseeds, there is no evidence that legumes
actively suppress root disease by releasing toxins into the soil.
There is a generally unstated assumption that absence of a host
explains the low levels of cereal root disease after legumes.
The dataset contains yields of wheat growing after field peas,
lupins, faba beans, chickpeas and lentils, not all growing in the
same experiments. Yield of wheat after the combined legumes
generally exceeded the yield of wheat after wheat (Fig. 4,
Table 4). Based on a linear model of the combined legume
data, the lupin relationship was significantly different from the
relationships of the other four species. For a wheat–wheat yield
of 4.0 t ha
–1
, the mean grain legume–wheat yield was 5.2 t ha
–1
.
The different effect of lupin may be due to different
experimental conditions, rather than an inherently greater
02468
Yield of wheat after flax or canola (t ha
–1
)
Yield of wheat after wheat (t ha
–1
)
02468
Yield of wheat after mustard or canola (t ha
–1
)
0
2
4
6
8
(a)
(b)
Fig. 3. Yield of wheat grown (a) after mustard or canola in the same experiments and (b) after flax or
canola growing together in a different set of experiments. The dashed lines represent equal yields and the
solid lines fi tted equations reported in Table 4. Symbol colours represent region where the data come from:
green, Australia; blue, Sweden; red, North America. Break crops:
&
, canola;
~
, mustard;
*
, flax.
Break crops and rotations for wheat Crop & Pasture Science 529

break-crop effect. To test these possibilities, we selected a subset
of 33 experiments that compared lupins and field peas and
compared the yield of wheat after these species with
wheat–wheat. In this subset, the differences in yields of
lupin–wheat and field pea–wheat were not statistically
different. The apparent superiority of lupins as a break crop
suggested in Fig. 4 does not appear to reflect a real effect and
may be driven by one value with high leverage.
Fallow
Fallowing, the removal of all plants from the soil surface by
cultivation or herbicides, affects the yield of the following wheat
crop by some of the same mechanisms as break crops (Connor
et al. 2011). Like break crops, clean fallows provide a disease
break and an opportunity to control intractable weeds of wheat.
Both fallows and legume break crops may supply mineral N to
the following crop; legumes may also provide hydrogen (H
2
)
fertilisation, as discussed below. Unlike most break crops, fallows
can conserve soil water, and the additional water is probably the
most consistent benefit of fallows, particularly in semi-arid
regions and on soils with high water-holding capacity (Sims
1977). Fallowing by cultivation in wet environments can cause
soil compaction, which is whyyields of crops after fallows are less
than after break crops in long-term experiments in Sweden
(L. Ohlander, unpubl. data).
The data set contains the results of 32 experiments comparing
yield of wheat after fallow with wheat after one or more break
crops. Where more than one break crop species was tested in an
experiment, the species selected was the one that gave the highest
wheat yield. Figure 5 shows that both fallow–wheat and break
crop–wheat generally gave higher yields than wheat–wheat.
A general linear model fitted to the data showed that fallow
and break crops had significantly different effects on wheat
yield. The lines in Fig. 5 and the regression equations in
Table 4 indicate that, at low yields, fallows boosted low wheat
yields more than did break crops, but break crops had a greater
effect when wheat–wheat yields were >1.7 t ha
–1
. At this yield
level, the increase in yield due to fallow was 0.95 t ha
–1
, almost
twice the average yield response to fallow for Australian-only
experiments reviewed by Oliver et al. (2010). Those authors
simulated the effect of the additional soil water provided by
fallow and showed that it was most beneficial on soils with high
water-holding capacity and if the seasons after fallow were dry.
The simulated effect of dry seasons supports the relatively greater
yield effects of fallow at low yield levels shown in Fig. 5. The
simulated effect of soil type may also explain why the average
yield responses to fallow were so high in our dataset, because
they included only sites that were suf ficiently favourable to
grow break crops.
Persistence of the break-crop effect
For this review, we found 223 comparisons of the yield of first
and second wheat crops after a break crop. Of these, 108 also
compared a third wheat crop. Most of the data were from series
of experiments conducted throughout Sweden and reported by
Olofsson and Wallgren (1984) and Nilsson (1985). The other
data are mostly from single experiments in Australia, Canada
and the UK. The break-crop species represented in the Swedish
experiments included 75 comparisons with oilseeds (63 canola,
12 other brassicas) and 93 comparisons with oats. Experiments in
other countries used a range of species including 27 comparisons
Yield of wheat after wheat (t ha
–1
)
0246810
Yield of wheat after legume (t ha
–1
)
0
2
4
6
8
10
Lupin
Other legumes
Field pea
Faba bean
Lupin
Chickpea
Lentil
Fig. 4. Yield of wheat grown after grain legumes compared with wheat
after wheat growing in the same experiments. The dashed lines represent
equal yields and the solid lines fitted equations reported in Table 4.
Yield of wheat after wheat (t ha
–1
)
01234567
Yield of wheat after break crop or fallow (t ha
–1
)
0
1
2
3
4
5
6
7
Break crop
Fallow
Fig. 5. Comparison of yields of fallow–wheat and break crop–wheat with
wheat–wheat in the same experiments. For experiments that tested more than
one break crop, the data refer to the highest wheat yield after any of these
crops. The dashed lines represent equal yields and the solid lines fitted
equations reported in Table 4.
530 Crop & Pasture Science J. F. Angus et al.

with grain legumes. Based on the data for all break-crop species
combined, the effect on the yield of the second wheat crop was
less than on the first, whereas the mean effect on the third wheat
crop was negligible (Table 5). In some individual experiments,
impacts on third and fourth wheat crops have been as high as
0.9 t ha
–1
when a series of dry seasons preserved the legacies of
high N and low disease through a crop sequence (Kirkegaard
and Ryan 2014). Persistent yield benefits were also reported by
Sieling and Christen (2015) in a more humid environment. Earlier
experiments in humid environments, by Hanley and Ridgman
(1978) and Olofsson and Wallgren (1984), showed no yield
benefit to the second or later wheat crops after a break crop.
The persistence of break-crop effects likely depends on seasonal
conditions, and it will be greatest in dry environments and more
variable in moist and semi-arid environments.
When the break crops were classified as oilseeds, oats and
grain legumes, effects on the yields of the first wheat crops were
similar, but not identical, to those shown in the larger datasets
presented earlier. The differences are partly due to the different
origins and sample sizes of the data. The most surprising result
was that oilseeds had a relatively smaller effect on the second
wheat crop than oats or grain legumes. The relative difference also
appeared in 32 comparisons of oats and canola pre-crops in the
same experiments reported by Nilsson (1985), where the second
wheat crops after canola yielded no more than the second wheat
crop after oats, despite a much greater response in the first crops.
The wheat-yield responses of first and second crops for
individual comparisons were poorly correlated (Fig. 6). The
correlation coefficients between yield response to the first and
second crops were 0.26 for legumes, 0.14 for oats and 0 for
oilseeds, none of which were significant. The low correlation for
crops growing after canola was associated with several examples
of yield increases >1tha
–1
for the first wheat crop but yield
reductions for the second wheat crop. A possible explanation is
that after a break crop, the first cereal extracts more soil water and
nutrients than after wheat and so leaves less of both resources for
subsequent crops. This explanation applies to dry environments
and if nutrient inputs are relatively low. It applies less to most of
the data in Table 5, which are from experiments in humid
environments with adequate fertiliser N. In such environments,
yield responses to non-legume break crops of >1tha
–1
in the
first wheat crop, in our experience, are usually found when
break crops suppress a particularly severe root disease. When
the yield in the following wheat crop is low, there may have been a
pathogen rebound because of reduced microbial antagonism
(Kreutzer 1965; Gardner et al. 1998).
Table 5. Mean effect of break crops on wheat growing 1, 2 or
3 years later
Additional yield values represent wheat after a break crop grown in year 0
compared with wheat after wheat. For each section of the table, within
columns, means followed by the same letter are not significantly different
(P > 0.05), based on a generalised linear model
Break crop No. of
comparisons
Additional wheat yield
(t ha
–1
)
Year 1 Year 2 Year 3
Two crops after break
Oats 93 0.40a 0.22a
Oilseeds 102 0.64b 0.16a
Legumes 28 0.63b 0.36a
Same experiments (Sweden)
Oats 32 0.40a 0.26a
Canola 32 0.72b 0.27a
Oilseeds in:
Sweden 91 0.71a 0.17a
Australia 8 0.29b 0.16a
Three crops (all species) 108 0.52 0.20 0.05
–2 –1 0 1 2 3 4
–2
–1
0
1
2
Oilseeds
–2 –1 0 1 2 3 4
Yield response of second wheat crop (t ha
–1
)
–2
–1
0
1
2
Yield response of first wheat crop (t ha
–1
)
–2 –1 0 1 2 3 4
–2
–1
0
1
2
Oats
Grain legumes
(a)
(b)
(c)
Fig. 6. Comparison of the yield response of wheat growing in the first and
second years after break crops: (a) oilseeds, (b) oats, or (c) grain legumes.
Each point represents the yield difference between wheat after a break crop
and wheat after wheat.
Break crops and rotations for wheat Crop & Pasture Science 531

There may also be differences in the persistence of the break
crop effect in different regions, as shown in Table 5 by the shorter
persistence of the oilseed break-crop effect in Sweden than in
Australia, admittedly based on a small number of experiments.
The most persistent yield bene fit of a break crop in the dataset is
for chickpeas as a winter crop in a summer-rainfall grain-growing
region in eastern Australia (Holford and Crocker 1997). In this
case, wheat after chickpeas yielded an extra ~1 t ha
–1
more than
wheat after wheat in each of the 3 years following. Studies
undertaken in dry seasons in Australian winter-rainfall regions
also reported residual benefits of break crops persisting for
3–4 years (Seymour et al. 2012; Kirkegaard and Ryan 2014).
We speculate that in generally dry environments, pathogens are
suppressed longer after a single break crop, or persist longer after
wheat, and that break-crop effects may therefore persist for longer
under dry conditions.
Double breaks
If a single break crop generally increases the yield in the following
wheat crop, will two break crops produce a greater increase?
The review of the publications in the Appendix showed 74
experiments that compared the yields of wheat, growing in the
same season, after either one or two break crops (Fig. 7). There are
several crop species in the datasets, but too few to distinguish
between species. Only 23 of the 75 sequences include legumes,
so the results are presumably not dominated by residual N.
As with the graphs comparing yield after a single break crop,
most of the points lie above the 1 : 1 line and form a relationship
that is almost parallel to it. Equation 1 is a linear regression fitted
to the yield of wheat after two break crops (Y
2b
) and one break
crop (Y
1b
):
Y
2b
¼ 0:321ð0:098Þþ0:969ð0:026ÞY
1b
; n ¼ 75, R
2
¼ 0:95
ð1Þ
The intercept indicates that the yield advantage of a second
break crop is larger at low yields than at high yields. For example,
where the yield of wheat was 2 t ha
–1
there was, on average, an
additional yield of 0.26 t ha
–1
after a break crop and where wheat
yield was 6 t ha
–1
, the additional yield was 0.14 t ha
–1
.
Although growing two break crops in sequence may not be
a common practice on commercial farms, it may be used more
widely in the future as a strategy to assist in controlling herbicide-
resistant grass weeds (Hunt et al. 2013; McBeath et al. 2015).
The main use of the data about double break crops in this review
is to test whether a single break crop provides potential wheat
yield in the absence of biological soil constraints.
Synergy of break crops with other management practices
Crop sequence and tillage
Break crops show strong synergies with tillage and stubble
management. This became important with the adoption of
direct-drilling/no-till, which in some situations led to lower
yields than for crops sown after ploughing with mouldboards
or tines. Olofsson (1993) measured the interaction between tillage
and previous crop on the yield of winter wheat in 46 experiments
in Sweden (Fig. 8). Yield of wheat after cereals (wheat, barley and
oats) when grown with mouldboard ploughing was greater than
when direct-drilled, but yields after break crops (turnip rape and
field peas) were higher and little affected by tillage method.
Part of the response was because cultivation removed the
surface residues of previous wheat crops and so reduced foliar
diseases. It is also possible that cultivation reduced the effect of
deleterious soil bacteria, which reduce root growth in topsoil,
particularly in wet environments (Cook and Haglund 1991).
Similar results were found in the wheatbelt of Western
Australia where the cultivation treatment was deep ripping and
Wheat yield after 1 break crop (t ha
–1
)
02468
Wheat yield after 2 break crops (t ha
–1
)
0
2
4
6
8
Harris et al. (2002)
Angus et al. (2011)
Evans et al. (2003)
Nilsson (1985)
Gan et al. (2003)
Krupinsky et al. (2006)
Fig. 7. Yield of wheat crops growing in the same year after either one
break crop or two successive break crops. The dashed line represents equal
yields and the solid line Eqn 1.
Wheat yield (t ha
–1
)
0
1
2
3
4
5
6
After cereals After break crops
Plough Zero-till Plough Zero-till
Fig. 8. Interaction of previous crop and tillage on wheat yield in 46
experiments in Sweden reported by Olofsson (1993). The error bar
represents least significant difference.
532 Crop & Pasture Science J. F. Angus et al.

the break crop was narrow-leaf lupin (Delroy and Bowden
1986). In this experiment the yield of direct-drilled wheat after
wheat was 1.2 t ha
–1
but cultivation, or a previous lupin crop, or
both, increased yield to ~3 t ha
–1
. An additional aspect of this
experiment was that N fertiliser also increased yield of direct-
drilled wheat after wheat to ~3 t ha
–1
.
Our hypothesis to explain these results is that there are two
biological processes operating. One involves control of the root-
rot complex of fungi and other microorganisms by break crops,
as discussed elsewhere in this review. The other involves separate
growth-inhibiting microbes that infect roots when high soil
strength restricts root growth in the topsoil, particularly in wet
environments (Cook and Haglund 1991). More recently, Watt
et al. (2005) showed that tillage and rapidly growing genotypes
enabled roots to minimise exposure to growth-inhibiting
microbes. It is possible that break crops and N fertiliser can
increase the rate of root elongation and consequently reduce
root exposure to these microbes, as can tillage and vigorous
genotypes.
Kirkegaard and Ryan (2014) reported another example of an
interaction between tillage and crop sequence during drought,
but in this case, root disease did not appear to be involved and
the effects of both tillage and previous crop were through soil-
water conservation. In contrast to these results, there was little or
no interaction between cultivation and rotation (field pea–wheat
and wheat–wheat) in a long-term experiment reported by Roget
(1995) despite strong effects of cultivation in controlling
Rhizoctonia bare patch.
Other reports of crop sequences and tillage come from the
maize–soybean–wheat system in North America (Hammel 1995;
Dill-Macky and Jones 2000). In these cases, the interaction of
crop sequence and tillage differs from those discussed above, in
that wheat after wheat yielded poorly and tillage produced little or
no improvement. Direct-drilled wheat after soybean produced
yields similar to wheat after wheat, yet mouldboard ploughing led
to a large yield increase. Fusarium head blight was the reason for
low yields in the experiment described by Dill-Macky and Jones
(2000), the effects of which were overcome by the combination of
a break crop and cultivation, which presumably buried infective
straw. Vyn et al. (1991) conducted a similar experiment and found
increased yield of wheat after soybean but no effect of tillage.
Lund et al. (1993) also investigated crop sequence–tillage
interactions over 3 years and found inconsistent results, direct
drilling giving higher yields in 2 years and lower yields in 1 year.
These results are not explained by break crops reducing the
deleterious effect of direct drilling. It may be that in these
cases, tillage reduced other soil constraints and enabled wheat
to respond to residual N from a previous soybean crop.
The role of the rhizosphere in the effects of tillage on crop
yield can be explored with molecular techniques that identify the
species. For example, Donn et al. (2014) showed that rhizosphere
bacterial populations in a continuous wheat sequence differed
from those in a fallow– wheat–wheat sequence. There were
population differences among wheat cultivars but these were
smaller than the sequence effects and changes through time. The
effects of sequence and cultivar were greater in the rhizoplane
(on and in the root) than in the rhizosphere and bulk soil, showing
the need for sampling protocols to take account of temporal and
spatial dynamics.
Crop sequence and nitrogen
An important question about break crops is whether the
requirement for N fertiliser by a wheat crop depends on the
previous crop. Where water supply limits yield, there is normally
no yield response to break crops or N fertiliser (Kirkegaard et al.
1997, 2008). An exception arises when wheat after a break crop is
able to extract more soil water than wheat –wheat and its yield
increases with N fertiliser (Angus et al. 1991). Another exception
arises when a combination of break crops and high soil N,
provided by fertiliser or legume residues, stimulates excess
seedling growth and leads to haying off and lower yield than
wheat–wheat with no applied N (Kirkegaard and Ryan 2014).
Where water supply is not limited, the potential yield of
wheat after a break crop is generally greater than wheat after
wheat. In this situation, is there a greater requirement for N
fertiliser? Based on 41 experiments, Vaidyanathan et al.
(1987) concluded that this was not the case in the UK, where
the optimum rate of N fertiliser for wheat after break crops
(canola or field peas) was 40 kg N ha
–1
less, and the yield
0.75 t ha
–1
more, than for wheat after wheat. The lower
requirement for N fertiliser was partly because break crops left
more residual N and partly because of more efficient N recovery
by wheat after break crops. However, in fields where the yield of
wheat after wheat was very low, apparently because of severe root
disease, the efficiency of N fertiliser was relatively high
(Sylvester-Bradley and Cross 1991). Other research conducted
in conditions of little or no water deficit also showed higher
efficiency of N fertiliser use by wheat after break crops than after
wheat, in addition to higher yield without added N (Claupein and
Zoschke 1987; Wallgren 1987). The yield response to N for
wheat after an oilseed (flax) was greater than after a grain legume
(field peas), and at high N rates, yields were similar after
both crops, i.e. their N-response curves converged (Beckie and
Brandt 1997).
In contrast to these experiments, there are examples of similar
N-use efficiencies by wheat after break crops or after wheat, i.e.
their N-response curves were parallel (Nilsson 1985; Angus et al.
1991; Sieling et al. 2005; Lindén and Engström 2006). In these
examples, the yield of wheat without N fertiliser after a break crop
was greater than after wheat. A possible explanation for the
similar N efficiencies in these examples is that low levels of
root disease enabled wheat–wheat to recover more fertiliser N
from the soil.
Variability of root disease among fields may be responsible for
other examples of unpredictable yield responses to N fertiliser.
Angus et al. (1989) measured the wheat yield response to N
fertiliser in experiments on 39 farm fields in south-eastern
Australia and found that for wheat following break crops
(mostly canola), the yield response to N was predictable, with
large positive responses for low-N crops and large negative
responses (i.e. haying off) for high-N crops, where shoot
density was measured as a proxy for N status. By contrast, for
wheat following wheat, the yield response to N was unpredictable,
apparently because the severity of root disease varied between
fields. The difference was because the N relations in some
paddocks of wheat after wheat were disrupted by root disease,
whereas there was no disruption for wheat after break crops.
The average use of N fertiliser on dryland wheat in southern
Break crops and rotations for wheat Crop & Pasture Science 533

Australia was <5kg N ha
–1
until the mid-1990s, despite many
decades of vigorous promotion by the fertiliser industry, and
increased to >30 kg N ha
–1
in the following 6 years, a period that
coincided with widespread adoption of break crops, particularly
canola, and rapidly increasing yield of commercial crops. Angus
(2001) suggested that break crops triggered the greater reliability
of the yield response to applied N and that the synergy between
break crops and N response was responsible for a large increase in
regional yield.
Crop sequence and fungicide
Because of the role of break crops in controlling soilborne disease,
it could be anticipated that application of fungicide to the soil or
wheat seed after a break crop would not increase yields further.
Malik et al. (2015) tested the combination of a lupin break crop
with application of the fungicide fluquinconazole (Jockey
®
), to
the seed of the following wheat crop. This fungicide produced an
additional wheat yield of 0.4 t ha
–1
on top of a break-crop benefit
of 0.6 t ha
–1
from the lupin. These results resemble those of
Kirkegaard et al. (1994) when methyl bromide was injected
into soil before sowing wheat after canola. The effect of the
break crop was to increase maximum wheat biomass from 6.8 to
8.6 t ha
–1
, and methyl bromide fumigation caused a further
increase to 9.3 t ha
–1
. The growing season rainfall of 288 mm
was not sufficient to support a wheat yield commensurate with the
largest biomass, so the crop hayed off and the yield responses did
not reflect the biomass responses. The activity of methyl bromide
is not confined to fungal pathogens and it may cause release of
nutrients. Crops in this experiment received optimum fertiliser
and otherwise good management, so it would be surprising if
additional nutrients increased yield.
Kirkegaard et al. (2008) reviewed other reports where
fungicides increased wheat yield more than break crops alone,
and did find evidence of consistent interactions between break
crops and fungicides. This topic deserves further research.
Crop sequence and inter-row sowing
Precision guidance of agricultural machinery using global
positioning and auto-steering enables farmers to place seeds in
exact positions or rows in the field. With no-till systems, crop
residues and any attached rootpathogens are not spread throughout
the soil; so, it is possible to separate residues from the current
crop by inter-row sowing. Experiments in South Australia and
northern New South Wales showed reduced incidence of crown
rot, common root rot and take all, and yield increases of 9–21%
by sowing no-till wheat between the rows of the previous wheat,
compared with sowing on the rows (Verrell et al. 2005; Cay
et al. 2012; Simpfendorfer et al. 2012). Later experiments tested
the interaction of inter-row sowing with the break crops canola
and chickpeas before wheat. The best strategy was to sow break
crops between the rows of wheat stubble and to sow the following
wheat on break-crop rows (Verrell 2014).
Why was the mean yield response an offset rather than
a percentage?
Much of the research reported on break crops refers to a
percentage increase in the yield of the following wheat crop.
This was an understandable interpretation because the yield
response might be expected to be greater in absolute terms at
high yield levels when there were fewer limitations on yield than
at low yields. The many comparisons presented here show that
rather than the mean yield benefit of oats and oilseeds being a
percentage increase, the mean yield response did not vary across
a wide range of wheat after wheat grain yields. This conclusion
is consistent with previous observations (Angus et al. 2011;
Seymour et al. 2012; Kirkegaard et al. 2014). We propose two
possible reasons for this response. One is that the most severe
incidence of root disease was concentrated among the lower
yielding wheat–wheat crops in the dataset presented in
Figs 1 –4. The implication is that control of severe diseases by
break crops would result in large yield increases. The other
explanation relates to capture of soil water and mineral N and/
or other nutrients. Wheat growing after break crops was able to
extract more of both resources than wheat after wheat, but the
absolute yield responses were limited by the availability of soil
resources.
The mean yield responses to legumes generally increased with
greater yields of subsequent wheat crops. The likely difference
with legumes is that N mineralisation during the growth of wheat
following legumes continued at a higher rate and for a longer
period when the following crop grew in favourable seasons.
Reports of the largest responses to legume break crops were
concentrated in a series of wet seasons in southern New South
Wales during the early 1990s. Another explanation is that some
legumes leave more residual water and N than cereals or other
break crops (Miller et al. 2002, 2003). Because subsequent wheat
yields are often co-limited by water and N, it is more likely,
especially under wetter, higher yielding conditions, that adequate
N will be available to support higher yields,whereas at loweryield
levels the additional N may cause haying off (Kirkegaard and
Ryan 2014).
A consequence of a similar mean yield response over a wide
range of yields is that at low yield levels, the profitability of a
break crop sequence depends relatively more on the extra returns
from the following crop than the break crop itself. At high yield
levels, profitability of the break crop is relatively more important
(Angus et al. 2011).
Variability of the yield response to break crops
Farmers and advisers in south-eastern Australia reported in a 2001
survey that about one-quarter of the wheat crops grown after
canola grew slowly and produced lower than expected yields
(Ryan et al. 2003). This result is consistent with the distribution of
yield responses in Figs 1
–4. The yield responses to break crops
varied among fields and the amount of variation appeared to be as
great within a region as among regions, confirming the conclusion
of Kirkegaard et al. (2008) that mean yield responses to break
crops did not vary greatly among Australia, North America and
Western Europe. One of the reasons for variable responses may
have been the erratic incidence of root disease, which Lawes et al.
(2013) showed was not closely related to the magnitude of the
yield benefit of break crops. Root disease may be unimportant on
disease-suppressive soils or if pathogen levels are low, and severe
where these conditions do not apply. Other reasons for variable
yield responses to break crops may have been environmental; for
example, on some sites, the soil water and N may have been
534 Crop & Pasture Science J. F. Angus et al.

inadequate for crops to express their full yield potential, or indeed
greater water and N use by some break crops may have left less
residual water or N in the soil.
The among-site variability of the break-crop effect can be
placed in context with a comparison of the among-site variability
of variety trials, expressed as standard error. A typical example is
a 24-site comparison of wheat varieties conducted in New South
Wales during 2010, in which the variety Spitfire (released in
2010) outyielded the variety Sunvale (released in 1995) by
0.02  0.07 t ha
–1
(NVT 2015). The wheat yield response to a
previous canola crop was 0.80  0.06 t ha
–1
(Table 4).
Most of the evaluation of break crops in south-eastern
Australia was in the 1990s, when large responses coincided
with generally favourable seasons. There is a perception that
yield responses to break crops have been less since 2000.
A comparison for two regions confirms the reduced responses
and that the growing-season rainfall during the later experiments
was less than during the earlier experiments (Table 6). The lower
rainfall on the more recent experiments may have reduced the
expression of the higher yield potential of crops after break
crops, or there may have been less disease incidence and
residual soil water because of a history of more break crops
and the generally lower rainfall since the mid-1990s.
Mechanisms of break-crop effects
Control of root diseases
Break crops reduce the incidence and severity of most root and
crown diseases of following cereal crops, and the level of wheat
foliar disease is often less for wheat after break crops (Olofsson
1993). The pathogens that cause most of these diseases are fungal
species listed in Table 2, which Butler (1961) referred to as the
root-rot complex. Fungi are not the only root pathogens controlled
by break crops; nematodes and bacteria are other groups of soil
organisms that can constrain cereal root growth and reduce yield
(Zillinsky 1983). Most of the diseases caused by these pathogens
are controlled by a single year of break crop. One exception is
Rhizoctonia bare patch, caused by Rhizoctonia solani,a
promiscuous pathogen of temperate crops that is partly
controlled by Brassica break crops (Gupta et al. 2012).
Another exception is common root rot, which can persist for
2 years without a host (Burgess et al. 2001); it is not clear whether
two successive break crops are needed for control. The cyst
nematodes of cereals are generally suppressed by a broadleaf
break crop; however, some root-lesion nematodes share hosts
with broadleaf crops (Hollaway et al. 2000).
Crop sequence is important for controlling root disease of
wheat because other methods are relatively ineffective (Gardner
et al. 1998). There is little or no effective host resistance to take-all
in wheat (Cook 2006) but good resistance to cyst nematodes
(Eastwood et al. 1991) and root-lesion nematodes (Vanstone et al.
1998). Some wheat cultivars have host resistance to crown rot but
they do not reliably yield more than susceptible cultivars in the
presence of the disease (Kirkegaard et al. 2004). Three
explanations have been offered for the control of cereal root
disease by crop sequences, as follows.
(1) Non-hosts. The earliest recognition that some crop sequences
control soilborne diseases is that the population of root
pathogens decreases when they lack a suitable host during
a break-crop phase (McAlpine 1904). Under this first
explanation, more closely related plant species are more
likely to host and pass on pathogens than distantly related
species. Although this is generally true, there are exceptions,
for example, carry-over of wheat take-all by soybean and
lucerne (Nilsson 1969). Absence of host certainly explains
the effect of fallows on pathogens, but it has generally been
assumed, rather than tested, for break crops. Butler (1961)
reviewed research questioning this explanation, based on a
lack of relationship between pathogen assays and yield loss.
This controversy was not resolved and absence of host is still
a plausible explanation for pathogen control when there is no
evidence for the following two possible control methods.
(2) Microbial antagonists. A second explanation is suppression
of root pathogens by microbial antagonists (Zogg 1969),
mainly fluorescent Pseudomonas bacteria (Cook and Rovira
1976; Haas and Défago 2005; Cook 2006; Gupta et al. 2011).
Soils vary in their ability to suppress soilborne disease
naturally. The best example of natural disease suppression
is the process of ‘take-all decline’, first reported by Slope and
Cox (1964) as a significant decrease in the symptoms of take-
all after several years of continuous cereals, but with no
significant yield increase. There is also evidence of more
suppression of foliar pathogens by continuous wheat than by
break crops (Fernandez et al. 1998).
(3) Biofumigation. A third process, biofumigation, occurs when
compounds released from break-crop roots actively suppress
cereal pathogens. The term biofumigation was coined to
describe the release of isothiocyanates (ITCs) and related
compounds formed from the hydrolysis of glucosinolates
contained in living roots or residues of Brassica species
(Angus et al. 1994; Brown and Morra 1995; Kirkegaard
et al. 1996b; Sarwar et al. 1998; Smith and Kirkegaard 2002).
Table 6. Mean wheat yield response (
þ
standard error) to canola break crops from comparisons in two regions of southern Australia in different
decades
Data sources are Angus et al. (1999) for the 1990s, McBeath et al. (2015) for Mallee after 2005, and M. Peoples (unpublished data) for the Western
Slopes after 2005. Growing-season rainfall (GSR) is for the years of the experiments at the nearest station of the Australian Bureau of Meteorology
(www.bom.gov.au)
Mallee Western Slopes
Yield response
(t ha
–1
)
No. of comparisons Mean GSR (mm) Yield response
(t ha
–1
)
No. of comparisons Mean GSR (mm)
1990s 0.62 ± 0.32 6 316 0.81 ± 0.19 8 350
Post-2005 0.08 ± 0.08 12 133 0.18 ± 0.10 11 219
Break crops and rotations for wheat Crop & Pasture Science 535

Many glucosinolates are contained in the roots and shoots of
the Brassicaceae and other families. As well as suppressing
some pathogens, their breakdown products suppress some
insects and germination of some seeds (van Dam et al. 2009).
Of the three possible mechanisms of root disease control,
biofumigation, although a ‘natural’ process, is most similar to
the action of synthetic pesticides. Many pest organisms have
evolved resistance to synthetic pesticides; so, it is possible that
root pathogens of wheat could evolve resistance to biofumigants.
There is indirect evidence that this can occur, from the effects
of garlic mustard (Alliaria petiolata), an invasive member of the
Brassicaceae family that is weakly suppressive to soil fungi in
its native Europe but highly suppressive in North America
(Callaway et al. 2008). How long it takes for root pathogens
or other soil fungi to evolve resistance to biofumigation is
unknown, but the possibility warrants vigilance.
Relative effectiveness of control methods
There is good evidence that microbial antagonists can reduce
pathogen populations and disease symptoms after several years of
continuous wheat, for example, in the process of take-all decline
(Hornby 1998). A good example of root disease control
comes from Cook (1981b), who showed that introducing
Gaeumannomyces graminis var. tritici (Ggt) inoculum into the
soil greatly reduced yield of wheat after break crops, but had less
effect on continuous wheat.
This suggests that wheat roots supported a more suppressive
population of soil microbes than the roots of break crops.
However, the decrease in disease symptoms that characterises
take-all decline does not necessarily translate into increased yield.
Slope and Etheridge (1971) and Bowerman and Jarvis (1982)
showed that wheat yield increased when the break crops oats
or faba beans interrupted a long sequence of continuous wheat
crops. In these cases, break crops were a more effective means of
controlling root disease than were microbial antagonists. Hanley
and Ridgman (1978) concluded that there was no recovery in
wheat yield several years after a break crop, based on observations
after 10 break crops (Fig. 9). That research has been largely
ignored in the literature, in contrast to the extensive research on
microbial antagonists that has not reported crop yield.
Rothrockand Cunfer (1986)also reported no recovery in wheat
yield several years after a break crop. Although many observations
of take-all decline suggest that continuous monoculture boosts
microbial antagonists, Olsson (1995) reported an exception when
abarley–canola rotation supported a larger population of microbial
antagonists to Pythium arrhenomanes than continuous barley.
Microbes that suppress root disease produce antibiotics, so far
identified as 2,4-diacetylphloroglucinol, pyoluteorin, pyronitrin,
phenazines, hydrogen cyanide and viscosinamide (Cook 2006;
Kwak and Weller 2013). Suppressive soils also contain more
diverse populations of bacteria (Mendes et al. 2011) and fungi
(Penton et al. 2014) than non-suppressive soils. In some cases
disease-suppressive soils also contain more organic matter than
adjacent, non-suppressive soils (Pentonet al. 2014),butitisunclear
whether addition of organic matter to soil promotes suppression.
Suppressiveness can be general, meaning suppression of several
diseases, or specific, meaning suppression of a particular disease,
such as take-all (Rovira and Wildermuth 1981).
Research on suppressive soils and microbial antagonists has
mostly been uncoupled from break-crop research, apart from
the report by Cook (1981b) referred to above, which combined
both. Break-crop research would be strengthened if there were an
ex ante guide to the disease suppressiveness of experimental soil.
Research on microbial antagonists should include break crops
to provide a yield benchmark. There are several unanswered
questions about how antagonists to wheat disease respond to
break crops. Does the size of the wheat-yield response to a break
crop depend on the disease suppressiveness of the soil, and
whether the suppressiveness is general or specific? Does the
persistence of microbial antagonists in the soil depend on the
continued presence of pathogens? If the pathogen population
decreases because of a break crop, does the population of
microbial antagonists also decrease and, if so, how rapidly?
Can a decrease in microbial antagonists then lead to a later
resurgence of pathogens?
The levels of several cereal root pathogens can be estimated
with commercial DNA soil tests, which provide a guide to the
disease risk and therefore decisions about growing wheat or a
break crop (Ophel-Keller et al. 2008). There are no equivalent
tests of microbial antagonistsor specific antibiotics.Development
of such tests may improve the prediction of disease risk. DNA
tests are also important in confirming that yield loss to root disease
and the yield response to break crops are indeed due to the
suspected pathogen.
The activity of microbial antagonists may play a part in
the rebound of root disease and low yield 1 year or 2 years
after applying a disease-control treatment. This effect was first
described by Kreutzer (1965) following fungicide and fumigation
treatments, and can be explained as a host–parasite relationship in
which a decrease in population of hosts (pathogens) provoked a
decrease in the population of parasite (microbial antagonists),
so that the host population could build up later. Gardner et al.
(1998) reported a similar rebound in Ggt after fumigation,
fungicide application and Brassica break crops. As discussed
earlier, Malik et al. (2015) investigated the combination of break
Years since a break crop
051015
Yield (t ha
–1
)
0
1
2
3
4
5
6
1965
1970
1966
1971
1967
1972
1968
1973
1969
1974
Fig. 9. Yield of wheat grown in continuous wheat sequence after a break
crop, showing no evidence of take-all decline. The different symbols represent
the year of yield measurement. Plotted from the UK data of Hanley and
Ridgman ( 1978).
536 Crop & Pasture Science J. F. Angus et al.

crop and fungicide applied to the seed of the (second) wheat
crop. They found that the fungicide increased wheat yield
in the year of application but reduced the yield of the third
wheat crop, confirming results by Bateman et al. (2008). In
both these cases it is not clear whether pathogen rebound was
responsible for the yield reduction. Environmental conditions
also play a part in pathogen rebound. For example, in southern
Australia, relatively wet spring seasons cause rapid growth
of pathogen populations on host plants, and dry summers
reduce decomposition of saprophytic mycelia. Dry spring and
wet summer seasons cause the opposite responses (Roget and
Rovira 1991).
Severity of root disease
In both the Northern Great Plains of North America and south-
eastern Australia, cereal root diseases broke out in newly
established wheat-growing regions where there were chronic
and severe symptoms of ‘whitehead’ patches (areas of stunted
plants with bleached and sterile spikes dying prematurely). The
earliest report of root disease control by break crops was in
Manitoba, Canada, where Bedford (1899) showed higher yield
of wheat following break crops than following wheat. Bolley
(1913) reported similar results in North Dakota, USA, and
McAlpine (1904) in Victoria, Australia, observed that wheat
growing after oats suffered less take-all and yielded more than
wheat after wheat. These early studies concluded, without
confirmatory microbiological measurements, that control of
root disease occurred because a break crop deprived pathogens
of a host.
In long-established wheat-growing regions, a renewed interest
in root disease and its control by break crops attracted attention
after outbreaks in the newer regions. Nilsson-Ehle (1920)
recommended a return to diverse rotations in southern Sweden
when take-all appeared in continuous wheat crops, but only
three severe outbreaks had occurred in 20 years rather than the
annual devastation in North America and Australia. More control
of root diseases by microbial antagonists was likely in established
wheat-growing regions than in new cropping regions during
the period between arrival of the pathogen and development of
microbial antagonists. The severity of take-all in the newly
established wheat-growing regions may also have been due to
root disease tending to produce more spectacular white-head
symptoms and reducing yield more in dry conditions than in
more humid regions. Many of the reports of take-all are from the
Northern Great Plains of North America, northern Europe and
southern Australia, where long periods of the wheat-growing
season have temperatures close to 158C when the disease is most
severe (Cook 1981a).
Control of take-all by break crops was so successful in New
South Wales that Sutton (1911) speculated that the disease could
be eradicated from a field by growing oats. This was a vain hope,
at least in a moist environment, as shown in a 17-year rotation
experiment in Sweden where wheat grown following a broadleaf
species consistently had less root disease and produced higher
yields than wheat following a cereal (Ebbersten 1980). There is
still no evidence from extensive experiments in Sweden that
cereal root diseases can be eliminated from a field by break crops
(S. Ebbersten, pers. comm. 2013).
The authors of many studies reviewed here assumed that the
yield benefit of break crops was mostly due to control of root
disease. An exception is Bullock (1992), who suggested that
disease control did not strongly contribute to yield increase after
break crops because microbial antagonists build up during
continuous wheat and control disease. This interpretation
probably exaggerates the relative importance of microbial
antagonists in monoculture.
Soil water supply, retention and demand
Different crop species require different amounts of water;
therefore, the amount of residual soil water at maturity is
likely to vary, with consequences for the water supply to the
following crop and its yield when water is limited. For example,
soil profiles after field peas can be wetter than after wheat,
although not as wet as after fallow (Fig. 10).
Residual soil water at maturity of a break crop can be used by
the following crop provided it is not first lost through evaporation
or transpired by weeds during the fallow period between crops
(Miller et al. 2002; Hunt and Kirkegaard 2011; Hunt et al.
2013). The depth of residual soil water affects its retention,
with more water retained deep in the soil profile. Break-crop
stubble can affect retention of soil water and can affect infiltration
and retention of precipitation between the time of harvesting a
break crop and sowing the following wheat crop (Hunt et al. 2013;
Kirkegaard and Ryan 2014).
Kirkegaard et al. (2001) presented an example of how break
crops could retain soil water for the following wheat crop in
water-limiting conditions. In this case, the stubble of dual-
purpose oats retained the largest amount of stored soil water,
50 mm more than under the stubble of narrow-leaf lupin. The
grain yield of wheat after the oats was 0.95 t ha
–1
more than after
the lupin, implying a water-use efficiency of 19 kg ha
–1
mm
–1
,
which is close to the benchmark proposed by French and Schultz
(1984).
Water content (% v/v)
0 5 10 15 20 25 30
Soil depth (m)
0.0
0.5
1.0
1.5
2.0
Fig. 10. Soil water profiles under wheat (
&
), field peas (
*
), and fallow (*)
sampled at the time of crop maturity. The amount of soil water after harvest
in the 1.6 m root zone after wheat was 75 mm less than after field peas and
150 mm less than after fallow. These unpublished data are from an experiment
in south-eastern Australia reported by Hocking et al. (1997).
Break crops and rotations for wheat Crop & Pasture Science 537

Additional yield after break crops is not always negatively
related to the extraction of soil water by the previous crop, as
shown by higher yields of barley following safflower than
following barley, despite the deep roots and high water use of
safflower (Yau and Ryan 2012). When a previous crop leaves
deep cracks in the soil after maturity, as often happens with
safflower growing on a Vertosol, the soil fills with water from the
bottom and relatively little is lost by soil evaporation. There are
analogous results for wheat after field peas that normally yield
no more than wheat after lupin despite less residual soil water
after lupin. In this case, other factors override water supply, as
discussed in the next section.
As well as affecting the supply of soil water to the following
crop, break crops can increase the extraction of soil water by the
following crop. The evidence for this comes from experiments in
dryland crop-sequence experiments in south-eastern Australia
where the soil was drier at maturity of wheat after a break crop than
after wheat (Angus et al. 1991; Kirkegaard et al. 1994). The soil-
water content at maturity of wheat after break crops was typically
drier by 2–4% (v/v) and the soil suction much higher than wheat
after wheat, with the largest differences in the middle of the
root-zone, at depths of 0.5–1.0 m from the soil surface. Although
these percentage differences in water content were relatively
small, they amount to 20–40 mm less water in the whole root-
zone of wheat after a break crop than after wheat. These amounts
represent the additional evapotranspiration by wheat after a break
crop and are consistent with additional yields of 0.4–0.8 t ha
–1
due
to break crops, based on water-use efficiency of 20 kg ha
–1
mm
–1
(French and Schultz 1984; Angus and van Herwaarden 2001).
Similar results were reported in the more humid environment of
northern Germany, where wheat yield responses to canola break
crops were larger in dry than in wet seasons (Sieling and Christen
2015). The probable reason was that residual soil water after
canola and the additional soil water extracted by wheat after
canola were relatively more important in dry than in wet seasons.
There is less evidence of additional extraction of soil water by
wheat after break crops in the south of Western Australia, where
Gregory (1998) found no difference in soil-water content at
maturity of wheat after break crops and after wheat. However,
Asseng et al. (1998) reported carryover of residual subsoil water
after lupin, which may have contributed to increased water use by
a following wheat crop. The cause of the different results from
those in south eastern Australia may have been the low water-
holding capacity of the soils in the experiments in the south of
Western Australia.
Additional soil-water extraction by the first wheat after break
crops may lead, in water-limited environments, to less soil water
and lower yields of second and later crops by depleting soil-water
reserves (Kirkegaard and Ryan 2014). When there is significant
rainfall between crops, the water deficit may be partly offset by
more water infiltration into drier soil. Drier soil may also provide
environmental benefits of less runoff and less deep drainage. It is
likely that growing diverse crops in sequence, along with stubble
retention and direct drilling, will result in more rainfall used
for transpiration rather than lost in evaporation, runoff or deep
drainage (Hunt and Kirkegaard 2011). Greater awareness of
soil-water status will be needed in managing dryland cropping
systems that do not have the buffer of wet soil provided by
continuous cereals.
Additional water extraction also has implications for
understanding the lower limit of soil water content, which has
been proposed as characteristic of a soil profile and crop species
(Ritchie 1981). These results suggest that the soil lower limit
could be controlled by crop management as well as plant species
and soil properties (Angus and van Herwaarden 2001). The
results also raise doubts about the precision of crop simulation
models that use a single value for the crop lower limit, irrespective
of crop management.
Nitrogen supply and demand
This section discusses break-crop effects on the supply of
plant-available soil N and the N requirement of crops.
Legumes can contribute to total N content of cropping soils
through biological N
2
fixation when the amount of N fixed
exceeds the N harvested in grain (Peoples et al. 2009).
Significant changes in soil N are often difficult to quantify
because the inputs of fixed N are small relative to the large
and variable background concentration of organic N (Chalk
1998). Nonetheless, increased wheat yields can often be
related to the amount of N remaining in the stubble of prior
legume crops (Asseng et al. 1998), and a portion of the N in the
nodulated roots and vegetative residues becomes available as
soil inorganic N, typically in a period of a few weeks (Angus et al.
2011). The microbial decomposition and mineralisation of
residue N into inorganic forms is accompanied by breakdown
of organic compounds, which provides soil microbes with a
carbon (C) source for respiration and growth. Much of the
organic N released is immobilised by the microbes (Murphy
et al. 1998), so mineral N accumulates only when the amounts
of N released from the organic residues exceed the microbial
growth requirements (i.e. when gross N mineralisation exceeds
microbial immobilisation). This is more likely to occur with
legume material than with cereal stubble because legume
organic matter has a higher N content and lower C : N ratio
(Kumar and Goh 2000). Environmental conditions, especially
soil water and a range of constituents in legume tissues (e.g.
lignin, polyphenols, soluble carbon and N compounds) can
modify microbial activity and affect the rate of release of
legume organic N (Kumar and Goh 2000; Bolger et al. 2003),
especially for senesced vegetative residues at the end of the
growth cycle (Bremer and van Kessel 1992; Russell and
Fillery 1999; Evans et al. 2003). Some of the additional soil
mineral N following legumes may also be derived from apparent
‘spared’ N, which can represent mineral N not utilised during
legume growth and the rapid mineralisation of organic N released
into the legume rhizosphere (Chalk 1998), or be due to a reduced
rate of immobilisation of available soil N by crop stubble, as
reported by Green and Blackmer (1995) in a soybean–maize
system.
Nitrogen benefits to following crops are not confined to
legumes. There can be more soil mineral N at the time of
sowing a crop after canola than after a cereal because net
mineralisation is greater (Kirkegaard et al. 1999; Ryan et al.
2006). Brown and Morra (2009) showed, using crop residues
added to soil, that Brassica tissue inhibited nitri fication as well
as stimulating net N mineralisation.
Typically, wheat growing after break crops depletes soil
mineral N to lower concentrations than wheat after wheat
538 Crop & Pasture Science J. F. Angus et al.

(Angus et al. 1991; Kirkegaard et al. 1994). Increasing yield by
controlling foliar disease with fungicides is capable of inducing
similar depletion of soil mineral N (Ishikawa et al. 2012), so the
additional N demand by higher yielding crops is analogous to the
greater demand for soil water, as discussed in the previous section.
A consequence of soil N depletion is that the level of soil mineral
N for a second cereal after a break crop is likely to be low and the
requirement of N fertiliser may be more than otherwise expected.
The importance of N supply in explaining the effect of break
crops is summarised in Fig. 11, which collates data from
published and unpublished Australian experiments. Within
each of these datasets, there were close relationships between
soil mineral N after pre-crops and the yield of the following
wheat crops; yields of wheat–wheat were generally the lowest
and legume–wheat the highest. For the combined data, each tonne
of wheat was associated with 57 kg ha
–1
of soil mineral N, an
amount that does not include in-crop mineralisation or the
effects of N on grain protein. The closeness of the relationship
for the combined data suggests that factors such as root disease
were unimportant for these crops. The data sets with the least
slopes, from Culcairn (M. B. Peoples, unpubl. data) and
Formartin (Strong et al. 1986), were from droughted crops.
Hydrogen fertilisation by legumes
Since the 1970s, nitrogenase enzyme in the root nodules of
legumes has been known to produce H
2
as a byproduct of the
process of biological N
2
fixation (Evans et al. 1981). Some
symbioses possess a hydrogenase uptake system within the
nodule (designated HUP-plus) that is able to recycle almost all
of the H
2
evolved, whereas others lack this capacity (designated
HUP-minus), and in this case, the H
2
diffuses from the nodule into
the soil. A legume’s HUP status is specific to the rhizobial strain
(Table 7; Evans et al. 1988), and the rate of H
2
emission from
nodules in HUP-minus associations appears to be influenced
by legume species, as shown in Table 7 by particularly high
emissions from faba bean and lupin nodules. The amount of
H
2
released over a growing season has been calculated to
represent up to 300 000 L ha
–1
for HUP-minus legumes fixing
200 kg N ha
–1
(Peoples et al. 2008; Golding et al. 2012).
The exposure of soil to H
2
results in changes in the
composition of the soil microbial population such that,
ultimately, all of the H
2
is consumed by specificH
2
-oxidising
bacteria (La Favre and Forcht 1983; Osborne et al. 2010). The
consequences of H
2
emission for crops following legumes has
been investigated in: (i) experiments comparing plants in soil
exposed to either air or H
2
under controlled conditions
(McLearn and Dong 2002; Dong et al. 2003); (ii) field studies
quantifying rotational bene fits following soybean inoculated with
HUP-plus and HUP-minus rhizobial strains (Dean et al. 2006;
Peoples et al. 2008); and ( iii) field and glasshouse trials using soil
extracts or microbial isolates from H
2
-treated soils (Dong et al.
2003; Golding et al. 2012). This research has demonstrated
improvement in crop growth and yield of 10–50%, implying
Soil mineral N after previous crops (kg ha
–1
)
0 100 200 300 400 500
Wheat yield (t ha
–1
)
0
2
4
6
8
Triticale
Wh
Ch
Wh
Ca
Pe
Oa
Wh
Ca
Lu
Fb
Fb
Ch
Ba
Wh
Ca
Pe
Wh
Oa
Lu
Pe
Ch
Ca
Wh
Lu
Ln
Ch
Wh
Ch
Fx
Ca
Kirkegaard and Ryan (2014) MB Peoples, unpubl. (Culcairn)
MB Peoples, unpubl. (Junee)
JR Hunt, unpubl. (Hopetoun)
Felton et al. (1998)
Dalal et al. (1998)
Khan, unpubl. (Breeza)
Strong et al. (1986)
Fig. 11. Relationship between residual soil mineral N in the root-zone (0.9 m) after different
crops and grain yield of the following wheat crop. Each set of symbols and the fitted line is from
a different field experiment. Wh, Wheat; Ba, barley; Oa, oats; Ca, canola; Fx, flax; Pe, field peas;
Ln, lentils; Ch, chickpeas; Fb, faba beans; Lu, lupins; Fa, fallow.
Break crops and rotations for wheat Crop & Pasture Science 539

that the changed soil microflora induced by H
2
includes plant-
growth-promoting bacteria.
None of these results provide a direct estimate of additional
wheat yield due to H
2
fertilisation by temperate legumes, and no
readily available rhizobial inocula for temperate grain legumes
are available with both forms of HUP that could be used to
make such an assessment. Both HUP forms have been found in
rhizobial cultures of field peas (Evans et al. 1981), so we screened
the CSIRO rhizobial library for strains that inoculate chickpeas,
faba beans, field peas, lentils and narrow-leaf lupins (Table 7).
All combinations of legume species and rhizobial strains
produced H
2
from their nodules, although the amounts varied
widely with individual strains. None of the measures of H
2
evolution were sufficiently low to be classified as HUP-plus
(i.e. <0.01 mmol H
2
h
–1
g
–1
nodule dry weight; Peoples et al.
2009). We conclude that the HUP-minus form is common among
rhizobial strains used for temperate grain legumes and that H
2
fertilisation is likely to be responsible for some of the increased
yield of cereals following temperate legumes. The quantity of that
increase cannot be directly estimated using the available rhizobial
strains, and the best estimates remain those from the experiments
with soybean discussed previously—a yield increase of at least
10%.
Research comparing both forms of HUP rhizobia for
temperate legumes is needed to provide a field estimate of the
contribution of H
2
fertilisation to wheat yield and of how this
compares with the effect of residual N. This information would be
useful in estimating the fertiliser N required by following crops.
Another question that should be resolved about H
2
fertilisation is
whether weeds growing in or after a legume will detract from the
increased yield of the following crop.
Non-nitrogen benefits of legumes
There has long been recognition that cereal yield after a legume
break crop tends to be greater than after other break crops. One
approach to partitioning the yield benefit into N and non-N
components has been to use experiments comparing yields of
a cereal (after cereal) supplied with different rates of N fertiliser
with yields of cereal after a legume (Chalk 1998). Studies reported
by Wright (1990), Rowland et al. (1988, 1994), Stevenson
and van Kessel (1996) and Beckie and Brandt (1997) show
that that non-N effects can dominate the break-crop benefitof
legumes.
Before the discovery of H
2
fertilisation, the non-N yield
benefit of legumes was assumed due to control of cereal
diseases and, in dry regions for some legume species, more
residual soil water. This assumption is no longer held and the
non-N benefit of legumes should now also be explained by
disease control plus H
2
fertilisation and may include additional
yield due to suppression of AMF when the legume is a lupin
species that is non-host to AMF. There are no reports of
experiments that unscramble these factors, but it may be
possible to separate or at least minimise the disease effect by
growing a prior non-legume break crop, and to separate the effect
of AMF by comparing legumes that are hosts of AMF and those
that are not.
Weed control
Experiments testing rotations and break crops usually aim to
control weeds in all treatments, by using non-commercial
methods if necessary. The results of such experiments are not
useful for quantifying the likely contribution of break crops to
weed control in commercial crops. Despite the lack of
experimental evidence from break-crop experiments, there is
no doubt that weed control is an important factor in the value
of break crops (Haramoto and Gallandt 2004). In an early report of
a rotation experiment, Lawes and Gilbert (1894) acknowledged
that weed control, presumably by harrowing and hand-weeding,
was easier in rotated crops than continuous monocultures.
Seymour et al. (2012) concluded that weed control was an
important contribution of lupin in the cropping systems of
Western Australia in the 1990s. In a survey in the Mallee
region of south-eastern Australia, grain growers ranked weed
control as the main benefit of break crops (Moodie 2012). This
highlights a gulf between research and commercial practice,
because break-crop experiments control for the factor that is of
most interest to farmers and advisors. We propose that future
research focus on the control in break crops of weeds that are
difficult to control in wheat crops. This includes control by
herbicides, combined with competition and suppression by
break crops used for green manure, brown manure or cutting
for hay or silage.
The breakdown products of glucosinolates contained in
brassicas suppress seed germination of other species when
they are released from decomposing roots (Brown and Morra
1995). Thus, they may have potential to control weeds in the
following crop. The chemical suppression of one species by
another is called allelopathy when the two species are growing
together. However, this term does not capture the effect of a
Brassica species on weeds in the following season. This effect
could be called ‘delayed allelopathy’. There is also evidence
that weeds in canola crops can be controlled by canola root
exudates containing allelopathic compounds. The most active
of these compounds are sinapyl alcohol, p-hydroxybenzoic
acid and 3,5,6,7,8-pentahydroxyflavone (Asaduzzaman et al.
2015). The form and concentration of allelopathic compounds
vary among canola cultivars, some of which contain high
Table 7. Evolution of hydrogen gas (H
2
, mmol h
–1
g
–1
nodule dry wt)
from roots of nodulated grain legumes inoculated with rhizobial strains
(M. Peoples, J. Brockwell and J. Angus, unpublished data)
Three replicate plants of each combination of legume species and one of a
numberofrhizobialstrainsmaintainedwithinCSIRO’scollectionweregrown
with N-free nutrients in a sand–vermiculite medium in pots in a glasshouse at
CSIRO Canberra. When plants reached leaf 4 stage, the nodulated roots were
washed, detached from the shoots and placed in a cuvette for measurement of
H
2
evolution from nodules using the methodologies described by Peoples
et al. (2008)
Legume species No. of rhizobial H
2
evolution
strains tested Mean Range
Soybeans 11 0.06 0.06–0.07
Chickpeas 6 0.09 0.05–0.15
Field peas 36 0.17 0.02–0.61
Lentil 23 0.25 0.02–0.56
Faba beans 19 0.51 0.15–1.20
Narrow-leaf lupins 10 0.51 0.21–0.80
540 Crop & Pasture Science J. F. Angus et al.

concentrations and provide strong control of several weeds in the
field. The variation suggests scope for improving weed control
by breeding and selecting canola for increased allelopathy
(Asaduzzaman et al. 2014). Further research is needed to
confirm that allelopathic weed control by canola extends to
weeds in following crops and to determine whether other
Brassica crop species can provide similar weed control. There
may also be scope for economising on herbicides, delaying the
onset of herbicide resistance and improving weed control by
combining allelopathy with herbicides.
Wheat crops following canola often have relatively few
weeds. Whether this is due to allelopathy or the persistent
herbicides used on canola is not known. The low weed
populations after brassicas often allow earlier sowing of a
following wheat crop. This benefit, however, will not be
detected in experiments where all plots are sown at the same time.
Some persistent herbicides used on broadleaf break crops,
such as triazines, imidazolinones, pyridines and amides, are
capable of reducing the burden of weeds growing during the
fallow period prior to planting of the subsequent wheat crop.
Controlling fallow weeds increases the yield of the subsequent
crop in dry environments, largely a result of retaining soil water
and mineral N (Hunt and Kirkegaard 2011; Hunt et al. 2013;
Kirkegaard et al. 2014; Kirkegaard and Ryan 2014), but possibly
also by indirectly controlling wheat pathogens hosted on the weed
roots (Vanstone and Russ 2001).
Post-emergent control of grass weeds in wheat relies upon
herbicides that inhibit the activity of acetyl CoA and acetolactate
synthase as their mode of action. When populations of genetically
diverse, outcrossing grass weed species such as annual ryegrass
(Lolium rigidum L.) are exposed to these herbicides, resistant
individuals are selected for and rapidly come to dominate (Tardif
et al. 1996). This has serious implications for the profitability and
sustainability of cereal crop production (Pannell and Gill 1994).
In broadleaf break crops, particularly triazine- and glyphosate-
tolerant canola, herbicides that have differing modes of action
and are less susceptible to evolution of resistance can contribute
significantly to farm profitability and sustainability by
maintaining weed seedbanks at low levels (Monjardino et al.
2005).
A related question is whether it is essential to control all grass
weeds in Brassica crops to break the life cycle of cereal
pathogens. If the only advantage of a break crop for disease
control consists of depriving pathogens of their host, then
presumably grass weeds in a break crop could carry pathogens
to the following wheat. However, if biofumigation contributes to
disease control and the break-crop benefit, then pathogens hosted
on grass weeds in a break crop may be suppressed and disease
carryover on weeds may be reduced. The only research on this
topic showed that introducing a low density of wheat plants into
canola break crops increased carryover of Ggt inoculum and
reduced yield of the following wheat (Kirkegaard et al. 2000).
This result supports the conclusion that host deprivation is more
important than biofumigation in controlling root disease.
Reduced parasitism by arbuscular mycorrhizal fungi
Arbuscular mycorrhizal fungi are obligate symbionts that
colonise the roots of most crop species (Thompson and
Wildermuth 1989). Colonisation by AMF is often reported to
enhance growth and yield by the host plant through enhanced
uptake of the relatively immobile nutrients phosphorus (P) and
zinc (Zn) when the external hyphae of AMF increase the volume
of soil explored by the host-plant root system under conditions
where these nutrients are limiting for plant growth (Smith and
Read 2008). In return for receiving mineral nutrients, the plant
provides the AMF with carbohydrate, and the balance of these
transfers determines the effect of AMF on yield. AMF are
commonly reported to use 4–20% of host photosynthate, the
variation being probably due to differences among species
of AMF and potential stimulation of host photosynthesis
(Lendenmann et al. 2011).
Testing of the net effect of AMF on wheat in the field is made
possible by varying their hyphal density with preceding host and
non-host break crops, the latter including brassicas (Schreiner and
Koide 1993), white lupins (Thompson and Wildermuth 1989) and
narrow-leaf lupins, which are only weakly colonised (Trinick
1977). In the temperate cropping region of Australia, wheat crops
following brassicas have repeatedly shown less colonisation by
AMF than wheat following wheat or other host crops, but there
is no evidence that the reduced colonisation has a negative
effect on nutrition or growth, irrespective of the level of soil P
or Zn (Ryan et al.
2002; Ryan and Angus 2003; Ryan and
Kirkegaard 2012). Indeed, Ryan et al. (2005) found negative
correlations between colonisation by AMF and shoot biomass
or concentration of water-soluble carbohydrate at tillering. The
possibility of confounding the effects of colonisation with
wheat root diseases was minimised by removing grass hosts of
these diseases in the year preceding the experiment. This involved
removing grass from a pasture with a selective herbicide, a
practice known as winter cleaning (Kidd et al. 2002).
A test of the possible yield cost of AMF through a single
break crop can be made by comparing the yield of canola–wheat
and flax–wheat for the 42 comparisons presented in Fig. 3b and
Table 4. In this case, the yield relationships of these two sequences
were identical, suggesting that reduced colonisation by AMFafter
a single, non-host crop had on average no positive or negative
effect on yield.
The effect of two non-hosts for AMF within a cropping
sequence has also been examined in field experiments. Harris
et al. (2002) reported that wheat following two non-hosts (i.e.
canola–wheat–lupin–wheat) yielded more than wheat following a
single non-host (i.e. wheat–wheat–lupin–wheat). An alternative
explanation to the involvement of AMF is that the second break
crop prevented resurgence of root disease that sometimes follows
2 years after a break crop, apparently because of slow recovery
in the level of microbial antagonists, as discussed above. In
another experiment in which there was less likelihood of
pathogen resurgence, Angus et al. (2008) compared the effect
on wheat yield of either two preceding hosts of AMF (flax and
faba beans) or two preceding non-hosts (canola and narrow-
leaf lupins) grown in succession. Wheat yield after two non-
hosts was greater than after two hosts. No measurements of
colonisation were made in this experiment, so the results
should be extrapolated cautiously. It should also be noted that
not all hosts would be highly colonised following wheat. For
instance, wheat had similar levels of colonisation following field
peas (host) and canola (non-host) in one experiment (Ryan et al.
Break crops and rotations for wheat Crop & Pasture Science 541

2005) and field peas and narrow-leaved lupins (non-host) in a
second (Kirkegaard and Ryan 2014). This may reflect field peas
causing proliferation of species of AMF not favoured by wheat
(Ryan and Kirkegaard 2012).
Overall, in southern Australia, it seems plausible that the
benefits of non-hosts for following wheat may in part be due
to their inducement of lower colonisation by AMF. However,
since 37 of the 42 comparisons of canola–wheat and flax–wheat,
and both of the experiments with 2 years of non-hosts, were
from southern Australia, the applicability of these results to other
regions needs to be tested. Positive effects from AMF on wheat
growth are reported in the subtropical cropping region of eastern
Australia (Thompson 1987; Owen et al. 2010; Thompson et al.
2013), and there is evidence for a positive role for AMF in growth
of field crops in other farming systems (e.g. Pellegrino and Bedini
2014). Conversely, examples of poor growth of crops following
non-hosts of AMF should not be automatically ascribed to
reduced colonisation by AMF without exploring other rotation
effects (Koide and Peoples 2012).
The evidence for suppression of AMF as a factor contributing
to the yield benefits of break crops is sufficiently encouraging to
justify further examination, particularly in separating the effects
of AMF and root pathogens. If the yield benefit of suppression of
AMF is confirmed, then breeding cereals with reduced potential
to host AMF in their roots may be a means to increase yield
(Ryan et al. 2005). Such an approach should be balanced against
a possible loss of ecosystem services such as improving soil
structure that may be provided by AMF, unless such services
are provided more economically by other means (Ryan and
Kirkegaard 2012).
Break crops solubilising soil phosphorus
The roots of most plant species excrete organic acids that
solubilise tightly held soil P into forms that are more available
for uptake, and this ability is most strongly expressed in plants
that form cluster roots (Pearse et al. 2006). Plants take up this P
for their own requirements, and after maturity, they may leave
available P in the soil that can be taken up by subsequent crops,
an effect observed in many pot experiments (Hocking 2001).
Among the break crops grown in wheat-based cropping systems,
white lupins most strongly solubilise P with their proteoid roots,
which can envelope small portions of soil. Pot experiments
showed that white lupins increased P uptake and growth of a
following wheat crop by this mechanism (Hocking 2001).
However, the most recent review of break crops had no reports
from field experiments where the yield of a following wheat crop
increased because of P solubilising (Kirkegaard et al. 2008) and
no such reports have since been made.
Although proteoid roots may enable white lupins to
solubilise significant amounts of P in a small volume of soil,
they may not be key to extracting P from a large root-zone in the
field. Funayama-Noguchi et al. (2014) showed that white lupins
have less ability than narrow-leaf lupins to increase the size of the
root system in conditions of P deficiency. This ability, known as
functional equilibrium, is a common reaction of plants, enabling
them to allocate growth selectively to organsthat acquire more of a
limitingresource (Connor et al. 2011). This limited ability of white
lupins to increase the size of their root system when P-deficient
seems to offset any benefits of proteoid roots. Other break crops
have shownabilitytosolubiliseP fora followingcrop in pots butno
effective P solubilisation in the field (Rick et al. 2011).
In the experiments reviewed, only 14 comparisons of crop
sequences included white lupins, too few to draw conclusions.
These experiments were managed with relatively high levels
of P fertiliser so, any yield benefi
ts from solubilising soil P
by white lupins would have been swamped by other effects
of break crops. In order to demonstrate that the process of P
solubilising is a useful contribution by break crops, its benefits
should be shown in the field rather than in the limited soil volume
of pots, and by growing break crops in soils that contain levels of
available P that are representative of the farming system.
Root observations
We have observed that roots of wheat growing after break
crops are usually white and clean, resembling those of plants
grown in steam-pasteurised potting mix, whereas roots of wheat
after wheat are stained brown (Fig. 12). The nature of the stain
is unknown and may indicate a microbiological or chemical
change that has value for diagnosing soil conditions that limit
yield.
Conclusions
Experiments testing the effects of break crops on the yield of
following wheat crops provide similar data to those from long-
term rotation experiments and with clearer conclusions about
productivity, in a shorter time and at lower cost. Break-crop
experiments do not provide information about the long-term
effects of farming systems and they are not a substitute for
long-term rotation experiments for assessing sustainability.
Break crops affect the yield of subsequent crops by their
effects on root pathogens and other soil organisms, on weeds,
soil water, soil N and possibly other nutrients, and on H
2
fertilisation. These factors can operate singly and in
combination; for example, the largest responses to break crops
occur when they control severe root disease in the presence of
high levels of N and non-limiting water supply. Conversely,
Fig. 12. Wheatplantsontheleft,grownaftercanola,havebrighterrootsthan
those on the right, grown after wheat. The plants were dug from the top 10 cm
of soil from a break-crop experiment in Junee, south-eastern Australia.
542 Crop & Pasture Science J. F. Angus et al.

control of root disease may not increase yield if soil N or water
is limiting.
Averaged over >900 comparisons reported in >70 reports in
this meta-analysis, wheat growing after break crops yielded
significantly more than wheat growing after wheat. The mean
additional yield varied among break crop species. The smallest
increase was 0.5 t ha
–1
after oats and the largest was 1.2 t ha
–1
after
grain legumes. The species ranking of break crops for increasing
yield response, based on overlapping experiments, was:
oats < canola  mustard  flax < field peas  faba beans 
chickpeas  lentils  lupins. The mean additional yield
derived from oats and oilseed break crops was not
proportional to yield, but was a constant amount across a
range of yields. Wheat yield responses following temperate
grain legume crops were generally larger at higher yield levels.
The yield effect of break crops varied among experimental sites
but the variability was no greater than in variety trials.
A succession of two break crops produced 0.2–0.4 t ha
–1
more wheat yield than a single break crop. There was additional
yield of a second wheat crop after a break crop, ranging from 20%
of the effect on a first wheat crop for canola, to 60% for legumes.
The effect on a third wheat crop was negligible except in dry
conditions. The variability of the break-crop effect on the yield
of a second wheat crop was larger than for a first wheat crop,
particularly after canola.
Diverse crop rotations have been recommended for many
centuries, and several gross margin analyses have shown that a
break crop–wheat sequence is more profitable than a succession
of two wheat crops (Angus et al. 2001; Lindén and Engström
2006; Miller et al. 2015; Preissel et al. 2015), and two break crops
before wheat can be highly profitable (Khakbazan et al. 2014).
Despite these advantages worldwide, many wheat crops are not
preceded by an alternative species.
Suggestions to improve the productivity of wheat-based
cropping systems are (1) reform of marketing and institutional
constraints to growing break crops, as discussed in the
Introduction;(2) fully accounting for all benefits and costs of
break crops; (3) further research to increase the break-crop benefit
and help predict when break-crop benefits are most reliable.
Full accounting of benefits and costs
Quantifying the benefits and costs of break crops on the
following crop or crops provides evidence on which farmers
can plan a crop sequence. It is important to distinguish between
a yield increase that is proportional to yield and, as found for
oats and oilseeds in this study, one that is similar over a range of
yields. In the latter case, the financial benefits of break crops
are relatively greater at low yield levels. It is also important to
include the effect, positive or negative, on the yield of second
and later crops after a break and consider the value of more than
one break crop.
Additional costs of break crops should be balanced against the
benefits. As well as variable costs, these may include capital costs
of machinery and storage as well as the less tangible cost of
marketing and a more complex cropping system. This review of
international data may be too general to apply to a particular
region, but it provides a start for anyone accumulating a larger
or regional dataset, such as those of Seymour et al. (2012)
and Preissel et al. (2015). Nevertheless, mean effects of break
crops were surprisingly similar in different regions of the world.
Future research
Research to understand and reduce the variability in wheat yield
responses may improve adoption of break crops by identifying
where the benefits are most likely to be greatest and provide
criteria to select a cropping sequence tactically. Possible reasons
for the variability are the interactions of break crops with soil
water and N supply, tillage and the natural disease
suppressiveness of the soil. There will inevitably be limitations
to tactical crop sequences because a break crop benefit will partly
depend on weather and grain prices for the break crop and the
subsequent wheat crops (Kirkegaard and Ryan 2014).
A topic that needs further research is the interaction of
microbial suppression of root pathogens with break crops. Are
these methods of similar effectiveness, can they be combined or
do break crops weaken natural soil suppressiveness? Other topics
that require further research are the effects of AMF, H
2
fertilisation, and on weed control by combining herbicides,
break crops and allelopathy.
An emerging problem is the low and variable yields of the
second crop after a break crop, particularly after canola. A
possible reason is depletion of soil water and/or nutrients by
the first wheat crop and this can be diagnosed with irrigation and
fertiliser. Another possible reason is rebound of soilborne disease,
which can be diagnosed with fungicide.
The estimate in the Introduction that 40% of the world’s wheat
crops are not preceded by a break crop (or break forage or fallow)
suggests scope for yield increases that could contribute
significantly to global food supply and security, although the
potential contribution of crop sequence has been generally
overlooked (Foley et al. 2011). The potential importance of
break crops can be estimated from the mean global wheat
yield of 3.3 t ha
–1
in 2013 (FAOSTAT 2014) and an average
wheat yield increase of 0.7 t ha
–1
on the 40% of wheat currently
grown in monoculture, based on an average of the responses
reported in Table 4. This represents an 8.5% yield increase, which
may be increased by synergies of break crops with inputs and with
additional yield from later crops. There is a special need for break
crop research in Africa, South America and the parts of Eurasia
other than Western Europe, where there are few reports of break-
crop research or adoption on farms.
The lessons from wheat-based systems in this review may
have application for other cropping systems where there has been
research into legume–cereal sequences but less emphasis on non-
legume break crops. Research into temperate maize systems
shows that yield benefits of soybean grown in rotation are
comparable to the benefits of grain legumes for wheat (Karlen
et al. 1994), and likewise, studies on crop sequences of legume
crops in tropical cropping systems that include rice, sorghum and
maize (Peoples and Craswell 1992). There is promising research
in south-east USA on introducing canola into a wheat–soybean
double-cropping system (Cunfer et al. 2006), which has not
extended widely in other wheat–soybean systems. Opportunity
exists for research on occasional break crops as an alternative
to wheat in rice –wheat systems. There are also largely unexplored
possibilities for the use of break crops in intercropping systems
Break crops and rotations for wheat Crop & Pasture Science 543

(Jensen and Hauggaard-Nielsen 2003), for example, by
alternating the spatial pattern of dissimilar species in
successive seasons.
Acknowledgements
We thank Göran Bergkvist, Con Campbell, Martin Entz, Sten Ebbersten,
Jeff Evans, Cindy Grant, Matt McCallum, Perry Miller and Hans Eric Nilsson
for helpful discussions and correspondence, John Brockwell for providing
access to his rhizobial collection and assistance with experimentation, Alec
Zwart for statistical advice, Tony Swan for the plant samples in Fig. 12
and Janet Angus for the photo. The Grains Research and Development
Corporation provided generous current and past financial support for
break-crop research.
References
Angus JF (2001) Nitrogen supply and demand in Australian agriculture.
Australian Journal of Experimental Agriculture 41, 277–288. doi:10.1071/
EA00141
Angus JF, van Herwaarden AF (2001) Increasing water use and water use
efficiency in dryland wheat. Agronomy Journal 93,290–298. doi:10.2134/
agronj2001.932290x
Angus JF, van Herwaarden AF, Fischer RA (1989) Predicting the yield
response of wheat to topdressed nitrogen. In ‘Proceedings 5th
Australian Agronomy Conference’. p. 550. (Australian Society of
Agronomy/The Regional Institute: Gosford, NSW) Available at: www.
regional.org.au/au/asa/1989/contributed/plant-nutrition/p-18.htm
Angus JF, van Herwaarden AF, Howe GN (1991)Productivity and break-crop
effects of winter-growing oilseeds. Australian Journal of Experimental
Agriculture 31, 669–677. doi:10.1071/EA9910669
Angus JF, Gardner PA, Kirkegaard JA, Desmarchelier JM (1994)
Biofumigation: Isothiocyanates released by Brassica roots inhibit
growth of the take-all fungus. Plant and Soil 162, 107–112. doi:10.1007/
BF01416095
Angus JF, Desmarchelier JM, Gardner PA, Green A, Hocking PJ, Howe GN,
Kirkegaard JA, Marcroft S, Mead AJ, Pitson GD, Potter TD, Ryan MH,
Sarwar M, van Herwaarden AF, Wong PTW (1999) Canola and Indian
mustard as break crops for wheat. In ‘Proceedings 10th International
Rapeseed Congress’. (The Regional Institute: Gosford, NSW) Available
at: www.regional.org.au/au/gcirc/2/333.htm#P0_0
Angus JF, Kirkegaard JA, Peoples MB (2001) Rotation, sequence and phase:
research on crop and pasture sequences. In ‘Proceedings 10th Australian
Agronomy Conference’. (Australian Society of Agronomy/The Regional
Institute: Gosford, NSW) Available at: www.regional.org.au/au/asa/
2001/plenary/4/angus.htm#TopOfPage
Angus JF, Peoples MB, Kirkegaard JA, Ryan MH, Ohlander L (2008) The
valueofbreakcropsforwheat.In ‘Proceedings 14th AustralianAgronomy
Conference’. (Australian Society of Agronomy/The Regional Institute:
Gosford, NSW) Available at: www.regional.org.au/au/asa/2008/
concurrent/rotations/5786_angusjf.htm
Angus J, Kirkegaard J, Peoples M, Ryan M, Ohlander L, Hufton L (2011) A
review of break-crop benefits of brassica. In ‘Proceedings 17th Australian
Research Assembly on Brassicas’. pp. 123–127. (Australian Oilseeds
Federation) Available at: www.australianoilseeds.com/__data/assets/
pdf_file/0005/8294/S6-P1-Angus.pdf
Armstrong EL, Heenan DP, Pate JS, Unkovich MJ (1997) Nitrogen benefits
of lupin, field pea, and chickpea to wheat production in south-eastern
Australia. Australian Journal of Agricultural Research 48,39–47.
doi:10.1071/A96054
Arshad MA, Soon YK, Azooz RH (2002) Modified no-till and crop sequence
effects on spring wheat production in northern Alberta, Canada. Soil &
Tillage Research 65,29–36. doi:10.1016/S0167-1987(01)00272-0
Asaduzzaman M, Luckett DJ, Cowley RB, An M, Pratley JE, Lemerle D
(2014) Canola cultivar performance in weed-infested field plots confirms
allelopathy ranking from in vitro testing. Biocontrol Science and
Technology 24, 1394–1411. doi:10.1080/09583157.2014.942596
Asaduzzaman M, Pratley JE, An M, Luckett DJ, Lemerle D (2015)
Metabolomics differentiation in canola genotypes: towards an
understanding of canola allelochemicals. Frontiers in Plant Science 5,
Art. No. 765. doi:10.3389/fpls.2014.00765
Asseng S, Fillery IRP, Gregory PJ (1998) Wheat responses to alternative
crops on a duplex soil. Australian Journal of Experimental Agriculture
38, 481–488. doi:10.1071/EA97152
Bateman GL, Gutteridge RJ, Jenkyn JF, Self MM (2008) Effects of
fluquinconazole and silthiofam, applied as seed treatments to single or
consecutive crops of wheat, on take-all epidemic development and
grain yields. Annals of Applied Biology 152, 243–254.
Beckie HJ, Brandt SA (1997) Nitrogen contribution of field pea in annual
cropping systems 1. Nitrogen residual effects. Canadian Journal of
Plant Science 77, 311–322. doi:10.4141/P96-161
Bedford SA (1899) ‘Experimental farm for Manitoba. Report.’ (Government
Printing Bureau: Ottawa)
Bolger TP, Angus JF, Peoples MB (2003) Nitrogen mineralisation from
root residues of subterranean clover and lucerne. Biology and Fertility of
Soils 38, 296–300. doi:10.1007/s00374-003-0629-y
Bolley HL (1913) Control of soil sickness in wheat lands and possible
methods of control. North Dakota Agricultural Experimental Station
Bulletin No. 107. Available at: http://hdl.handle.net/2027/uiug.301120
19777124
Bourgeois L, Entz MH (1996) Influence of previous crop type on yield of
spring wheat: analysis of commercial field data. Canadian Journal of
Plant Science 76, 457–459. doi:10.4141/cjps96-080
Bowerman P, Jarvis RH (1982) Breaking a run of continuous wheat crops.
Experimental Husbandry 38,20–24.
Brandt SA, Zentner RP (1995) Crop production under alternative rotations on
a dark brown chernozemic soil at Scott, Saskatchewan. Canadian Journal
of Plant Science 75, 789–794. doi:10.4141/cjps95-133
Bremer E, van Kessel C (1992) Plant-available nitrogen from lentil and
wheat residues during a subsequent growing season. Soil Science
Society of America Journal 56, 1155–1160. doi:10.2136/sssaj1992.
03615995005600040025x
Brown PD, Morra MJ (1995) Glucosinolate-containing plant tissues as
bioherbicides. Journal of Agricultural and Food Chemistry 43,
3070–3074. doi:10.1021/jf00060a015
Brown PD, Morra MJ (1997) Control of soil-borne plant pests using
glucosinolate-containing plants. Advances in Agronomy 61, 167–231.
doi:10.1016/S0065-2113(08)60664-1
Brown PD, Morra MJ (2009) Brassicaceae tissues in inhibitors of nitrification
in soil. Journal of Agricultural and Food Chemistry 57, 7706–7711.
doi:10.1021/jf901516h
Bullock DG (1992) Crop rotation. Critical Reviews in Plant Sciences 11,
309–326. doi:10.1080/07352689209382349
Burgess LW, Backhouse D, Summerell BA, Swan LJ (2001) Crown rot of
wheat. In ‘ Fusarium’. (Eds BA Summerell, JF Leslie, D Backhouse, WL
Bryden, LW Burgess) pp. 271 –294. (APS Press: St. Paul, MN, USA)
Butler FC 1961. Root and foot-rot diseases of cereals. Scientific Bulletin No.
77. NSW Department of Agriculture, Wagga Wagga, NSW.
Callaway RM, Cipollini D, Barto K, Thelen GC, Hallett SG, Prati D, Stinson
K, Klironomos J (2008) Novel weapons: invasive plant suppresses fungal
mutualists in America but not in its native Europe. Ecology 89,
1043–1055. doi:10.1890/07-0370.1
Calviño P, Monzon J (2009) Farming systems of Argentina: yield constraints
and risk management. In ‘Crop physiology’. (Ed. VO Sadras) pp. 55–70.
(Academic Press: Burlington, MA)
Campbell CA, Zentner RP, Janzen HH, Bowen KE (1990) ‘Crop rotation
studies on the Canadian prairies.’ (Agriculture Canada: Ottawa)
Cay E, Williams B, McCallum M (2012) ‘Inter-row sowing. PA in practice II.’
Ch. 3. pp. 52–53. (SPAA Precision Agriculture Australia and GRDC)
Available at: www.grdc.com.au/PAinPractice2
544 Crop & Pasture Science J. F. Angus et al.

Chalk PM (1998) Dynamics of biologically fixed N in legume–cereal
rotations: a review. Australian Journal of Agricultural Research 49,
303–316.
Christen O, Sieling K, Hanus H (1992) The effect of different preceding crops
on the development, growth and yield of winter wheat. European Journal
of Agronomy 1,21–28. doi:10.1016/S1161-0301(14)80058-0
ClaupeinW, Zoschke M (1987) Influence of winter wheat growncontinuously
on grain yield, foot rot infection and nematodes compared with winter
wheat in rotations. Journal of Agronomy & Crop Science 158, 227–235.
doi:10.1111/j.1439-037X.1987.tb00267.x
Connor DJ, Loomis RS, Cassman KG (2011) ‘Crop ecology.’ (Cambridge
University Press: Cambridge, UK)
Cook RJ (1981a) The effect of soil reaction and physical condition. In
‘Biology and control of take-all’. (Eds MJC Asher, PJ Shipton)
pp. 343–352. (Academic Press: London)
Cook RJ (1981b) The influence of rotation crops on take-all decline
phenomenon. Phytopathology 71, 189–192. doi:10.1094/Phyto-71-189
Cook RJ (2006) Towards cropping systems that enhance productivity and
sustainability. Proceedings of the National Academy of Sciences of the
United States of America 103, 18389–18394. doi:10.1073/pnas.06059
46103
Cook RJ, Haglund WA (1991) Wheat yield depression associated with
conservation tillage caused by root pathogens, not phytotoxins from
the straw. Soil Biology & Biochemistry 23, 1125–1132. doi:10.1016/
0038-0717(91)90024-E
Cook RJ, Rovira AD (1976) The role of bacteria in the biological control of
Gaeumannomyces graminis by suppressive soils. Soil Biology &
Biochemistry 8, 269–273. doi:10.1016/0038-0717(76)90056-0
Cook RJ, Weller DM (2004) In defense of crop monoculture. In ‘Proceedings
4th International Crop Science Congress’. (The Regional Institute Ltd:
Gosford, NSW) Available at: www.cropscience.org.au/icsc2004/
symposia/2/1/1128_cookrj.htm
Crook DG, Ellis RH, Summerfield RJ (1999) Winter-sown lentil and its
impact on a subsequent cereal crop. Aspects of Applied Biology 56,
241
–247.
Cunfer BM, Buntin GD, Phillips DV (2006) Effect of crop rotation on take-
all of wheat in double-cropping systems. Plant Disease 90, 1161–1166.
doi:10.1094/PD-90-1161
Curl EA (1963) Control of plant diseases by crop rotation. Botanical Review
29, 413–479. doi:10.1007/BF02860813
Dalal RC, Strong WM, Weston EJ, Cooper JE, Wildermuth GB, Lehane KJ,
King AJ, Holmes CJ (1998) Sustaining productivity of a Vertisol at
Warra, Queensland, with fertilisers, no-tillage, or legumes 5. Wheat
yields, nitrogen benefits and water use efficiency of chickpea-wheat
rotation. Australian Journal of Experimental Agriculture 38, 489–501.
doi:10.1071/EA98027
Daubeny C (1845) Memoir on the rotation of crops, and on the quantity
of inorganic matters abstracted from the soil by various plants under
different circumstances. Philosophical Transactions of the Royal
Society of London 135, 179–252. doi:10.1098/rstl.1845.0007
Dean CA, Sun W, Dong Z, Caldwell CD (2006) Soybean nodule hydrogen
metabolism affects soil hydrogen uptake and growth of rotation crops.
Canadian Journal of Plant Science 86, 1355–1359. doi:10.4141/P06-082
Delroy ND, Bowden JW (1986) Effect of deep ripping, the previous crop,
and applied nitrogen on the growth and yield of a wheat crop. Australian
Journal of Experimental Agriculture 26, 469–479. doi:10.1071/
EA9860469
Dill-Macky R, Jones RK (2000) The effect of previous crop residues and
tillage on Fusarium head blight of wheat. Plant Disease 84,71–76.
doi:10.1094/PDIS.2000.84.1.71
Dong Z, Wu L, Kettlewell B, Caldwell CD, Layzell DB (2003) Hydrogen
fertilization of soils—is this a benefit of legumes in rotation? Plant, Cell
& Environment 26, 1875–1879. doi:10.1046/j.1365-3040.2003.01103.x
Donn S, Kirkegaard JA, Perera G, Richardson AE, Watt M (2014) Evolution
of bacterial communities in the wheat crop rhizosphere. Environmental
Microbiology 17, 610–621. doi:10.1111/1462-2920.12452
Doyle AD, Moore KJ, Herridge DF (1988) The narrow-leafed lupin (Lupinus
angustifolius L.) as a nitrogen-fixing rotation crop for cereal production.
III. Residual effects of lupins on subsequent cereal crops. Australian
Journal of Agricultural Research 39, 1029–1037. doi:10.1071/AR988
1029
Dyke GV, Slope DB (1978) Effects of previous legume and oat crops on
grain yield and take-all in spring barley. Journal of Agricultural
Science, Cambridge 91, 443–451.
Eastwood RF, Lagudah ES, Appels R, Hannah M, Kollmorgen JF (1991)
Triticum tauschii: a novel source of resistance to cereal cyst nematode
(Heterodera avenae). Australian Journal of Agricultural Research 42,
69–77.
Ebbersten S (1980) Optimised plant husbandry III. Skogs och
Lantbruksakademiens Tidskrift 119, 367–389.
Evans LT (1998) ‘Feeding the ten billion.’ (Cambridge University Press:
Cambridge, UK)
Evans HJ, Purohit K, Cantrell MA, Eisbrenner G, Russell SA, Hanus FJ, Lepo
JE (1981) Hydrogen losses and hydrogenases in nitrogen-fixing
organisms. In ‘Current perspectives in nitrogen fixation’. (Eds AH
Gibson, WE Newton) pp. 84–96. (Australian Academy of Science:
Canberra, ACT)
Evans HJ, Russell SA, Hanus FJ, Ruiz-Argueso T (1988) The importance of
hydrogen recycling in nitrogen fixation by legumes. In ‘World crops: cool
season food legumes’. (Ed. RJ Summerfield) pp. 777–791. (Kluwer
Academic Publishers: Boston, MA)
Evans J, Fettell NA, Coventry DR, O’Connor GE, Walsgott DN, Mahoney J,
Armstrong EL (1991) Wheat responses after temperate crop legumes in
south-eastern Australia. Australian Journal of Agricultural Research 42,
31–43. doi:10.1071/AR9910031
EvansJ, Scott G, Lemerle D, Kaiser A, Orchard B, Murray GM, ArmstrongEL
(2003) Impact of legume ‘break’ crops on the residual amount and
distribution of soil mineral nitrogen. Australian Journal of Agricultural
Research 54, 763–776. doi:10.1071/AR02149
FAOSTAT (2014) FAOSTAT3. Available at: www.faostat3.fao.org
Felton WL, Marcellos H, Alston C, Martin RJ, Backhouse D, Burgess LW,
Herridge DF (1998) 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, 401–407. doi:10.1071/A97067
Fernandez MR, Zentner RP, McConkey BG, Campbell CA (1998) Effects
of crop rotations and fertilizer management on leaf spotting diseases
of spring wheat in southwestern Saskatchewan. Canadian Journal of
Plant Science 78, 489–
496. doi:10.4141/P97-082
Fischer RA, Byerlee D, Edmeades GD (2014) ‘ Crop yields and global food
security.’ (ACIAR: Canberra, ACT)
Foley JA, Ramankutty N, Brauman KA, Cassidy ES, Gerber JS, Johnston M,
Mueller ND, O’Connell C, Ray DK, West PC, Balzer C, Bennett EM,
Carpenter SR, Hill J, Monfreda C, Polasky S, Rockstrom J, Sheehan J,
Siebert S, Tilman D, Zaks DPM (2011) Solutions for a cultivated planet.
Nature 478, 337–342. doi:10.1038/nature10452
French RJ, Schultz JE (1984) Water use efficiency in wheat in a
Mediterranean type environment: II. Some limitations to efficiency.
Australian Journal of Agricultural Research 35,765–775. doi:10.1071/
AR9840765
Funayama-Noguchi S, Noguchi KO, Terashima I (2014) Comparison of the
response to phosphorus deficiency in two lupin species, Lupinus albus
and L. angustifolius, with contrasting root morphology. Plant, Cell &
Environment 38, 399–410. doi:10.1111/pce.12390
Gan YT, Miller PR, McConkey BG, Zentner RP, Stevenson FC, McDonald
CL (2003) Influence of diverse cropping sequences on durum yield
Break crops and rotations for wheat Crop & Pasture Science 545

and protein in the semiarid northern Great Plains. Agronomy Journal 95,
245–252. doi:10.2134/agronj2003.0245
Gardner PA, Angus JF, Pitson GD, Wong PWT (1998) A comparison of six
methods to control take-all in wheat. Australian Journal of Agricultural
Research 49, 1225–1240. doi:10.1071/A98050
Glass GV (1976) Primary, secondary and meta-analysis of research.
Educational Researcher 5,3–8. doi:10.3102/0013189X005010003
Golding A-L, Zou Y, Yang X, Flynn B, Dong Z (2012) Plant growth
promoting H
2
-oxidizing bacteria as seed inoculants for cereal crops.
Agricultural Sciences 3, 510–516. doi:10.4236/as.2012.34060
Green A (1986) A mutant genotype of flax (Linum usitiatissimum L.)
containing very low levels of linolenic acid in its seed oil. Canadian
Journal of Plant Science 66, 499–503. doi:10.4141/cjps86-068
Green CJ, Blackmer AM (1995) Residue decomposition effects on nitrogen
availability following corn or soybean. Soil Science Society of America
Journal 59, 1065–1070. doi:10.2136/sssaj1995.03615995005900040016x
Gregory PJ (1998) Alternative crops for duplex soils: growth and water use
of some cereal, legume and oilseed crops and pastures. Australian
Journal of Agricultural Research 49,21–32. doi:10.1071/A97053
Gupta VVSR, Rovira AD, Roget DK (2011) Principles and management
of soil biological factors for sustainable rainfed farming systems. In
‘Rainfed farming systems’. (Eds P Tow, I Cooper, I Partridge, C
Birch) pp. 149–184. (Springer: Dordrecht, The Netherlands)
Gupta VVSR, McKay A, Diallo S, Smith D, Cook A, Kirkegaard J, Ophel-
Keller K, Davoren W, Llewellyn R, Roget DK, MacLeod WJ (Eds) (2012)
Rhizoctonia solani AG8 inoculum levels in Australian soils are influenced
by crop rotation and summer rainfall. In ‘Proceedings Seventh Australian
Soilborne Diseases Conference’. Fremantle, W. Aust. (Australasian Plant
Pathology Society)
Haas D, Défago G (2005) Biological control of soil-borne pathogens by
fluorescent pseudomonads. Nature Reviews. Microbiology 3, 307–319.
doi:10.1038/nrmicro1129
Hamblin J, Delane R, Bishop A, Adam G (1993) The yield of wheat following
lupins: effects of different lupin genotypes and management. Australian
Journal of Agricultural Research 44, 645–659. doi:10.1071/AR9930645
Hammel JE (1995) Long-term tillage and crop rotation effects on winter wheat
production in northern Idaho. Agronomy Journal 87,16–22. doi:10.2134/
agronj1995.00021962008700010004x
Hanley F, Ridgman WJ (1978) Some effects of growing winter wheat
continuously. The Journal of Agricultural Science 90, 517–521.
doi:10.1017/S0021859600056033
Haramoto ER, Gallandt ER (2004) Brassica cover cropping for weed
management. Renewable Agriculture and Food Systems 19, 187–198.
doi:10.1079/RAFS200490
Harris HC (1995) Long-term trials on soil and crop management at
ICARDA. In ‘Soil management: experimental basis for sustainability
and environmental quality’. (Eds R Lal, BA Stewart) pp. 447–469. (Lewis
Publishers: Boca Raton, FL, USA)
Harris RH, Scammell GJ, Müller W, Angus JF (2002) Wheat productivity
after break crops and grass-free annual pasture. Australian Journal of
Agricultural Research 53, 1271–1283. doi:10.1071/AR02021
Heenan DP (1995) Effect of broad-leaf crops and their sowing time on
subsequent wheat production. Field Crops Research 43,19–29.
doi:10.1016/0378-4290(95)00033-M
Heenan DP, Taylor AC, Cullis BR, Lill WJ (1994) Long term effects of
rotation, tillage and stubble management on wheat production in southern
N.S.W. Australian Journal of Agricultural Research 45,93–117.
doi:10.1071/AR9940093
Herridge DF (2006) Chickpea increases productivity and profitability of
farming systems. GRDC Update. pp. 62–63. Available at: www.grdc.
com.au/growers/res_upd/north/n06/herridge2.htm
Hocking PJ (2001) Organic acids exuded from roots in phosphorus uptake and
aluminum tolerance of plants in acid soils. Advances in Agronomy 74,
63–97. doi:10.1016/S0065-2113(01)74031-X
Hocking PJ, Kirkegaard JA, Angus JF, Gibson AH, Koetz EA (1997)
Comparisons of canola, Indian mustard and Linola in two contrasting
environments. I. Effects of nitrogen fertiliser on dry-matter production,
seed yield and seed quality. Field Crops Research 49, 107–125.
doi:10.1016/S0378-4290(96)01063-5
Holford ICR, Crocker GJ (1997) A comparison of chickpeas and pasture
legumes for sustaining yields and nitrogen status of subsequent wheat.
Australian Journal of Agricultural Research 48, 305–315. doi:10.1071/
A96072
Hollaway GJ, Taylor SP, Eastwood RF, Hunt CH (2000) Effect of field crops
on the density of Pratylenchus in South Eastern Australia; Part 2:
P. thornei. Journal of Nematology 32, 600–608.
Hornby D (Ed.) (1998) ‘Take-all disease of cereals: a regional perspective.’
(CAB International: Wallingford, UK)
Hort AF (1926) ‘Theophrastus: Enquiry into Plants II. ’ (Harvard University
Press: Cambridge MA, USA)
Hunt JR, Kirkegaard JA (2011) Re-evaluating the contribution of summer
fallow rain to wheat yield in southern Australia. Crop & Pasture Science
62, 915–929. doi:10.1071/CP11268
Hunt JR, BrowneC, McBeathTM, VerburgK, Craig S, Whitbread AM (2013)
Summer fallow weed control and residue management impacts on
winter crop yield through soil water and N accumulation in a winter-
dominant, low rainfall region of southern Australia. Crop & Pasture
Science 64, 922–934. doi:10.1071/CP13237
Ishikawa S, Hare MC, Kettlewell PS (2012) Effects of strobilurin fungicide
programmes and fertilizer nitrogen rates on winter wheat: leaf area,
dry matter yield and nitrogen yield. The Journal of Agricultural
Science 150, 427–441. doi:10.1017/S0021859611000724
Jensen ES, Hauggaard-Nielsen H (2003) How can increased use of
biological N
2
fixation in agriculture benefit the environment? Plant
and Soil 252, 177–186. doi:10.1023/A:1024189029226
Karlen DL, Varvel GE, Bullock DG, Cruse RM (1994) Crop rotations for
the 21st century. Advances in Agronomy 53,1–45. doi:10.1016/S0065-
2113(08)60611-2
Khakbazan M, Grant CA, Huang HC, Smith EG, O’Donovan JT, Blackshaw
RE, Harker KN, Lafond GP, Johnson EN, Gan Y, May WE, Turkington
TK, Lupwayi NZ (2014) Economic effects of preceding crops and
nitrogen application on canola and subsequent barley. Agronomy
Journal 106, 2055–2066. doi:10.2134/agronj14.0253
Kidd CR, Murray GM, Pratley JE, Leys AR (2002) Effect of timing of pasture
grass removal on subsequent take-all incidence and yield in wheat in
southern New South Wales. Australian Journal of Experimental
Agriculture 42, 1087–1094. doi:10.1071/EA01097
King PM (1984) Crop and pasture rotations at Coonalpyn, South Australia:
effects on soil-borne diseases, soil nitrogen and cereal production.
Australian Journal of Experimental Agriculture and Animal
Husbandry 24, 555–564. doi:10.1071/EA9840555
Kirkegaard JA, Ryan MH (2014) Magnitude and mechanisms of persistent
crop sequence effects on wheat. Field Crops Research 164, 154–165.
doi:10.1016/j.fcr.2014.05.005
Kirkegaard JA, Gardner PA, Angus JF, Koetz E (1994) Effect of Brassica
break crops on the growth and yield of wheat. Australian Journal of
Agricultural Research 45, 529–545. doi:10.1071/AR9940529
KirkegaardJA, Angus JF, Howe GN, Gardner PA, Creswell HP (1996a) Grain
oats—a poor break crop for wheat. In ‘Proceedings Australian Agronomy
Conference’. pp. 349–352. (Australian Society of Agronomy/The
Regional Institute: Gosford, NSW) Available at: www.regional.org.au/
au/asa/1996/contributed/349kirkegaard.htm#P1_40
Kirkegaard JA, Wong PTW, Desmarchelier JM (1996b) In vitro suppression
of fungal root pathogens of cereals by Brassica tissues. Plant Pathology
45, 593–603. doi:10.1046/j.1365-3059.1996.d01-143.x
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
546 Crop & Pasture Science J. F. Angus et al.

crops. Field Crops Research 52, 179–191. doi:10.1016/S0378-4290(96)
01057-X
Kirkegaard JA, Howe GN, Mele PM (1999) Enhanced accumulation of
mineral-N following canola. Australian Journal of Experimental
Agriculture 39, 587–593. doi:10.1071/EA98163
Kirkegaard JA, Sarwar M, Wong PTW, Mead AJ, Howe GN, Newell M
(2000) Field studies on the biofumigation of take-all by Brassica break
crops. Australian Journal of Agricultural Research 51, 445–456.
doi:10.1071/AR99106
Kirkegaard JA, Howe GN, Pitson G (2001) Agronomic interactions between
drought and crop sequence. In ‘Proceedings 10th Australian Agronomy
Conference’. (Australian Society of Agronomy/The Regional Institute:
Gosford, NSW) Available at: www.regional.org.au/au/asa/2001/4/c/
kirkegaard1.htm#TopOfPage
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. doi:10.1071/AR03178
Kirkegaard JA, Christen O, Krupinsky J, Layzell DB (2008) Break crop
benefits in temperate wheat production. Field Crops Research 107,
185–195. doi:10.1016/j.fcr.2008.02.010
Kirkegaard JA, Hunt JR, McBeath TM, Lilley JM, Moore A, Verburg K,
Robertson MJ, Oliver YM, Ward PR, Milroy DF, W hitbread AM
(2014) Improving water productivity in the Australian grains
industry—a nationally coordinated approach. Crop & Pasture
Science 65,583–601.
Koide RT, Peoples MS (2012) On the nature of temporary yield loss in
maize following canola. Plant and Soil 360, 259–269. doi:10.1007/
s11104-012-1237-0
Kollmorgen JF, Griffiths JB, Walsgott DN (1983) The effect of various
crops on the survival and carryover of the wheat take-all fungus
Gaeumannomyces graminis var. tritici. Plant Pathology 32,73–77.
doi:10.1111/j.1365-3059.1983.tb01303.x
Kreutzer WA (1965) The reinfestation of treated soil. In ‘Ecology of soil-
borne plant pathogens’. (Eds KF Baker, WC Synder) pp. 495–508.
(University of California Press: Berkeley, CA, USA)
Krupinsky JM, Tanaka DL, Merrill SD, Liebig MA, Hanson JD (2006) Crop
sequence effects of 10 crops in the northern Great Plains. Agricultural
Systems 88, 227
–254. doi:10.1016/j.agsy.2005.03.011
Kumar K, Goh KM (2000) Crop residues and management practices:
effects of soil quality, soil nitrogen dynamics, crop yield and nitrogen
recovery. Advances in Agronomy 68, 197 –319. doi:10.1016/S0065-2113
(08)60846-9
Kwak Y-S, Weller DM (2013) Take-all of wheat and natural disease
suppression: a review. The Plant Pathology Journal 29, 125–135.
La Favre JS, Forcht DD (1983) Conservation in soil of H
2
liberated from N
2
fixation by Hup– nodules. Applied and Environmental Microbiology 46,
304–311.
Lafond GP, Loeppky H, Derksen DA (1992) The effect of tillage systems
and crop rotations on soil water conservation, seedling establishment
and crop yield. Canadian Journal of Plant Science 72, 103–115.
doi:10.4141/cjps92-011
Larney FJ, Lindwall CW (1994) Winter wheat performance in various
cropping systems in southern Alberta. Canadian Journal of Plant
Science 74,79–86. doi:10.4141/cjps94-014
Lawes JB, Gilbert JH (1894) Rotation of crops. Journal of the Royal
Agricultural Society of England 55, 585–646.
Lawes RA, Gupta VVSR, Kirkegaard JA, Roget DK (2013) Evaluating the
contribution of take-all control to the break-crop effect in wheat. Crop &
Pasture Science 64, 563–572. doi:10.1071/CP13151
Ledingham RJ (1961) Crop rotations and common root rot in wheat.
Canadian Journal of Plant Science 41, 479–486. doi:10.4141/cjps61-072
Leighty CE (1938) Crop rotation. In ‘Soil and men: yearbook of agriculture
1938’. (US Government Printing Office: Washington, DC)
Lendenmann M, Thonar C, Barnard RL, Salmon Y, Werner RA, Frossard E,
Jansa J (2011) Symbiont identity matters: carbon and phosphorus fluxes
between Medicago truncatula and different arbuscular mycorrhizal fungi.
Mycorrhiza 21, 689–702. doi:10.1007/s00572-011-0371-5
Lindén B, Engström L (2006) ‘Winter oilseed rape, oats and field peas as
crops preceding winter wheat—effect on nitrogen dynamics in the soil
and on wheat yields.’ (Swedish University of Agricultural Sciences:
Uppsala, Sweden)
Lund MG, Carter GA, Oplinger ES (1993) Tillage and crop rotation affect
corn, soybean,and winter wheat yields. Journal of Production Agriculture
6, 207–213. doi:10.2134/jpa1993.0207
Malik RS, Seymour M, French RJ, Kirkegaard JA, Lawes RA, Liebig MA
(2015) Dynamic crop sequencing in Western Australian cropping
systems. Crop & Pasture Science 66, 594–609.
Marcellos H (1984) Influence of prior crops of chickpeas, fababeans and
lupinson wheat. Journal of the Australian Institute of Agricultural Science
50
, 111–113.
McAlpine D (1904) Take-all and white-heads in wheat. Agriculture Journal
of Victoria 2, 410–427.
McBeath TM, Gupta VVSR, Llewellyn RS, Davoren CW, Whitbread AM
(2015) Break-crop effects on wheat production across soils and seasons
in a semi-arid environment. Crop & Pasture Science 66, 566–579.
McEwen J, Darby RJ, Hewitt MV, Yeoman DP (1989) Effects of field beans,
fallow, lupin, oats, oilseed rape, peas, ryegrass, sunflower and wheat on
nitrogen residues in the soil and on the growth of a subsequent wheat crop.
Journal of Agricultural Science, Cambridge 115, 209–219.
McLearn N, Dong Z (2002) Microbial nature of the hydrogen-oxidizing agent
in hydrogen-treated soil. Biology and Fertility of Soils 35, 465–469.
doi:10.1007/s00374-002-0495-z
Mendes R, Kruijt M, de Bruijn I, Dekkers E, ven der Voort M, SchneiderJHM,
Piceno YM, DeSantis TZ, Andersen GL, Bakker PAHM, Raaijmakers JM
(2011) Deciphering the rhizosphere microbiome for disease-suppressive
bacteria. Science 332, 1097–1100. doi:10.1126/science.1203980
Miller PR, Holmes JA (2005) Cropping sequence effects of four broadleaf
crops on four cereal crops in the northern Great Plains. Agronomy Journal
97, 189–200. doi:10.2134/agronj2005.0189
Miller PR, Waddington J, McDonald CL, Derksen DA (2002) Cropping
sequence affects wheat productivity on the semiarid northern Great Plains.
Canadian Journal of Plant Science 82, 307–318. doi:10.4141/P01-116
Miller PR, Gan Y, McConkey BG, McDonald CL (2003) Pulse crops for
the northern Great Plains. 1. Grain productivity and residual effects on
soil water and N. Agronomy Journal 95, 972–979. doi:10.2134/
agronj2003.0972
Miller PR, Engel RE, Holmes JA (2006) Cropping sequence effect of pea
and pea management on spring wheat in the northern Great Plains
Agronomy Journal 98, 1610–1619. doi:10.2134/agronj2005.0302
Miller PR, Bekkerman A, Jones CA, Burgess MA, Holmes JA, Engel RE
(2015) Pea in rotation with wheat reduced uncertainty of economic returns
in southwest Montana. Agronomy Journal 107, 541–550. doi:10.2134/
agronj14.0185
Monjardino M, Pannell DJ, Powles SB (2005) The economic value of
glyphosate-resistant canola in the management of two widespread crop
weeds in a Western Australian farming system. Agricultural Systems 84,
297–315. doi:10.1016/j.agsy.2004.06.018
Moodie M (2012) ‘Production and environmental impacts of broad leaved
break crops in the Mallee.’ (Mallee Catchment Management Authority:
Mildura, Vic.)
Moyer JR, Blackshaw RE, Doram RC, Huang HC, Entz T (2005) Effect of
previous crop and herbicides on weed growth and wheat yield. Canadian
Journal of Plant Science 85, 735–746. doi:10.4141/P04-170
Murphy DV, Fillery IRP, Sparling GP (1998) Seasonal fluctuations in gross
N mineralisation, ammonium consumption, and microbial biomass in
a Western Australian soil under different land uses Australian Journal of
Agricultural Research 49, 523–535. doi:10.1071/A97096
Break crops and rotations for wheat Crop & Pasture Science 547

Nilsson HE (1969) Studies of root and foot rot diseases of cereals and grasses.
Lantbrukshogskolans Annaler 35, 275 –807.
Nilsson G (1985) Crops preceding spring wheat. Swedish University of
Agricultural Sciences, Uppsala, Sweden.
Nilsson G, Wallgren B (1981) Yields of cereals in different crop rotations
in trials R4-006-1 and R4-006-2 in Lanna. Swedish University of
Agricultural Sciences, Uppsala, Sweden.
Nilsson-Ehle H (1920) Winter wheat and crop rotation in Scania Skånsk
Jordbrukstidskrift 36, 561–565. [in Swedish]
NVT (2015) Available at: www.nvtonline.com.au/google-maps/2010/
WheatMaNSW.xls
O’Donovan JT, Grant CA, Blackshaw RE, Harker KN, Johnson EN, Gan Y,
Lafond GP, May WE, Turkington TK, Lupwayi NZ, Stevenson FC,
McLaren DL, Khakbazan M, Smith EG (2014) Rotational effects of
legumes and non-legumes on hybrid canola and malting barley.
Agronomy Journal 106, 1921–1932. doi:10.2134/agronj14.0236
Oliver YM, Robertson MJ, Weeks C (2010) A new look at an old practice:
benefits from soil water accumulation in long fallows under Mediterranean
conditions. Agricultural Water Management 98,291–300. doi:10.1016/j.
agwat.2010.08.024
Olofsson S (1993) Influence of preceding crop and crop residue on stand and
yield of winter wheat (Triticum aestivum L.) in different tillage systems,
including zero tillage. Swedish University of Agricultural Sciences,
Uppsala, Sweden.
Olofsson S, Wallgren B (1984) Winter wheat in the crop rotation. Results
from two trial series with various crops preceding winter wheat: R-1711
and 1712. Department of Plant Husbandry, Swedish University of
Agricultural Sciences, Uppsala, Sweden.
Olsson S (1995) On barley monoculture soils. Swedish University of
Agricultural Sciences, Uppsala, Sweden.
Olsson S, Kadir S (1994) Characteristics of a Pythium arrhenomanes with
high frequence in barley soils. Journal of Phytopathology 140, 335–345.
doi:10.1111/j.1439-0434.1994.tb00616.x
Ophel-KellerK, McKay A, Hartley D, Herdina , Curran J (2008)Development
of a routine DNA-based testing service for soilborne diseases in Australia.
Australasian Plant Pathology 37, 243–253. doi:10.1071/AP08029
Osborne CA, Peoples MB, Janssen PH (2010) Detection of reproducible,
single-member shift in soil bacterial communities exposed to low levels of
hydrogen. Applied and Environmental Microbiology 76, 1471–1479.
doi:10.1128/AEM.02072-09
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.
doi:10.1071/CP09345
Pannell DJ, Gill GS (1994) Mixtures of wild oats (Avena fatua) and ryegrass
(Lolium rigidum) in wheat—competition and optimal economic control.
Crop Protection 13, 371
–375. doi:10.1016/0261-2194(94)90052-3
Patterson HD (1964) Theory of cyclic rotation experiments. Journal of the
Royal Statistical Society. Series A (General) 31,1–45.
Pearse SJ, Veneklaas EJ, Cawthray GR, Bolland MDA, Lambers H (2006)
Carboxylate release of wheat, canola and 11 grain legume species as
affected by phosphorus status. Plant and Soil 288, 127–139. doi:10.1007/
s11104-006-9099-y
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 &
Biochemistry 68, 429–439. doi:10.1016/j.soilbio.2013.09.030
Penton CR, Gupta VVSR, Tiedje JM, Neate SM, Ophel-Keller K, Gillings M,
Harvey P, Pham A, Roget DK (2014) Fungal community structure in
disease suppressive soils assessed by 28S LSU gene sequencing. PLOS
One 9. doi:10.1371/journal.pone.0093893
Peoples MB, Craswell ET (1992) Biological nitrogen fixation: investments,
expectations and actual contributions to agriculture. Plant and Soil 141,
13–39. doi:10.1007/BF00011308
Peoples MB, McLennan PD, Brockwell J (2008) Hydrogen emission from
nodulated soybeans [Glycine max (L.) Merr.] and consequences for the
productivity of a subsequent maize (Zea mays L.) crop. Plant and Soil
307,67– 82. doi:10.1007/s11104-008-9582-8
Peoples MB, Brockwell J, Herridge DF, Rochester IJ, Alves BJR, Urquiaga S,
BoddeyRM, Dakora FD, Bhattarai S, Maskey SL, Sampet C, Rerkasem B,
Khan DF, Hauggaard-Nielsen H, Jensen ES (2009) The contributions
ofnitrogen-fixing crop legumesto the productivity of agriculturalsystems.
Symbiosis 48,1–17. doi:10.1007/BF03179980
Powlson DS, Johnston AE (1994) Long-term field experiments: their
importance in understanding sustainable land use. In ‘Soil resilience
and sustainable land use’. (Eds DJ Greenland, I Szabolcs)
pp. 367–394. (CAB International: Wallingford, UK)
Preissel S, Reckling M, Schläfke N, Zander P (2015) Magnitude and farm-
economic value of grain legume pre-crop benefits in Europe. Field Crops
Research 175,64–79. doi:10.1016/j.fcr.2015.01.012
Prew RD, Dyke GV (1979) Experiments comparing ‘break crops’ as a
preparation for winter wheat followed by spring barley. Journal of
Agricultural Science, Cambridge 92, 189–
201.
Reeves TG (1982) Improved soil management—the key to more dollars
with less dust. Western Australian Department of Agriculture, Perth,
W. Aust.
Reeves TG, Ellington A, Brooke HD (1984) Effect of lupin-wheat rotations
on soil fertility, crop disease and crop yields. Australian Journal of
Experimental Agriculture and Animal Husbandry 24, 595–600.
doi:10.1071/EA9840595
Rick TL, Jones CA, Engel RE, Miller PR (2011) Green manure and
phosphate rock effects on phosphorus availability in a northern Great
Plains dryland organic cropping system. Organic Agriculture 1,81–90.
doi:10.1007/s13165-011-0007-2
Ridgman WJ, Walters DE (1982) A comparison of growing wheat
continuously with growing wheat in a four-course rotation. The
Journal of Agricultural Science 99, 139–143. doi:10.1017/S0021859
600055131
Ripley PO (1941) The influence of crops upon those which follow.
Scientific Agriculture 21, 522–584.
Ritchie JT (1981) Soil water availability. Plant and Soil 58, 327–338.
doi:10.1007/BF02180061
Robertson MJ, Lawes RA, Bathgate A, Byrne F, White P, Sands R (2010)
Determinants of the proportion of break crops on Western Australian
broadacre farms. Crop & Pasture Science 61, 203–213. doi:10.1071/
CP09207
Roget DK (1995) Decline in root rot (Rhizoctonia solani AG-8) in wheat
in a tillage and rotation experiment at Avon, South Australia. Australian
Journal of Experimental Agriculture 35, 1009–1013. doi:10.1071/EA99
51009
Roget DK, Rovira AD (1991) The relationship between incidence of take-
all infection of wheat, rainfall and yield. Australian Journal of
Experimental Agriculture 31, 509–513. doi:10.1071/EA9910509
Rothrock CS, Cunfer BM (1986) Absence of take-all decline in double-
cropped fields. Soil Biology & Biochemistry 18, 113–114. doi:10.1016/
0038-0717(86)90112-4
Rothrock CS, Cunfer BM (1991) Influence of small grains rotations on take-
all in a subsequent wheat crop. Plant Disease 75, 1050–1052. doi:10.1094/
PD-75-1050
Rovira AD (1969) Plant root exudates. Botanical Review 35,35
–57.
doi:10.1007/BF02859887
Rovira AD (1988) Soilborne root diseases—a major constraint on
productivity of dryland cereals. In ‘Proceedings, International
Conference on Dryland Farming’.15–19 August 1988, Amarillo/
Bushland, TX, USA.
Rovira AD, Wildermuth GB (1981) The nature and mechanisms of
suppression. In ‘Biology and control of take-all’. (Eds MJC Asher, PJ
Shipton) pp. 385–415. (Academic Press: London)
548 Crop & Pasture Science J. F. Angus et al.

Rowland IC, Mason MG, Hamblin J (1988) Effect of lupin and wheat on the
yield of subsequent wheat crops grown at several rates of applied nitrogen.
Australian Journal of Experimental Agriculture 28,91–97. doi:10.1071/
EA9880091
Rowland IC, Mason MG, Pritchard IA, French RJ (1994) Effect of field
peas and wheat on the yield and protein content of subsequent wheat
crops grown at several rates of applied nitrogen. Australian Journal of
Experimental Agriculture 34, 641–646. doi:10.1071/EA9940641
Russell RC (1934) Studies in cereal diseases X. Studies of take-all and its
causal organism, Ophiobolis graminis Sacc. Bulletin No 170 New Series.
Department of Agriculture, Ottawa. Available at: https://archive.org/
details/studiesoftakeall170russ
Russell CA, Fillery IRP (1999) Turnover of nitrogen from compounds of
lupin stubble to wheat in a sandy soil. Australian Journal of Soil Research
37, 575–592. doi:10.1071/S98071
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. doi:10.1023/
A:1022839930134
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. doi:10.1016/
j.agee.2012.03.011
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. doi:10.1071/AR02005
Ryan MH, Taylor G, Kirkegaard JA (2003) Poor wheat crops following
canola—a survey of farmers and agronomists. In ‘Proceedings 11th
Australian Agronomy Conference’. (Australian Society of Agronomy/
The Regional Institute: Gosford, NSW) Available at: www.regional.org.
au/au/asa/2003/c/6/ryan.htm
Ryan MH, van Herwaarden AF, Angus JF, Kirkegaard JA (2005) Reduced
growth of autumn-sown wheat in a low-P soil is associated with high
colonisation by arbuscular mycorrhizal fungi. Plant and Soil 270,
275–286. doi:10.1007/s11104-004-1611-7
Ryan MH, Kirkegaard JA, Angus JF (2006) Brassica crops stimulate soil
mineral N accumulation. Australian Journal of Soil Research 44,
367–377. doi:10.1071/SR05143
Ryan J, Singh M, Pala M (2008) Long-term cereal-based rotation trials in
the Mediterranean region: implications for cropping sustainability.
Advances in Agronomy 97,273–319. doi:10.1016/S0065-2113(07 )00007-7
Sarwar M, Kirkegaard JA, Wong PTW, Desmarchelier JM (1998)
Biofumigation potential of brassicas. III. In-vitro toxicity of
isothiocyanates to soil-borne fungal pathogens. Plant and Soil 201,
103–112. doi:10.1023/A:1004381129991
Schreiner RP, Koide RT (1993) Mustards, mustard oils and mycorrhizas. New
Phytologist 123, 107–113. doi:10.1111/j.1469-8137.1993.tb04536.x
Schultz JE (1995) Crop production in a rotation trial at Tarlee, SouthAustralia.
Australian Journal of Experimental Agriculture 35, 865–876.
doi:10.1071/EA9950865
Selman M (1969) Continuous wheat: the problem and prospects. Report of
Boxworth Experimental Husbandry Farm for 1968, Cambridge, UK.
pp. 19–26.
Selman M (1975) Experiments in continuous wheat. Part I. The effects
of break crops introduced into a run of continuous wheat (Sykes’s
Field). Experimental Husbandry 29,1–7.
Seymour M, Kirkegaard JA, Peoples MB, White PF, French RJ (2012)
Break-crop benefits to wheat in Western Australia—insights from over
three decades of research. Crop & Pasture Science 63,1–16. doi:10.1071/
CP11320
Sieling K, Christen O (2015) Crop rotation effects on yield of oilseed
rape, wheat and barley and residual effects on the subsequent wheat.
Archives of Agronomy and Soil Science, doi:10.1080/03650340.2015.
1017569
Sieling K, Stahl C, Winkelmann C, Christen O (2005) Growth and yield of
winter wheat in the first 3 years of a monoculture under varying N
fertilization in NW Germany. European Journal of Agronomy 22,
71–84. doi:10.1016/j.eja.2003.12.004
Simpfendorfer S, Brettell RIS, Nicol JM (Eds) (2012) Inter-row sowing
reduces crown rot in winter cereals. In ‘First International Crown Rot
Workshop for wheat improvement’. Narrabri. (Organising Committee of
the 1st International Crown Rot Workshop)
Sims HJ (1977) Cultivation and fallowing practices. In ‘Soil factors in
crop production in a semi-arid environment’. (Eds JS Russell, EL
Greacen) pp. 243–261. (University of Queensland Press: Brisbane, Qld)
Slope DB, Cox J (1964) Continuous wheat growing and the decline of take-
all. Rothamsted Experiment Station Report for 1963. p. 108.
Slope DB, Etheridge J (1971) Grain yield and incidence of take-all
(Ophiobolis graminis (Sacc.) in wheat grown in different crop
sequences. Annals of Applied Biology 67,13–22. doi:10.1111/j.1744-
7348.1971.tb02904.x
Smith BJ, Kirkegaard JA (2002) In vitro inhibition of soil microorganisms
by 2-phenylethyl isothiocyanate. Plant Pathology 51, 585–593.
doi:10.1046/j.1365-3059.2002.00744.x
Smith SE, Read DJ (2008) ‘
Mycorrhizal symbiosis.’ (Academic Press:
Oxford, UK)
Smith BJ, Kirkegaard JA, Howe GN (2004) Impacts of Brassica break-crops
on soil biology and yield of following wheat crops. Australian Journal
of Agricultural Research 55,1–11. doi:10.1071/AR03104
Stevenson FC, van Kessel C (1996) The nitrogen and non-nitrogen benefits
of pea to succeeding crops. Canadian Journal of Plant Science 76,
735–745. doi:10.4141/cjps96-126
Strong WM, Harbison J, Nielsen RGH, Hall BD, Best EK (1986) Nitrogen
availability in a Darling Downs soil following cereal, oilseed and grain
legume crops. 2. Effects of residual soil nitrogen and fertiliser nitrogen on
subsequent wheat crops. Australian Journal of Experimental Agriculture
26, 353–359. doi:10.1071/EA9860353
Sutton GL (1911) Take-all. Practical methods or its eradication and control.
Agricultural Gazette of NSW 22, 161–163.
Sylvester-Bradley R, Cross RB (1991) Nitrogen residues from pea and beans
and the response of the following cereal to applied nitrogen. Aspects of
Applied Biology 27, 293–298.
Tardif FJ, Preston C, Holtum JAM, Powles SB (1996) Resistance to
acetyl-coenzyme A carboxylase-inhibiting herbicides endowed by a
single major gene encoding a resistant target site in a biotype of Lolium
rigidum. Australian Journal of Plant Physiology 23,15–23. doi:10.1071/
PP9960015
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.
doi:10.1071/AR9870847
Thompson JP, Wildermuth GB (1989) Colonisation of crop and pasture
species with vesicular-arbuscular mycorrhizal fungi and infection by
Bipolaris sorokiniana. Canadian Journal of Botany 67, 687–693.
doi:10.1139/b89-092
Thompson JP, Clewett TG, Fiske ML (2013) Field inoculation with
arbuscular-mycorrhizal fungi overcomes phosphorus and zinc
deficiencies of linseed (Linum usitatissimum) in a vertisol subject to
long-fallow disorder. Plant and Soil 371, 117–137. doi:10.1007/
s11104-013-1679-z
Timsina J, Connor DJ (2001) Productivity and management of rice–wheat
cropping systems: issues and challenges. Field Crops Research 69,
93–132. doi:10.1016/S0378-4290(00)00143-X
Trinick MJ (1977) Vesicular-arbuscular infection and soil phosphorus
utilization in Lupinus spp. New Phytologist 78, 297–304. doi:10.1111/
j.1469-8137.1977.tb04833.x
Turner EM (1961) An enzymatic basis for pathogenic specificity in
Ophiobolus graminis. Journal of Experimental Botany 12, 169–175.
doi:10.1093/jxb/12.1.169
Break crops and rotations for wheat Crop & Pasture Science 549

Vaidyanathan LV, Sylvester-Bradley R, Bloom TM, Murray AWA (1987)
Effects of previous cropping and applied nitrogen on grain nitrogen
content in winter wheat. Aspects of Applied Biology 15, 227–237.
van Dam NM, Tytgat TOG, Kirkegaard JA (2009) Root and shoot
glucosinolates: a comparison of their diversity, function and
interactions in natural and managed ecosystems. Phytochemistry
Reviews 8, 171–186. doi:10.1007/s11101-008-9101-9
Vanstone VA, Russ MH (2001) Ability of weeds to host the root lesion
nematodes Pratylenchus neglectus and P. thornei. II. Broad-leaf weeds.
Australasian Plant Pathology 30, 251–258. doi:10.1071/AP01026
Vanstone VA, Rathgen AJ, Ware AH, Wheeler RD (1998) Relationship
between root lesion nematodes (Pratylenchus neglectus and P. thornei)
and performance of wheat varieties. Australian Journal of Experimental
Agriculture 38, 181–188. doi:10.1071/EA97109
Verrell A (2014) Managing crown rot through crop sequencing and row
placement. GRDC. Available at: http://grdc.com.au/Research-and-
Development/GRDC-Update-Papers/2014/07/Managing- crown-rot-
through-crop-sequencing-and-r ow-placement
Verrell A, Simpfendorfer S, Nash P, Moore K (2005) Inter-row planting and
stubble management affect crown rot, common root rot and grain yield
in durum wheat. GRDC. Available at: http://grdc.com.au/Research-and-
Development/GRDC-Update-Papers/2005/03/Inter-row-planting-and-
stubble-management#sthash.zASkeB89.dpuf
Vyn TJ, Sutton JC, Raimbault BA (1991) Crop sequence and tillage effects on
winter wheat development and yield. Canadian Journal of Plant Science
71, 669–676. doi:10.4141/cjps91-099
Wallenhammar AC, Pettersson B (2003) Management of take all in spring
wheat by different precrops and seed treatments. In ‘Proceedings 22nd
Nordic Association of Agricultural Scientists Congress’. (MTT Agrifood
Research: Jokioinen, Finland)
Wallgren B (1987) Spring wheat, barley and oats in crop rotation and in
monoculture. Department of Plant Husbandry, Swedish University of
Agricultural Sciences, Uppsala, Sweden. [In Swedish]
Watt M, Kirkegaard JA, Rebetzke GJ (2005) A wheat genotype developed for
rapid leaf growth copes well with physical and biological constraints of
unploughed soil. Functional Plant Biology 32, 695–706. doi:10.1071/
FP05026
Whitbread AM, Davoren CW, Gupta VVSR, Llewellyn R, Roget D (2015)
Long-term cropping system studies support intensive and responsive
cropping systems in the low-rainfall Australian Mallee. Crop &
Pasture Science 66, 553–565.
White NH (1947) The etiology of take-all disease of wheat 4. The effect of
agronomic practices on the incidence and severity of take-all symptoms in
wheat. Journal. Council for Scientific and Industrial Research (Australia)
20,82– 86.
White KD (1970) Fallowing, crop rotation and crop yield in Roman times.
Agricultural History 44, 281–290.
Widdowson FV, Penny A, Gutteridge RJ, Darby RJ, Hewitt MV (1985)
Tests of amounts and times of application of aphicides and fungicides
on winter wheat, following either beans or wheat, and the effects of
take-all (Gaeumannomyces graminis var. tritici), on two varieties as
Saxmundham, Suffolk 1980-3. Journal of Agricultural Science,
Cambridge 105,97–122.
Williams CM, King JR, Ross SM, Olson MA, Hoy CF, Lopetinsky KJ (2014)
Effects of three pulse crops on subsequent barley, canola, and wheat.
Agronomy Journal 106, 343–350. doi:10.2134/agronj2013.0274
Wilson JM, Hamblin J (1990) The effect of fumigation, rotation with lupins
and nitrogen application on plant growth and common root rot of wheat
at Geraldton, Western Australia. Australian Journal of Agricultural
Research 41, 619–631. doi:10.1071/AR9900619
Wright AT (1990) Yield effects of pulses on subsequent cereal crops in the
northern prairies. Canadian Journal of Plant Science 70, 1023–1032.
doi:10.4141/cjps90-125
Yau S-K, Ryan J (2012) Does growing safflower before barley reduce
barley yields under Mediterranean conditions. Agronomy Journal 104,
1493–1500. doi:10.2134/agronj2012.0110
Zentner RP, Campbell CA (1988) First 18 years of a long-term crop rotation
study in southwestern Saskatchewan - yields, grain protein and economic
performance. Canadian Journal of Plant Science 68,1–21. doi:10.4141/
cjps88-001
Zentner RP, Spratt ED, Reisdorf H, Campbell CA (1987) Effect of crop
rotation and N and P fertilizer on yields of spring wheat grown on a Black
Chernozemic clay. Canadian Journal of Plant Science 6, 965–982.
Zillinsky FJ (1983) ‘Common diseases of small grain cereals.’ (CIMMYT:
Mexico, D.F.)
Zogg H (1969) Crop rotation and biological soil disinfection. Qualitas
Plantarum et Materiae Vegetabiles 18, 256–273. doi:10.1007/BF011
00225
550 Crop & Pasture Science J. F. Angus et al.

Appendix 1. Sources of experimental data and the crop species grown before wheat in the evaluation of the break-crop (BC) effects on
wheat yield
Abbreviations: Wh, wheat; Ba, barley; Oa, oats; Ca, canola; Mu, musta rd, Fx, flax; Pe, field peas; Ln, lentils; Ch, chickpeas; Fb, faba beans; Lu, lupins;
Fa, fal low
Reference Location Expt type Wh Ba Oa Ca Mu Fx Pe Le Ch Fb Lu Fa
Eastern Australia
Holford and Crocker 1997 Tamworth Persistence x x x
Marcellos 1984 Tamworth Single BC x x x x
Doyle et al. 1988 N NSW Single BC x x
Strong et al. 1986 Dalby Single BC x x x x x x x x
Dalal et al. 1998 Warra Rotation x x
Felton et al. 1998 N NSW Single BC x x
Herridge 2006 N NSW Single BC x x
Kirkegaard et al. 2004 N NSW Single BC x x x x x
Owen et al. 2010 Formartin Single BC x x x x x x
South-eastern Australia
White 1947 Canberra Single BC x x x
King 1984 Coonalpyn Single BC x x x
Heenan et al. 1994 Wagga Wagga Rotation
Heenan 1995 Wagga Wagga Single BC x x x x x x x
Armstrong et al. 1997 Wagga Wagga Persistence x x x x x
Reeves 1982 Rutherglen Single BC x x x x
Reeves et al. 1984 Rutherglen Persistence x x
Kollmorgen et al. 1983 Walpeup Single BC x x x x x x x
Angus et al. 1991 Barellan Single BC x x x x x
A. Green and J. Angus, unpubl. Dirnaseer Single BC x x x x
Kirkegaard et al. 1994 S NSW Single BC x x x x x
Gardner et al. 1998 S NSW Single BC x x x x
Kirkegaard et al. 1997 S NSW Persistence x x x x
Kirkegaard et al. 2000 S NSW Single BC x x x x x
S. J. Marcroft, unpubl. Minnipa Single BC x x x
T. D. Potter, unpubl. SE South Australia Single BC x x x x
J. Evans, unpubl. Wagga Wagga Single BC x x x
Evans et al. 1991 S NSW Single BC x x x x x
Evans et al. 2003 Wagga Wagga Double BC x x x
Ryan et al. 2005 Junee Reefs Single BC x x x
Harris et al. 2002 Rutherglen Double BC x x
Ryan et al. 2002 NSW and Vic. Single BC x x x x x x
Rovira 1988 South Australia Single BC x x x
Schultz 1995 Tarlee Rotation x x x x x x
Western Australia
Hamblin et al. 1993 3 locations Single BC x x
Wilson and Hamblin 1990 East Chapman Single BC x x
Delroy and Bowden 1986 Wongan Hills Single BC x x
Rowland et al. 1988 Various Single BC x x
Rowland et al. 1994 Various Single BC x x
Gregory 1998 Beverley Single BC x x x x x x x x
Asseng et al. 1998 Beverley Single BC x x x x x x
Western Europe
Olofsson and Wallgren 1984 Sweden Double BC, persistence x x x x
Lindén and Engström 2006 S Sweden Single BC x x x
Nilsson 1985 Sweden Persistence x x x x
Wallgren 1987 Sweden Rotation x x
Wallenhammar and Pettersson 2003 Örebro Single BC x x x x x
Selman 1969 Boxworth Single BC x x
Selman 1975 Boxworth Double BC, persistence x x
Christen et al. 1992 Kiel Rotation x x x
Sieling and Christen 2015 Kiel Rot., persist., double BC x x x x x
Claupein and Zoschke 1987 Geissen Rotation
McEwen et al. 1989 Rothamsted Rotation x x x x x
Prew and Dyke 1979 Rothamsted Persistence x x x
(continued next page)
Break crops and rotations for wheat Crop & Pasture Science 551

Appendix 1. (continued )
Reference Location Expt type Wh Ba Oa Ca Mu Fx Pe Le Ch Fb Lu Fa
Widdowson et al. 1985 Saxmundham Single BC
Dyke and Slope 1978 Rothamsted Single BC x x x x
Crook et al. 1999 Reading Single BC x x x
North America
Miller et al. 2002 Southwest SK Single BC x x x x x
Rothrock and Cunfer 1991 Georgia Single BC x x
Krupinsky et al. 2006 Mandan, ND Double BC, persistence x x x x x
Stevenson and van Kessel 1996 NW SK Single BC
Wright 1990 SK Single BC x x x x x
Miller et al. 2006 Montana Single BC x x x
Miller et al. 2003 Stewart Valley Single BC x x x x x
Miller and Holmes 2005 Montana Single BC x x x x x
Lafond et al. 1992 Indian Head, SK Single BC x x x x
Gan et al. 2003 SK Double BC, persistence x x x
Larney and Lindwall 1994 Lethbridge Rotation x x x
Arshad et al. 2002 Beaverlodge Persistence x x x
Moyer et al. 2005 Lethbridge Single BC x x x x x x
Zentner and Campbell 1988 Swift Current Rotation x x x
Zentner et al. 1987 Melfort, SK Rotation x x
Brandt and Zentner 1995 Scott, SK Rotation x x x
Beckie and Brandt 1997 MelfortandScott, SK Single BC x x x
West Asia
Harris 1995 Tel Hadya Rotation x x x x
Ryan et al. 2008 Tel Hadya Rotation x x x x
552 Crop & Pasture Science J. F. Angus et al.
www.publish.csiro.au/journals/cp