Journal of Experimental Botany, Vol. 54, No. 389, pp. 1931±1939, August 2003
DOI: 10.1093/jxb/erg192
RESEARCH PAPER
Volatile communication between barley plants affects
biomass allocation
Velemir Ninkovic*
Department of Entomology, Swedish University of Agricultural Sciences, Box 7044, SE-750 07 Uppsala, Sweden
Received 15 July 2002; Accepted 15 April 2003
Abstract
Introduction
Patterns of biomass allocation between different
plant organs have often been used to explain the
response of plants to variations in resource availability. This paper reports how aerial allelopathy (plant±
plant communication) affects biomass allocation, that
is the trade-off between root, stem and leaves, and
also relative growth rate (RGR, increase in biomass
per unit biomass per unit of time, mg g±1 d±1) and its
components. Based on previous experiments, communication between two barley (Hordeum vulgare L.)
cultivars (Alva and Kara) was used for the present
study. Kara exposed to volatiles from Alva allocated
signi®cantly more biomass to roots compared with
Kara exposed to volatiles from Kara or to clean air.
There was no signi®cant difference between plants of
Kara exposed to volatiles from Kara and those
exposed to clean air. Changes in total dry weight
(TDW), RGR and unit leaf rate (ULR, increase in biomass per unit time and leaf area, kg m±2 d±1) were not
signi®cantly affected by plant±plant communication.
However, there was a signi®cant increase in speci®c
leaf area (SLA, leaf area per leaf dry weight, m2 kg±1)
in Kara when exposed to volatiles from Alva. The
results show that aerial plant±plant communication
does not affect total biomass production but does
signi®cantly affect biomass allocation in individual
plants. There may be differences in the volatile pro®les of Kara and Alva that induce increased biomass
allocation to roots in the Kara plants exposed to
volatiles from Alva.
Plants live together in communities composed of one or
more species, with the possibility for allelopathic communication between individuals (Rice, 1984). This means that
the growth of plants and their organs may be affected by a
variety of compounds released from other plants into the
environment. The release of active substances can be the
result of at least four different processes: volatilization,
leaching, decomposition of plant residues in the soil, and
root exudation (Muller, 1965, 1969; Rice, 1984; Putnam
and Tang, 1986). These released metabolites can inhibit or
delay germination and also inhibit or stimulate the growth
of roots and shoots of neighbouring plants (Rice, 1984;
Mizutani, 1999). Still a subject of continued debate is how
signi®cant allelopathy is to the average competitive
interaction between plants growing in natural stands
(Harborne, 1997).
One explanation of the role of competition in plants
attributes a primary role to trade-offs between root and
shoot allocation (Grime, 2001), which makes it possible
for plants to minimize the negative effects of competition.
By allocating biomass among various plant organs, plants
modify their growth in response to environmental conditions in a way that maximizes plant growth. There are
some experiments in which alterations in the pattern of
biomass allocation between shoots and roots were induced
by exposing plants either to low mineral nutrient concentrations or to shade±light treatments (Brouwer, 1962a, b;
CorreÂ, 1983; Van der Werf et al., 1993; GlimskaÈr and
Ericsson, 1999; McConnaughay and Coleman, 1999).
According to Ryser and Lambers (1995), there must be a
trade-off in the allocation of plant biomass to either above
or below-ground organs, and there may be considerable
plasticity in this allocation pattern. The plant's ability to
modify its morphology or physiology, so-called phenotypic plasticity, is assumed to have genetic basis.
Key words: Aerial allelopathy, mass fractions, relative growth
rate, speci®c leaf area, unit leaf ratio.
* Fax: +46 18 67 28 90. E-mail: [email protected]
Journal of Experimental Botany, Vol. 54, No. 389, ã Society for Experimental Biology 2003; all rights reserved
1932 Ninkovic
An additional source of variation in a plant's ability to
allocate biomass between different organs arises from the
fact that plant exudates differ in the extent to which they
can in¯uence the trade-off in the allocation of the plant
biomass to be expressed. Most studies of allelopathic
effects have focused on early vegetative plant growth, i.e.
seed germination, a period with high metabolic activity.
But it is not known how allelopathic interaction between
plants affects their biomass allocation patterns in later
stages of vegetative growth.
Allelopathic interactions have been identi®ed in all
major temperate cereal crops, wheat (Triticum aestivum
L.), barley (Hordeum vulgare L.), oats (Avena sativa L.),
and rye (Secale cereale L.) (Lovett and Hoult, 1995). For
instance, barley has proved to be allelopathically active
against some weed species, inhibiting seed germination
and growth of selected plant species (Overland, 1966; Liu
and Lovett, 1990, 1993). In aerial allelopathic interactions,
volatiles released from leaves of Artemisia tridentata Nutt.
var. vaseyana (Weaver and Klovich, 1977) and sasa, Sasa
cernua Makino (Li et al., 1992) inhibited the growth of
barley seedlings and decreased the respiration rate of
germinating seeds. Intraspeci®c aerial allelopathy interactions in barley have been demonstrated. Volatile compounds that were released after pruning of barley leaves
induced systemic resistance against powdery mildew
fungus in intact barley seedlings (Fujiwara et al., 1987).
Another study has shown the effects of aerial allelopathy
on aphid acceptance and leaf temperature (Pettersson et al.,
1999). Signi®cant effects were obtained when certain
cultivars were combined.
The aim of the present study was to investigate whether
the aerial allelopathy shown to affect aphid settling and
leaf temperature in previous studies (Pettersson et al.,
1999; Ninkovic et al., 2002) also affects biomass allocation in roots, stems and leaves. Growth responses following exposure to volatiles were monitored during the
vegetative growth period, i.e. the plant stage during
which an allelopathy-induced effect on aphid settling
was demonstrated. Biomass allocation was measured as a
function of time, plant size and allometric relationship
among various biomass components. As a model system,
the interaction between the cultivars Alva and Kara was
used, based on their performance described in previous
studies (Pettersson et al., 1999; Ninkovic et al., 2002).
When Kara plants were exposed to volatiles emitted from
Alva plants, aphid acceptance and leaf temperature were
signi®cantly decreased when compared with Kara plants
exposed to volatiles from the same cultivar.
Materials and methods
Plant material
Two spring barley cultivars (Kara, Alva) used in previous studies
(Pettersson et al., 1999; Ninkovic et al., 2002) were chosen for these
experiments. Alva was chosen as the inducing cultivar and Kara as
the responding cultivar, these being the most inducing and most
responsive cultivar, respectively, with regard to changes in aphid
acceptance and leaf temperature. Furthermore, Kara was the most
inert to self-induction (Pettersson et al., 1999).
Exposure of plants to volatiles from other plants
Exposure of one cultivar to volatiles from another was done in a
series of `twin-chamber cages' (Pettersson et al., 1999), each of
which was considered to be a replicate. Air taken into one of these
chambers (inducing chamber, IC) passed through a hole in the wall
to the next chamber (responding chamber, RC) and was then drawn
out from the top of this cage to a vacuum tank before being vented
outside the room (Fig. 1). When plants reached the top of the cage,
another cage was placed on top of the ®rst and air was drawn out
through the upper RC. Plant roots were grown in 13 0.05 m
perforated plastic bags (each hole was 1 cm2) ®lled with clean sand.
The upper, open end of the bag was pulled through a hole (532 cm)
made in the table top, to which the bag was ®xed by a staple. The
hole was placed in the middle of the bottom of each cage. The lower
end of the bags had two drainage holes (3 mm diameter) connected
to draining-pipes. To prevent light from above the table affecting the
roots below, and to minimize the in¯ux of air into the cage from
beneath the table, a foam plastic plug was placed between the plastic
tube and the wall of the hole. To provide darkness for the roots a
black plastic curtain was ®xed around the table.
Growing conditions
The experiments were conducted in a greenhouse with a 16/8 h light/
dark cycle, a temperature of 2063 °C and no control of relative
humidity. Seeds were placed ®rst on moistened ®lter paper in Petri
dishes in a climate room (20 °C, 80% relative humidity) to
germinate. After 24 h the seeds that had started to germinate were
transferred to the cage system. One seed per bag was placed at 1 cm
depth in the sand in the plastic bags. During the experiment, a
commercial plant nutrition solution (102 mg N, 20 mg P, 86 mg K,
8 mg S, 6 mg Ca, 8 mg Mg, plus micro nutrients, l±1) (Wallco,
MoÈlnlycke, GoÈteborg, Sweden) was supplied through an automatic
watering system to give a speci®ed water and nutrient regime. The
pH of the nutrient solution was adjusted automatically to pH 6.5±7
with a pH-regulator that was connected to the water pump (DGT
Volmatic).
Experimental treatments
The phenotypic expression throughout the vegetative growth phase
of a barley cultivar was registered when exposed to different odour
sources. The observations started 2 weeks after germination and
ended when plants were at the end of the four and beginning of the
®ve-leaf stage, that is during the ®rst part of the vegetative stage.
This is the period during which the newly-differentiated leaves are
able to contribute substantially to carbon assimilation (Hunt, 1987),
and is characterized by an exponential increase in total dry weight.
The Kara plants were exposed to three different treatments: air
from Alva, air from Kara and clean air. To avoid pseudoreplication,
each treatment was represented by 30 separate `twin-chamber cages'
in each of which a single Kara plant was exposed to a treatment: a
single Alva plant (Fig. 1A), a single Kara plant (Fig. 1B) or clean air
(Fig. 1C). The twin-cages were randomly allocated to positions on
the table in the greenhouse at the beginning of the experiment to
compensate for any spatial variation in irradiance in the greenhouse.
The growth of plants was followed for 34 d, from the one leaf stage
to the ®ve leaf stage. During this period, ®ve destructive plant
samplings were made, with 3±5 d between each sampling. At each
sampling time, ®ve plants were chosen per treatment from RC
(responding chamber) and separated into roots, stems and leaves.
Volatile communication between barley plants 1933
Fig. 1. System of twin-chamber cages for exposure of one plant to volatiles from another. (A) The responding plant B was exposed to volatiles
emitted from inducing plant A. (B) Plant B was exposed to volatiles from the same cultivar. (C) Plant B was not exposed to plant volatiles.
The ®rst sampling was done 15 d after sowing. The exposed plant in
the responding cage represented one replicate which means that each
experimental treatment consisted of 30 replicates (plants), that is the
whole experiment was carried out in 90 separate cages. Directly after
separation, the green leaf area of individual plants was measured
using a Li-3100 (Li-Cor, Lincoln, Nebraska, USA) leaf area meter.
Roots were washed carefully with water. After drying for 48 h at
70 °C to constant mass dry weights, plant parts were weighed. These
data were used for calculation of integral morphological indices,
namely the leaf mass fraction (LMF), the root mass fraction (RMF)
and the stem mass fraction (SMF). Total dry weight (TDW) was the
sum of these plant parts.
1934 Ninkovic
lines was tested by ANOVA as the interaction between two tested
factors (Littell et al., 1991). For post-ANOVA comparisons of
signi®cant differences in distance between lines of target allometric
parameters, Tukey's tests were used. The GLM procedure of the
SAS Institute Inc. (2000) package was used for the numerical
evaluations.
Other morphological parameters and growth
Differences between treatments in relative growth rate (RGR,
increase in biomass per unit biomass per unit of time, mg g±1 d±1)
and unit leaf rate (ULR, increase in biomass per unit time and leaf
area, kg m±2 d±1) were tested as an interaction effect in a two-way
ANOVA with treatment and time as the factors. The dependent
variable was ln-transformed dry weight (Poorter and Lewis, 1986).
The series of RGRs and ULRs were calculated by pairing plants: the
largest plant at time tn with the largest plant at time tn+1; and second
largest at time tn with the largest at time tn+1, and so on (Hunt, 1987).
However, RGR can also be applied as a model that relates a plant's
relative growth rate to the morphological and physiological factors
(Lambers et al., 1989). According to Poorter and Nagel (2000), RGR
can be factorized into two components. A `morphological
component' that can be further subdivided into speci®c leaf area
(SLA, leaf area per leaf dry weight, m2 kg±1) that re¯ects aspects of
leaf morphology, and the LMF that re¯ects biomass allocation to
leaves. A `physiological component' is ULR, which is generally
strongly correlated with the rate of photosynthesis per unit leaf area.
RGR=ULR3SLA3LMF
Fig. 2. The effects of plant volatiles on biomass allocation between
different plant organs when allocation is characterized as the fraction
of total biomass allocated to leaves (A) and roots (B). Fitted lines of
exposed plants of Kara to different volatiles sources: to volatiles from
plants of different cultivar (Alva) (®lled circles and solid line), to
volatiles from plants of the same cultivar (Kara) (open circles and
dashed line) and to plain air (open triangles and dotted line).
Calculations and statistical analyses
Data were transformed to their natural logarithms (ln). Transformed
data and their residuals were determined to be normally distributed
and homoscedastic using histograms and plots. In the cases where
LMF, SMF and RMF were plotted against TDW, normality and
homoscedasticity were achieved without data transformation.
Patterns of biomass allocation were tested in two different ways.
First, an allometric model expressed in two different ways was used
for analysis of the allometric relationship between shoot and root
(Parsall, 1927; Troughton, 1960) and shoot:root ratio (S:R) over a
substantial period of time. This is an entirely empirical method and
does not attempt to explain growth processes (Poorter and Nagel,
2000; Farrar and Gunn, 1998). Secondly, the LMF, SMF and RMF
were used for analysis of the biomass allocation. According to
Poorter and Nagel (2000), the use of biomass fractions has two
advantages: biomass fractions are less sensitive to small changes in
allocation than are S:R values (shoot:root), and also RMF and LMF
are integral parts of growth analyses.
The linearity of the allometric relationships examined here was
used for comparisons of (1) RMF versus TDW, (2) LMF versus TDW,
(3) SMF versus TDW, (4) ln total shoot dry biomass versus ln root
dry biomass ratio, and (5) S:R versus time. Linearity for each of the
three treatments (30 points per curve) along the gradient was studied
using an ANOVA goodness-of-®t test (Draper and Smith, 1966).
Assuming linearity for each treatment, the hypothesis of parallel
Differences in SLA and TDW between treatments were tested by a
two-way ANOVA with treatment and time as factors. When F-tests
were signi®cant, pairwise comparisons of means of target plant
parameters were performed using Tukey mean signi®cant difference
tests.
Results
Biomass allocation
The approach to analyse biomass allocation consisted of
two calculation steps. The ®rst was to examine the
linearity, which was a precondition to test parallelism of
the allometric relationships for these comparisons: (1)
RMF versus TDW, (2) LMF versus TDW, (3) SMF versus
TDW, (4) ln total shoot dry biomass versus ln root dry
biomass ratio, and (5) S:R versus time. The null hypothesis
was that these lines were parallel. The obtained P-values
for the F-test were in the interval 0.50±0.81. Hence, the
assumption of parallelism was not rejected.
The second step was comparisons of signi®cant differences (intercept) in the distance between the lines of the
target allometric parameters. Exposure of Kara plants to
different sources of plant volatiles had signi®cant effects
on partitioning of biomass between different plant organs.
When RMF was plotted against TDW a highly signi®cant
difference was found between treatments (F2,86=14.9; P
<0.0001). Plants exposed to volatiles from Alva showed a
signi®cant increase in root allocation when compared with
plants exposed to volatiles from the same cultivar (selfresponding) (P=0.001) and those exposed to clean air (P
<0.0001) (Fig. 2B) using a Tukey test. No signi®cant
difference in RMF was found between plants exposed to
Volatile communication between barley plants 1935
Fig. 4. The effect of aerial allelopathy between plants on total dry
weight. Symbols are as in Fig. 3B. Mean TDW for the ®ve plants
sampled at each harvest time. Error bars indicate s.e.m.
Fig. 3. The effect of plant volatiles on the allometric relationship
between shoot and root (A) and S:R ratio (B). (A) Symbols and line
®tting are as in Fig. 2. (B) Exposed plants of Kara to volatiles of Alva
plants (AK), volatiles of Kara plants (KK), or to plain air (OK). Mean
S:R ratio for the ®ve plants sampled at each harvest time. Error bars
indicate s.e.m.
volatiles from the same cultivar and those exposed to clean
air (Fig. 2B). LMF plotted against TDW showed the same
trend as RMF (F2,86=15.99; P <0.0001), but in the opposite
direction (Fig. 1A). Kara plants exposed to volatiles from
Alva showed a signi®cant reduction in LMF compared
with those exposed to Kara volatiles (P=0.0001) and those
exposed to clean air (P <0.0001), while the change in LMF
did not differ between the Kara±Kara combination and
plants exposed to clean air (Fig. 2A). SMF, plotted against
TDW, did not differ signi®cantly among treatments
(F2,86=1.09; P=0.34).
Allometric analyses of shoot:root balance (S:R) revealed
that Kara plants altered patterns in response to exposure to
volatiles from Alva, but not in response to volatiles from
the same cultivar (Fig. 3A). Even though all three
regression lines between root and shoot weights had
similar slopes, that is `allometric constant' k (Fig. 3A),
testing the parallelism by ANOVA revealed signi®cant
differences between the intercepts of these lines
(F2,86=11.69; P <0.0001). Comparisons of distance
between lines with a Tukey test showed that Kara plants
exposed to volatiles from Alva were signi®cantly different
Fig. 5. The effect of aerial allelopathy between plants on speci®c leaf
area. Symbols and line ®tting are as in Fig. 2. Mean SLA for the ®ve
plants sampled at each harvest time. Error bars indicate s.e.m.
from lines of self-responding plants (P=0.003) and plants
exposed to clean air (P <0.0001) which clearly demonstrates the difference in allocation in the plants exposed to
volatiles from Alva, compared to self-exposed plants and
those exposed to clean air (Fig. 3A). When the S:R was
compared for plants of the same age, a signi®cant
difference in biomass allocation between treatments was
found (F2,86=20.27; P <0.0001) (Fig. 3B). The combination Alva/Kara showed a signi®cantly lower S:R in
comparison with self-responding (P=0.006) and with nonexposed plants (P <0.0001) (Fig. 3B).
Other morphological parameters and growth
The TDW differed slightly among treatments (F2,72=3.42;
P=0.04). However, post-ANOVA comparisons of mean
TDW between treatments throughout the experimental
period did not reveal any signi®cant differences (Fig. 4).
The Treatment3Time interaction was used to look for
differences in RGR and ULR. The interaction effects of
exposing plants to different sources of volatiles were not
1936 Ninkovic
signi®cant for RGR (Treatment3Time F8,60=5.6; P=0.5)
and ULR (F8,60=0.57; P=0.8). This indicated that exposure
to the different volatile sources did not in¯uence RGR, and
that ULR, the `physiological component' of RGR, was
unaffected.
There were signi®cantly different changes in the second
morphological component, SLA (F2,72=3.68; P=0.03)
(Fig. 5). SLA was signi®cantly increased when plants
were exposed to volatiles from Alva when compared with
those exposed to volatiles from Kara (P=0.025), but were
not signi®cantly different from those exposed to clean air.
No signi®cant difference was found between plants
exposed to volatiles from the same cultivar (Kara) and
plants exposed to clean air.
Discussion
Aerial allelopathy and biomass allocation
The results show that exposure to volatiles from plants of
one barley cultivar, Alva, induces changes in the pattern of
biomass allocation in plants of a second cultivar, Kara.
Plants exposed to volatiles from Alva allocated more
biomass to roots compared with plants exposed to volatiles
from Kara, or to clean air (Fig. 2B). However, the TDW did
not differ between those treatments (Fig. 4), that is the
principal effect shown in the experiment is that this type of
plant interaction affects the allocation of biomass but not
the total biomass. It has been reported that plant individuals sharing rhizosphere space with another plant produce
more root mass than when one plant `owns' that space
(Gersani et al., 2001). However, in the present experiment
root growth was also stimulated when communication
between the two barley plant individuals was limited to an
exchange of air/volatiles. This means that the allelopathic
response had an effect on the whole plant, rather than on
speci®c parts such as roots, leaves etc.
This ®nding raises the question of whether a plant's
ability to respond to volatiles from a neighbouring plant by
changes in biomass allocation is of importance in
optimizing the intake of the plant's requirements from
the environment. It may be that biomass allocation is a
result of plant plasticity released by volatiles from another
plant. Theory predicts that plant allocation to organs that
are close to limiting resources will favour the growth of
these organs (Tilman, 1988). The investment in roots is
usually correlated with nutrient supply and/or demand
Ê gren, 1991), while the effect on allocation
(Ingestad and A
pattern of a limited water supply is less pronounced
(Poorter and Nagel, 2000). However, when nutrients, light
and water are equally abundant and root communication
between plants is prevented, as was the case in the current
study, differences in allocation to root biomass should
depend upon other factors, such as plant±plant communication.
Relative growth rate (RGR) and its components
Several plants are capable of considerable modi®cations of
their root:shoot ratio. It has been suggested that there are
effective mechanisms that modify allocation between roots
and shoot, thus indirectly regulating the balance between
photosynthesis and nutrient capture (Grime, 2001). Villar
et al. (1998) showed that the variation in RGR in Aegilops
and Amblyopyrum spp. (Poaceae) is associated mainly with
variation in biomass allocation and not with variation in
SLA. However, the results of the current study showed that
RGR was unchanged, even when allocation was higher to
roots and lower to leaves for plants exposed to volatiles
from Alva (Fig. 2A, B). The present results are in line with
a study where the allocation pattern of 10 wheat cultivars
was tested at two soil moisture levels in a growth chamber
(Van den Boogaard et al., 1997). Thus, the variation in
biomass allocation cannot always be used as prediction of
a change in RGR or vice versa.
According to studies of plant plasticity, SLA in plants
may show plasticity to variation in the light intensity.
There are species differences in the effect of lightdependent changes in SLA (Veneklaas and Poorter,
1998). Hence, a low allocation of biomass to the leaves
can be compensated for by a high SLA (Aerts et al., 1991;
Boot and Den Dubbelden, 1990). Such a compensatory
effect was also found in this study where the lower
allocation of biomass to the leaves of Kara (Fig. 2A), when
it was exposed to volatiles of Alva, is compensated for by
higher SLA (Fig. 5). Similarly, the compensation of lower
biomass allocation to the leaves by higher SLA was found
by Pegtel (1976) comparing two ecotypes of Sonchus
arvensis. Differences in SLA can probably be caused by
increases in water content and cell size (Andersen et al.,
1993) and also by altered thickness of leaves and vein
structure (Dijkstra, 1990). Apart from these morphological
factors, leaf physiology can also be changed, which is
re¯ected in the altered chemical composition of leaf
biomass (Dijkstra, 1990). According to Van der Werf
(1996), fast-growing plants bene®t from high SLA only if
leaves increase photosynthetic activity at the highest level
under certain environmental conditions. A possible reason
why plants exposed to volatiles from another cultivar have
a higher SLA is to enable them to maintain the same RGR.
A study of aerial allelopathy with combinations of four
barley cultivars showed that, in certain combinations, the
leaf temperatures of some cultivars (e.g. Kara) were
lowered when exposed to volatiles from certain other
cultivars (Pettersson et al., 1999). The leaf temperature
was measured by an infrared camera in the same two-cage
system. This effect links plant±plant communication with
the increased transpiration rate of the receiving plants. A
high transpiration rate is generally assumed to correlate
with greater conductance and the enhanced in¯ux of CO2.
High in¯ux of CO2 is required to maintain the higher
Volatile communication between barley plants 1937
photosynthetic activity, that is water use ef®ciency (WUE)
expressed as the mass of carbon assimilated per mass of
water transpired is reduced. The consequent need for
increased water extraction may require greater allocation
to roots. The lower leaf temperature and greater biomass
allocation to roots are also in line with the assumption of a
strong positive correlation between transpiration and root
allocation (Sims and Pearcy, 1994). By contrast, the
transpiration rate is lower when root weight per unit total
plant weight is larger in environments where the amount of
water is a limiting factor for plant growth (Van den
Boogaard et al., 1997). However, under environmental
conditions with no limiting factors, for example, nutrients,
light and water, and where root communication between
plants is prevented, differences in transpiration rate and
allocation pattern should depend on plant±plant communication.
Ecological implications
The current results add a new dimension to plant
coexistence, showing that aerial allelopathy can cause
changes in the biomass allocation in neighbouring plants.
To what extent this gives a competitive advantage to the
emitting (Alva) or responding/listening (Kara) plant
remains to be shown with further experimentation. Under
competitive growing conditions, increased root biomass
may contribute to plant ®tness by increasing the capacity
for below-ground competition, that is for water and
mineral uptake, which would favour the Kara-type
response. However, reducing green leaf biomass may
decrease the competitive capacity for light (Didon, 2002)
of Kara.
It has been hypothesized that the increased ability of a
plant to maximize growth over a range of different
available resources may be the result of morphological
plasticity (Tilman, 1988). It can be concluded that Kara,
under the current experimental conditions, shows this
capacity. However, to what extent this is a commonly
occurring capacity in other barley genotypes and in grasses
in general remains to be tested. The importance of
environmental conditions for the expression of this allocation capacity is more dif®cult to predict. GlimskaÈr and
Ericsson (1999) postulated that biomass allocation between root and shoot can be treated as a direct function of
the relative resource limitation of the plant. This raises the
question of how plant±plant interaction affects biomass
allocation patterns when plants grow under the optimum
resource availability and also how this reaction depends on
the neighbouring genotype/species. The competition for
resources water/nutrients (below ground) and light (above
ground) increases under conditions of higher nutrition
availability (Campbell et al., 1991). In response to
volatiles from Alva, Kara plants allocated more biomass
to the roots than when they were exposed to volatiles of the
same cultivar (Fig. 2B). The response of Kara seedlings to
the presence of Alva seedlings was increased biomass
allocation to roots, and indicates that some sort of coadaptation to the presence of the other cultivar may occur.
This can be expected to have further implications for
adaptive advantage under different environmental conditions.
The present experiments show that intraspeci®c plant±
plant communication via aerial allelopathy may occur. The
experiments do not give a basis for speculation about the
character of the active chemical compounds. However, it is
not obvious that only unique chemical compounds are to
be expected. It is important not to overlook the possibility
that trivial plant compounds or even the ratio between
some of those might be the active operating agent.
To what extent the plant capacities demonstrated in this
and previous investigations of aerial allelopathy, to
respond and to promote responses, have developed during
the breeding processes or whether they have survived in
spite of this, is an interesting but open question. However,
similar ®ndings in other plant species indicate that it
is a primitive plant ecological trait (Callaway, 2002).
Primarily, the experimental results contribute to an understanding of the ecology of plant individuals in different
types of plant communities. Obviously these mechanisms,
although less recognized, have been operating during all
efforts that have been made to improve production
systems, and it is not obvious if the effects under ®eld
conditions are of suf®cient magnitude to have agronomic
value.
Conclusions
In barley, volatiles from plants of cultivar (Alva) induced
changes in the pattern of biomass allocation in plants of a
second cultivar (Kara). Exposed plants allocated more
biomass to roots. The allelopathic effect on the pattern of
biomass allocation is active only if there are differences in
the volatile pattern of inducing cultivars, which may
consist of unique substances or unique ratios of volatile
compounds.
Analysis of RGR and its components was used to explain
likely morphological and physiological changes that can
be induced by plant±plant communication. RGR is
unchanged when a plant of one barley cultivar is exposed
to volatiles from a plant of a second barley cultivar. Even if
ULR, the physiological component of RGR, was unchanged it cannot be concluded that plant±plant communication is unable to affect other physiological processes in
the exposed plant. Both morphological components (LMF
and SLA) of plants exposed to volatiles from plants of
another cultivar were signi®cantly changed but in opposite
directions. This may explain why there was no change in
RGR.
1938 Ninkovic
Acknowledgements
For valuable statistical support I am indebted to Ulf Olsson, Henrik
Poorter and Lenart Norell. I am grateful to Bengt Torsell, Tom
Ê hman, Robert Glinwood, Richard Hopkins, and Jan
Ericsson, Inger A
Pettersson for providing helpful comments and linguistic corrections. Elham Ahmed is thanked for skilled technical assistance. The
study was made possible by grants from the Swedish Farmers'
Foundation for Agricultural Research, the Strategic Fund for
Environmental Research (MISTRA) and the Foundation of Oscar
and Lili Lamms Memory.
References
Aerts R, Boot RGA, Van der Aart PJM. 1991. The relation
between above- and below-ground biomass allocation patterns
and competitive ability. Oecologia 87, 551±559.
Andersen MN, Jensen CR, Losch R. 1993. The interaction effects
of potassium and drought in ®eld-grown barley. II. Nutrient
relation, tissue water content and morphological development.
Acta Agriculturae Scandinavica. Section B, Soil and Plant
Science 42, 45±56.
Boot RGA, Den-Dubbelden K. 1990. Effects of nitrogen supply on
growth allocation and gas exchange characteristics of two
perennial grasses from inland dunes. Oecologia 85, 115±121.
Brouwer R. 1962a. Distribution of dry matter in the plant.
Netherlands Journal of Agricultural Science 10, 361±376.
Brouwer R. 1962b. Nutrient in¯uence on the distribution of dry
matter in the plant. Netherlands Journal of Agricultural Science
10, 399±408.
Callaway RM. 2002. The detection of neighbors by plants. Trends
in Ecology and Evolution 17, 104±105.
Campbell BD, Grime JP, Mackey JML, Jalili A. 1991. The quest
for a mechanistic understanding of resource competition in plant
communities: the role of experiments. Functional Ecology 5,
241±253.
Corre WJ. 1983. Growth and morphogenesis of sun and shade
plants. I. The in¯uence of light intensity. Acta Botanica
Neelandica 32, 49±62.
Didon UME. 2002. Growth and development of barley cultivars in
relation to weed competition. PhD thesis, Swedish University of
Agriculture Science, Sweden, 7±22.
Dijkstra P. 1990. Cause and effect of differences in speci®c leaf
area. In: Lambers H, Cambridge ML, Konings H, Pons TL, eds.
Causes and consequences of variation in growth rate and
productivity of higher plants. The Hague, The Netherlands:
SPB Academic Publishing, 125±140.
Draper NR, Smith H. 1966. Applied regression analysis. New
York, USA: John Wiley and Sons, Inc.
Farrar J, Gunn S. 1998. Allocation: allometry, acclimationÐand
alchemy? In: Lambers H, Poorter H, Van Vuuren MMI, eds.
Inherent variation in plant growth. Physiological mechanisms
and ecological consequences, Leiden, The Netherlands:
Backhuys Publishers, 183±198.
Fujiwara M, Oku H, Shiraishi T. 1987. Involvement of volatile
substances in systemic resistance of barley against Erysiphe
graminis f. sp. hordei induced by pruning of leaves. Journal of
Phytopathology 120, 81±84.
Gersani M, Brown JS, O'Brien EE, Maina GM, Abramsky Z.
2001. Tragedy of the commons as a result of root competition.
Journal of Ecology 89, 660±669.
GlimskaÈr A, Ericsson T. 1999. Relative nitrogen limitation at
steady-state nutrition as a determinant of plasticity in ®ve
grassland plant species. Annals of Botany 84, 413±420.
Grime JP. 2001. Plant strategies, vegetation processes, and
ecosystem properties. Chichester: John Wiley and Sons, Ltd, 3±
115.
Harborne JB. 1997. Plant secondary metabolites. In: Crawley MJ,
ed. Plant ecology, Vol. V. London: Blackwell Science,
132±155.
Hunt R. 1987. Plant growth analysis. The Institute of Biology's
studies in biology, no. 96. London: Edward Arnold, 1±5.
Ê gren GI. 1991. The in¯uence of plant nutrition on
Ingestad T, A
biomass allocation. Ecological Applications 1, 168±174.
Lambers H, Cambridge ML, Konings H, Pons TL. 1989. Causes
and consequences of variation in growth rate and productivity of
higher plants. The Hague: SPB Academic Publishing.
Li HH, Nishimura H, Hasegawa K, Mizutani J. 1992.
Allelopathy of Sasa cernua. Journal of Chemical Ecology 18,
1785±1796.
Littell RC, Freund RJ, Spector PC. 1991. SAS/system for linear
models, 3rd edn. North Carolina, USA: SAS Cary, Institute Inc,
243±246.
Liu DL, Lovett JV. 1990. Allelopathy in barley: potential for
biological suppression of weeds. In: Basset C, Whitehouse LJ,
Zabkiewicz JA, eds. Alternatives to the chemical control of
weeds. Proceedings of an International Conference, Rotorua, New
Zealand, July 1989. FRI Bulletin 155, 85±92.
Liu DL, Lovett JV. 1993. Biologically active secondary
metabolites of barley. I. Developing techniques and assessing
allelopathy in barley. Journal of Chemical Ecology 19, 2217±
2230.
Lovett JV, Hoult AHC. 1995. Allelopathy and self-defense in
barley. In: Inderjit, Dakshini, KMM, Einhellig FA, eds.
Allelopathy: organisms, processes, and applications, ACS
Symposium Series 582, Washington, DC: American Chemical
Society, 170±183.
McConnaughay KDM, Coleman JS. 1999. Biomass allocation in
plants: ontogeny or optimality? A test along three resource
gradients. Ecology 80, 2581±2593.
Mizutani J. 1999. Selected allelochemicals. Critical Reviews in
Plant Sciences 18, 653±671.
Muller CH. 1965. Inhibitory terpenes volatilized from Salvia
shrubs. Bulletin of the Torrey Botanical Club 92, 38±45.
Muller CH. 1969. Allelopathy as factor in ecological process.
Vegetation 18, 348±357.
Ninkovic V, Olsson U, Pettersson J. 2002. Mixed barley cultivars
affects aphid host plant acceptance in ®eld experiments.
Entomologia Experimentalis et Applicata 102, 177±182.
Overland L. 1966. The role of allelopathic substances in the
smother crop barley. American Journal Botany 53, 423±432.
Parsall WH. 1927. Growth studies. VI. On the relative size of
growing plant organs. Annals of Botany 41, 549±556.
Pegtel BM. 1976. On the ecology of two varieties of Sonchus
arvenisis L. PhD thesis, University of Groningen, The
Netherlands, 61±90.
Pettersson J, Ninkovic V, Ahmed FE. 1999. Volatiles from
different barley cultivars affect aphid acceptance of neighbouring
plants. Acta Agriculturae Scandinavica. Section B, Soil and Plant
Science 49, 152±157.
Poorter H, Lewis C. 1986. Testing differences in relative rate: a
method avoiding curve ®tting and paring. Physiologia Plantarum
67, 223±226.
Poorter H, Nagel O. 2000. The role of biomass allocation in the
growth response of plants different levels of light, CO2, nutrients
and water: a quantitative review. Australian Journal of Plant
Physiology 27, 595±607.
Putnam AR, Tang CS. 1986. The science of allelopathy. New
York: John Wiley & Sons.
Rice EL. 1984. Allelopathy. Orlando, Florida: Academic press,
Inc.
Volatile communication between barley plants 1939
Ryser P, Lambers H. 1995. Root and leaf attributes accounting for
the performance of fast and slow-growing grasses at different
nutrient supply. Plant and Soil 170, 251±265.
SAS Institute Inc. 2000. SAS/stats user's guide, Version 8. North
Carolina, USA: SAS Cary, Institute Inc.
Sims DA, Pearcy RW. 1994. Scaling sun and shade photosynthetic
acclimation of Alocasia macrorrhiza to whole-plant performance.
I. Carbon balance and allocation at different daily photon ¯ux
densities. Plant, Cell and Environment 17, 881±887.
Tilman D. 1988. Plant strategies and the dynamics and structure of
plant communities, Vol. IV. Princeton: Princeton University
Press, 98±135.
Troughton A. 1960. Further studies on the relationship between
shoot and root system of grasses. Journal of the British Grassland
Society 15, 41±47.
Van den Boogaard R, Alewijnse D, Veneklaas EJ, Lambers H.
1997. Growth and water-use ef®ciency of 10 Triticum aestivum
cultivars at different water availability in relation to allocation of
bimass. Plant, Cell and Environment 20, 200±210.
Van der Werf A. 1996. Growth analysis and photoassimilate
partitioning. In: Zamski E, Schaffer AA, eds. Photoassimilate
distribution in plants and crops: source-sink relationships. New
York: Marcel Dekker Inc, 1±20.
Van der Werf A, Vissar AJ, Schieving F, Lambers H. 1993.
Evidence for optimal partitioning of biomass and nitrogen at a
range of nitrogen availabilities for a fast- and slow-growing
species. Functional Ecology 7, 63±74.
Veneklaas EJ, Poorter L. 1998. Growth and carbon partitioning of
tropical tree seedlings in contrasting light environments. In:
Lambers H, Poorter H, Van Vuuren MMI, eds. Inherent variation
in plant growth. Physiological mechanisms and ecological
consequences. Leiden: Backhuys Publishers, 337±361.
Villar R, Veneklaas EJ, Jordano P, Lambers H, 1998. Relative
growth rate and biomass allocatio in 20 Aegilops (Poaceae)
species. New Phytologist 140, 425±437.
Weaver TW, Klovich D. 1977. Allelopathic effects of volatile
substances from Artemisia tridentata Nutt. American Midland
Naturalist 97, 508±512.