Ecology Letters, (2008) 11: 727–739 doi: 10.1111/j.1461-0248.2008.01183.

REVIEW AND
SYNTHESIS Plant behaviour and communication

Abstract
Richard Karban* Plant behaviours are defined as rapid morphological or physiological responses to
Department of Entomology, events, relative to the lifetime of an individual. Since Darwin, biologists have been aware
University of California, Davis, that plants behave but it has been an underappreciated phenomenon. The best studied
CA 95619, USA plant behaviours involve foraging for light, nutrients, and water by placing organs where
*Correspondence: E-mail:
they can most efficiently harvest these resources. Plants also adjust many reproductive
[email protected]
and defensive traits in response to environmental heterogeneity in space and time. Many
plant behaviours rely on iterative active meristems that allow plants to rapidly transform
into many different forms. Because of this modular construction, many plant responses
are localized although the degree of integration within whole plants is not well
understood. Plant behaviours have been characterized as simpler than those of animals.
Recent findings challenge this notion by revealing high levels of sophistication previously
thought to be within the sole domain of animal behaviour. Plants anticipate future
conditions by accurately perceiving and responding to reliable environmental cues.
Plants exhibit memory, altering their behaviours depending upon their previous
experiences or the experiences of their parents. Plants communicate with other plants,
herbivores and mutualists. They emit cues that cause predictable reactions in other
organisms and respond to such cues themselves. Plants exhibit many of the same
behaviours as animals even though they lack central nervous systems. Both plants and
animals have faced spatially and temporally heterogeneous environments and both have
evolved plastic response systems.

Keywords
Conditioning, environmental heterogeneity, foraging, integration, movement, phenotypic
plasticity.

Ecology Letters (2008) 11: 727–739

(Silvertown & Gordon 1989). Behaviour ultimately has a

INTRODUCTION
physiological basis, which is mediated by chemical reactions.
The idea that plants exhibit complicated behaviours in However, response to a stimulus, relative speed, and non-
response to environmental stimuli is not new. Charles permanence distinguish behaviour from other physiological
Darwin (1880) published a comprehensive description of and chemical reactions. Behaviour does not include onto-
the widespread movements of plant tissues excited by light, genetic changes that are programmed to proceed during the
gravity and contact. Since then the repertoire of behaviours course of development, such as the changes that necessarily
that have been catalogued has increased greatly, and plants occur as a seed germinates and transforms into a seedling
occasionally get brief mention in modern textbooks about (Silvertown 1998). In animals, behaviour usually refers to
behaviour (e.g. Krebs & Davies 1997). Plant behaviour has movements generated by muscles, typically the result of
been defined as a response to an event or environmental nervous action although this restrictive definition precludes
change during the course of the lifetime of an individual a consideration of similar phenomena in plants. Silvertown
(Silvertown & Gordon 1989; Silvertown 1998). and GordonÕs (1989) definition of plant behaviour may be
This definition is similar to commonly used descriptions confusing to some animal behaviourists and perhaps that is
of phenotypic plasticity in plants (Bradshaw 1965). Behav- one reason that plant behaviour is still an uncommon term.
iour is a form of phenotypic plasticity in response to a My aim in using the term in this review is to draw attention
stimulus that is relatively rapid and potentially reversible to the fact that plants have many complicated responses that

 2008 Blackwell Publishing Ltd/CNRS

728 R. Karban Review and Synthesis

were unappreciated until recently. In addition, as a more small herbivores and expose larger ones to protective
mature field, animal behaviour has been successful in ways thorns. Other tropical legumes lower their leaves in
that plant biologists can emulate. response to heavy rain, but not light rain or alighting
Plant behaviour has recently been the subject of intense insects, and this response accelerates leaf surface drying
research interest and several reviews (e.g. Novoplansky (Dean & Smith 1978). Carnivorous plants rapidly move to
2002; de Kroon & Mommer 2005; de Kroon et al. 2005; catch insects that stimulate trigger hairs, setting off a series
Trewavas 2005). These more narrowly focused reviews of changes in expansion of cells that ultimately results in a
highlighted the ways in which plants forage for resources. meal (Darwin 1893; Braam 2005). This behaviour allows
Plants place leaves and roots non-randomly within their carnivorous plants to thrive in resource-poor environments.
heterogeneous environments and this placement allows Stamens and stigmas of many plant species move in
them to actively modify their acquisition of essential response to insect visitation which increases the likelihood
nutrients, water and light. The growth or abscission of of outcrossing (Braam 2005). The mechanisms responsible
organs such as leaves or roots is not strictly reversible for these rapid plant movements are varied and include
although a plant may reverse its commitment to invest in changes in turgor, osmotic changes in ionic concentrations,
one direction, recover some of its investment, and redirect action potentials and electrical signals instead of the actin-
its growth elsewhere. Resource foraging is probably the best myosin system common in animals.
studied plant behaviour; for example, we know something In addition to these very rapid movements, Darwin
about the physiological mechanisms employed by plants (1880) argued that all plant organs undergo subtle move-
foraging for light (e.g. Pearcy & Sims 1994; Ballare 1999; ments around their axes of elongation which he called
Smith 2000) as well as its evolutionary consequences circumnutation. By modification of this phenomenon, many
(Schmitt et al. 1999). Without broadening SilvertownÕs different plant tissues have developed conspicuous and
definition of plant behaviour, information about many directed movements in response to light, gravity and other
other types of plant behaviours, in addition to foraging, have environmental stimuli. These movements allow plants to be
been elaborated recently and will be described in the next more efficient at capturing resources such as light compared
section of this review. with individuals that are prevented from moving.
For at least a century, plant biologists have observed that
roots become more abundant in soil that contains higher
PLANT RESPONSES TO ENVIRONMENTAL
levels of nutrients compared with soil with lower levels
HETEROGENEITY AND CHANGE
(Weaver 1926). This prompted the hypothesis that roots
Environments vary over space and over time and plants grew selectively in favourable patches in order to increase
respond to this variation by adjusting their phenotypes to resource acquisition. Experiments mapping the growth of
match current conditions. This statement implicitly assumes barley roots in soil compartments that contained different
that all responses are adaptive. We know that this assump- nutrient levels confirmed this hypothesis (Drew et al. 1973).
tion is sometimes violated and that some changes produce Those parts of a single root that contacted soil containing
mismatches with the environment. However, the assump- high nitrate concentrations grew many more lateral roots
tion of adaptation has been used successfully by behavioural relative to those root parts that contacted soil with less
ecologists in many situations. Phenotypic adjustments may nitrate.
take the form of plant movement, physiological acclimation Morphological plasticity also allows plants to forage
and change, growth of new tissue, or shedding of existing efficiently for light. Vertical shoots elongate less and branch
tissue. These categories are not mutually exclusive, as for more when they are exposed to favourable light conditions
instance, all phenotypic plasticity has a physiological basis. compared with shoots with reduced light. Light transmitted
The range of plant responses is fairly large in terms of the through leaves has a lower ratio of red : far red than
behaviours that they exhibit, the conditions that they unfiltered light (Smith 2000). Plants sense light availability
respond to, and their ecological and evolutionary conse- using phytochrome photoreceptors that detect the ratio of
quences (Table 1). red : far red radiation. Individuals that are shaded by
neighbours undergo a reprogramming of their morpholog-
ical development causing them to grow taller and in the
Plant movement and foraging
direction of canopy gaps (Ballare 1999). More and larger
Some of the most impressive plant behaviours involve rapid buds develop on branches in sunny patches than on
movements in response to physical stimuli. Leaflets of branches in shady patches, resulting in crown asymmetry.
Ôsensitive legumesÕ rapidly fold up if disturbed by insects and This growth pattern allows greater capture of light (Schmitt
neighbouring leaves fold up, as well, following wounding et al. 1999). This often translates into fitness benefits
(Eisner 1981; Braam 2005). This behaviour may scare away accruing to the plants that respond to light cues although

 2008 Blackwell Publishing Ltd/CNRS

Review and Synthesis Plant behaviour and communication 729

Table 1 Plant behaviours, their causes and consequences

Tissue
Behaviour Stimulus responding Consequences Reference

Foraging
Movement Contact, Many Improved Darwin (1880), Braam (2005)
light, gravity resource
acquisition
Root Nutrients, water Root Improved Hutchings and de Kroon (1994)
growth ⁄ shedding meristems resource
acquisition
Shoot R ⁄ FR light Shoot Improved Smith (2000)
growth ⁄ shedding meristems resource
acquisition
Enzyme deployment, Light Leaves Increased Pearcy and Sims (1994)
gas exchange photosynthesis,
reduced
photoinhibition
Parasitism Cues from Haustoria Successful parasitism Kelly (1992), Yoder (2001)
favourable
hosts
Reproduction
Reproductive Environ. Shoot Outcrossing rates, Bradshaw (1965),
strategy conditions, meristems reproductive Paige and Whitham (1987)
pollinators success
Functional Stress Reproduct. Maleness Freeman et al. (1980)
gender Resources tissues Femaleness Hendrix and Trapp (1981)
Floral damage Femaleness
Seed germination Light, temp, Seed coat, Seedling grows in Baskin and Baskin (1998)
other physical embryo favourable
and biotic environment
conditions
Defence
Production of Microbes Many Improved defence Hammerschmidt (1999)
phytoalexins
Accumulation of Herbivores & Many Improved defence Karban and Baldwin (1997)
secondary microbes
metabolites
Attraction of Herbivore attack Systemic response Improved defence Dicke and van Loon (2000)
predators and
parasites of
herbivores

these fitness consequences depend upon the broader may involve subtle and coordinated changes in the
selective environment that the plant experiences (Huber photosynthetic apparatus (Pearcy & Sims 1994) or shedding
et al. 2004). Taken to its extreme, the tendency to grow of entire leaf canopies in drought-deciduous species
towards light can be detrimental. For example, competition (Comstock et al. 1988).
for light near forest gaps can promote unbalanced growth so Plants that are obligate or facultative parasites must
that trees eventually fall into the gaps (Young & Hubbell forage in a manner more similar to animals than autotrophs.
1991). Parasitic plants locate and invade the tissues of other plants
Plants experience vastly different light and water levels to rob their hosts of nutrients and water (Yoder 2001;
over space and time and they acclimate to optimize Runyon et al. 2006). Haustoria of the parasitic plants
photosynthetic and gas exchange rates, and to avoid respond to chemicals released by other species, allowing
photoinhibition (Pearcy & Sims 1994). These physiological the parasite to recognize and attack appropriate host plants.
adjustments may occur over seconds at the chloroplast and Indeed, parasitic dodder is more likely to grow towards and
cellular level or over longer times at the branch level. They accept hosts of higher nutritional quality compared with

 2008 Blackwell Publishing Ltd/CNRS

730 R. Karban Review and Synthesis

those of lesser quality (Kelly 1992). Even for facultative teristics. Some induced responses make plants less suscep-
plant parasites, fitness is closely linked to host quality (Adler tible to, or less preferred by, attackers and thus may increase
2003). plant fitness relative to attacked individuals that do not
induce. Induced responses to pathogens and herbivores
have generally not been included in reviews of plant
Mating and germination behaviour
behaviour. This is partly a historical oversight and partly
The examples listed above all involve resource acquisition. because plant defences have been assumed to be primarily
Plants also display responses to environmental cues that are chemical traits that do not involve movement or positional
reflected in their reproductive behaviours. Individuals that changes. In this regard, most plant responses to attacks are
fail to get pollinated may increase their investment in less similar to behaviour from a zoological perspective and
rewards that attract insect visitors (Ladio & Aizen 1999). more similar to physiological plasticity that is also well
Plants that experience conditions that are unfavourable for described for animals (Tollrian & Harvell 1999). However,
pollination may respond by producing cleistogamous the responses of plants to attackers are often relatively rapid
flowers that do not open and are self-pollinated (Bradshaw and reversible and fit the definition of plant behaviour
1965). Individuals of Ipomopsis aggregata that failed to get outlined above. Below, I give three examples of induced
pollinated shifted from being semelparous (flowering once plant responses to pathogens and herbivores that have been
and dying) to being iteroparous (flowering again; Paige & particularly well studied.
Whitham 1987). These responses allow successful repro- Plants that are attacked by microbes undergo a variety
duction to occur under suboptimal conditions. of physical and chemical changes in an attempt to prevent
Plants also may adjust their functional gender in response infection, curtail the growth of the pathogen, and survive
to the conditions that they experience. Environmental the attack (Ferreira et al. 2007). One induced plant
stresses generally cause plants to invest disproportionately in response that has been well documented is the accumu-
male flowers whereas access to water and other nutrients lation of phytoalexins (Hammerschmidt 1999). These are
causes plants to invest more in female reproduction low molecular weight secondary metabolites that exhibit
(Freeman et al. 1980). Herbivory to reproductive tissues antimicrobial and antifungal properties. Phytoalexins are
causes plants to selectively abort and ⁄ or regrow reproduc- often synthesized de novo in response to infection and they
tive organs; flower damage generally shifts plants towards degrade rapidly so that they are undetectable in unchal-
more femaleness (Hendrix & Trapp 1981; Krupnick & Weis lenged tissue. Treatments that increase or decrease
1998). concentrations of phytoalexins in planta cause correlated
The timing of seed germination for many species is increases or decreases in resistance to many disease
strongly affected by environmental conditions (Baskin & threats.
Baskin 1998). The decision to germinate is a conditional Herbivores also induce the synthesis of secondary
response in contrast to the specific programs of develop- compounds in many plants (Karban & Baldwin 1997).
mental changes that occur during the germination process, One well-studied example is nicotine accumulation in
which are fixed and therefore not considered as plant damaged tobacco plants. This alkaloid is produced by the
behaviour. For instance, the number of hours of daylight roots of tobacco and is transported in the xylem stream up
determines whether some species will germinate or remain to the leaves following herbivory (Baldwin 1999). Nicotine
dormant and later, whether vegetative growth is determinate is deadly to most leaf-feeding herbivores and reduces
or indeterminate. The spectral quality of light (red : far red feeding by even those species that can tolerate it. It is
ratio), temperature, fire, exposure to water, oxygen, CO2, difficult to determine the costs and benefits of nicotine
ethylene and other chemicals, passage through animal guts accumulation (or other secondary chemicals) as experimen-
and attack by insects can all be important in determining tal treatments that induce one plant response also induce a
whether seeds geminate. These conditional responses allow large number of correlated responses. In the field, trans-
seeds to germinate into environments that are favourable for formed plants that lacked the ability to respond to herbivory
growth and to remain dormant when conditions are (including the nicotine response plus many others) were
unfavourable. more vulnerable to insects that specialize on tobacco and
were also attacked by generalist herbivores that do not
usually feed on tobacco (Kessler et al. 2004). Artificially
Induced plant responses to pathogens and herbivory
induced responses, including nicotine accumulation,
Plants respond to the attacks of pathogens and herbivores increased seed production of tobacco in those field
by changing many phenotypic traits (Karban & Baldwin situations where plants were likely to be attacked by
1997; Agrawal et al. 1999b). These induced responses herbivores (Baldwin 1998). Tobacco plants that specifically
include chemical, physiological, and morphological charac- lacked the nicotine response experienced more damage in

 2008 Blackwell Publishing Ltd/CNRS

Review and Synthesis Plant behaviour and communication 731

the field although plant fitness has not yet been evaluated from physiological adjustments at the subcellular level up to
(Steppuhn et al. 2004). larger scale alterations of plant morphology. Despite this
Some plant responses to herbivore attack are used as cues diversity, many generalities in plant behaviour seem to
by the predators and parasites of herbivores as they forage emerge. Below I attempt to list some of these recurring
for food (Dicke & Sabelis 1988; Dicke & van Loon 2000). patterns that have not been appreciated widely and to
Plants release complex volatile blends that differ in contrast them to similar behaviours exhibited by animals.
composition depending upon whether they have been
attacked and upon the specific nature of the attacker. The
Plant behaviours can be complex
predators and parasitoids that are attracted by herbivore-
induced volatiles have been shown to increase rates of Because plants lack central nervous systems, their behav-
predation and parasitism and decrease levels of damage iours have previously been categorized as relatively simple
inflicted by herbivores under natural field conditions in compared with the behavioural repertoires of animals
some, though not all, instances (Thaler 1999; Kessler & (Silvertown & Gordon 1989). The argument that plant
Baldwin 2001; Heil 2004a,b; Karban 2007a,b). When behaviours are relatively simple was based on the observa-
predators and parasites are attracted, this mechanism is tions that plants responded to thresholds, gradients, or
considered an indirect plant defence because the induced changes in the magnitude of environmental variables but
plant behaviour causes a behavioural shift in a third trophic not to more complex patterns in those variables. Evidence
level. Although the three plant responses described above all concerning induced responses to herbivores indicates that
involve chemical changes, plants also adjust their morpho- plants can distinguish among a large set of cues and respond
logies (spines, trichomes, stature, leaf toughness, etc.) as well appropriately. For example, plants emit different blends of
as where they allocate resources (above or below ground) in volatile chemicals in response to attack by closely related
response to attack (Karban & Baldwin 1997). caterpillars (De Moraes et al. 1998). These cues provide
detailed information that allows species-specific parasitoid
wasps to locate their particular hosts but do not attract them
Conditional mutualisms
to non-host caterpillars. Selective plant emissions and
When predators and parasites increase plant fitness by parasitoid responses require precision both by the plants
reducing plant damage caused by herbivores, the plant- and the parasitoids.
predator interaction can be considered as beneficial to both Plants are also capable of responding to complex cues
plants and predators. Plants adjust the rewards that they about their current and future environments. For example,
provide to these mutualists, increasing extrafloral nectar sagebrush plants respond to volatile cues released by their
production when risk of herbivory is great (Heil et al. 2001, neighbours to increase their levels of resistance after a
2004a) and reducing nectar and other rewards when risk is neighbour has been attacked (Karban et al. 2004, 2006).
low (Huntzinger et al. 2004). Plants adjust their investments These same volatile cues are required for an individual
in other mutualisms as well. Some plants house N-fixing sagebrush to coordinate its defences among several
bacteria in root nodules and receive nitrogen in return. branches as vascular integration is limited. These examples
When the benefits to the host plant in terms of nitrogen of recognition of complicated patterns and other examples
supplied by bacteria were experimentally reduced, the host described below now indicate that plant behaviours are
plant responded by reducing the oxygen supply to the much more sophisticated than the authors of previous
bacteria (Kiers et al. 2003). This plant-imposed sanction reviews were aware.
decreased reproductive success of the bacteria by c. 50%.
Plants also adjust rewards depending upon the quality of
Plant behaviours can be rapid
service provided by pollinating insects. For example, yucca
moths both pollinate and oviposit into the seed heads of Plant behaviours have been depicted in previous reviews as
their yucca hosts. Plants are more likely to allow fruits with relatively slow (Harper 1985; Silvertown & Gordon 1989).
low egg loads and high pollen loads to mature (Pellmyr & However, plant movements can be as rapid as those of
Huth 1994). Selective maturation increases seed production many animals, exemplified by carnivorous plants, although
and polices moths that lay many eggs or provide low quality the spatial extent of these movements is more limited
pollination. compared with more motile animals. Previous authors have
had foraging plants as their models of behaviour and these
rely on morphological plasticity, requiring either growth or
PATTERNS IN PLANT BEHAVIOUR
death of plant organs. However, many of the plant
Plants respond to a great variety of environmental stimuli behaviours listed in Table 1 such as the responses to light
and conditions. Plants also respond at a diversity of levels or to herbivores occur within seconds, much more rapidly

 2008 Blackwell Publishing Ltd/CNRS

732 R. Karban Review and Synthesis

than most morphological changes (Pearcy & Sims 1994; Plastic responses are often portrayed as norms of
Karban & Baldwin 1997). As most plants are rooted to a reactions for the entire organism (e.g. Via 1987; Harvell
substrate, their opportunities for distant movement are 1990). This view implicitly assumes systemic integration and
more limited than those of unattached animals. masks the reality of localized plant responses to fine scale
The difference between many plant and animal behav- environmental heterogeneity. Instead of shifting the phe-
iours is largely one of spatial and temporal scale. Indeed, the notype of the entire plant, localized plasticity tends to
fastest motion yet observed in biology is the pollen release, increase phenotypic variation among leaves or other ramets
at velocities that exceed half the speed of sound, of white within an individual (Stout et al. 1996; de Kroon et al. 2005).
mulberry flowers in response to dry air (Taylor et al. 2006). Indeed, the extent to which plastic responses are integrated
Of course, the spatial range of pollen movement in this and the consequences of integration or sectoriality (lack of
example is relatively small. Similarly, the movement of plant integration) are important areas in which our understanding
roots through soils of varying nutritional value is less is very incomplete (Harper 1985; Hutchings & de Kroon
extensive (and less rapid) than the foraging trails of ants in 1994; Silvertown 1998; de Kroon et al. 2005; Orians 2005).
spatially variable habitat patches. However, both of these
foraging behaviours are highly conditional and highly
Plant behaviours may be correlated and context-
reversible, as fine roots turn over rapidly.
dependent
Plant responses to heterogeneous and changing environ-
Modular construction enables many behaviours
ments often occur as correlated suites of behaviours or
Morphological plasticity, exemplified by behaviours such as syndromes (Agrawal & Fishbein 2006). These correlated
foraging, is possible because plants are made up of largely changes produce different consequences than the responses
autonomous modular units (Harper 1985; Silvertown & would have if they occurred independently. One of the best
Gordon 1989). These modular units arise from iterative studied induced responses to herbivory is the synthesis of
active meristems that have the ability to grow into organs of proteinase inhibitors in solanaceous plants (Ryan 1990).
undetermined characteristics, varying in type, size, shape and Proteinase inhibitors produced by plants interfere with the
number. As a result, plants can transform in radically action of digestive enzymes of herbivores and decrease
different ways in response to different environmental cues. performance of many herbivores. However, the effective-
Many plant behaviours rely on modular organization to ness of proteinase inhibitors depends critically upon other
respond to environmental heterogeneity in much the same induced plant changes. Many wound-induced changes
way as many animal behaviours rely on mobility. deactivate proteinase inhibitors as do high concentrations
of nutrient levels in the herbivore gut (Duffey & Felton
1989). On the other hand, induced nicotine acts synergis-
Plant behaviours are often localized
tically with induced proteinase inhibitors to affect herbivore
A consequence of independent modular construction is performance more adversely than either response does in
that plants are generally less well integrated than are isolation (Steppuhn & Baldwin 2007). The effect of any
animals (Hutchings & de Kroon 1994; de Kroon et al. given plant behaviour is highly context-dependent and the
2005). Plants typically respond to fine-grained heterogene- context ranges from cellular to environmental.
ity in their environments with localized adjustments. For
example, in a classic experiment, Drew (1975) grew barley
Tradeoffs in plant behaviours
in soils containing different distributions of phosphate.
Those portions of the root system in contact with higher The effects of behaviours depend upon the current state of
concentrations of phosphate grew more and longer lateral the plant. Responses to one environmental condition may
rootlets, indicating independence of modules. However, affect and constrain the plantÕs responses to other condi-
this response was influenced to some extent by the levels tions. This phenomenon is widespread and becomes
of nutrients available to the whole plant, indicating some particularly visible as tradeoffs in opposing plant responses
degree of integration. The localized foraging response was to different conditions. Plants appear to use a relatively
stronger when the whole plant was growing in a small number of hormones to perceive and respond to
phosphate-poor environment compared with one that many different environmental stimuli. The signals that
supplied relatively more phosphate to the entire root induce one behaviour may lessen the plantÕs ability to induce
system. This result exemplifies a situation in which a different behaviour. For example, the signals used to
responses are partly localized and independent but also induce defence against some chewing insects may interfere
partly integrated with the conditions experienced by the with the signals used to induce defence against pathogens
whole plant. (Felton et al. 1999; Bostock 2005). Many other tradeoffs in

 2008 Blackwell Publishing Ltd/CNRS

Review and Synthesis Plant behaviour and communication 733

plant behaviours have been noted. Plants that have induced nicate with other organisms, including conspecifics and this
a shade avoidance response were less able to induce defence communication alters plant behaviour. Below I will expand
against herbivory (Cipollini 2004; Kurashige & Agrawal on these three themes.
2005; Izaguirre et al. 2006). Plant responses to salt and water
stress may interfere with their ability to defend against
PLANTS ANTICIPATE FUTURE ENVIRONMENTAL
pathogens and herbivores (Thaler & Bostock 2004). These
CONDITIONS
tradeoffs are often attributed to the fact that plants have
limited resources available to invest in multiple functions Plants behave in ways that make them appear to anticipate
although this underlying mechanism has not been clearly future conditions in many of the cases listed in Table 1,
established and alternatives have not been adequately although these behaviours do not involve true cognition.
considered. This is most often accomplished by responding to cues that
are good correlates of future conditions (Karban et al. 1999).
Many deciduous plants drop leaves in the autumn in
Consequences of plant behaviours are poorly known
response to shortening photoperiod as if they are anticipat-
We know much less about the ecological consequences of ing winter conditions that have a high probability of
plant behaviours than we do about the phenomena damaging leaves and branches and are suboptimal for
themselves (Callaway et al. 2003). Changes in behavioural photosynthesis. Shortened photoperiod is a cue that
traits may have various and far reaching direct and indirect proceeds, and is highly correlated with, the onset of
effects on other members of the community. For example, conditions that can be suboptimal or even harmful.
we know that there is additive genetic variance in wild radish Similarly, seeds use cues to break dormancy, such as heat
plants for plasticity of defensive traits (Agrawal et al. 2002). from fire for fire-adapted species. These cues are good
Defence in this case includes inducible glucosinolates, leaf predictors of conditions, such as limited competition and
toughness and density of trichomes. Plasticity in these traits abundant light and nutrients, that are uncommon but will be
affects a diverse herbivore community and increases plant very favourable for the growth of seedlings (Baskin &
fitness in a variety of different herbivore environments Baskin 1998).
(Agrawal 1998). But even in the best known systems, it is Plant foraging may place them in situations that will be
still unclear how these plastic behavioural traits influence favourable in the future. The mechanism of this anticipatory
interactions with the other plant, animal, and microbial behaviour is particularly well studied in the case of the shade
species (Agrawal 2001). In a second example, growth rates avoidance response that occurs before plants actually
and the resulting density of oak roots respond to the become shaded. Plants use phytochrome receptors to sense
availability of water and other nutrients. Our current an increase in far red radiation generated by neighbouring
knowledge suggests that this root foraging behaviour can green leaves (Smith 2000). Responding plants grow away
have profound effects on the other plants that grow in the from neighbours well before those neighbours diminish
oak woodland community, although these effects are still their actual acquisition of light (Ballare et al. 1990). Removing
poorly resolved (Callaway et al. 2003). These two examples the far red cue with filters abolishes the response. Plants
have been more closely examined than most, and future respond more strongly to the red : far red quality (the cue)
work examining the broader ecological consequences of than to diminished photon flux density (amount of
plant behaviours will shape our understanding of species photosynthetically active light) (Novoplansky 1991). Other
interactions in general. examples of foraging indicate that plants can anticipate
future conditions although the mechanisms are less well
resolved. For example, parasitic dodder accepts nutritious
Plants appear to anticipate, remember and communicate
hosts and rejects less nutritious hosts before taking up any
We are now realizing that plant behaviours are often quite food (Kelly 1992). The cues used in this decision making
sophisticated and possess attributes that were long consi- process are not known although recent work suggests that
dered the exclusive domain of animals with central nervous volatiles released by host plants allow dodder to preferen-
systems (Borges 2005; Trewavas 2005). Plants engage in tially grow in the direction of high quality hosts (Runyon
three behaviours that we intuitively associate with cognition. et al. 2006).
(i) Plants often anticipate environmental changes that have Induced responses to herbivory are only effective if
not yet occurred. (ii) Plants often become conditioned by current herbivory is a good predictor of future risk (Karban
experiences that they or their parents have had and this et al. 1999). Surprisingly few studies have documented that
conditioning alters their behaviours. In other words, plants past or current herbivory is actually an information-rich cue
appear to have a ÔmemoryÕ that influences their responses that predicts future risk. Early season damage to wild cotton
based on past experience. (iii) Plants use cues to commu- plants was a good predictor of the risk that plants were likely

 2008 Blackwell Publishing Ltd/CNRS

734 R. Karban Review and Synthesis

to encounter during the remainder of the season (Karban that are still not understood. Priming appears less costly
and Adler 1996). Some phytoplankton form protective than responses that change the plant resistance immediately
colonies in response to waterborne cues associated with after pathogen attack (Van Hulten et al. 2006).
their herbivores. Herbivore cues reliably predict when it is Priming has recently been reported for plants attacked by
safe to remain unspined and solitary and when these herbivores as well (Engelberth et al. 2004). Plants primed by
behaviours are dangerous (Van Donk et al. 1999; Verschoor a previous attack became less suitable as hosts for
et al. 2004). In a situation where two herbivores have herbivores than plants responding for the first time (Ton
opposing preferences for solitary and clumped colonies, the et al. 2007). Once plant workers became aware of priming,
specific cues of each herbivore species induces opposite but several groups rapidly reported it from their systems
adaptive responses in phytoplankton (Long et al. 2007). By although the sensitizing agents responsible for the effect
responding to cues that reliably predict future conditions, appear to be species-specific (reviewed by Ton et al. 2007).
plants can anticipate risk and behave preemptively. Priming may also be involved in plant responses to a variety
of other stresses such as drought and salt (Conrath et al.
2006).
CONDITIONING AND MEMORY
These examples involve plants responding more quickly
One of the hallmarks of animal behaviour is that behaviours and more strongly when they themselves have experienced
are influenced by experience and learning. Previous expe- an attack. There is also evidence that the maternal
rience also greatly influences plant behaviour through environment can influence the traits displayed by offspring.
conditioning. Note that the term ÔconditioningÕ implies Maternal conditions that influence seed germination include
learning associated with a particular favourable or unfa- CO2 level, competition, day length, pathogen infection,
vourable outcome to an animal behaviourist, but has no nutrition, water stress, among others (Baskin & Baskin
such connotations in the lexicon of the plant biologist. I use 1998). Maternal radish plants that had been damaged by
the term ÔconditioningÕ to mean the reversible changes in caterpillars became more resistant themselves and also
behaviour (plasticity) that have been altered by experience. produced genetically distinct offspring that were less suitable
Plant responses are shaped by the plantÕs own experiences for herbivores than seedlings from undamaged mothers
(conditioning) or even by the experiences of its parents (Agrawal et al. 1999a). Seedlings from mothers that had
(preconditioning). survived herbivory induced higher levels of glucosinolates
Our awareness of plant conditioning dates at least as far and more trichomes following herbivory than seedlings
back as Darwin (1880: pp. 460–461), who observed that from comparable control mothers. It is not clear how
responses of cotyledons to light were influenced by their widespread preconditioning or maternal effects on plant
previous exposure. Many foraging decisions are affected behaviour will prove to be, although recent evidence
both by current conditions and past experiences. For indicates that plants also respond adaptively to the light
example, the growth of a clover branch depends upon its environments experienced by their mothers (Galloway &
current neighbours and also upon the neighbours that it Etterson 2007).
encountered over the past year (Turkington et al. 1991).
Reproductive decisions are also affected by past events. For
COMMUNICATION
instance, some biennials respond to cues to flower only after
they first experience a prolonged period of cold, known as Plants emit cues that cause other organisms to change their
vernalization. Recently, the genetic control of this process behaviours. They also respond to stimuli emitted by other
has been elaborated for winter wheat; prolonged exposure organisms to alter their own traits. Workers in the nascent
to cold inhibits the gene that represses flowering (Yan et al. field of plant communication would benefit by agreeing
2004). upon a set of definitions and I offer a dichotomous key of
One of the hallmarks of animal immune responses is the terms that includes a comparison to definitions used in
capacity for immunological memory (Harvell 1990). Animals studying animal communication (Table 2). For communi-
that recover from an attack respond more rapidly and cation to occur, receivers must respond rapidly to cues or
effectively the second time that they encounter the parasite. stimuli produced by other organisms (emitters; step 1 in
Priming or conditioning is also proving to be common in Table 2). Animal behaviourists have found this criterion to
plant responses to pathogens and herbivores. Plants that be insufficient because it includes so many phenomena and
have been primed by an initial attack respond more rapidly consider that it specifies an induced response but not
and more effectively to a second exposure to that same communication. I will restrict my use of the term Ôplant
pathogen or to a variety of novel attackers (Conrath et al. communicationÕ to situations in which the emission or
2006). This phenomenon has been described for many plant display of the cue is plastic and the response of the emitter is
species and appears to involve several different mechanisms conditioned on receiving the cue (step 2 in Table 2). For

 2008 Blackwell Publishing Ltd/CNRS

Review and Synthesis Plant behaviour and communication 735

Table 2 A dichotomous key for plant communication Arimura et al. 2000; Kost & Heil 2006; Paschold et al. 2006).
The precise cue has yet to be determined for most of the
Does the cue cause a rapid response in a receiver organism?
1 No: not communication
examples of plant communication that have been described.
1¢ Yes: induced response (proceed to 2) As previously mentioned, plants respond to cues that
Is emission of cue plastic and conditional? indicate the presence of other individuals and develop a
2 No: receiver responds to environment, not communication morphology that increases fitness in the competitive
2¢ Yes: plant communication (proceed to 3) environment that they are likely to experience (Ballare
Is emission of cue (signal) intentional and beneficial? 1999; Schmitt et al. 1999). Some plants also adjust their
3 No: not Ôtrue communicationÕ sensu animal behaviourists defences to develop a phenotype that matches the herbivore
3¢ Yes: true communication environment that they are likely to experience. For example,
several species respond to volatile cues emitted by sagebrush
example, a plant is not necessarily ÔcommunicatingÕ with neighbours that have been experimentally damaged by
insect visitors if its appearance does not vary conditionally. herbivory (Karban et al. 2000, 2004). Again, responding to
In contrast, a plant is communicating when it alters its floral such cues can be beneficial under some circumstances
display in response to its circumstances and this causes (Karban & Maron 2002). Although the benefits to plants
visitors to respond accordingly. that respond to reliable cues are easier to understand, it is
Some animal behaviourists reserve the term Ôtrue com- also possible that damaged plants might benefit by emitting
municationÕ to refer more narrowly to the intentional cues. Volatile cues emitted by damaged branches may allow
transfer of signals (rather than cues) that, by definition, sagebrush individuals to integrate their own defences
benefit both the emitter and the receiver (step 3 in Table 2; (Karban et al. 2006) and reduce the germination of
Bradbury & Vehrencamp 1998). Workers have difficulty competitors (Karban 2007b).
determining benefit or intent even for animals. Intent is In addition to influencing other plant organs or individ-
impossible for plants (except in an ultimate or evolutionary uals, cues emitted by damaged plants are used by various
sense) and this requirement seems overly restrictive. Some eavesdroppers to adjust their behaviours. For instance,
animal behaviourists have reached similar conclusions; for volatile organic chemicals emitted when plants are damaged
example, most of the communication that occurs within a by herbivores may alter the behaviours of herbivores that
colony of honey bees involves cues rather than signals subsequently visit. There are many examples in which
(Seeley 1995). As such, I will use the term plant commu- damaged plants became either less attractive to herbivores
nication to refer to transfer of cues from one individual to (De Moraes et al. 2001; Heil 2004b) or more attractive
another without any assumptions about intent or benefit for (Dicke & van Loon 2000) relative to undamaged controls.
the emitter or receiver. The emitter may not have intended As discussed earlier, plants that are under attack by
to communicate with the receiver and the emitter may herbivores release volatile cues that help predators locate
benefit from the communication, experience no fitness and consume the herbivores and these cues may reduce
consequence, or suffer as a result. The first step in studying plant losses to herbivory (Dicke & Sabelis 1988). In some
potential communication involves determining what the cases, the plant cues that attract parasitoids are produced de
cues are and how they affect responses in other organisms novo and cannot be detected in unattacked plants (Turlings
(steps 1 and 2). However, the most interesting questions et al. 1990). Generally, artificial damage was not as attractive
involve determining how communication affects the fitness to the predators and parasitoids as actual damage by
of the emitter and receiver (step 3) and students of plant herbivores (Dicke & van Loon 2000).
communication should vigorously pursue this question. Plants present cues that are used by their mutualists to
Plants communicate with other plants, with their herbi- provide beneficial services. Volatile and visual cues com-
vores, mutualists, and parasites, as well as the predators and municate the location, quality and abundance of nectar and
parasites of these interactors. Well-documented cues that pollen rewards for animals that visit flowers (Dobson 1994;
plants emit include light quality and volatile chemicals. For Chittka & Raine 2006). Animals that visit flowers for these
example, both far red and blue light have been implicated in rewards may move pollen from one individual plant to
shade avoidance responses (Pierik et al. 2004). In addition, another of the same species. Diverse groups of animals use
transgenic plants that were insensitive to ethylene were less these cues to adjust the amount of time they spend at
responsive to shading, suggesting that volatile cues are also flowers and to avoid flowers that will be unrewarding. These
involved (Pierik et al. 2003). Two other volatile cues, methyl cues are also used by other plant visitors that rob nectar and
jasmonate and green leaf volatiles, have been implicated in consume floral tissue without necessarily providing pollina-
plant responses following wounding although demonstrat- tion (Irwin et al. 2004). Plants may adjust their flowering
ing the activity of any particular cue under natural strategies and the cues presented depending upon their
conditions has been very difficult (Farmer & Ryan 1990; pollination status (Paige & Whitham 1987; Weiss 1991;

 2008 Blackwell Publishing Ltd/CNRS

736 R. Karban Review and Synthesis

Ladio & Aizen 1999; Nuttman & Willmer 2003). Many of tition with the others. This approach has rarely been applied
the classes of behaviours listed in Table 1 have been found to plant behaviours. In an early and seminal work, Rhoades
to be triggered by cues from other individuals in at least (1979, 1983, 1985) presaged many of the plant defences
some instances. Essentially all plants produce cues that serve covered in this review by taking an adaptationist approach
to communicate with other organisms, whether production and asking how and where plants should respond. Animal
of these cues has been favoured by selection or has been the behaviour has also pursued questions about the costs and
incidental consequence of other processes. benefits of particular behaviours for the individual display-
ing them as well as their consequences for other individuals.
This very successful approach has rarely been applied to
CONCLUSION
plant behaviours, although early attempts suggest that
Plant behaviours facilitate effective foraging, reproduction, behaviour is likely to have important and far reaching
and defence in a spatially heterogeneous and constantly consequences on plant communities (Callaway et al. 2003).
changing environment. Behaviour is critically important in In the future, plant biologists more generally would benefit
the struggle of plants to pass on their genes although we by borrowing the techniques that have been developed over
currently have a poor understanding of the consequences of the past century to document and understand animal
most plant behaviours. The plant behaviours included in behaviour.
Table 1 are more widespread, more diverse, and more
sophisticated than commonly acknowledged. Plants are
ACKNOWLEDGEMENTS
capable of responding to complex cues that involve multiple
stimuli. Responses show considerable specificity in terms of This review grew out of discussions with Judy Stamps and
recognition and reaction. By responding to reliable predic- Louie Yang. I thank Anurag Agrawal, Mikaela Huntzinger,
tive cues, plant behaviours often anticipate future environ- Rebecca Irwin, Andy McCall, Judy Stamps, Louie Yang, and
mental conditions. Plants also become conditioned by their several anonymous referees for improving the manuscript.
past experiences and appear to have memories. Plants not
only respond to reliable cues in their environments but also
REFERENCES
produce cues that communicate with other plants and with
other organisms, such as pollinators, seed dispersers, Adler, L.S. (2003). Host species affects herbivory, pollination, and
herbivores, and enemies of those herbivores. reproduction in experiments with parasitic Castilleja. Ecology, 84,
2083–2091.
We are now realizing that plants display many of the
Agrawal, A.A. (1998). Induced responses to herbivory and
behaviours that have been long thought to reside within the increased plant performance. Science, 279, 1201–1202.
exclusive domain of animals. That this is possible is Agrawal, A.A. (2001). Phenotypic plasticity in the interactions and
remarkable, particularly because plants lack a central organ evolution of species. Science, 294, 321–326.
to coordinate these sophisticated behaviours. On the other Agrawal, A.A. & Fishbein, M. (2006). Plant defense syndromes.
hand, plants have faced many of the same selective Ecology, 87, S132–S149.
pressures as animals over evolutionary time. Like animals, Agrawal, A.A., Laforsch, C. & Tollrian, R. (1999a). Transgenera-
tional induction of defences in animals and plants. Nature, 401,
they live in heterogeneous and varying environments.
60–63.
Presumably the same general selective pressures that Agrawal, A.A., Tuzin, S. & Bent, E. (eds) (1999b). Induced Plant
favoured the evolution of animal behaviours also favoured Defenses Against Pathogens and Herbivores. APS Press, St Paul, MN.
the evolution of conditional, plastic plant responses to Agrawal, A.A., Conner, J.K., Johnson, M.T.J. & Wallsgrove, R.
current and anticipated conditions. Decentralized responses (2002). Ecological genetics on an induced plant defense against
that can occur at any meristem allow plants to exhibit herbivores: additive genetic variance and costs of phenotypic
behaviours that have outwardly converged with those plasticity. Evolution, 56, 2206–2213.
Arimura, G., Ozawa, R., Shimoda, T., Nishioka, T., Boland, W. &
exhibited by animals with complex central nervous systems.
Takabayashi, J. (2000). Herbivory-induced volatiles elicit defence
This review reframes many of the examples from plant genes in lima bean leaves. Nature, 406, 512–515.
physiology and ecology as plant behaviours. Some of the Baldwin, I.T. (1998). Jasmonate-induced responses are costly but
ideas elaborated here overlap with those developed to benefit plants under attack in native populations. Proc. Natl Acad.
explain phenotypic plasticity. By reconsidering these traits as Sci. USA, 95, 8113–8118.
behaviours, one goal has been to compare them with animal Baldwin, I.T. (1999). Inducible nicotine production in native
behaviours. Behavioural ecology as a discipline has been Nicotiana as an example of adaptive phenotypic plasticity.
J. Chem. Ecol., 25, 3–30.
successful in aggressively erecting and testing optimality and
Ballare, C.L., Scopel, A.L., Sánchez, R.A. (1990). Far-red radiation
adaptive hypotheses (Parker 1984; Reeve & Sherman 1993). reflected from adjacent leaves: an early signal of competition in
This approach compares alternative behaviours or allocation plant canopies. Science, 247, 329–332.
decisions to determine which ones will persist in compe-

 2008 Blackwell Publishing Ltd/CNRS

Review and Synthesis Plant behaviour and communication 737

Ballare, C.L. (1999). Keeping up with the neighbors: phytochrome J.R. & Sonnet, P.E.). American Chemical Society, Washington,
sensing and other signalling mechanisms. Trends Plant Sci., 4, DC, pp. 289–313.
97–102. Eisner, T. (1981). Leaf folding in a sensitive plant: a defensive
Baskin, C.C. & Baskin, J.M. (1998). Seeds. Ecology, Biogeography, and thorn-exposure mechanism? Proc. Natl Acad. Sci. USA, 78, 402–
Evolution of Dormancy and Germination. Academic Press, San Die- 404.
go, CA. Engelberth, J., Alborn, H.T., Schmelz, E.A. & Tumlinson, J.H.
Borges, R.M. (2005). Do plants and animals differ in phenotypic (2004). Airborne signals prime plants against insect herbivore
plasticity? J. Biosci., 30, 41–50. attack. Proc. Natl Acad. Sci. USA, 101, 1781–1785.
Bostock, R.M. (2005). Signal crosstalk and induced resistance: Farmer, E.F. & Ryan, C.A. (1990). Interplant communication: air-
straddling the line between cost and benefit. Annu. Rev. Phyto- borne methyl jasmonate induced synthesis of proteinase inhibi-
pathol., 43, 545–580. tors in plant leaves. Proc. Natl Acad. Sci. USA, 87, 7713–7716.
Braam, J. (2005). In touch: plant responses to mechanical stimuli. Felton, G.W., Korth, K.L., Wesley, S.V., Huhman, D.V., Mat-
New Phytol., 165, 373–389. thews, M.C., Murphy, J.B. et al. (1999). Inverse relationship
Bradbury, J.W. & Vehrencamp, S.L. (1998). Principles of Animal between systemic resistance of plants to microorganisms and to
Communication. Sinauer, Sunderland, MA. insect herbivory. Curr. Biol., 9, 317–320.
Bradshaw, A.D. (1965). Evolutionary significance of phenotypic Ferreira, R.B., Monteiro, S., Freitas, R., Santos, C., Chen, Z.,
plasticity in plants. Adv. Genet., 13, 115–155. Batista, L.M. et al. (2007). The role of plant defence proteins in
Callaway, R.M., Pennings, S.C. & Richards, C.L. (2003). Phe- fungal pathogenesis. Mol. Plant Pathol., 8, 677–700.
notypic plasticity and interactions among plants. Ecology, 84, Freeman, D.C., Harper, K.T. & Charnov, E.L. (1980). Sex change
1115–1128. in plants: old and new observations and new hypotheses. Oeco-
Chittka, L. & Raine, N.E. (2006). Recognition of flowers by poll- logia, 47, 222–232.
inators. Curr. Opin. Plant Biol., 9, 428–435. Galloway, L.F. & Etterson, J.R. (2007). Transgenerational plasticity
Cipollini, D. (2004). Stretching the limits of plasticity: can a plant is adaptive in the wild. Science, 318, 1134–1136.
defend against both competitors and herbivores? Ecology, 85, Hammerschmidt, R. (1999). Phytoalexins: what have we learned
28–37. after 60 years? Annu. Rev. Phytopathol., 37, 285–306.
Comstock, J.P., Cooper, T.A. & Ehleringer, J.R. (1988). Seasonal Harper, J.L. (1985). Modules, branches, and the capture of
patterns of canopy development and carbon gain in nineteen resources. In: Population Biology and Evolution of Clonal Organisms
warm desert shrub species. Oecologia, 75, 327–335. (eds Jackson, J.B.C., Buss, L.W. & Cook, R.E.). Yale University
Conrath, U. et al. (2006). Priming: getting ready for battle. Mol. Press, New Haven, CT, pp. 1–33.
Plant Microbe Interact., 19, 1062–1071. Harvell, C.D. (1990). The ecology and evolution of inducible
Darwin, C. (1880). The Power of Movement in Plants. John Murray, defenses. Q. Rev. Biol., 65, 323–340.
London. Heil, M., Koch, T., Hilpert, A., Fiala, B., Boland, W. & Linsenmair,
Darwin, C. (1893). Insectivorous Plants. John Murray, London. K.E. (2001). Extrafloral nectar production of the ant-associated
De Moraes, C.M., Lewis, W.J., Pare, P.W., Alborn, H.T. & plant, Macaranga tanarius is an induced, indirect, defensive
Tumlinson, J.H. (1998). Herbivore-infested plants selectively response elicited by jasmonic acid. Proc. Natl. Acad. Sci. USA, 98,
attract parasitoids. Nature, 393, 570–573. 1083–1088.
De Moraes, C.M., Mescher, M.C. & Tumlinson, J.H. (2001). Cat- Heil, M. (2004a). Induction of two indirect defences benefits lima
erpillar-induced nocturnal plant volatiles repel conspecific fe- bean (Phaseolus lunatus, Fabaceae) in nature. J. Ecol., 92, 527–536.
males. Nature, 410, 577–580. Heil, M. (2004b). Direct defense or ecological costs: responses of
Dean, J.M. & Smith, A.P. (1978). Behavioral and morphological herbivorous beetles to volatiles released by wild lima beans.
adaptations of a tropical plant to high rainfall. Biotropica, 10, 152– J. Chem. Ecol., 30, 1289–1295.
154. Hendrix, S.D. & Trapp, E.J. (1981). Plant-herbivore interactions:
Dicke, M. & Sabelis, M.W. (1988). How plants obtain predatory insect induced changes in host plant sex expression and fecun-
mites as bodyguards. Neth. J. Zool., 38, 148–163. dity. Oecologia, 49, 119–122.
Dicke, M. & van Loon, J.J.A. (2000). Multitrophic effects of her- Huber, H., Kane, N.C., Heschel, M.S., von Wettberg, E.J., Banta,
bivore-induced plant volatiles in an evolutionary context. Ento- J., Leuck, A. et al. (2004). Frequency and microenvironmental
mol. Exp. Appl., 97, 237–249. pattern of selection on plastic shade-avoidance traits in a natural
Dobson, H.E.M. (1994). Floral volatiles in insect biology. In: Insect– population of Impatiens capensis. Am. Nat., 163, 548–563.
Plant Interactions, Vol. 5 (ed. Bernays, E.A.), CRC Press, Boca Huntzinger, M., Karban, R., Young, T.P. & Palmer, T.M. (2004).
Raton, FL, 5, 47–81. Relaxation of induced indirect defenses of acacias following
Drew, M.C. (1975). Comparison of the effects of a localized supply exclusion of mammalian herbivores. Ecology, 85, 609–614.
of phosphate, nitrate, ammonium and potassium on the growth Hutchings, M.J. & de Kroon, H. (1994). Foraging in plants: the role
of the seminal root system, and the shoot, of barley. New Phytol., of morphological plasticity in resource acquisition. Adv. Ecol.
111, 479–490. Res., 25, 159–238.
Drew, M.C., Saker, L.R. & Ashley, T.W. (1973). Nutrient supply Irwin, R., Adler, L.S. & Brody, A.K. (2004). The dual role of floral
and the growth of the seminal root system in barley. I. The traits: pollinator attraction and plant defense. Ecology, 85, 1503–
effect of nitrate concentration on the growth of axes and laterals. 1511.
J. Exp. Bot., 24, 1189–1202. Izaguirre, M.M., Mazza, C.A., Biodini, M., Baldwin, I.T. & Ballare,
Duffey, S.S. & Felton, G.W. (1989). Plant enzymes in resistance to C.L. (2006). Remote sensing of future competitors: impacts on
insects. In: Biocatalysis in Agricultural Biotechnology (eds Whitaker, plant defenses. Proc. Natl Acad. Sci. USA, 103, 7170–7174.

 2008 Blackwell Publishing Ltd/CNRS

738 R. Karban Review and Synthesis

Karban, R. (2007a). Damage to sagebrush: attracts predators but Novoplansky, A. (2002). Developmental plasticity in plants:
this does not reduce herbivory. Entomol. Exp. Appl., 125, 71–80. implications of non-cognitive behavior. Evol. Ecol., 16, 177–188.
Karban, R. (2007b). Experimental clipping of sagebrush inhibits Nuttman, C. & Willmer, P. (2003). How does insect visitation
seed germination of neighbors. Ecol. Lett., 10, 791–797. trigger floral colour change? Ecol. Entomol., 28, 467–474.
Karban, R. & Adler, F.R. (1996). Induced resistance to herbivores Orians, C. (2005). Herbivores, vascular pathways, and systemic
and the information content of early season attack. Oecologia, induction: facts and artifacts. J. Chem. Ecol., 31, 2231–2242.
107, 379–385. Paige, K.N. & Whitham, T.G. (1987). Flexible life history traits:
Karban, R. & Baldwin, I.T. (1997). Induced Responses to Herbivory. shifts by scarlet gilia in response to pollinator abundance. Ecol-
University of Chicago Press, Chicago. ogy, 68, 1691–1695.
Karban, R. & Maron, J. (2002). The fitness consequences of Parker, G.A. (1984). Evolutionarily stable strategies. In: Behavioural
interspecific eavesdropping between plants. Ecology, 83, 1209– Ecology: An Evolutionary Approach, 2nd edn (eds Krebs, J.R. &
1213. Davies, N.B.). Blackwell, Oxford, pp. 30–61.
Karban, R., Agrawal, A.A., Thaler, J.S. & Adler, L.S. (1999). In- Paschold, A., Halitschke, R. & Baldwin, I.T. (2006). Using ÔmuteÕ
duced plant responses and information content about risk of plants to translate volatile signals. Plant J., 45, 275–291.
herbivory. Trends Ecol. Evol., 14, 443–447. Pearcy, R.W. & Sims, D.A. (1994). Photosynthetic acclimation to
Karban, R., Baldwin, I.T., Baxter, K.J., Laue, G. & Felton, G.W. changing light environments: scaling from the leaf to the whole
(2000). Communication between plants: induced resistance in plant. In: Exploitation of Environmental Heterogeneity by Plants (eds
wild tobacco plants following clipping of neighboring sagebrush. Caldwell, M.M. & Pearcy, R.W.). Academic Press, San Diego,
Oecologia, 125, 66–71. CA, pp. 145–174.
Karban, R., Huntzinger, M. & McCall, A.C. (2004). The specificity Pellmyr, O. & Huth, C.J. (1994). Evolutionary stability of mutu-
of eavesdropping on sagebrush by other plants. Ecology, 85, alism between yucca and yucca moths. Nature, 372, 257–260.
1846–1852. Pierik, R., Visser, E.J.W., de Kroon, H. & Voesenek, L.A.C.J.
Karban, R., Shiojiri, K., Huntzinger, M. & McCall, A.C. (2006). (2003). Ethylene is required in tobacco to successfully com-
Damage-induced resistance in sagebrush: volatiles are key to pete with proximal neighbors. Plant Cell Environ., 26, 1229–
intra- and interplant communication. Ecology, 87, 922–930. 1234.
Kelly, C.K. (1992). Resource choice in Cuscuta europaea. Proc. Natl Pierik, R., Whitelam, G.C., Voesnek, L.A.C.J., de Kroon, H. &
Acad. Sci. USA, 89, 12194–12197. Visser, E.J.W. (2004). Canopy studies on ethylene-insensitive
Kessler, A. & Baldwin, I.T. (2001). Defensive function of herbi- tobacco identify ethylene as a novel element in blue light and
vore-induced plant volatile emissions in nature. Science, 291, plant–plant signalling. Plant J., 38, 310–319.
2141–2144. Reeve, H.K. & Sherman, P.W. (1993). Adaptation and the goals of
Kessler, A., Halitschke, R. & Baldwin, I.T. (2004). Silencing the evolutionary research. Q. Rev. Biol., 68, 1–32.
jasmonate cascade: induced plant defenses and insect popula- Rhoades, D.F. (1979). Evolution of plant chemical defense against
tions. Science, 305, 665–668. herbivores. In: Herbivores: Their Interaction with Secondary Plant
Kiers, E.T., Rousseau, R.A., West, S.A. & Denison, R.F. (2003). Metabolites (eds Rosenthal, G.A. & Janzen, D.H.). Academic,
Host sanctions and the legume-rhizobium mutualism. Nature, New York, pp. 3–54.
425, 78–81. Rhoades, D.F. (1983). Herbivore population dynamics and plant
Kost, C. & Heil, M. (2006). Herbivore-induced plant volatiles induce chemistry. In: Variable Plant and Herbivores in Natural and Managed
an indirect defence in neighboring plants. J. Ecol., 94, 619–628. Systems (eds Denno, R.F. & McClure, M.S.). Academic, New
Krebs, J.R. & Davies, N.B. (1997). Behavioral Ecology, 4th edn. York, pp. 155–220.
Blackwell, Oxford. Rhoades, D.F. (1985). Offensive–defensive interactions between
de Kroon, H. & Mommer, L. (2005). Root foraging theory put to herbivores and plants: their relevance in herbivore population
the test. Trends Ecol. Evol., 21, 113–116. dynamics and ecological theory. Am. Nat., 125, 205–238.
de Kroon, H., Huber, H., Stuefer, J.F. & van Groenendael, J.M. Runyon, J.B., Mescher, M.C. & De Moraes, C.M. (2006). Volatile
(2005). A modular concept of phenotypic plasticity in plants. chemical cues guide host location and host selection by parasitic
New Phytol., 166, 73–82. plants. Science, 313, 1964–1967.
Krupnick, G.A. & Weis, A.E. (1998). Floral herbivore effect on the Ryan, C.A. (1990). Proteinase inhibitors in plants: genes for
sex expression of an andromonoecious plant, Isomeris arborea improving defenses against insects and pathogens. Annu. Rev.
(Capparaceae). Plant Ecol., 134, 151–162. Phytopathol., 28, 425–450.
Kurashige, N.S. & Agrawal, A.A. (2005). Phenotypic plasticity to Schmitt, J., Dudley, S.A. & Pigliucci, M. (1999). Manipulative
light competition and herbivory in Chenopodium album (Cheno- approaches to testing adaptive plasticity: phytochrome-medi-
podiaceae). Am. J. Bot., 92, 21–26. ated shade-avoidance responses in plants. Am. Nat., 154, S43–
Ladio, A.H. & Aizen, M.A. (1999). Early reproductive failure in- S54.
creases nectar production and pollination success of late flowers Seeley, T.D. (1995). The Wisdom of the Hive. Harvard University
in south Andean Alstroemeria aurea. Oecologia, 120, 235–241. Press, Cambridge, MA.
Long, J.D., Smalley, G.W., Barsby, T., Anderson, J.T. & Hay, M.E. Silvertown, J. (1998). Plant phenotypic plasticity and non-cognitive
(2007). Chemical cues induce consumer-specific defenses in a behaviour. Trends Ecol. Evol., 13, 255–256.
bloom-forming marine phytoplankton. PNAS, 104, 10512– Silvertown, J. & Gordon, D.M. (1989). A framework for plant
10517. behavior. Annu. Rev. Ecol. Syst., 20, 349–366.
Novoplansky, A. (1991). Developmental responses of portulaca Smith, H. (2000). Phytochromes and light signal perception by
seedlings to conflicting spectral signals. Oecologia, 88, 138–140. plants – an emerging synthesis. Nature, 407, 585–591.

 2008 Blackwell Publishing Ltd/CNRS

Review and Synthesis Plant behaviour and communication 739

Steppuhn, A. & Baldwin, I.T. (2007). Resistance management in a Inducible Defenses (eds Tollrian, R. & Harvell, C.D.). Princeton
native plant: nicotine prevents herbivores from compensating University Press, Princeton, NJ, pp. 89–103.
for plant protease inhibitors. Ecol. Lett., 10, 499–511. Van Hulten, M., Pelser, M., van Loon, L.C., Pieterse, C.M.J. &
Steppuhn, A., Gase, K., Krock, B., Halitschke, R. & Baldwin, I.T. Ton, J. (2006). Costs and benefits of priming for defense in
(2004). NicotineÕs defensive function in nature. PLoS Biol., 2, Arabidopsis. Proc. Natl Acad. Sci. USA, 103, 5602–5607.
1074–1080. Verschoor, A.M., van der Stap, I., Helmsing, N.R., Lurling, M. &
Stout, M.J., Workman, J. & Duffey, S.S. (1996). Identity, spatial Van Donk, E. (2004). Inducible colony formation within the
distribution, and variability of induced chemical responses in Scenedesmaceae: adaptive responses to infochemicals from two
tomato plants. Entomol. Exp. Appl., 79, 255–271. different herbivore taxa. J. Phycol., 40, 808–814.
Taylor, P.E., Card, G., House, J., Dickinson, M.H. & Flagan, R.C. Via, S. (1987). Genetic constraints on the evolution of phenotypic
(2006). High-speed pollen release in the white mulberry tree, plasticity. In: Genetic Constraints on Adaptive Evolution (eds Loes-
Morus alba L. Sex. Plant Reprod., 19, 19–24. chcke, V.). Springer-Verlag, Berlin, pp. 47–69.
Thaler, J.S. (1999). Jasmonate-inducible plant defenses cause in- Weaver, J.E. (1926). Root Development of Field Crops. McGraw Hill,
creased parasitism of herbivores. Nature, 399, 686–688. New York.
Thaler, J.S. & Bostock, R.M. (2004). Interactions between abscisic- Weiss, M.R. (1991). Floral colour changes as cues for pollinators.
acid-mediated responses and plant resistance to pathogens and Nature, 354, 227–229.
insects. Ecology, 85, 48–58. Yan, L., Loukoianov, A., Blechl, A., Tranquilli, G., Ramakrishna,
Tollrian, R. & Harvell, C.D. (1999). The Ecology and Evolution of W., SanMiguel, P. et al. (2004). The wheat VRN2 gene is a
Inducible Defenses. Princeton University Press, Princeton, NJ. flowering repressor down-regulated by vernalization. Science, 303,
Ton, J., DÕAlessandro, M., Jourdie, V., Jakab, G., Karlen, D., Held, 1640–1644.
M. et al. (2007). Priming by airborne signals boosts direct and Yoder, J.I. (2001). Host-plant recognition by parasitic Scrophu-
indirect resistance in maize. Plant J., 49, 16–26. lariaceae. Curr. Opin. Plant Biol., 4, 359–365.
Trewavas, A. (2005). Plant intelligence. Naturwissenschaften, 92, 401– Young, T.P. & Hubbell, S.P. (1991). Crown asymmetry, treefalls,
413. and repeat disturbance of broad-leaved forest gaps. Ecology, 72,
Turkington, R., Hamilton, R.S. & Gliddon, C. (1991). Within- 1464–1471.
population variation in localized and integrated responses of
Trifolium repens to biotically patchy environments. Oecologia, 86,
183–192. Editor, Jonathan Chase
Turlings, T.C.J., Tumlinson, J.H. & Lewis, W.J. (1990). Exploita- Manuscript received 8 January 2008
tion of herbivore-induced plant odors by host-seeking parasitic First decision made 4 February 2008
wasps. Science, 250, 1251–1253.
Van Donk, E., Lurling, M. & Lampert, W. (1999). Consumer-
Second decision made 25 February 2008
induced changes in phytoplankton: inducibility, costs, benefits, Manuscript accepted 27 February 2008
and the impacts on grazers. In: The Ecology and Evolution of

 2008 Blackwell Publishing Ltd/CNRS