Abstract

Flowers exhibit characteristics through which they exploit the sensory biases of pollinating insects, and both signaler and receiver benefit from this interaction, either through reproductive service or food reward. However, the preferences of pollinators for certain flower traits such as color or odor might be exploited by predators that target pollinating insects. Crab spiders, Thomisus spectabilis, position themselves on flowers to prey on pollinators such as honeybees, Apis mellifera. We gave both honeybees and crab spiders the choice between two randomly chosen white Chrysanthemum frutescens, including olfactory signals in one experiment and excluding odor in a second experiment. When olfactory signals were included, crab spiders and honeybees clearly preferred the same flower out of a pair. However, agreement level was at chance in the absence of olfactory signals. We also analyzed the visual flower characteristics that might influence the decision of the animals. Neither the size of flowers (diameter of flower and diameter of reproductive flower center) nor the reflectance properties (receptor excitation values in ultraviolet, blue, and green; overall brightness) influenced the choices of crab spiders and honeybees. Therefore, odor seems to be the floral signal that bees use to identify high-quality flowers and that crab spiders exploit to encounter honeybees.

In communication, signals modulate the behavior of their receivers (Bradbury and Vehrencamp, 1998). To maximize the effectiveness of signals, and also to make them detectable among the background noise, many signalers exploit preexisting sensory biases in their receivers, a process called sensory exploitation (Basolo, 1995; Johnstone, 1997). The flowers of many plants provide an excellent example of sensory exploitation, as the floral color matches the insect visual system, which predates the evolution of flowering plants by approximately 400 million years (Chittka, 1996; Chittka et al., 1999; Kevan and Baker, 1983; Lunau, 2000; von Helversen and von Helversen, 1999). The obvious benefit for the plants of being frequently visited by animals is reproductive service, whereas pollinators gain nectar or pollen rewards (Harder et al., 2001). Other well-studied systems of sensory exploitation include intersexual signals designed to increase mating success (Hebets and Uetz, 1999; Rosenthal and Evans, 1998; Sun et al., 2000).

However, signaling systems are often manipulated or exploited, with the result that either the signaler or the receiver does not benefit from the signal. An example of signal manipulation is the nonnectar-producing orchid Disa pulchra, with visual signals that are similar to those of the sympatric and nectar-producing iris Watsonia lepida. The pollinating fly Philoliche aethiopica is unable to distinguish between the two, thus pollinating the orchid without a nectar reward (Johnson, 2000). Predators and parasites also often exploit the sexual signals and sensory biases of potential prey to enhance foraging success (for review, see Zuk and Kolluru, 1998).

Spiders have evolved several strategies to exploit the responses of flying insects to visual signals. One strategy for increasing foraging success is to identify habitats that attract prey. For instance, the nocturnal orb-web spider, Larinioides sclopetarius, locates artificial light sources and constructs webs close to these lights, which also attract high numbers of flying insects (Heiling, 1999). A number of different orb-web spider taxa (Blackledge 1998b; Edmunds and Edmunds, 1986; Herberstein et al., 2000) add conspicuous silk decorations to their completed webs. Various correlational and experimental studies show that webs with decorations attract more prey (Bruce et al., 2001; Craig, 1994; Hauber, 1998; Herberstein, 2000; Tso, 1998a,b; Watanabe 1999). The ultraviolet (UV) reflectance of web decorations may play an important role in sensory exploitation. Craig and Bernard (1990) showed that insects avoided orb-webs without UV reflectance but were attracted to orb-webs that reflected UV light. Watanabe (1999) showed that flies' preference for web decorations disappeared when UV wavelengths of light were removed. Furthermore, Argiope argentata attracts insects by a UV-reflecting body surface (Craig and Ebert, 1994). In contrast, one experimental study with Argiope aurantia found that decorated webs attracted less prey but also suffered less damage from birds (Blackledge and Wenzel 1999). The investigators suggest that the coloration of web decorations embodies a trade-off to make them conspicuous to vertebrates (in order to warn them off, an “advertisement” function) while maximizing crypticity to potential invertebrate prey (Blackledge 1998a; Blackledge and Wenzel 2000). Another experimental study, however, found that isolated web decorations of the same species attract prey (Tso 1998a). The issue remains controversial, and web decorations may have multiple functions, quite possibly differing across species (Eberhard 1990; Herberstein et al., 2000).

Crab spiders do not build webs but ambush pollinating insects on flowers with their raptorial forelimbs (Schmalhofer, 1999). They are attracted by flower odors (Aldrich and Barros, 1995; Krell and Krämer, 1998) and use visual and tactile cues for selecting hunting sites (Greco and Kevan, 1994; Morse, 1988). Field observations of male Misumena vatia revealed that they used cues from both prey and substrate (i.e., flower) as indicators of high-quality hunting sites. They remained longer on flowers in peak condition than on senescent flowers and responded in the absence of female spiders to floral cues indicating the presence of prey (Chien and Morse, 1998). Furthermore, it is generally assumed that crab spiders adjust their color (white, bright yellow, or pink) to match the flower color in an attempt to camouflage (Chittka, 2001; Foelix, 1996; Schmalhofer, 2000; Théry and Casas, 2002). However, some crab spiders are not cryptic to pollinators but actively attract their prey by manipulating floral signals in the UV spectrum of light (Heiling et al., 2003). This unexpected twist challenges our current understanding of this predator/prey system. Rather than passively and randomly selecting flowers to hunt on, crab spiders may identify floral signals aimed at pollinating insects, thus exploiting the signaling communication between flower and pollinator.

The Australian crab spider, Thomisus spectabilis, is an ambush hunter that often captures honeybees (Apis mellifera; A.M. Heiling, K. Cheng, and M.E. Herberstein, personal observations). We tested whether T. spectabilis exploits the communication between flower and pollinator by selecting flowers that honeybees prefer. Pairs of flowers were offered to crab spiders and honeybees for choice. We predicted that the two species tend to choose the same members of a pair. We also investigated the sensory bases for these choices.

METHODS

Study species

Thomisus spectabilis (Thomisidae) were collected from the urban area of Brisbane, Australia, in October and November 2001. Females of this species belong to the largest thomisid crab spiders, and their body color ranges from white to yellow (A.M. Heiling, K. Cheng, and M.E. Herberstein, personal observations). In our experiments, however, we used only white individuals. The spiders were maintained in the laboratory on a 12-h light/12-h dark cycle, the temperature ranging from 20–25°C. They were fed live crickets (Acheta domestica) and Drosophila flies every week and watered daily. Honeybees, Apis mellifera (Apidae), were introduced into Australia from Europe about 200 years ago. We maintained them in an outdoor hive on campus and trained them to visit a feeding station where the experiments were also performed. Between the experimental trials, sucrose solution (25%) was dispensed ad libitum from the feeding station.

Experimental design

The experiments tested whether crab spiders can identify and exploit floral signals intended for pollinators. We performed experiments in which crab spiders and honeybees were given a choice between two randomly selected white daisies. Importantly, the same pair of flowers was offered to both spiders and honeybees. We used Chrysanthemum frutescens of the variety “Summer Angel,” which are a natural substrate of T. spectabilis. Each daisy, crab spider, and honeybee was used only once in the experiments. In the first experiment (N = 60 choice trials), olfactory cues were included. A transparent foil was used to exclude olfactory cues (N = 29 choice trials). The foil (Glad Wrap) is permeable to light greater than 300 nm, with less than 5% attenuation of all wavelengths. Thus, the foil had no effect on the spectral shape of the flowers. Daisies from plants maintained in growth cabinets were randomly selected and placed in black plastic lids (diameter = 4 cm) to provide a constant black background to both spiders and honeybees.

Procedure

Crab spiders were anesthetized by using carbon dioxide and placed into the center of a test surface, consisting of a circular black cardboard (diameter = 17 cm) within an enclosed plastic arena (height = 14.5 cm). The two daisies were arranged vertically with a distance of 4 cm between the daisy centers and their corollas enclosing an angle of 120 degrees. The spider was placed in front of and in between the two flowers at a distance of 5 cm from the flower centers. We considered the spider to have made a choice if it approached and touched a daisy, often positioning itself on the petals of the flower. Smell is known to influence the choice behavior of honeybees (see Laska et al., 1999; Pelz et al., 1997). Thus, after having made a choice, we placed the spider on the rejected daisy for an equal amount of time to exclude any influence of smell generated by the spider before the flower pair was re-tested using honeybees. All spiders made a choice within 30 min of being placed in the enclosed arena. We presented the daisy pair to the honeybees horizontally on a rectangular piece of black cardboard (18 × 13 cm), replacing the feeding station. We classified the landing of a honeybee on one of the two daisies as a choice. All honeybee choices were recorded within 30 min of presenting the flowers.

Analysis of visual cues

To identify potential floral signals that may drive the flower choice of crab spiders and honeybees, we considered chromatic and achromatic cues as well as morphological characteristics. Flower measurements cause damage to flowers, and we thus performed them after the choice experiments were completed. We measured the overall flower diameter and the diameter of the pollen center to the nearest millimeter. Moreover, we measured the spectral reflectance of daisy petals and daisy centers six times, respectively, by using a USB 2000 spectrometer with a PX-2-pulsed xenon light source attached to a PC running OODBase32 software (Ocean Optics Inc.). The flowers were measured without the foil, as the foil only reduced the amount of light reflected without changing the spectral shape. We averaged the six measurements from each area and calculated the receptor excitation values (E) for the photoreceptors (UV, blue, and green) of honeybees (for methods, see Chittka, 1996). The calculations generate the proportion of maximum potential excitation in the UV, blue, and green receptors. Moreover, we calculated the overall brightness of daisy petals and flower centers (Chittka, 1996). These values refer to the visual system of honeybees, as the calculations incorporate the spectral sensitivity functions of animals, which are available for honeybees but not for crab spiders.

RESULTS

Flower choice of crab spiders and honeybees

When given the choice between two randomly selected Chrysanthemum frutescens in the presence of olfactory cues, spiders and bees chose the same daisy in 75% of the cases (n = 60, exact binomial p =.00013, two-tailed) (Figure 1). However, when olfactory cues were excluded, crab spiders and honeybees did not choose the same daisy from a pair at significant frequencies (n = 29, exact binomial p = 1.0, two-tailed; power analysis: 1 − b = 0.768).

Effect of floral cues

The measurements of the spectral reflectance properties of C. frutescens daisies are depicted in Figure 2. The white petals of this variety of daisies reflect light at wavelengths greater than 380 nm, whereas the yellow centers reflect light at wavelengths from slightly less than 500 nm upward. Honeybees possess spectral receptors most sensitive near 345 nm (UV receptors), 440 nm (blue receptors), and 535 nm (green receptors; Briscoe and Chittka, 2001). We have little understanding of the color vision of crab spiders, although various studies on the visual system of diurnal spider species revealed that the spectral sensitivity of their eyes ranged from ultraviolet to green (Barth et al., 1993; De Voe, 1975; Yamashita and Tadeda, 1976, 1983). The daisy center clearly appears green to the honeybees and probably to the spiders, as the receptor excitation values are highest in this region of the light spectrum (Table 1). The UV receptor of bees is more sensitive than is the green receptor by more than an order of magnitude (Chittka et al., 1994). Therefore, although the petals of daisy do not seem to reflect strongly in the UV, the photoreceptor of honeybees is excited in this region of the light spectrum (Table 1). However, the receptor excitation values are higher in the blue and the green, making the petals of daisies strongly chromatic blue-green signals for bees (Table 1).

We analyzed the flower characteristics of chosen and rejected flowers in the presence or absence of olfactory cues, in all cases in which spiders and bees chose the same flower from a pair. The results revealed a remarkable variance in both experiments in the size of daisies both in terms of flower diameter and diameter of the flower center (Table 1). However, a comparison of these size measures between chosen and rejected daisies revealed no differences, indicating that neither honeybees nor crab spiders chose daisies based on flower size.

A calculation of the excitations (E) of the ultraviolet, the blue and the green receptors of honeybees by daisy petals and flower centers revealed that there was variation in the reflectance properties between daisies (Table 1). In both experiments, however, there was no difference between the daisies chosen by spiders and honeybees and the rejected daisies in the receptor excitation values of daisy petals and flower centers for each of the three colors (UV, blue, and green) (Table 1). Moreover, the brightness of petals and centers did not differ between chosen and rejected daisies (Table 1). This indicates that the choice of honeybees and spiders was not affected by the visual signals provided by the daisies regardless of whether olfactory cues were available or not.

DISCUSSION

Our study provides evidence that the crab spiders, Thomisus spectabilis, respond to floral signals in the same manner as honeybees, Apis mellifera, do, thus exploiting the communication between flower and pollinator. When given the choice between two white Chrysanthemum frutescens daisies, crab spiders and honeybees clearly preferred the same flower from a pair. However, this agreement disappeared when olfactory cues were not available to the test animals. Crab spiders use a variety of cues or signals when selecting hunting sites (Chien and Morse, 1998; Greco and Kevan, 1994; Morse, 1988). However, it was not clear whether crab spiders are able to respond to floral traits a priori or whether they arrive at flowers by chance and decide a posteriori to remain on a good flower or to leave a bad flower.

Flowers are used in signaling between the plant and the pollinating insect. Because of the energetic demands of pollinators, their preferences should result in a directional selection for floral signals associated with predictable rewards (Heinrich and Raven, 1972). Pollinating insects are known to use a wide range of signals such as flower size (Neal et al., 1998; Young and Stanton, 1990), preferring large flowers (Ohara and Higashi, 1994) and even large flower models (Møller and Sorci, 1998) over smaller ones. Pollinator rewards (Dafni, 1992), odor (Galen and Kevan, 1983; Wells and Wells, 1985), the symmetry patterns of flowers (Møller and Eriksson, 1995; Møller and Sorci, 1998), and flower color (Chittka and Menzel, 1992; Kevan and Baker, 1983; Real, 1981; Waser and Price, 1981) also affect the decision of pollinating insects, although the brightness of objects does not seem to have any influence (Giurfa et al., 1995; Lunau et al., 1996).

Our experiments revealed that the flower preferences of spiders and honeybees were not influenced by morphological characteristics such as the overall flower size and the size of the yellow reproductive flower center. Similarly, although there was variation in the spectral reflectance of individual flowers, flower reflectance properties did not influence flower choice. Odor seems to be the floral signal that bees use to identify high-quality flowers and that crab spiders exploit to encounter bees. In addition to visual signals indicating the presence (and/or the quality) of pollen and nectar, flowers also produce scents that attract pollinators (Weiss and Lamont, 1997). It is likely that these scents correspond to the presence or absence of nectar or pollen rewards in our case. However, we do not suggest that smell is exclusively or even mainly determining choice behavior in honeybees and crab spiders. In recent literature there is a growing appreciation that flower-visiting animals perceive their foraging grounds as multimodal sensory entities (Gegear and Laverty, 2001), and a visual/olfactory synergism determines the flower choice of bees (Lunau, 1992). For example, in bumblebees (Bombus terrestris), olfactory signals facilitate color discrimination and memory formation (Kunze and Gumbert, 2001; but see Chittka et al., 2001).

We suggest that in the presence of olfactory cues, crab spiders and honeybees perceive the whole set of information on flower characteristics that enables them to generate an indicator of flower quality. Their notion of quality obviously corresponds, resulting in an agreement in flower choice. For the spider, flower quality is expressed by the ability of the flower to attract prey, which is likely to result in a greater chance of capturing a honeybee. For the honeybee, flower quality often translates into a reward but, in our case, may also be linked to a higher risk of predation. Therefore, honeybees suffer apparently from responding to the same floral characteristics as crab spiders do.

In the coevolution of pollinating insects and their predators, selection by predators has been strong enough to maintain antipredatory adaptations in pollinators, and these traits range from stings in bees and noxiousness in butterflies to complex forms of mimicry (Gilbert 1983; Plowright and Owen, 1980; Schmidt, 1990). However, European honeybees that did not coevolve with T. spectabilis do not appear to possess any antipredatory behavior when confronted with their spider predators (Heiling et al., 2003). In contrast, pollinating prey that have coevolved with T. spectabilis demonstrate antipredatory strategies (Heiling and Herberstein, 2004). Furthermore, although learning greatly contributes to the behavior of honeybees (Hammer, 1997; Menzel, 1990), their strong innate preferences for floral traits (Giurfa et al., 1995) facilitate their exploitation by T. spectabilis. It may be that the density of signal exploitation by crab spider is low, and thus, it is unlikely that honeybees have evolved a counter strategy (Grafen, 1990; Hasson, 1994), especially when the benefits for responding to floral signals in terms of rewards are high.

Figure 1

Percentage of flowers that were chosen by both crab spiders and honeybees (solid bars) or chosen only by crab spiders or by honeybees (shaded bars) in the presence or absence of olfactory cues

Figure 1

Percentage of flowers that were chosen by both crab spiders and honeybees (solid bars) or chosen only by crab spiders or by honeybees (shaded bars) in the presence or absence of olfactory cues

Figure 2

Spectral reflectance functions (median values) of the white petals and the yellow flower centers of Chrysanthemum frutescens, measured over the range of 300–700 nm. Reflectance varies from zero (no reflectance) to one (all light is reflected). The data from experiments in which olfactory signals were either included or excluded are pooled (N = 164), as there is no difference across experiments in reflectance properties of daisy petals (Wilcoxon test; Z81 = −1.672, p =.095) and daisy centers (Wilcoxon test; Z81 = −0.887, p =.375)

Figure 2

Spectral reflectance functions (median values) of the white petals and the yellow flower centers of Chrysanthemum frutescens, measured over the range of 300–700 nm. Reflectance varies from zero (no reflectance) to one (all light is reflected). The data from experiments in which olfactory signals were either included or excluded are pooled (N = 164), as there is no difference across experiments in reflectance properties of daisy petals (Wilcoxon test; Z81 = −1.672, p =.095) and daisy centers (Wilcoxon test; Z81 = −0.887, p =.375)

Table 1

Flower characteristics of Chrysanthemum frutescens chosen and rejected by crab spiders and honeybees in the presence or absence of olfactory signals.

 Chosen flowers Rejected flowers Paired t test 
Smell included 
    Diameter (mm) 39.13 ± 5.05 39.4 ± 4.98 t44 = −0.389 
    Center diameter (mm) 10.42 ± 1.31 10.62 ± 1.42 t44 = −0.976 
    Petals 
    EUV 0.708 ± 0.032 0.704 ± 0.028 t44 = 1.21 
    EB 0.906 ± 0.005 0.906 ± 0.005 t44 = 1.358 
    EG 0.877 ± 0.002 0.877 ± 0.002 t44 = 1.308 
    Brightness 2.491 ± 0.038 2.486 ± 0.036 t44 = 1.286 
    Center 
    EUV 0.146 ± 0.14 0.148 ± 0.131 t44 = −0.089 
    EB 0.327 ± 0.14 0.358 ± 0.145 t44 = −1.68 
    EG 0.815 ± 0.008 0.816 ± 0.007 t44 = −1.501 
    Brightness 1.288 ± 0.273 1.322 ± 0.269 t44 = −0.894 
Smell excluded 
    Diameter (mm) 41.67 ± 0.98 41.2 ± 1.78 t14 = 0.301 
    Center diameter (mm) 11.2 ± 1.08 11.13 ± 1.3 t14 = 0.86 
    Petals 
    EUV 0.72 ± 0.01 0.716 ± 0.011 t11 = 0.396 
    EB 0.908 ± 0.002 0.908 ± 0.002 t11 = 0.834 
    EG 0.877 ± 0.002 0.877 ± 0.002 t11 = 0.353 
    Brightness 2.505 ± 0.013 2.501 ± 0.013 t11 = 0.554 
    Center 
    EUV 0.167 ± 0.137 0.132 ± 0.119 t11 = 0.413 
    EB 0.337 ± 0.149 0.332 ± 0.128 t11 = 0.854 
    EG 0.812 ± 0.01 0.812 ± 0.007 t11 = 0.893 
    Brightness 1.316 ± 0.277 1.277 ± 0.236 t11 = 0.463 
 Chosen flowers Rejected flowers Paired t test 
Smell included 
    Diameter (mm) 39.13 ± 5.05 39.4 ± 4.98 t44 = −0.389 
    Center diameter (mm) 10.42 ± 1.31 10.62 ± 1.42 t44 = −0.976 
    Petals 
    EUV 0.708 ± 0.032 0.704 ± 0.028 t44 = 1.21 
    EB 0.906 ± 0.005 0.906 ± 0.005 t44 = 1.358 
    EG 0.877 ± 0.002 0.877 ± 0.002 t44 = 1.308 
    Brightness 2.491 ± 0.038 2.486 ± 0.036 t44 = 1.286 
    Center 
    EUV 0.146 ± 0.14 0.148 ± 0.131 t44 = −0.089 
    EB 0.327 ± 0.14 0.358 ± 0.145 t44 = −1.68 
    EG 0.815 ± 0.008 0.816 ± 0.007 t44 = −1.501 
    Brightness 1.288 ± 0.273 1.322 ± 0.269 t44 = −0.894 
Smell excluded 
    Diameter (mm) 41.67 ± 0.98 41.2 ± 1.78 t14 = 0.301 
    Center diameter (mm) 11.2 ± 1.08 11.13 ± 1.3 t14 = 0.86 
    Petals 
    EUV 0.72 ± 0.01 0.716 ± 0.011 t11 = 0.396 
    EB 0.908 ± 0.002 0.908 ± 0.002 t11 = 0.834 
    EG 0.877 ± 0.002 0.877 ± 0.002 t11 = 0.353 
    Brightness 2.505 ± 0.013 2.501 ± 0.013 t11 = 0.554 
    Center 
    EUV 0.167 ± 0.137 0.132 ± 0.119 t11 = 0.413 
    EB 0.337 ± 0.149 0.332 ± 0.128 t11 = 0.854 
    EG 0.812 ± 0.01 0.812 ± 0.007 t11 = 0.893 
    Brightness 1.316 ± 0.277 1.277 ± 0.236 t11 = 0.463 

The data presented here only represent the subset of trials in which spiders and honeybees chose the same flowers. E measures the proportion of the maximum physiological receptor voltage signals for each photoreceptor (ultraviolet, blue, and green) in the hymenopteran visual system (for methods, see Chittka, 1996) and does not carry a unit. All values are given as mean ± SD, and all reported paired t values indicate p >.05.

We thank Anne Gaskett for help with the experiments, Mark Peterson for help with beehive maintenance, Lars Chittka for help with the analyses of reflectance data, and Justin Marshall for help with the spectrometric measurements. We thank Marc Théry for helpful comments on the manuscript. This work was supported by an Austrian Science Foundation (Fond zur Förderung der Wissenschaftlichen Forschung) grant no. J2249 to A.M.H.

REFERENCES

Aldrich JR, Barros TM,
1995
. Chemical attraction of male crab spiders (Araneae, Thomisidae) and kleptoparasitic flies (Diptera, Milichiidae, and Chloropidae).
J Arachnol
 
23
:
212
-214.
Barth FG, Makagawa T, Eguchi E,
1993
. Vision in the ctenid spider Cupiennius salei: spectral range and absolute sensitivity (ERG).
J Exp Biol
 
181
:
187
-193.
Basolo AL,
1995
. Phylogenitic evidence for the role of a pre-existing bias in sexual selection.
Proc R Soc Lond B
 
259
:
307
-311.
Blackledge TA,
1998
. Signal conflict in spider webs driven by predators and prey.
Proc R Soc Lond B
 
265
:
1991
-1996.
Blackledge TA,
1998
. Stabilimentum variation and foraging success in Argiope aurantia and Argiope trifasciata (Araneae, Araneidae).
J Zool
 
246
:
21
-27.
Blackledge TA, Wenzel JW,
1999
. Do stabilimenta in orb webs attract prey or defend spiders?
Behav Ecol
 
10
:
372
-376.
Blackledge TA, Wenzel JW,
2000
. The evolution of cryptic spider silk: a behavioral test.
Behav Ecol
 
11
:
142
-145.
Bradbury JW, Vehrencamp SL,
1998
. Principles of animal communication. Sunderland, Massachusetts: Sinauer.
Briscoe AD, Chittka L,
2001
. The evolution of colour vision in insects.
Ann Rev Entomol
 
46
:
471
-510.
Bruce MJ, Herberstein ME, Elgar MA,
2001
. Signalling conflict between prey and predator attraction.
J Evol Biol
 
14
:
786
-794.
Chien SA, Morse DH,
1998
. The role of prey and flower quality in the choice of hunting sites by adult male crab spiders Misumena vatia (Araneae, Thomisidae).
J Arachnol
 
26
:
238
-243.
Chittka L,
1996
. Optimal sets of colour receptors and opponent processes for coding of natural objects in insect vision.
J Theor Biol
 
181
:
179
-196.
Chittka L,
2001
. Camouflage of predatory crab spiders on flowers and the colour perception of bees.
Entomol Gen
 
25
:
181
-187.
Chittka L, Menzel R,
1992
. The evolutionary adaptation of flower colours and the insect pollinators' colour vision.
J Comp Physiol A
 
171
:
171
-181.
Chittka L, Shmida A, Troje N, Menzel R,
1994
. Ultraviolet as a component of flower reflections, and the colour perception of hymenoptera.
Vision Res
 
34
:
1489
-1508.
Chittka L, Spaethe J, Schmidt A, Hickelsberger A.,
2001
. Adaptation, constraint and chance in the evolution of flower color and pollinator color vision. In: Cognitive ecology of pollination (Chittka L, Thomson JD, eds). Cambridge: Cambridge University Press; 106–126.
Chittka L, Thomson JD, Waser NM,
1999
. Flower constancy, insect psychology, and plant evolution.
Naturwissenschaften
 
86
:
361
-377.
Craig CL,
1994
. Limits to learning: effects of predator pattern and colour on perception and avoidance-learning by prey.
Anim Behav
 
47
:
1087
-1099.
Craig CL, Bernard GD,
1990
. Insect attraction to ultraviolet-reflecting spiders and web decorations.
Ecology
 
71
:
616
-623.
Craig CL, Ebert K,
1994
. Colour and pattern in predator-prey interactions: the bright body colours and patterns of a tropical orb-spinning spider attract flower-seeking prey.
Funct Ecol
 
8
:
616
-620.
Dafni A,
1992
. Pollination ecology: a practical approach. Oxford: Oxford University Press.
De Voe R,
1975
. Ultraviolet and green receptors in principal eyes of jumping spiders.
J Gen Physiol
 
66
:
193
-207.
Eberhard WG,
1990
. Function and phylogeny of spider webs.
Annu Rev Eco Syst
 
21
:
341
-372.
Edmunds J, Edmunds M,
1986
. The defensive mechanisms of orb weavers (Araneae: Araneidae) in Ghana, West Africa. In: Proceedings of the 9th International Congress of Arachnology, Panama, 1983 (Eberhard WG, Lubin YD, Robinson BC, eds). Washington: Smithsonian Institute Press; 73–89.
Foelix RF,
1996
. Biology of spiders. New York: Oxford University Press.
Galen C, Kevan PG,
1983
. Bumblebee foraging and floral scent dimorphism: Bombus kirbyellus Curtis (Hymenoptera: Apidae) and Polemonium viscosum Nutt. (Polemoniaceae).
Can J Zool
 
61
:
1207
-1213.
Gegear RJ, Laverty TM,
2001
. The effect of variation among floral traits on the flower constancy of pollinators. In: Cognitive ecology of pollination (Chittka L, Thomson JD, eds). Cambridge: Cambridge University Press; 1–20.
Gilbert LE,
1983
. Coevolution and mimicry. In: Coevolution (Futuyma DJ, Slatkin M, eds). Sunderland, Massachusetts: Sinauer; 263–281.
Giurfa M, Núñez J, Chittka L, Menzel R,
1995
. Colour preferences of flower-naive honeybees.
J Comp Physiol A
 
177
:
247
-259.
Grafen A,
1990
. Biological signals as handicaps.
J Theor Biol
 
144
:
517
-546.
Greco CF, Kevan PG,
1994
. Contrasting patch choosing by anthophilous ambush predators: vegetation and floral cues for decision by a crab spider (Misumena vatia) and males and females of an ambush bug (Phymata americana).
Can J Zool
 
72
:
1583
-1588.
Hammer M,
1997
. The neural basis of associative reinforcement learning in honeybees.
Trends Neurosci
 
20
:
245
-252.
Harder LD, Williams NM, Jordan CY, Nelson WA,
2001
. The effects of floral design and display on pollinator economics and pollen dispersal. In: Cognitive ecology of pollination (Chittka L, Thomson JD, eds). Cambridge: Cambridge University Press; 297–317.
Hasson O,
1994
. Cheating signals.
J Theor Biol
 
167
:
223
-238.
Hauber ME,
1998
. Web decorations and alternative foraging tactics of the spider Argiope appensa.
Ethol Ecol Evol
 
10
:
47
-54.
Hebets EA, Uetz GW,
1999
. Female responses to isolated signals from multimodal male courtship displays in the wolf spider genus Schizocosa (Araneae: Lycosidae).
Anim Behav
 
57
:
865
-872.
Heiling AM,
1999
. Why do nocturnal orb-web spiders (Araneidae) search for light?
Behav Ecol Sociobiol
 
46
:
43
-49.
Heiling AM, Herberstein ME,
2004
. Predator–prey coevolution: Australian native bees avoid their spider predators.
Proc R Soc Lond B
 
(Biol Lett)
:
(in press)
.
Heiling AM, Herberstein ME, Chittka L,
2003
. Crab spiders manipulate flower signals.
Nature
 
421
:
334
.
Heinrich B, Raven P,
1972
. Energetics and pollination ecology.
Science
 
176
:
597
-602.
Herberstein ME,
2000
. Foraging behaviour in orb-web spiders (Araneidae): do web decorations increase prey capture success in Argiope keyserlingi Karsch, 1878?
Aust J Zool
 
48
:
217
-223.
Herberstein ME, Craig CL, Elgar MA,
2000
. Foraging strategies and feeding regimes: web and decoration investment in Argiope keyserlingi Karsch (Araneae: Araneidae).
Evol Ecol Res
 
2
:
69
-80.
Johnson SD,
2000
. Batesian mimicry in the non-rewarding orchid, Disa pulchra and its consequences for pollinator behaviour.
Biol J Linn Soc
 
71
:
119
-132.
Johnstone RA,
1997
. The tactics of mutual mate choice and competitive search.
Behav Ecol Sociobiol
 
40
:
51
-59.
Kevan PG, Baker HG,
1983
. Insects as flower visitors and pollinators.
Annu Rev Entomol
 
28
:
407
-453.
Krell FT, Krämer F,
1998
. Chemical attraction of crab spiders (Araneae, Thomisidae) to a flower fragrance component.
J Arachnol
 
26
:
117
-119.
Kunze J, Gumbert A,
2001
. The combined effect of color and odor on flower choice behavior of bumble bees in flower mimicry systems.
Behav Ecol
 
12
:
447
-456.
Laska M, Galizia CG, Giurfa M, Menzel R,
1999
. Olfactory discrimination ability and odor structure-activity relationships in honeybees.
Chem Senses
 
24
:
429
-438.
Lunau K,
1992
. Innate recognition of flowers by bumblebees: orientation of antennae to visual stamen signals.
Can J Zool
 
70
:
2139
-2144.
Lunau K,
2000
. The ecology and evolution of visual pollen signals.
Plant Syst Evol
 
222
:
89
-111.
Lunau K, Wacht S, Chittka L,
1996
. Colour choices of naive bumble bees and their implications for colour perception.
J Comp Physiol A
 
178
:
477
-489.
Menzel R,
1990
. Learning, memory and “cognition” in honey bees. In: Neurobiology of comparative cognition (Kesner RP, Olton DS eds). Hilldale, New Jersey: Erlbaum; 237–292.
Møller AP, Eriksson M,
1995
. Pollinator preference for symmetrical flowers and sexual selection in plants.
Oikos
 
73
:
15
-22.
Møller AP, Sorci G,
1998
. Insect preference for symmetrical artificial flowers.
Oecologia
 
114
:
37
-42.
Morse DH,
1988
. Cues associated with patch-choice decisions by foraging crab spiders Misumena vatia.
Behaviour
 
107
:
297
-313.
Neal PR, Dafni A, Giurfa M,
1998
. Floral symmetry and its role in plant-pollinator systems: terminology, distribution, and hypotheses:.
Ann Rev Ecol Syst
 
29
:
345
-373.
Ohara M, Higashi S,
1994
. Effects of inflorescence size in visits from pollinators and seed set of Corydalis ambigua (Papaveraceae).
Oecologia
 
98
:
25
-30.
Pelz C, Gerber B, Menzel R,
1997
. Odorant intensity as a determinant for olfactory conditioning in honeybees- roles in discrimination, overshadowing and memory consolidation.
J Exp Biol
 
200
:
837
-847.
Plowright RC, Owen RE,
1980
. The evolutionary significance of bumblebee color patterns: a mimetic interpretation.
Evolution
 
34
:
622
-637.
Real LA,
1981
. Uncertainty and pollinator-plant interactions: the foraging behavior of bees and wasps on artificial flowers.
Ecology
 
62
:
20
-26.
Rosenthal GG, Evans CS,
1998
. Female preference for swords in Xiphophorus helleri reflects a bias for large apparent size.
Proc Natl Acad Sci USA
 
95
:
4431
-4436.
Schmalhofer VR,
1999
. Thermal tolerances and preferences of the crab spiders Misumenops asperatus and Misumenoides formosipes (Araneae, Thomisidae).
J Arachnol
 
27
:
470
-480.
Schmalhofer VR,
2000
. Diet-induced morphological color changes in juvenile crab spiders (Araneae, Thomisidae).
J Arachnol
 
28
:
56
-60.
Schmidt JO,
1990
. Hymenopteran venoms: striving towards the ultimate defense against vertebrates. In: Insect defenses (Evans DL, Schmidt JO, eds). Albany, New York: State University of New York Press; 387–419.
Sun L, Wilczynski W, Rand AS, Ryan MJ,
2000
. Trade-off in short- and long-distance communication in túngara (Physalaemus pustulosus) and cricket (Acris crepitans) frogs.
Behav Ecol
 
11
:
102
-109.
Théry M, Casas J,
2002
. Predator and prey views of spider camouflage.
Nature
 
415
:
133
.
Tso IM,
1998
. Isolated spider web stabilimentum attracts insects.
Behav
 
135
:
311
-319.
Tso IM,
1998
. Stabilimentum-decorated webs spun by Cyclosa conica (Araneae, Araneidae) trapped more insects than undecorated webs.
J Arachnol
 
26
:
101
-105.
von Helversen D, von Helversen O,
1999
. Acoustic guide in bat-pollinated flowers.
Nature
 
398
:
759
-760.
Waser NM, Price MV,
1981
. Pollinator choice and stabilizing selection for flower color in Delphinium nelsonii.
Evolution
 
35
:
376
-390.
Watanabe T,
1999
. Prey-attraction as a possible function of the silk decoration of the ulobroid spider Octonoba sybotides.
Behav Ecol
 
10
:
607
-611.
Weiss MR, Lamont BB,
1997
. Floral color change and insect pollination: a dynamic relationship.
Israel J Plant Sci
 
45
:
185
-199.
Wells PH, Wells H,
1985
. Ethological isolation of plants, 2: odour selection by honeybees.
J Apicult Res
 
24
:
86
-92.
Yamashita S, Tadeda H,
1976
. Spectral sensitivities of jumping spider eyes.
J Comp Physiol
 
105
:
29
-41.
Yamashita S, Tadeda H,
1983
. Cerebral photosensitive neurons in the orb weaving spiders, Argiope bruennichii and A. amoena.
J Comp Physiol
 
150
:
467
-472.
Young HJ, Stanton ML,
1990
. Influences of floral variation on pollen removal and seed production in wild radish.
Ecology
 
71
:
536
-547.
Zuk M, Kolluru GR,
1998
. Exploitation of sexual signals by predators and parasitoids.
Q Rev Biol
 
73
:
415
-438.