Predator and prey views of spider camouflage

Both hunter and hunted fail to notice crab-spiders blending with coloured petals.

Abstract

Crab-spiders (Thomisus onustus) positioned for hunting on flowers disguise themselves by assuming the same colour as the flower, a strategy that is assumed to fool both bird predators and insect prey. But although this mimicry is obvious to the human observer, it has never been examined with respect to different visual systems. Here we show that when female crab-spiders mimic different flower species, they are simultaneously cryptic in the colour-vision systems of both bird predators and hymenopteran prey.

Main

In animal communication, colouring is a compromise between being conspicuous to conspecifics and being poorly visible to predators or prey¹. Female crab-spiders adapt their entire body colour to that of the flowers on which they are trying to hide, a behaviour that is presumed to conceal them from predators and from the visiting pollinators that constitute their main prey².

To appear cryptic to both predators and prey, these spiders must precisely match the flower colour in their respective ranges of colour vision: four cone types, corresponding to ultraviolet (UV)–blue–green–red, for birds, and three (UV–blue–green) for insects^3,4. However, these visual systems differ markedly in their range of sensitivity and number of photoreceptors, making a precise colour match for both unlikely.

We used spectroradiometry to measure the degree of matching (crypsis) of T. onustus collected on the corollae of two flower species (Senecio jacobea and Mentha spicata) growing around Tours, France. To analyse spider crypsis with respect to potential predators and prey on each flower, we computed and compared colour distances in the respective colour spaces of birds and Hymenoptera (see supplementary information).

For birds, spiders matched pink Mentha corollae when viewed through the four-cone colour-vision system (Fig. 1; χ²=9, d.f.=1, P<0.01). Likewise, on Senecio, each spider matched the individual yellow corolla on which it was waiting for prey (χ²=7.5, d.f.=1, P<0.01). But on the Senecio UV–yellow petal background, spiders produced a strong colour contrast that birds were likely to detect (χ²=2.7, d.f.=1, P>0.05). For Hymenoptera, spiders matched the blue-green colour of Mentha corollae in the three-cone colour-vision system (χ²=4, d.f.=1, P<0.05). They also effectively mimicked the individual blue-green colour of Senecio corollae (χ²=7.5, d.f.=1, P<0.01), but contrasted on UV–green petals when seen by Hymenoptera (χ²=2.7, d.f.=1, P>0.05).

Figure 1: Colour contrast of spiders (mean euclidean distances ± s.e.m.) against Mentha corollae (pink bars), Senecio corollae (yellow bars) and Senecio petals (white bars) when viewed by birds and Hymenoptera.

Dashed lines indicate thresholds for colour-contrast detection calculated for birds and Hymenoptera. For details of modelling calculations, see supplementary information.

To detect small targets, birds and bees can use achromatic vision instead of colour contrast^5,6. Spiders were significantly brighter than corollae of Mentha (repeated-measures ANOVA, birds: F_1,17=8.5, P<0.001; Hymenoptera: F_1,17=8.6, P<0.001) and Senecio (birds: F_1,14=35.7, P<0.001; Hymenoptera: F_1,14=46.2, P<0.001), but were darker than Senecio petals (birds: F_1,14=133.7, P<0.001; Hymenoptera: F_1,14=157.3, P<0.001).

Our results indicate that crab-spiders' colour mimicry works successfully on the visual systems of both predator and prey, achromatic vision being potentially more efficient under particular viewing conditions. This aggressive mimicry may vary from species to species, as shown by Misumena vatia⁷, a crab-spider that reduces its chromatic contrast to bees on white flowers (as in Fig. 2), as does T. onustus, but is also able to reduce its achromatic contrast on yellow flowers.

Figure 2

Master of disguise: a crab-spider (left), concealed against a flower's white petals, preys on an unsuspecting bee.

Competing financial interests: declared none.