Arthropod-Plant Interactions (2007) 1:3–16 DOI 10.1007/s11829-006-9000-1 REVIEW PAPER Visual ecology of aphids—a critical review on the role of colours in host finding Thomas Felix Do¨ ring Æ Lars Chittka Received: 10 November 2006 / Accepted: 15 December 2006 / Published online: 2 March 2007 Ó Springer Science+Business Media B.V. 2007 Abstract We review the rich literature on behavio- far-reaching assumptions on aphid responses to colours ural responses of aphids (Hemiptera: Aphididae) to that are not likely to hold. Finally we also discuss the stimuli of different colours. Only in one species there implications for developing and optimising strategies are adequate physiological data on spectral sensitivity of aphid control and monitoring. to explain behaviour crisply in mechanistic terms. Because of the great interest in aphid responses to Keywords Aphid Á Aphididae Á Autumn colouration Á coloured targets from an evolutionary, ecological and Behaviour Á Colour Á Colour opponency Á Hemiptera Á applied perspective, there is a substantial need to Host finding Á Pest control Á Vision expand these studies to more species of aphids, and to quantify spectral properties of stimuli rigorously. We show that aphid responses to colours, at least for some Introduction species, are likely based on a specific colour opponency mechanism, with positive input from the green domain Everyone who cares for plants knows aphids (Hemip- of the spectrum and negative input from the blue and/ tera: Aphididae). These small and gentle insects with or UV region. We further demonstrate that the usual famously powerful reproductive potential are of im- yellow preference of aphids encountered in field mense importance both in agriculture and horticulture experiments is not a true colour preference but in- (Miles 1989), as well as in non-agricultural ecosystems volves additional brightness effects. We discuss the (Stadler et al. 1998; Wimp and Whitham 2001). They implications for agriculture and sensory ecology, with are major pests in many crop and fruit species, because special respect to the recent debate on autumn leaf they remove plant assimilates (Miles 1989), induce galls colouration. We illustrate that recent evolutionary (e.g. Brown et al. 1991), transmit plant viruses (Sylvester theories concerning aphid–tree interactions imply 1989), and excrete honey dew that acts as a growing medium for unwanted fungi (Rabbinge et al. 1981; Fokkema et al. 1983). However, as producers of honey- Handling Editor: Robert Glinwood dew, some aphid species also provide a resource eagerly sought by bee-keepers for the production of premium T. F. Do¨ ring Á L. Chittka forest honey (Bauer-Dubau and Scheurer 1993). School of Biological and Chemical Sciences, Queen Mary University of London, Mile End Road, E1 4NS London, UK The interest for the host finding behaviour of aphids, and for the biotic and abiotic factors that drive it, was L. Chittka e-mail: [email protected] often rooted in the area of virus vector control. For example, Volker Moericke, who in the 1950s and 1960s T. F. Do¨ ring (&) was the most productive researcher in investigating Aphid Biology Group, Division of Biology, Faculty of aphid responses to colours and the role of colours for Natural Sciences, Imperial College London, Silwood Park Campus, Ascot, Berkshire SL5 7PY, UK host finding in aphids, had begun his career with a e-mail: [email protected] thesis on the colonisation of potato by the aphid 123 4 Arthropod-Plant Interactions (2007) 1:3–16 Myzus persicae with the motivation to contribute to the The physiological basis for the perception of colours progress of potato virus control (Moericke 1941). A in aphids later paper on the response of alighting aphids to colours (Moericke 1952) was embedded in a potato The basic receptor units for the perception of light are virus control project. photoreceptor cells, which, in insects, are located in the Host finding in alate (winged) aphids is a complex retina of the compound eye and in the ocelli (Menzel behaviour that is closely linked to migration and the 1979; Menzel and Backhaus 1991; Briscoe and Chittka function of dispersal. The classic and often-cited con- 2001). Additional extraocular photoreceptors which cept of host finding behaviour in aphids (Moericke serve circadian clocks have also been found in aphids 1955a) distinguishes four overlapping behavioural (Hardie and Nunes 2001) but are disregarded in this stages (the teneral period; the distance flight or paper. A photoreceptor acts as a photon counter, so migration flight; the attacking flight, when the aphid that it cannot distinguish between photons of different repeatedly lands and probes on plants; and the final wavelengths. However, the light absorption of photo- settling period), each corresponding to a certain receptor pigments depends on the wavelength, so that behavioural ‘mood’ (motivation). For a different con- the strength of the response from a cell containing the cept of aphid host finding behaviour, see Kennedy pigment varies with wavelength for stimuli of equal (1966) and works cited there. intensity. This wavelength dependency of the photo- Many stimuli and environmental conditions have receptor’s capability to count photons can be plotted as been found to influence flight (Broadbent 1949; John- its spectral sensitivity function (Fig. 1). Thus, a bright son 1958; Kring 1972), and landing or probing response light with a high number of photons at a wavelength far during the ‘attacking flight’, including tactile (Hennig away from the sensitivity peak may cause the same 1963), visual (see below) and olfactory cues. Olfactory physiological response in the photoreceptor cell as a stimuli, such as plant volatiles, had long been consid- dim light at the peak sensitivity wavelength. A system ered to be of low importance (Kennedy 1950; Kennedy based on only one type of receptor could therefore not et al. 1959a, b), but it is now clear that odours play an distinguish colours. Many insects studied so far have important role in host finding of aphids (e.g., Petterson three types of photoreceptor cells in their compound 1970; Chapman et al. 1981; Hardie et al. 1994; Powell eyes, with one type showing maximal sensitivity in the et al. 1995; Park et al. 2000). Interactions between green, a second type with the peak in the blue and the olfactory and visual stimuli have also been reported third type with a peak in the ultraviolet (Briscoe and (Dilawari and Atwal 1989; Hardie et al. 1996) and this Chittka 2001). In fact, it has been suggested that the area clearly deserves further exploration. ancestor of pterygote insects was equipped with these Additional interest in the role of colours in host three types of photoreceptors (Chittka 1996a). Many selection of aphids was recently created by the debate species of insects, however, show variations from this on autumn leaf colouration as a potential signal or cue basic trichromatic system, with some having four or to aphids (e.g., Sinkkonen 2006), initiated by a paper more spectral receptor types (Arikawa et al. 1987; from Hamilton and Brown (2001); for a review see Briscoe and Chittka 2001). Manetas (2006). However, in this debate, the per- spective of colour perception by the aphids appears to have been largely neglected. Unfortunately, the rich 1.0 literature on behavioural responses of aphids to col- ours has not entered the discussion of the adaptive 0.8 significance of autumn leaf colouration yet. Moreover, the sensory aspects, especially concerning the progress 0.6 made in physiology and conceptualisation of colour 0.4 vision have largely been ignored in the agricultural (as well as the evolutionary) literature on aphid responses 0.2 to colour. We therefore describe the theoretical and normalised sensitivity(%) technical concepts necessary when setting up or inter- 0.0 preting colour vision experiments with aphids. Thus, 300 400 500 600 700 this review may serve as a bridge between the agri- wavelength (nm) cultural and the biological shore and will hopefully give Fig. 1 Tentative spectral sensitivities of three modelled types of both ecologists and agricultural entomologists new photoreceptors of Myzus persicae. Model after Stavenga et al. insights into the intriguing visual world of aphids. (1993) 123 Arthropod-Plant Interactions (2007) 1:3–16 5 There is still very limited information on photore- From photoreceptor spectral sensitivity to behavioural ceptor sensitivities in herbivorous insects, as already responses to colour lamented by Prokopy and Owens (1983). A reason for the scarcity of physiological information on aphids, in When the spectral sensitivities of an animal’s photo- particular, is that the appropriate techniques are diffi- receptors are known, it is possible to quantitatively cult to apply because the animals are so small and soft, predict the signal that these receptors will send to the which makes inserting microelectrodes into single cells brain when viewing a particular target. When light of their eyes exceptionally difficult. reflected from an object (a stimulus s) meets the The green peach aphid (Myzus persicae Sulzer, aphid’s eye, the excitation E of each photoreceptor R Hemiptera: Aphididae) is the only aphid species can be calculated, if the reflectance spectrum Is(k)of that has been physiologically tested for spectral the stimulus; the sensitivity function SR(k) of the sensitivity so far (Kirchner et al. 2005), using extra- photoreceptor; the illumination spectrum D(k); and the cellular recordings (ERG). The overall peak sensi- reflectance spectrum Ib(k) of the background b against tivity of the eye was found at 530 nm. This work which the stimulus appears are known; then clearly showed that there are more than two photo- receptors in this aphid species. Besides a putative ER ¼ PR=ðPR þ 1Þ with ð1Þ green receptor with a maximum sensitivity at Z Z 530 nm, a UV receptor with a peak at 320–330 nm P ¼ I ðkÞS ðkÞDðkÞdk= I ðkÞS ðkÞDðkÞdk ð2Þ was found, and a blue receptor with a peak sensi- R s R b R tivity somewhere between 440 nm and 480 nm was also necessary to explain the obtained results.