S45.4: UV vision and its functions in birds

School of Biological Sciences, University of Bristol, Woodland Road, Bristol BS8 1UG, UK, fax 44 117 925 7374, e-mail I.Cuthill@bris.ac.uk; J.C.Partridge@bris.ac.uk; Andy.Bennett@bris.ac.uk

Humans are generally blind to ultraviolet (UV) wavelengths, due to the high absorption of wavelengths below 400 nm by the ocular media preceding the retina . As we cannot see UV, there has been a tendency to regard species with UV vision as unusual, but there is an increasing realisation that UV vision is taxonomically widespread in both vertebrates and invertebrates . The consequences of this are twofold. First, the diversity raises the question of the adaptive significance and evolutionary origins of different colour vision systems . Second, it suggests that investigations of the evolution of animal and plant coloration, and of visually guided behaviours, must be addressed with reference to the colour cognition of the species responding to the visual stimuli . In such species human colour vision is largely irrelevant. In this review, we focus on birds, on account of their long popularity in tests of evolutionary hypotheses involving colour and their excellent colour vision including the UV waveband.

Colour vision is a mechanism for discriminating between objects based on differences in the relative amounts of different wavelengths of light emitted, transmitted, or reflected from their surfaces . However, a colour is not a pure wavelength, nor can it be equated directly with spectral composition; it is a psychophysical abstraction, entirely dependent upon the receptors and nervous system of the organism receiving the light. The sensation of colour stems from comparisons of the output from photoreceptors sensitive to different wavelengths of light. Therefore, species with different receptor sensitivities, or comparing receptors in different ways, will have differing perceptions of the same surfaces, under the same lighting conditions . There are three main differences between avian and human colour vision: (1) a wider spectral range (the near-UV, as well as human-visible, wavebands), (2) more retinal cone types, and (3) the possession of coloured oil droplets which filter the light entering the cones. The differences between the design of the avian and human eye suggest that it is unwise to extrapolate from human colour experience to that of birds . However, this is precisely what has been done in the majority of behavioural and evolutionary studies involving the colour world of birds . This is worrying, as birds have probably been the most popular taxon in tests of major evolutionary theories about sexual selection and signalling (feather colours), as well as mimicry, crypsis and warning coloration (e.g. in insects, where birds are the predators). So, understanding avian colour vision is not only an important task for the visual physiologist, it is essential for the behavioural and evolutionary ecologist.

The first behavioural evidence for UV vision in birds came from operant discrimination experiments on hummingbirds and pigeons . It took 20 more years for the first avian UV-sensitive visual pigment to be characterised by microspectrophotometry , but electrophysiological and behavioural experiments had already suggested UV vision to be widespread . To date, positive behavioural, electrophysiological or microspectrophotometric evidence exists for over 35 species of birds , with negative evidence for two owl species , perhaps not surprising given their nocturnal habits . Microspectrophotometry indicates that most birds possess retinal cones with peak absorption in the UV-violet region of the spectrum (see below) which will confer high UV sensitivity. In addition, all bird species so far investigated have been shown to have three types of single cone spanning the human-visible waveband (shortwave-, mediumwave- and longwave- sensitive; see below) in addition to double cones . The latter are the most common cone class in diurnal birds , but their function is still disputed and they may not be involved in colour vision (see below). Nevertheless, with more than three types of single cone, birds may have the potential for higher dimensional colour vision than our own . The sensation of ‘brightness’, the achromatic dimension of colour, stems from summing output from the different retinal receptors, whilst the perception of ‘hue’ results from contrasts in output between receptor types. Trichromatic humans compare long and medium wavelength cone outputs (the ‘red-green’ colour channel), and short wavelength cone output is compared to the summed output of long and medium wavelength cones (the ‘yellow-blue’ channel). Bee vision, being likewise trichromatic, differs from our own only in the actual wavelengths of the short, medium and long wavelength receptors being compared. Bees also have two orthogonal hue dimensions, although these differ from our own . However birds, with more than three receptor types, have the potential to see hues unknown to any trichromat , and, with four types of single cone most species are likely to be tetrachromatic. However, experiments demonstrating this unequivocally are lacking (see below).

The adaptive advantages of tetrachromacy are at present poorly understood and mammals, for instance, are generally dichromatic. Trichromacy has only recently evolved in various Old World primate groups by duplication of the ancestral green cone pigment to form the red . Indeed, it is possible that human trichromacy is suboptimal, even for the visual tasks for which it probably evolved . Thus there is no compelling evidence to suggest that human colour vision is particularly good, few studies which have investigated correlations between avian and human-perceived colours and, a priori, a case for believing that there could be marked human-bird differences .

As in most vertebrates, the visual photoreceptors of the avian retina are divisible into two distinct types which can be distinguished both on morphological and physiological grounds: rods which subserve scotopic vision at low light levels, and cones for photopic vision during daylight . Most birds are strongly diurnal and have retinae dominated by cones. In the dozen or so neognathus species that have been examined in detail, this cone compliment usually consists of four types of single cone, and one type of double cone in which the dissimilar members of the pair are in close contact . Such diversity of photoreceptors is not unusual among vertebrates and equally complex retinae are found in many fishes, and, superficially at least, the retinae of birds are virtually indistinguishable from those of some diurnal reptiles such as freshwater turtles .

All vertebrate visual photoreceptors contain visual pigments, coloured compounds which occur at extremely high concentrations (e.g. 2.2 x10^-3 mol l^-1 , equivalent to 1.3 x 10⁸ molecules per cell, in the rods of the European eel; Wood & Partridge 1993) in the extensive lipid membranes of their most scleral parts, the outer segments. Visual pigments consist of a protein moiety (opsin), all of which are members of the large class of G-protein-linked cell membrane receptors, and a chromophore that, in birds, is always 11-cis-retinal. All visual pigments have bell-shaped absorbance spectra and in birds the spectral location of peak absorbance (l _max) is determined only by the amino acid sequence of the different opsins present in the rod or cone outer segments. Phylogenetic analyses of opsin sequences have revealed that vertebrates have four classes of visual pigment: a longwave sensitive (LWS), a midwave sensitive (MWS), a shortwave sensitive (SWS), and an extreme shortwave sensitive (UVS/VS) pigment. All four types are found in cones, and avian rods contain pigments that group with the MWS cone pigments . In contrast to the mammals, which have lost some representatives of the ancestral vertebrate cone pigments, all four types of opsin are found in birds. The taxon has thus retained a plesiomorphic feature common to the ancestors of all tetrapods and this complexity of retinal visual pigments presumably arose early in the evolution of fishes .

Absorption of light by visual pigments initiates the transduction of photon energy into the movement of ions in the central nervous system and thus constitutes the first step in the process of vision. For this reason the absorption spectra of visual pigments are fundamental to their function; only photons that are absorbed play any part in the visual process. The measurement of visual pigment absorption spectra relies on microspectrophotometry (MSP) of the outer segments of single cells which in birds can be extremely small (e.g. 1.5 m m diameter, by 5 m m long), thus imposing severe technical difficulties. For this reason avian visual pigments have been measured in very few species and it is likely that at least some of the existing data will be revised as more measurements are made. Nevertheless, it seems clear that the retinae of the least derived avian species have a rod with a l _max value close to 505 nm; double cones both containing a visual pigment based on the LWS opsin and having a l _max close to 565 nm; and single cones with the same LWS pigment, a MWS pigment (l _max the same as the rod pigment), a SWS pigment with a l _max between 430 and 460 nm, and a UVS/VS pigment with a l _max between 355 and 420 nm. Values for the European Starling Sturnus vulgaris appear typical of passerines, this species having a rod with a l _max of 503 nm, double cones with l _max at 563 nm, and four types of single cone with l _max at 563, 504, 449, and 362 nm .

Although this pattern may be representative of an ancestral avian retina, there is considerably interspecific variation and, for instance, some bird species apparently lack the full complement of avian visual pigments. Even among those birds with a "typical" set of visual pigments, there is interspecific variation in l _max, and this is particularly clear in the case of pigments conferring shortwave sensitivity, including that to ultraviolet (UV) wavelengths. Available data suggest that the UVS/VS class of visual pigments have a considerable range of l _max but there is some indication that the values fall into two groups : species having l _max values in the UV (e.g. Budgerigar Melopsittacus undulatus, l _max 371 nm; Zebra Finch Taeniopygia guttata, l _max ca 370 nm; Pekin Robin Leiothrix lutea l _max 355 nm; European Starling, l _max 362 nm), and other species having UVS/VS pigments with l _max values in the range 400-420 nm (e.g. Humboldt Penguin Spheniscus humboldti l _max 403 nm; Manx Shearwater Puffinus puffinus 402 nm, Mallard Duck Anas platyrhynchos 420 nm; Domestic Fowl Gallus gallus 418 nm; Japanese Quail Coturnix japonica 419 nm; Pigeon Columba livia 409 nm).

It is tempting to propose ecological or phylogenetic explanations for the apparent division of the UVS/VS visual pigments but this may well prove to be merely an artefact of the limited available data. Nevertheless, it is clear that bird species will vary in their sensitivity to UV wavelengths due to their visual pigments alone. There are, however, potentially more important factors limiting sensitivity to the UV, including intra-ocular pigments in the cornea, lens and ocular media, and intracellular filters, in the form of coloured oil droplets in the photoreceptors themselves . All visual pigments, even those with a l _max at long wavelengths have considerable absorption at short wavelengths due to the b -peak of the visual pigment absorption spectrum , and thus all vertebrates have the potential for UV vision. That many (e.g. humans) lack UV vision is due to the evolution of UV-blocking pigments in, particularly, the lens, rather than an inherent insensitivity of retinal photoreceptors to the UV. In birds the most obvious spectral filters are in the form of carotenoid-containing oil droplets located in the vitreal part of the inner segments of cones. Photoreceptor oil droplets occur in many vertebrates, including some fish, amphibians, reptiles and even marsupial mammals , but it is only in the reptiles and birds that they are densely pigmented. A typical avian retina, like that of a freshwater turtle, thus has several types of oil droplet, each corresponding to the different cone types, and which appear variously as colourless, yellow, orange or red to the human observer . These droplets act as cut-off filters, transmitting only long wavelengths, and the spectral location of the cut-off is both highly correlated with visual pigment l _max and shows considerable interspecific variation which can be related to the visual ecology of different avian species . Coloured oil droplets alter the spectral sensitivity of the cone as a whole, narrowing the waveband to which the cone responds, and reducing overlap in the wavelengths to which different cone types are sensitive (see below). The adaptive significance of oil droplets is still debated, but both tetrachromacy and coloured oil droplets may improve discriminability of the spectra reflected from natural objects and may also improve colour constancy .

Except for the oil droplets associated with the UVS/VS cone pigments, which MSP has shown contain no carotenoids and are transparent into the UV, all oil droplets block UV wavelengths more or less completely . UV vision in birds can thus only be mediated via the UVS/VS single cones, or by the accessory member of the double cones (which contains a LWS pigment but generally lacks a functioning oil droplet and which are therefore potentially UV sensitive), although scotopic UV sensitivity by the rods is entirely possible.

Some birds, e.g. the Tawny Owl , appear to lack the UVS/VS cone class, but in the case of most avian species it is a reasonable assumption that the breadth of their visible spectrum will extend from the UV to wavelengths in the far-red as least as long as human vision allows. The latter has no definite spectral limit, vision being determined by intensity as well as wavelength (e.g. humans can beyond 700 nm, which is often stated to be the longwave limit to human vision). In contrast, the shortwave cut-off is well defined due to the absorbing properties of intra-ocular tissue, particularly the lens. Proteins in the lens cut off steeply at short wavelengths, 50% absorptance occurring between 320 and 350 nm (depending on lens diameter) due to the presence of aromatic amino acids in lenticular proteins, and thus no birds are likely to be able to see wavelengths shorter than ca. 310 nm . In many vertebrates, shortwave sensitivity is further reduced by UV-absorbing pigments in the ocular tissue, particularly the lens . Very few measurements of the spectral transmissions of avian ocular media and pre-retinal tissue have been made, but such short-wave absorbing pigments appear to be generally absent in birds. Thus even species such as the chicken, which has a UVS/VS pigment with a

l _max at 418 nm, will have considerable UV sensitivity because its lens is transparent in the UV (wavelength of 50% transmission 335 nm; R. H. Douglas pers. comm.). Indeed, to date, the only bird that appears to have lenticular shortwave transmission reduced by pigments is the Mallard Duck, in which the wavelength of 50% transmission of the ocular media is ca. 380 nm . It is, however, very likely that other cases will be encountered with further study, and in such species sensitivity to the UV will be considerably lower than that of most birds.

Birds are well known to be able to make visual discriminations on the basis of the spectral radiance of observed objects independent of brightness, and they thus have colour vision . Colour vision requires that at least two photoreceptor types, with different spectral sensitivities, are present in the retina, and birds, typically with six types of retinal cone in addition to a rod, clearly have retinae capable of supporting a complex colour vision system. Indeed, six types of retinal cone confers the potential for hexachromatic (six-dimensional) colour space. To demonstrate the dimensionality of a colour vision system requires that appropriate colour mixing experiments be conducted and, to date, published studies of these are rare in birds (e.g. Palacios & Verela 1992). However, more studies are underway (Butler & Goldsmith, pers. comm.). It is completely erroneous to assume that the number of photoreceptors directly indicates the dimensionality of colour space, this being a property of the neural interactions of photoreceptor output, as well as the number of photoreceptor types , and it is highly unlikely that birds will be shown to have hexachromatic vision. Indeed, hue discrimination and spectral sensitivity experiments indicate that only the single cones are involved in avian colour vision , suggesting that the avian visual system is, at most, tetrachromatic. This conclusion is supported by recent models of colour vision which predict spectral sensitivity from photoreceptor absorptance properties which, notably, do not require the involvement of double cones. (The role of double cones is currently one of the most intriguing mysteries in avian vision, particularly since they often constitute 40-50% of a bird’s retinal cones; Cuthill et al. 1999; Wilkie et al. 1998). However, there remains the possibility that the UV cone’s output is only compared with that of one or other cone class, and that the dimensionality of avian colour space is less than tetrachromatic. Indeed, the dimensionality may change with changes in illumination, as it does in goldfish . Until colour mixing experiments are conducted, the dimensionality of avian vision will remain unknown, but all indications are that UV sensitivity will prove to be an important part of the colour vision of most birds.

It is important to remember the distinction between origins and maintenance in discussions of the adaptive significance of avian UV vision . Whatever benefits the ability to see UV wavelengths confers on extant bird species, it is probable that possession of a UV-sensitive cone class, and tetrachromacy, are the ancestral states for tetrapods . As UV cones are also found in some fish we should not look to factors peculiar to avian ecology, or even the terrestrial environment, for the selective pressures favouring the initial evolution of a UV photoreceptor. The latter issue is beyond the scope of this review, but a consideration of the role which UV vision serves in modern birds may shed light on its function in other groups, as well as being of central importance to avian behavioural ecology. Two extreme possibilities can be considered. The UV cone may have a dedicated function, acting in isolation to evoke what would be classed ‘wavelength-dependent behaviours’ rather than a ‘true colour’ response, as the cone output is not being compared at the neural level with the output of any other receptor type . Such a role need not even involve resolution of a visual image, for example if, as has been suggested, UV wavelengths are involved in the entrainment of circadian rhythms . That said, as we have discussed, available evidence indicates that the UV cone is part of a ‘true’ colour vision system. The other extreme is the, quite likely, possibility that UV vision serves no special role, but is simply part of a general purpose colour vision system that happens to be tetrachromatic rather than trichromatic. In between these extremes, there is the possibility that the UV cone is involved in a colour channel used for specific visual tasks.

Because we humans lack UV vision, there is tendency to assume it has some special, extraordinary function. It may do, as wavelength-dependent properties of light predispose the ultraviolet for use in some visual tasks . But it may actually be no more meaningful to talk of ‘UV vision’ in birds than to single out ‘blue vision’ in humans. Of the plausible functions, attention has focused on foraging (prey detection), signalling (e.g. in mate choice) and orientation. We have reviewed these in detail before , so shall concentrate here on recent experimental evidence in support of these hypotheses. Perhaps surprisingly, given the research effort directed at avian navigation and orientation, there have been few experiments to investigate the relative importance of different wavelengths of light. In bees, UV and polarisation vision are linked, as it is the intricate arrangement of UV receptors in the eye which allow detection of the pattern of polarised light in the sky . In birds, most experimental effort has been devoted to determining whether polarisation patterns per se in the sky are used in orientation, and how these interact with other compass information . But polarisation vision need not be linked to UV receptors and the UV cones could be involved in orientation irrespective of polarisation sensitivity, through detection of colour gradients in the sky which relate to the sun’s position . In support of this, it has been noted that there are more short-wavelength receptors in the ventral (skyward-looking) surface the pigeon’s retina and the cones here are relatively dispersed , which suggests panoramic vision rather than fine spatial resolution . But not all birds show this pattern; indeed starlings have more UV cones in the posterior-dorsal (down and forward-looking) region of the retina.

Light is also implicated in magnetic orientation and its effects seem wavelength-dependent in a variety of taxa, vertebrate and invertebrate . Whilst magnetite has long been invoked in magnetoreception , there is also evidence of a light-dependent magnetic compass, presumably involving the visual system. Amongst vertebrates, this was first shown to be wavelength-dependent in newts , where orientation follows the normal pattern under blue, but not red, light. Red light has also recently been shown to interfere with the magnetic compass in birds . Whether UV is as, or more, effective in enabling correct magnetic orientation than blue light, and whether there are differences between species with violet or UV visual pigments, remains to be investigated, but this is an exciting time for orientation research. In the following sections we concentrate on those hypothesized functions of UV vision which have been more thoroughly investigated experimentally, namely foraging and signalling.

Some of the earliest demonstrations of UV vision in birds were in hummingbirds , perhaps natural choices given the importance of UV vision for pollinating bees . Surprisingly, though, the role of UV in natural foraging by hummingbirds has not been investigated. This may be because red is common amongst bird-pollinated flowers, and Raven postulated that red was perhaps a ‘private channel’ for signalling to birds and not bees. However, bees are not truly ‘red-blind’ , many bird-pollinated flowers are not red, and some reflect UV . Thus, although there may be no innate preferences for particular flower colours , the role of UV in flower visitation by birds needs to be reassessed.

Many bird-dispersed fruits are red or black . Burkhardt noted that the waxy bloom that develops on some fruits enhances UV reflectance. He postulated that this might enhance their contrast against green, UV-absorbing, foliage. This may be so, but such blooms tend to enhance reflectance at all wavelengths, making the fruit brighter, rather than a specific UV-rich hue. Indeed, because the bloom adds ‘white’ to the underlying hue of the fruit, the resulting colours are less saturated than the fruit without its bloom . In the only experimental studies to investigate whether the bloom enhances detection by, or affects the preferences of, wild birds, there was no obvious effect . However, as the authors pointed out, the fruit in these trials were presented as artificial clusters, and not against leaves. Despite the widespread interest in the role of colour in avian frugivory , hardly any studies have systematically investigated the influence of hue, brightness, saturation or background contrast in fruit selection , far less the role of UV.

The first study clearly demonstrating that wild birds use UV cues whilst foraging, came when Viitala et al. showed that Common Kestrels Falco tinnunculus used UV to detect active vole trails They suggested that vole urine absorbs strongly in the UV and this provides a strong contrast against the background. Field experiments showed that raptors tended to hunt near artificial vole trails which had experimentally added, urine and faeces-soaked straw . Lab experiments showed that kestrels spent more time above arenas with vole urine, but only under illumination that included UV wavelengths and not when illumination was restricted to human-visible light. The implication is that the birds see the vole trails better when UV information was present. No bird of prey has had its UV sensitivity determined, either microspectrophotometrically or electrophysiologically, but this behavioural evidence would suggest that they can see UV.

Insects are important prey items for many bird species, and it has long been known that some moths and butterflies reflect in the UV , but whether birds utilise these UV patterns in prey detection or recognition is unknown. Likewise, despite the interest in the design features of aposematic colours which make them effective warning signals , to date colour patterns have only been assessed from a human perspective. It is likely that aposematic patterns that are conspicuous and highly contrasting to us (e.g. red or yellow on black) will be equally, or more, conspicuous to birds, as the avian eye is well designed for long wavelength discriminations . But there is the possibility of UV-aposematic patterns in species that are cryptic to humans and the colour match between model and mimic species may differ for avian and human viewers . As green leaves and brown leaf litter or bark reflect little UV, it might be expected that cryptic insects should match their backgrounds throughout the avian-visible spectrum. Where reflectance spectra have been measured this is often, but not always, the case . However, even if neither the green caterpillar, nor the green leaf on which it sits, reflect much UV, the difference in UV reflectance between object and background may still be biologically important for prey detection. In the first study to experimentally investigate the importance of UV for birds hunting cryptic prey, Church et al. found this to be the case. When Blue Tits Parus caeruleus were trained to forage for green Mamestra (Cabbage Moth) or Operophtera (Winter Moth) caterpillars in an arena, their latency to find the first prey item was increased in trials where UV wavelengths were removed from the illuminating light. This effect was most pronounced when the UV contrast between prey and background was greatest, namely the Mamestra against a cabbage leaf. However, the reduction in performance was transient, as the birds soon appeared to switch to other foraging cues than colour-matching between prey and background . It is perhaps unwise to interpret this experiment as demonstrating that blue tits locate prey using ‘UV cues’ per se. The detrimental effect of removing UV could be via alteration of the hue of both prey and background, and it is the combined colour change which renders the task more difficult. Furthermore, removal of UV reduces overall light levels as well as changing the composition of the illuminating light, so the nature of the influence of UV on prey detection remains to be determined.

It is perhaps the very colourfulness of birds, to us, which has hindered the acceptance of the idea that the UV forms an important part of the avian colour world . This has maybe been fuelled by the apparent failure to find spectacular hidden UV-monochromatic patterns in bird plumage . But this expectation, fostered by the familiar UV photographs of ‘hidden’ honeyguides in certain plants and patterns on some lepidopteran wings, is perhaps unrealistic. A UV black-and-white photograph is typically juxtaposed with a human-visible black-and-white photograph, and the contrast appears striking . But the human-visible image averages reflectance across all human-visible colours, so any patches which differ in hue but not brightness, will not be apparent. The patterns which appear to contrast strikingly only in the UV may also be apparent to humans as patches of different hues, say yellow and white, but these are not apparent in the black-and-white photographs (pers. obs.). This is not to say that the marked UV contrast is not important to the insect eye , but we should not be surprised to find that much avian plumage which differs in the UV also has correlated human-visible differences. This does not devalue to relevance of the UV component to birds.

The fact that a species’ plumage reflects in the UV is no guarantee of a role in visual signalling. Many integumentary structures and pigments will vary in their reflection (or absorption) of UV wavelengths, but this may be an unavoidable by-product of their chemical or physical structure, and no more relevant for signalling than differential reflection or absorption of infra-red radiation . Perhaps more suggestive of design for signalling is high UV reflection from plumage that is otherwise dark, as in Asian Whistling Thrushes , or sexual dimorphism that is pronounced in the UV but not in human-visible wavelengths, as in Blue Tits . There appears to be regional variation in Blue Tits, with a British study finding peak reflectance from the ‘blue’ crest firmly in the UV and a Swedish study finding peak reflectance at 430 nm, in the human-visible blue . In British tits, therefore, the human-perceived blue is thus generated by merely the tail of a peak which would strongly stimulate the Blue Tit’s UV cone . In both studies, male crests were brighter than those of females and there was significant between-sex variation in both spectral shape and the wavelength of peak reflectance. Mate choice trials indicated that this region of plumage may play a role in female mate choice since all females tested chose the male with the brightest crest . This is consistent with the finding of Andersson et al. that there is assortative mating in the field between males and females with similar crest reflectances.

The evidence that the UV reflection from the blue tit’s crest is used in sexual signalling may appear strong , but it is still circumstantial. Manipulation of UV cues is the only way to determine the relevance to signalling of the UV component of plumage reflections. Two techniques have been employed, both involving measurement of the association preference of birds in choice chambers: use of UV-blocking filters between the test bird and stimuli , and application of UV-blocking chemicals directly to the plumage . Each technique has its advantages and disadvantages. Filters do not interfere directly with the bird, but change the appearance of both stimuli and background. Chemical application is much more selective, allowing manipulation of specific plumage areas, but may not be homogeneous in effect and alters the plumage’s physical structure and appearance. However, whichever method has been used, the results to date have been the same: removal of UV cues reduces the preference for stimulus birds so treated. Maier was the first to use such behavioural choice tests, showing that the association preference of Pekin Robins for roost-mates, of same or opposite sex, was lower when a UV-blocking filter was placed between test and stimulus bird. Bennett et al. used a more extensive series of experiments to show that a similar effect in Zebra Finch Taeniopygia guttata mate choice was due to a hue change in the plumage colours. Simple removal of UV could act either by altering the spectral composition (and hence perceived hue) or reducing the total amount of light (and hence perceived brightness), so Bennett et al. also had choices involving neutral density filters which change only the overall intensity, and not spectral composition, of the light. They found no detectable effect on preference for stimulus birds seen behind such neutral density filters, but a strong reduction in preference for birds behind UV-blocking filters which transmitted similar total number of quanta; this suggests that blocking the UV is altering the perceived hue of stimulus birds. In a similar attempt to separate the effects of UV-removal on hue and brightness, Andersson and Amundsen compared pairwise preferences of female Bluethroats Luscinia svecica svecica for males whose UV-blue throat patches had been treated with either preen oil+sunblock (no UV) or preen oil+graphite (UV present, but reduced overall reflectance); test birds preferred stimulus birds with the latter treatment. Although one cannot be sure that the brightness of the two treatments matched to avian eyes, the mixtures were chosen so that total quantal flux (amount of reflected light) was as similar as possible, so that it seems that here too, the effect of removing UV reflectance was to alter the perceived plumage hue.

It has been suggested that blocking the UV simply makes the stimulus bird look ‘odd’, perhaps even unrecognisable as a conspecific. For example, Hunt et al. showed that the mate-choice preferences for particular colour bands shown by zebra finches disappeared in the absence of UV cues. Further experiments showed that this was not due to a change in the colour of the leg bands, but an effect of UV-blocking filters on the overall appearance of stimulus birds . But this simple interpretation would still have important implications, as it could explain the inconsistent results in this sort of experiment where different types of artificial light or one-way glass have been used . More generally, experiments where plumage colours have been manipulated using paints or dyes have invariably matched colours using the human eye, with no guarantee of a good match to the avian eye . However, a few experiments have shown removal of UV can have a more subtle effect than possible abolition of species recognition. Bennett et al. showed that UV was important in assessment of a specific sexual ornament, without changing the overall appearance of the bird, by taking advantage of the Zebra Finch’s preference not just for coloured leg bands, but for symmetrical arrangements of the bands . By showing that females preferred males with symmetrical combinations of UV-reflecting and UV-blocking leg bands (identical in appearance to the human eye), Bennett et al. could be sure that the preference was related to the ‘UV appearance’ of this artificial ornament, because all stimulus birds had normal avian-coloured plumage and, as all birds wore the same number of each type of band, were equally ‘odd-looking’. In a different type of experiment, showed that the free choice of female European Starlings for males in a choice chamber, correlated with the shape of the reflectance spectra from the males’ throat and covert feathers, regions displayed in mate attraction . Furthermore, females preferred males viewed through UV-transmitting filters over those behind UV-blocking filters (A. T. D. Bennett, I. C. Cuthill, J. C. Partridge, unpublished data), in a similar way to Zebra Finches .

The reflectance spectra of iridescent European Starling feathers are complex, with peaks of reflectance in both the UV and human-visible wavebands , so do mate choice patterns with or without UV cues actually differ? Bennett et al. investigated this by comparing the preference rankings of females for groups of males when the males were viewed through UV-transmitting filters, with the preference rankings of females viewing the same males through UV-blocking filters. Under illumination including UV wavelengths, female preferences for the observed males were highly correlated, but even under UV-deficient conditions females still had consistent preferences, indicating that species-recognition and motivation for choice were intact. However, the preference rankings under the two filter conditions were significantly different; that is, removal of UV simply changed the criteria for choice. This experiment shows that natural variation in UV reflectance is important in starling mate assessment and that the prevailing light environment can have profound effects on observed mating preferences. The latter result reinforces the conclusions of Hunt et al. : behavioural experiments conducted under UV deficient conditions can produce repeatable results which are, none-the-less, unrelated to the patterns of choice under other lighting conditions, including natural daylight.

None of the above experiments address the question of whether UV is in any way a ‘special’ waveband for avian sexual signalling . We would not claim, or expect, that removal of UV wavelengths was any more detrimental to visually-based mate choice in starlings than, say, removal of ‘green’ light. Experiments like the filter or ‘sunblock’ type describe above, but where other wavebands are systematically removed, will be required to investigate this. Meanwhile, one should be mindful about ignoring the UV waveband when considering avian mate choice, plumage coloration or display behaviour, and when designing experiments where the light environment could affect results.

Birds can see the ultraviolet waveband by virtue of a class of single cones maximally sensitive to violet or ultraviolet light (depending on the species). The available psychophysical evidence indicates that the output from these cones is compared with one or more of their three other cone classes, which span the human-visible waveband, and thus UV-sensitivity is part of a, probably, tetrachromatic colour vision system. With very few exceptions, when experiments have been performed to test whether the UV information is utilised in behavioural decisions, such as in foraging or signalling, the conclusion has been that it is. The obvious exceptions have concerned species such as owls, which may not possess UV-sensitive cones and, in any case, probably do not rely on colour vision when hunting at night. It is notable that, when sexual signalling has been investigated, removal of UV wavelengths has affected mate-choice even in species, such as the Zebra Finch or Bluethroat, which are colourful and sexually dichromatic to us . This is a reminder that we should not think of UV as a colour, because it is the relative amounts of different wavelengths (UV and human-visible) that a visual system interprets as a colour. Across different species, we can find examples of red, yellow, green, blue, black and white feathers that either do or do not reflect UV, and thus will be perceived as quite different colours to birds, whilst being classed as similar by humans . Even within the human-visible waveband, because the l _max values and spectral overlap of the avian cones differ from those of humans (most noticeably the red and green cones), the discriminability of different coloured objects will differ between humans and birds . Species which are ‘colourful’ to us may well still be colourful to birds , but it is also clear that there are colour differences which have gone unnoticed by humans, for example in sexual dimorphism . This has implications not only for any comparative analyses which attempt to score sexual dichromatism, but for taxonomic and conservation purposes, where racial and sub-species differences may have been underestimated.

We must take into account the sensory system of the animal(s) concerned when investigating or testing evolutionary hypotheses involving colour . This means that behavioural ecologists should pay attention to sensory physiologists. The increasing realisation that, previously ignored, UV information is important in avian ecology, provides an excellent example of this general message. But also, as a taxon with a highly developed vision system, involving double and single cones, four single cone types and highly pigmented oil droplets, birds may help understand the selective forces that lead to the evolution of different types of colour vision. To accomplish this rather more ambitious aim, the sensory physiologists need to interact with behavioural ecologists.

Our research has been supported by grants from the BBSRC, NERC, Royal Society and Nuffield Foundation. We are grateful to other members of the Bristol vision group for their contributions, notably Stuart Church, Sarah Hunt and Nathan Hart, and to many others for their support, ideas and encouragement. Amongst these we would particularly like to thank Ron Douglas, Justin Marshall, John Endler, Dietrich Burkhardt, Klaus Lunau, Erhard Maier, Daniel Osorio, Misha Vorobyev and Staffan Andersson.

Able, K.P. & Able, M.A. 1993. Daytime calibration of magnetic orientation in a migratory bird requires a view of skylight polarization. Nature 364: 523-525.

Able, K.P. & Able, M.A. 1995. Manipulations of polarized skylight calibrate magnetic orientation in a migratory bird. Journal of Comparative Physiology A, Sensory Neural and Behavioral Physiology 177: 351-356.

Allen, R.B. & Lee, W.G. 1992. Fruit selection by birds in relation to fruit abundance and appearance in the naturalized shrub Berberis darwinii. New Zealand Journal of Botany 30: 121-124.

Andersson, S. 1996. Bright ultraviolet coloration in the Asian Whistling Thrushes (Myiophonus spp.). Proceedings of the Royal Society of London B, Biological Sciences 263: 843-848.

Andersson, S. & Amundsen, T. 1997. Ultraviolet colour vision and ornamentation in Bluethroats. Proceedings of the Royal Society of London B, Biological Sciences 264: 1587-1591.

Andersson, S., Örnborg, J. & Andersson, M. 1998. Ultraviolet sexual dimorphism and assortative mating in blue tits. Proceedings of the Royal Society of London B, Biological Sciences 265: 445-450.

Avery, J.A., Bowmaker, J.K., Djamgoz, M.B.A. & Downing, J.E.G. 1983. UV sensitive receptors in a fresh-water fish. Journal of Physiology, London 334: 23-24.

Avery, M.L., Decker, D.G., Humphrey, J.S., Hayes, A.A. & Laukert, C.C. 1995. Color, size, and location of artificial fruits affect sucrose avoidance by Cedar Waxwings and European Starlings. Auk 112: 436-444.

Bennett, A.T.D. & Cuthill, I.C. 1994. Ultraviolet vision in birds: what is its function? Vision Research 34: 1471-1478.

Bennett, A.T.D., Cuthill, I.C. & Norris, K.J. 1994. Sexual selection and the mismeasure of color. American Naturalist 144: 848-860.

Bennett, A.T.D., Cuthill, I.C., Partridge, J.C. & Lunau, K. 1997. Ultraviolet plumage colors predict mate preferences in starlings. Proceedings of the National Academy of Sciences U.S.A. 94: 8618-8621.

Bennett, A.T.D., Cuthill, I.C., Partridge, J.C. & Maier, E.J. 1996. Ultraviolet vision and mate choice in zebra finches. Nature 380: 433-435.

Bowmaker, J.K. 1977. The visual pigments, oil droplets and spectral sensitivity of the pigeon. Vision Research 17: 1129-1138.

Bowmaker, J.K. 1991a. The evolution of vertebrate visual pigments and photoreceptors. In: J. R. Cronly-Dillon & R. L. Gregory (eds) Vision and visual dysfunction. Vol. 2; Boston; CRC Press Inc.: 63-81.

Bowmaker, J.K. 1991b. Visual pigments, oil droplets and photoreceptors. In: P. Gouras (eds). Vol. 6; Boston; CRC Press Inc.: 108-127.

Bowmaker, J.K., Heath, L.A., Wilkie, S.E. & Hunt, D.M. 1997. Visual pigments and oil droplets from six classes of photoreceptor in the retinas of birds. Vision Research 37: 2183-2194.

Bowmaker, J.K. & Hunt, D.M. In press. Molecular biology of photoreceptor spectral sensitivity. In: S. N. Archer, M. B. A. Djamgoz, E. R. Loew, J. C. Partridge & S. Valerga (eds) Adaptive Mechanisms in the Ecology of Vision; London; Chapman & Hall.

Bowmaker, J.K. & Martin, G.R. 1978. Visual pigments and colour vision in a nocturnal bird, Strix aluco (Tawny Owl). Vision Research 18: 1125-1130.

Bowmaker, J.K., Wilkie, S.E. & Hunt, D.M. 1999. Photoreceptors and molecular genetics of visual pigments. In: Adams, N.J. & Slotow, R.H. (eds) Proc. 22 Int. Ornithol. Congr., Durban: 2729-2742. Johannesburg: BirdLife South Africa.

Burkhardt, D. 1982. Birds, berries and UV. Naturwissenschaften 69: 153-157.

Burkhardt, D. 1996. Ultraviolet perception by bird eyes and some implications. Naturwissenschaften 83: 492-497.

Burkhardt, D. & Maier, E. 1989. The spectral sensitivity of a passerine bird is highest in the UV. Naturwissenschaften 76: 82-83.

Burley, N., Krantzberg, G. & Radman, P. 1982. Influence of colour-banding on the conspecific preferences of zebra finches. Animal Behaviour 30: 444-455.

Cameron, D.A. & Pugh Jr., E.N. 1991. Double cones as a basis for a new type of polarization vision in vertebrates. Nature 353: 161-164.

Chen, D. & Goldsmith, T.H. 1986. Four spectral classes of cone in the retinas of birds. Journal of Comparative Physiology A 159: 473-479.

Chen, D.M., Collins, J. S. & Goldsmith, T.H. 1984. The ultraviolet receptor of bird retinas. Science 225: 337-340.

Chen, D.M. & Stark, W.S. 1994. Electroretinographic analysis of ultraviolet sensitivity in juvenile and adult goldfish retinas. Vision Research 34: 2941-2944.

Chittka, L. 1996a. Does bee color vision predate the evolution of flower color? Naturwissenschaften 83: 136-138.

Chittka, L. 1996b. Optimal sets of colour receptors and colour opponent systems for coding of natural objects in insect vision. Journal of Theoretical Biology 181: 179-196.

Chittka, L. 1997. Bee color vision is optimal for coding flower color, but flower colors are not optimal for being coded. Why? Israel Journal of Plant Sciences 45: 115-127.

Chittka, L., Beier, W., Hertel, H., Steinmann, E. & Menzel, R. 1992. Opponent colour coding is a universal strategy to evaluate the photoreceptor inputs in Hymenoptera. Journal of Comparative Physiology A, Sensory, Neural, and Behavioral Physiology 170: 545-563.

Chittka, L. & Menzel, R. 1992. The evolutionary adaptation of flower colours and the insect pollinators' colour vision. Journal of Comparative Physiology A, Sensory, Neural, and Behavioral Physiology 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 Research 34: 1489-1508.

Chittka, L. & Waser, N.M. 1997. Why red flowers are not invisible to bees. Israel Journal of Plant Sciences 45: 169-183.

Church, S.C., Bennett, A.T.D., Cuthill, I.C., Hunt, S., Hart, N.S. & Partridge, J.C. 1998. Does lepidopteran larval crypsis extend into the ultraviolet? Naturwissenchaften : in press.

Coemans, M.A.J.M., Vos Hzn, J.J. & Nuboer, J.F.W. 1994a. The orientation of the e-vector of linearly polarized light does not affect the behaviour of the pigeon Columba livia. J. exp. Biol. 191: 107-123.

Coemans, M.A.J.M., Vos Hzn, J.J. & Nuboer, J.F.W. 1994b. The relation between celestial colour gradients and the position of the sun, with regard to the sun compass. Vision Research 34: 1461-1470.

Collins, S.A. & Ten Cate,C. 1996. Does beak colour affect female preference in zebra finches? Animal Behaviour 52: 105-112.

Cuthill, I.C. & Bennett, A.T.D. 1993. Mimicry and the eye of the beholder. Proceedings of the Royal Society of London B, Biological Sciences 253: 203-204.

Cuthill, I.C., Partridge, J.C. & Bennett, A.T.D. 1999. UV vision and its functions in birds. In: Adams, N.J. & Slotow, R.H. (eds) Proc. 22 Int. Ornithol. Congr., Durban: 2743-2758.   Johannesburg: BirdLife South Africa.

Cuthill, I.C., Bennett, A.T.D., Partridge, J.C. & Maier, E.H. In press. Plumage reflectance and the objective assessment of avian sexual dichromatism. American Naturalist, in press.

Darwin, C. 1859. On the origin of species. London; John Murray.

Derim-Oglu, E.N. & Maximov, V.V. 1994. Small passerines can discriminate ultraviolet surface colors. Vision Research 34: 1535-1539.

Dittrich, W., Gilbert, F., Green, P., McGregor, P. & Grewcock, D. 1993. Imperfect mimicry: a pigeons perspective. Proceedings of the Royal Society of London B, Biological Sciences 251: 195.

Douglas, R.H. & Marshall, N.J. In press. A review of vertebrate and invertebrate ocular filters. In: S. N. Archer, M. B. A. Djamgoz, E. R. Loew, J. C. Partridge & S. Valerga (eds) Adaptive Mechanisms in the Ecology of Vision; London; Chapman & Hall.

Eens, M., Pinxten, R. & Verheyen, R.F. 1990. On the function of singing and wing-waving in the European starling Sturnus vulgaris. Bird Study 37: 48-52.

Eisner, T., Silberglied, R.E., Aneshansley, D., Carrell, J.E. & Howland, H.C. 1969. Ultraviolet video viewing: the television camera as an insect eye. Science 166: 1172-1174.

Emmerton, J. & Delius, J.D. 1980. Wavelength discrimination in the "visible" and UV spectrum by pigeons. Journal of Comparative Physiology A 141: 47-52.

Endler, J.A. 1990. On the measurement and classification of colour in studies of animal colour patterns. Biological Journal of the Linnean Society 41: 315-352.

Endler, J.A. 1992. Signals, signal conditions, and the direction of evolution. American Naturalist 139: S 125-S 153.

Endler, J.A. 1993. Some general comments on the evolution and design of animal communication systems. Philosophical Transactions of the Royal Society of London B, Biological Sciences 340: 215-225.

Engström, K. 1958. On the cone mosaic in the retina of Parus major. Acta Zoologica (Stockholm) 39: 65-69.

Feare, C. 1984. The Starling. Oxford; Oxford University Press.

Fischer, K.E. & Chapman, C.A. 1993. Frugivores and fruit syndromes: differences in patterns at the genus and species level. Oikos 66: 472-482.

Fratzer, C., Dorr, S. & Neumeyer, C. 1994. Wavelength discrimination of the goldfish in the ultraviolet spectral range. Vision Research 34: 1515-1520.

Giurfa, M., Nunez, J., Chittka, L. & Menzel, R. 1995. Color preferences of flower-naive honeybees. Journal of Comparative Physiology A, Sensory, Neural, and Behavioral Physiology 177: 247-259.

Goldsmith, T.H. 1980. Hummingbirds see near ultraviolet light. Science 207: 786-788.

Goldsmith, T.H. 1990. Optimization, constraint, and history in the evolution of eyes. Quarterly Review of Biology 65: 281-322.

Goldsmith, T.H. 1994. Ultraviolet receptors and color-vision: evolutionary implications and a dissonance of paradigms. Vision Research 34: 1479-1487.

Goldsmith, T.H., Collins, J.S. & Licht, S. 1984. The cone oil droplets of avian retinas. Vision Research 24: 1661-1671.

Goldsmith, T.H., Collins, J.S. & Perlman, D.L. 1981. A wavelength discrimination function for the hummingbird Archilochus alexandri. Journal of Comparative Physiology A, Sensory, Neural, and Behavioral Physiology 143: 103-110.

Griswold, M.S. & Stark, W.S. 1992. Scotopic spectral sensitivity of phakic and aphakic observers extending into the near ultraviolet. Vision Research 32: 1739-1743.

Guilford, T. 1986. How do warning colours work? Conspicuousness may reduce recognition errors in experienced predators. Animal Behaviour 34: 286-288.

Guilford, T. & Dawkins, M.S. 1991. Receiver psychology and the evolution of animal signals. Animal Behaviour 42: 1-14.

Hárosi, F.I. & Hashimoto, Y. 1983. Ultraviolet pigments in a vertebrate: A tetrachromatic cone system in the dace. Science 222: 1021-1023.

Hart, N.S., Partridge, J.C. & Cuthill, I.C. 1998. Visual pigments, oil droplets and cone photoreceptor distribution in the European starling (Sturnus vulgaris). Journal of Experimental Biology 201: 1433-1446.

Hawryshyn, C.W. & McFarland, W.N. 1987. Cone photoreceptor mechanisms and the detection of polarized light in fish. Journal of Comparative Physiology A Sensory Neural and Behavioural Physiology 160: 459-465.

Hunt, S., Bennett, A.T.D., Cuthill, I.C. & Griffiths, R. 1998. Blue tits are ultraviolet tits. Proceedings of the Royal Society of London B, Biological Sciences 265: 451-455.

Hunt, S., Cuthill, I.C., Swaddle, J.P. & Bennett, A.T.B. 1997. Ultraviolet vision and band colour preferences in female zebra finches, Taeniopygia guttata. Animal Behaviour 54: 1383-1392.

Huth, H.H. & Burkhardt, D. 1972. Der Spektrale sehbereich eines Violetta Kolibris. Naturwissenschaften 59: 650.

Jacobs, G.H. 1981. Comparative color vision. New York; Academic Press.

Jacobs, G.H. 1992. Ultraviolet vision in vertebrates. American Zoologist 32: 544-554.

Jacobs, G.H. 1996. Primate photopigments and primate color vision. Proceedings of the National Academy of Sciences, U.S.A. 93: 577-581.

Jane, S.D. & Bowmaker, J.K. 1988. Tetrachromatic colour vision in the duck (Anas platyrhynchos L.): microspectrophotometry of visual pigments and oil droplets. Journal of Comparative Physiology A, Sensory, Neural, and Behavioral Physiology 162: 225-235.

Kevan, P., Giurfa, M. & Chittka, L. 1996. Why are there so many and so few white flowers? Trends in Plant Science 1: 280-284.

Koivula, M., Korpimaki, E. & Viitala, J. 1997. Do Tengmalm's owls see vole scent marks visible in ultraviolet light? Animal Behaviour 54: 873-877.

Kreithen, M.L. & Keeton, W.T. 1974. Detection of polarized light by the homing pigeon, Columba livia. Journal of Comparative Physiology 89: 83-92.

Lee, W.G., Weatherall, I.L. & Wilson, J.B. 1994. Fruit conspicuousness in some New Zealand Coprosma (Rubiaceae) species. Oikos 69: 87-94.

Lunau, K. & Maier, E.J. 1995. Innate color preferences of flower visitors. Journal of Comparative Physiology A, Sensory, Neural, and Behavioral Physiology 177: 1-19.

Lunau, K., Wacht, S. & Chittka, L. 1996. Colour choices of naive bumble bees and their implications for colour perception. Journal of Comparative Physiology A, Sensory, Neural, and Behavioral Physiology 178: 477-489.

Lythgoe, J.N. 1979. The ecology of vision. Oxford; Oxford University Press.

Maier, E.J. 1992. Spectral sensitivities including the ultraviolet of the passeriform bird Leiothrix lutea. Journal of Comparative Physiology A, Sensory, Neural, and Behavioral Physiology 170: 709-714.

Maier, E.J. 1993. To deal with the invisible: on the biological significance of ultraviolet sensitivity in birds. Naturwissenschaften 80: 476-478.

Maier, E.J. & Bowmaker, J.K. 1993. Colour vision in the passeriform bird, Leiothrix lutea: correlation of visual pigment absorbency and oil droplet transmission with spectral sensitivity. Journal of Comparative Physiology A, Sensory, Neural, and Behavioral Physiology 172: 295-301.

Marples, N.M. & Roper, T.J. 1996. Effects of novel colour and smell on the response of naive chicks towards food and water. Animal Behaviour 51: 1417-1424.

Mason, J.R., Arzt, A.H. & Reidinger, R.F. 1984. Comparative assessment of food preferences and aversions acquired by blackbirds via observational learning. Auk 101: 796-803.

Mollon, J. 1995. Seeing colour. In: T. Lamb & J. Bourriau (eds) Colour: Art and science; Cambridge; Cambridge University Press: 127-150.

Munro, U., Munro, J.A., Phillips, J.B. & Wiltschko, W. 1997. Effect of wavelength of light and pulse magnetisation on different magnetoreception systems in a migratory bird. Australian Journal of Zoology 45: 189-198.

Munro, U. & Wiltschko, R. 1995. The role of skylight polarization in the orientation of a day migrating bird species. Journal of Comparative Physiology A, Sensory Neural and Behavioral Physiology 177: 357-362.

Muntz, W.R.A. 1972. Inert absorbing and reflecting pigments. In: H. J. A. Dartnall (eds) Photochemistry of Vision. Vol. VII/1; Berlin, Heidelberg, New York; Springer-Verlag: 529-565.

Murray, K.G., Winnettmurray, K., Cromie, E.A., Minor, M. & Meyers, E. 1993. The influence of seed packaging and fruit color on feeding preferences of American Robins. Vegetatio 108: 217-226.

Neumeyer, C. & Arnold, K. 1989. Tetrachromatic colour vision in the goldfish becomes trichromatic under white adaptation light of moderate intensity. Vision Research 29: 1719-1727.

Osche, G. 1983. Optical (visual) signals in the coevolution of plants and animals. Berichte Der Deutschen Botanischen Gesellschaft 96: 1-27.

Osorio, D. & Vorobyev, M. 1996. Colour vision as an adaptation to frugivory in primates. Proceedings of the Royal Society of London B, Biological Sciences 263: 593-599.

Owens, E.D. & Prokopy, R.J. 1986. Relationship between reflectance spectra of host plant surfaces and visual detection of host fruit by Rhagoletis pomonella flies. Physiological Entomology 11: 297-307.

Palacios, A.G. & Varela, F.J. 1992. Color mixing in the pigeon (Columba livia). 2. A psychophysical determination in the middle, short and near-UV wavelength range. Vision Research 32: 1947-1953.

Parrish, J.W., Ptacek, J.A. & Will, K.L. 1984. The detection of near-ultraviolet light by nonmigratory and migratory birds. Auk 101: 53-58.

Partridge, J. C. 1989. The visual ecology of avian cone oil droplets. Journal of Comparative Physiology A, Sensory, Neural, and Behavioral Physiology 165: 415-426.

Peiponen, V.A. 1992. Colour discrimination of two passerine bird species in the Munsell system. Ornis Scandinavica 23: 143-151.

Phillips, J.B. & Borland, S.C. 1992. Behavioural evidence for use of a light-dependent magnetoreception mechanism by a vertebrate. Nature 359: 142-144.

Phillips, J.B. & Borland, S.C. 1994. Use of a specialized magnetoreception system for homing by the Eastern Red-spotted Newt Notophthalmus viridescens. Journal of Experimental Biology 188: 275-291.

Phillips, J.B. & Moore, F.R. 1992. Calibration of the sun compass by sunset polarized light patterns in a migratory bird. Behavioral Ecology and Sociobiology 31: 189-193.

Phillips, J.B. & Sayeed, O. 1993. Wavelength-dependent effects of light on magnetic compass orientation in Drosophila melanogaster. Journal of Comparative Physiology A Sensory Neural and Behavioural Physiology 172: 303-308.

Pohl, H. 1992. Ultraviolet radiation: a zeitgeber for the circadian clock in birds. Naturwissenschaften 79: 227-229.

Puckey, H.L., Lill, A. & Odowd, D.J. 1996. Fruit color choices of captive Silvereyes (Zosterops lateralis). Condor 98: 780-790.

Radwan, J. 1993. Are dull birds still dull in UV? Acta Ornithologica 27: 125-130.

Raven, P.H. 1972. Why are bird-visited flowers predominately red? Evolution 76: 674.

Roper, T.J. & Marples, N.M. 1997a. Colour preferences of domestic chicks in relation to food and water presentation. Applied Animal Behaviour Science 54: 207-213.

Roper, T.J. & Marples, N.M. 1997b. Odour and colour as cues for taste avoidance learning in domestic chicks. Animal Behaviour 53: 1241-1250.

Rowe, C. & Guilford, T. 1996. Hidden color aversions in domestic clicks triggered by pyrazine odors of insect warning displays. Nature 383: 520-522.

Sallabanks, R. 1993. Hierarchical mechanisms of fruit selection by an avian frugivore. Ecology 74: 1326-1336.

Sanders, M.J., OwenSmith, R.N. & Pillay, N. 1997. Fruit selection in the olive thrush: the importance of colour. South African Journal of Zoology 32: 21-23.

Shyue, S.K., Hewettemmett, D., Sperling, H.G., Hunt, D.M., Bowmaker, J.K., Mollon, J.D. & Li, W.H. 1995. Adaptive evolution of color vision genes in higher primates. Science 269: 1265-1267.

Silberglied, R.E. 1979. Communication in the ultraviolet. Annual Review of Ecology & Systematics 10: 373-398.

Silberglied, R.E. 1984. Visual communication and sexual selection among butterflies. In: R. I. Vane-Wright & P. E. Ackery (eds) The biology of butterflies. Symposium of the Royal Society of London, Number 11.; London; Academic Press: 207-223.

Snow, B. & Snow, D. 1988. Birds and berries : a study of an ecological interaction. Calton; Poyser.

Stark, W.S. 1987. Photopic sensitivities to ultraviolet and visible wavelengths and the effects of the macular pigments in human aphakic observers. Current Eye Research 6: 631-638.

Swaddle, J.P. & Cuthill, I.C. 1994. Preference for symmetrical males by female zebra finches. Nature 367: 165-166.

Thompson, E., Palacios, A. & Varela, F.J. 1992. Ways of coloring: Comparative color vision as a case study for cognitive science. Behavioral and Brain Sciences 15: 1-74.

Viitala, J., Korpimaki, E., Palokangas, P. & Koivula, M. 1995. Attraction of kestrels to vole scent marks visible in ultraviolet light. Nature 373: 425-427.

Vorobyev, M., Gumbert, A., Kunze, J., Giurfa, M. & Menzel, R. 1997. Flowers through insect eyes. Israel Journal of Plant Sciences 45: 93-101.

Vorobyev, M. & Osorio, D. 1998. Receptor noise as a determinant of colour thresholds. Proceedings of the Royal Society of London B, Biological Sciences 265: 351-358.

Vorobyev, M., Osorio, D., Bennett, A.T.D., Marshall, N.J. & Cuthill, I.C. 1998. Tetrachromacy, oil droplets and bird plumage colours. Journal of Comparative Physiology A, Sensory, Neural, and Behavioral Physiology : in press.

Vos Hzn, J.J., Coemans, M.A.J M. & Nuboer, J.F.W. 1994. The photopic sensitivity of the yellow field of the pigeon's retina to ultraviolet light. Vision Research 34: 1419-1425.

Vos Hzn, J.J., Coemans, M.A.J.M. & Nuboer, J.F.W. 1995. No evidence for polarization sensitivity in the pigeon electroretinogram. Journal of Experimental Biology 198: 325-335.

Walls, G.L. 1963. The vertebrate eye and its adaptive radiation. New York; Hafner.

Wehner, R. & Bernard, G.D. 1993. Photoreceptor twist: a solution to the false color problem. Proceedings of the National Academy of Sciences of the United States of America 90: 4132-4135.

Wheelwright, N.T. & Janson, C.H. 1985. Colors of fruit displays of bird-dispersed plants in two tropical forests. American Naturalist 126: 777-799.

Whelan, C.J. & Willson, M.F. 1994. Fruit choice in migrating North American birds: field and aviary experiments. Oikos 71: 137-151.

Willson, M.F., Graff, D.A. & Whelan, C.J. 1990. Color preferences of frugivorous birds in relation to the colors of fleshy fruits. Condor 92: 545-555.

Willson, M.F. & Whelan, C.J. 1989. Ultraviolet reflectance of fruits of vertebrate-dispersed plants. Oikos 55: 341-348.

Willson, M.F. & Whelan, C.J. 1990. The evolution of fruit color in fleshy-fruited plants. American Naturalist 136: 790-809.

Wilkie, S.E., Vissers, P.M.A.M., Das, D., DeGrip, W.J., Bowmaker, J.C. & Hunt, D.M. 1998. The molecular basis for UV vision in birds: spectral characteristics, cDNA sequence and retinal localization of the UV-sensitive visual pigment of the budgerigar (Melopsittacus undulatus). Biochemical Journal 330: 541-547.

Wiltschko, R. & Wiltschko, W. 1998. Pigeon homing: Effect of various wavelengths of light during displacement. Naturwissenschaften 85: 164-167.

Wiltschko, W., Munro, U., Ford, H. & Wiltschko, R. 1993. Red light disrupts magnetic orientation of migratory birds. Nature 364: 525-527.

Wiltschko, W. & Wiltschko, R. 1993. Navigation in birds and other animals. Journal of Navigation 46: 174-191.

Wiltschko, W. & Wiltschko, R. 1995. Migratory orientation of European robins is affected by the wavelength of light as well as by a magnetic pulse. Journal of Comparative Physiology A, Sensory Neural and Behavioral Physiology 177: 363-369.

Wood, P. & Partridge, J. C. 1993. Opsin substitution induced in retinal rods of the eel (Anguilla anguilla (L.)): a model for G-protein-linked receptors. Proceedings of the Royal Society of London B, Biological Sciences 254: 227-232.

Wright, A. A. 1972. Psychometric and psychophysical hue discrimination functions for the pigeon. Vision Research 12: 1447-1464.

Yokoyama, S., Radlwimmer, F. B. & Kawamura, S. 1998. Regeneration of ultraviolet pigments of vertebrates. FEBS Letters 423: 155-158.