ABSTRACT
Diffraction gratings were believed to be a rare cause of colour in nature. However, surface gratings, which include diffraction gratings and Bragg gratings, appear to occur in many forms within a diversity of animals. Structural colours in general have implications in the study of animal behaviour and evolution.
‘The finely colour’d feathers of some birds, and particularly those of the peacocks’ tail, do in the very same part of the feather appear of several colours in several positions of the eye, after the very same manner that thin plates were found to do.’ (Isaac Newton, 1704, Optiks)
There is a vast body of literature documenting functional colour in animals as the result of chemical pigmentation and bioluminescence. There is, however, a third extensive category of animal light display: structural coloration. Structural colours result from the selective reflectance of incident light due to the physical nature of a structure. Employment of structural colours involves relatively little energetic expenditure compared with bioluminescence and possibly the synthesis of pigment.
Newton (1704) was the first to identify an animal structural colour, and Goureau (1843) and Rayleigh (1919) made subsequent fundamental discoveries. Süffert (1924) and Mason (1926, 1927a,b) later described the general differentiation of structural colours and pigment colours in animals, providing a classification of the former. The detailed study of the anatomical basis of structural colours accelerated following the introduction by Anderson and Richards (1942) of the electron microscope to the subject.
In animals, structural colours are commonly known to result from (i) structures causing random scattering of light waves (e.g. Fig. 1A) or (ii) single or multiple thin-layer reflectors (e.g. Fig. 1B). Notable review papers are those by Denton (1970), Land (1972) and Herring (1994). However, the field may be expanded with the inclusion of a third type of mechanism responsible for structural colour: surface gratings, which include diffraction gratings and Bragg gratings.
Surface gratings, like thin-layer reflectors, produce ‘iridescent’ effects (some changing colour with orientation, others appearing only as one colour). The shared specular (‘directional’)-type reflectance of multilayer reflectors and diffraction gratings is indicative of the similarity of the physical nature of the two reflector types. However, there are two significant features that set these two systems apart. First is the issue of the direction of the modified wavefront. A multilayer reflector separates the incident wave into two component waves: one that is transmitted and one that is reflected (Fig. 1B). These waves propagate according to the mirror-reflection laws. A diffraction grating, in contrast, separates the incident wave into many waves that propagate in various directions set by the grating laws (Fig. 1C). The second feature involves the orientation of the plane containing the periodicity relative to the surface plane of the complete structure. In a diffraction grating, the plane with the periodicity is parallel to the surface plane (Fig. 1C); in a multilayer reflector, it is perpendicular to the surface plane (Fig. 1B). These features give rise to a general classification that breaks down a large complex subject into convenient categories. However, the physical processes cannot be ordered into one category or another and, indeed, a structure with a reflection mechanism which lies between that of a diffraction grating and a multilayer reflector is known in animals, the Bragg grating (a term recently created to describe an analogous structure designed for optical fibres). Bragg gratings will be considered in the category ‘surface gratings’, along with diffraction gratings, because they are surface structures, i.e. the plane containing the periodicity is parallel to the surface plane (Fig. 1D). However, the mechanism of reflection is most similar to that of a multilayer reflector. A Bragg grating consists of a series of ridges where the ridge width, or period, is γ/2n (where γ is wavelength and n is refractive index) and is almost like the edge of a multilayer reflector. One period can be subdivided into two regions; one with a mean refractive index that is close to that of the surrounding medium (air or water), and one with a mean refractive index that is close to that of the material of the animal structure (e.g. chitin). Colour is observed in retroreflection when the illumination and observation directions form a grazing geometry with respect to the surface plane (Fig. 1D).
Under the same incident white light, iridescent colours can appear much brighter than colour pigments when viewed from the appropriate direction. This difference appears even more pronounced against a natural background where colour pigments are present. Here, the chromatic effect of animal pigmentation is modified by background radiation to a greater extent than the effect of iridescence. However, structures causing scattering produce effects rather comparable with those of pigments because the reflections are diffuse. A comparison between diffuse and specular reflections can be made when viewing a credit card; the coloured print (pigment) produces a diffuse reflection while the reflection from the hologram (diffraction grating) is specular.
‘True diffraction by natural gratings occupies a place of relatively minor importance in the production of iridescence in animals.’ (D. L. Fox, 1976, Animal Biochromes and Structural Colors)
Over the last 30 years or so, many cases of multilayer reflectors have been discovered in nature, but until recently discoveries of natural diffraction gratings were rare. Few examples had ever been suggested (Fox and Vevers, 1960; Fox, 1976; Nassau, 1983), and all the gratings hypothesised in pre-electron microscope analyses have since been disproved as such, with the iridescence observed resulting from multilayer reflectors. In fact, gratings have been established in only three groups of animals: (i) in the beetle Serica sericea (Anderson and Richards, 1942) and other scarabaeids (Hinton and Gibbs, 1969; Hinton, 1973), although the periodicities involved here are sometimes questionably large (up to 5 μm) and/or irregular; (ii) in wasps in the family Mutillidae (Hinton and Gibbs, 1969); and (iii) in cypridinid ostracod crustaceans (Parker, 1995). It has recently been found that natural surface gratings (including diffraction gratings) are represented in animals in a variety of forms. These occur in a diversity of animals living on land and at the shallowest and deepest parts of the aquatic photic zone, from surface water in bright sunlight to 1000 m oceanic depths (the effective limit of sunlight penetration in the sea; Denton, 1990).
Surface gratings may be suitably accommodated on fine setae/setules, from which an extremely small and precise reflection can result. Some surface gratings can reduce the reflection of white light more efficiently than a smooth surface. Such an antireflection grating occurs on the corneas of certain flies and serves to increase photon capture by the eyes (Parker et al. 1998a). This diffraction grating is known as a zero-order grating because the optical behaviour is seen primarily in the zero diffraction order. Zero-order gratings typically have periodicities smaller than the wavelength of violet light. However, gratings that reflect first and higher diffraction orders under illumination normal to the surface plane may also become antireflection structures when the angle of illumination changes.
‘Whenever colour has been modified for some special purpose, this has been, as far as we can judge, either for direct or indirect protection, or as an attraction between sexes.’ (Charles Darwin, 1859, On the Origin of Species by Means of Natural Selection)
Colour is displayed in nature when the advantages of the colour-producing structure outweigh its disadvantages. Advantages, however, may be associated with properties of the colour-producing structure other than the colour itself, where the display of colour is a by-product (such as the red colour of blood). Because needlessly attracting attention to oneself carries obvious disadvantages, a structure that produces iridescence from the external surface of an animal as a by-product may become modified, through selection, to prevent such a high external reflection. The effect of a diffraction grating fades significantly when the grating period exceeds twice the wavelength of light or becomes irregular. Therefore, an almost colourless grating may be selected for when a reflection has no behavioural advantage for the host. Selection of a colourless multilayer reflector occurs in Ovalipes crabs (Parker et al. 1998b). In all species of Ovalipes, iridescence as a result of multilayer reflectors is a potentially useful means of intraspecific communication. However, only those species of Ovalipes that live at depths where only the blue portion of sunlight remains can display iridescence in a plane parallel to the sea floor and thus appear relatively inconspicuous to their predators in the water column. These deep-water species possess efficient multilayer reflectors in their cuticles. In shallow waters, other colours exist, and these would be reflected up into the water column where predators live. Consequently, the shallow-water species possess exoskeletons where the component layers are too thick to form a multilayer reflector in the visible region throughout most of their body.
In many invertebrates, the shells or exoskeletons display iridescence internally (e.g. Fig. 2C), but possess an opaque external surface which prevents light reaching the reflector (e.g. Fig. 2B). The internal layers of these structures contribute structural strength and comprise an ‘incidental’ multilayer reflector (see Land, 1972). A similar case occurs in cylindroleberidid ostracods (Crustacea), where the fine setae of a comb (‘food filter’) form the ridges of a diffraction grating, but its iridescence (Fig. 2E) cannot be displayed external to the bivalved carapace that encloses the body of the ostracods. Additionally, iridescence which has a biological function can be hidden from external view when not required by the host. For example, some ostracods possess iridescent structures where the light display has a function, and these can be withdrawn into their carapace (Parker, 1995). Also, the iridescence of tanaid and callianassid crustaceans (Fig. 2A,D) is only visible when they are out of their tubes and burrows respectively.
Animal structures producing iridescence externally are often so efficient optically that colour display appears highly selected for (e.g. Parker et al. 1998c). Certain ostracods which possess surface gratings show an increase in the spectral purity and reflectivity characteristics of their gratings through evolution; the development of these gratings parallels phylogeny (correlating positively with a conventionally constructed phylogeny) (Parker, 1995). In this case, iridescence was originally incidental to the primary mechanical function (Parker, 1998) in low-light regimes, becoming functional in high-light regimes. Additionally, there appears to be a relationship between structural colour and the morphology/evolution of compound eyes in ostracod crustaceans (Parker, 1995). Another interesting point is that some animals and plants have independently evolved the same intricate structural designs to produce the same iridescent effect. For example, scarab beetles (Neville and Caveney, 1969) and the fern Danaea nodosa (Neville and Levy, 1986) possess microfibrils, which appear helicoidal on an oblique face, in their cuticles and fronds respectively. This forms a reflector that is analogous to a ‘liquid crystal’ structure, where each 180 ° rotation of the microfibrils forms a row of arcs or a ‘layer’, and the whole structure effectively comprises a ‘multilayer reflector’ (Land, 1972) or ‘three-dimensional diffraction grating’ (Nassau, 1983). Considering the parallelism, this design may be highly selected for as a reflector.
The functions of structural reflectors include conspecific recognition, such as during courtship and aggregation (e.g. Parker, 1995), predator evasion (Denton, 1970; Hinton, 1973), colour filtering in eyes (Bernard, 1971), mirror-like optical reflection (e.g. in eyes, photophores and iridophores; Herring, 1994) and even reduction of reflection, such as in the moth eye (Miller et al. 1966) or in the eyes of certain flies (Parker et al. 1998a). Determination that an animal reflector is highly efficient, physically, provides evidence towards it having a function. However, this information is more useful as an indicator of the potential for behavioural study than for speculation on function, considering the arduous task of deducing a function from experimentation with living hosts.
ACKNOWLEDGEMENTS
This work was funded by an Australian Research Council grant.