... Electro-magnetic materials > Iridescence in butterfly wings > Classification of butterfly scales

School of Physics

 School       Teaching       Admissions       Research     
  Overview     Astrophysics     Biomedical Physics     Electromagnetic materials     Quantum Systems  


Published in Photonics Science News, 6, 61-66 (2000).

Optical Classification of Microstructure in Butterfly Wing-scales

P. Vukusic1, J.R.Sambles1, and H. Ghiradella2

(1Thin Film Photonics, School of Physics, Exeter University, Exeter EX4 4QL, UK and  2 Department of Biology, The University at Albany, Albany, NY 12222, USA).


Colouration in nature may be broadly classified into two categories; colouration through pigmentation and colouration through the interaction of light with microstructures. While the former category represents the most common origin of colour in plants, insects and animals, it is the optical effects associated with microstructures that have been of most interest to workers in photonics.

Recent work (1-3) with structural colouration in butterflies has highlighted the very broad range of microstructures that can be responsible for different colourations and optical effects. It is in these butterflies, as with several other taxa (4-7), that a range of different systems of microstructure has evolved to interact with light to produce vivid colouration.

With a few exceptions the seat of colouration in butterflies, whether through pigmentation or structure, lies in the scales that cover the surface of the wings and body. Each scale is a flattened projection of cuticle from a single epidermal cell within the epithelial layer that makes up the surface of the wing. The arrangement of scales on the wing resembles that of shingles on a roof (figure 1) with, in most species, there being two distinct layers of different scales present. Typical scale dimensions are of the order of 75 m m by 200 m m. While the underside of scales are rather planar and featureless, their interiors and their externally visible top-surfaces exhibit intricate microstructure. The cuticle material itself is a composite of rods of chitin set in a matrix of proteins. It has been shown to have a refractive index of approximately 1.56-1.58 at visible wavelengths (8). The presence and concentration of melanin pigment dictates the associated amount of optical absorption (9).

classification_fig1.jpg (25965 bytes)

Figure 1. The arrangement of type Ia scales on the wing of M. rhetenor.

All scales, whether iridescent or not, display one of a variety of forms of ridging extending longitudinally from one end of the scale to the other. Within the ridging, fine structure may be developed in the form of lamellae or microribbing. Often the ridges may be connected at intervals by a series of arched structures referred to as crossribs. Spacing between ridges on a single scale is often quite uniform, lying in the range 0.5 m m - 5.0 m m depending on species and scale type.

In the scales of some species the regions between the ridges and crossribs are hollow, exposing the scale interior with its pillar-like trebaculae which run from the base of the ridges to the scale substrate beneath. As far as is known, pigment other than melanin is laid into pigment granules in this region, while melanin tends to be distributed in the scale structures themselves.


For the purpose of this overview, we choose to group the microstructure of butterfly scales that display structural colouration into three primary categories. Figure 2 illustrates this schematically. Categorisation is based on the type and the location in the wing-scale of structural reflecting and scattering elements. Type I comprises multilayered systems incorporated into scale ridging. Type II comprises multilayered systems incorporated into the scale body. Colour is generated in these systems through multilayer interference in the normal way (8). Type III does not comprise multilayering, but instead other specialised diffraction and scattering systems.

The structures associated with all three of these categories are variations on the same general template of scale design (shown in figure 2). Developmental processes leading to the formation of each class of scale are addressed in other texts (10).

wpe6.gif (100881 bytes)

Figure 2. a) SEM image overview of the classification of scale microstructure described in this overview . (Schematic images reprinted with permission from Wiley-Liss, copyright 1998)

Type I Scales.

 Figure 2.b) SEM and TEM images showing scale and ridge cross-sections for type I and type II species. (Species; Ia M. rhetenor, Ib A. meliboeus, Ic T. magellanus).

wpe9.gif (59068 bytes)    wpeA.gif (89145 bytes)


In structurally coloured butterflies with type I scales, multilayering is incorporated within the ridging of scales. However, this may then be categorised in a secondary way that is associated with the angle made by the multilayering with the base of the scale. Type Ia, layering by so-called ridge-lamellae, is defined as a category in which lamellae run parallel or near-parallel to the base of the scale. The brightest structurally coloured butterflies are associated with this category of scale for which there may be as many as twelve ridge-lamellae of cuticle for the multilayer system. Lower brightness in these scales may be attributed to ridges with fewer lamellae.

The visibility of certain Morpho butterflies is an excellent example of the effect of type Ia scales. The 10-12 cuticle-layer stack within a single M. rhetenor scale yields reflectivity at blue wavelengths of up to 80% (3). Lateral (side-side) tilting of the ridging and of lamellae surfaces within the ridging provide, in one plane, an additional broad angle spread to the reflection. For another Morpho species, M. didius, not only does lateral ridge tilting provide angle spread in the reflected light, but this is also assisted by diffraction through the ridges of a superficial second layer of near-transparent scales (3). In this way for these species, the type Ia nature of their systems facilitates simultaneous bright and broad-angle visibility.

Type Ib scales comprise ridge-lamellae that have moderate inclination to the scale base. High numbers of ridge-lamellae (greater than five or six) are generally not found within these systems. This limit, together with the steeper angle of inclination facilitates a larger range of structurally reflected colours than is possible with type Ia scales (11). Additionally, the steeper angle of inclination limits the portion of the viewing hemisphere above the scale from which the iridescent colour may be observed even under diffuse illumination. The butterfly Ancyluris meliboeus is adorned with patches of type Ib scales and exhibits deep blue to orange structural colouring from the same region at different scale orientations. Figure 3 illustrates this broad range of possible colours from different wing regions, all comprising type Ib scales. Figures 4a and 4b illustrate how a very small change in wing orientation may extinguish this colour.

Figure 3. When a narrow beam of collimated white light is incident normally on a single iridescent region of A. meliboeus wing, a range of colour is produced, with each colour reflected at different angles.

Type Ic scales comprise cases where this angle of multilayer inclination is extremely steep, to the extent that in some cases it approaches orthogonality with the base of the scale. In such cases, the layering is often formed from an extremely developed system of microribs rather than from the ridge-lamellae. With such steep multilayering, one would expect the portion of the viewing hemisphere above the wing from which the associated structural colour is observable to be extremely limited. Indeed, it is only from near-grazing incidence to the scale that the structural colour becomes visible. Troides magellanus is a species that exhibits type Ic scales. It displays bright blue-green iridescence only in back-reflection at grazing incidence. This colour appears in addition to the overall yellow colouration derived from its papiliochrome pigmentation, which also exhibits strong UV absorption and green-fluorescence.

Type II Scales

 Figure 2.b) SEM and TEM images showing scale and ridge cross-sections for  type II species. (Species; IIa U. leilus, IIb P. palinurus).

wpe11.gif (40700 bytes)   wpeD.gif (37179 bytes)



Type II scales comprise multilayering (referred to as body-lamellae) that is incorporated into the body of the scale itself. For this category, ridging is present on the surface of the scales, but it is usually a much simpler non-specialised form of ridging than found in type I. As many as ten cuticle layers have been shown to exist in some species with iridescent type II scales. As with type I, we choose to categorise type II scales in a secondary way.

Type IIa comprise the simplest form of such body-lamellae systems in which the lamellae are parallel and flat across the entire area of the scale (allowing for any curvature of the scale itself). Additionally, not only does ridging sit above the system of body-lamellae, but the ridges are often orthogonally interconnected by structures known as crossribs. The spacing of these crossribs is usually quite periodic and is of the order of the spacing of the ridges themselves. It is believed that these and other surface structures associated with type II systems provide one or both of the following effects. They form impedance matching elements that reduce broadband reflection from the top-most surface of the scale (12), and they may constitute diffraction elements that diffuse and spread in angle the selectively reflected colour from the multilayer over which they rest. Urania leilus is a species of diurnal moth, the scales of which are of type IIa. Its appearance is less striking than that of Morpho; its iridescent regions are less highly reflective and its hue is rather softer in appearance (13).

Type IIb scales represent a category in which the flat body-lamellae of type IIa have additional orthogonal modulations imposed upon them. These modulations cause the appearance of an array of shallow flat-bottomed concavities across the surface of scale. The body-lamellae follow the profile of the resulting surface. Interestingly, in certain type IIb species, the physical dimension of each layer appears to remain constant in the direction orthogonal to the scale surface (12). This creates the effect of a sculpted multilayer with constant layer dimensions, rather than a patterned multilayer of variable dimensions. The particular sculpting associated with some type IIb systems has been shown to produce several unique optical effects, such as the simultaneous production of two structural colours, which combine additively to give the stimulus of a third colour (14). Furthermore, through a double reflection from the opposite sides of each depression, the polarisation of one of the colours may be rotated by 90 degrees. Additionally with type IIb scales, crossribs are generally absent, ridging appears to play much less of an optical role and the scale surface exhibits one of a variety of forms of impedance matching surface texture.

Type III Scales

 Figure 2.b) SEM images showing scale and ridge cross-sections for type III species. (Species; IIIa C. rubi, IIIb P. zalmoxis).

   wpe13.gif (39012 bytes)


Species with type III scales display structural colouration that does not arise from multilayering but from scattering structures within and on the body of the scale. For this reason, the characteristic angle-dependence of colour, associated with layered structures, is absent and there is a broadly constant colour at all angles of observation.

Type IIIa scales comprise an ordered 3-dimensional lattice of cuticle within the body of the scale. In effect, it is an array of spherical holes in a matrix of cuticle. The order imposed by such a system leads to strong diffraction of favoured wavelengths in certain directions. Analogous, but inverse, structures lead to the structural colouration in the mineral opal (15), for which arrays of hydrated silica spheres surrounded by air lead to diffraction colours. The lattices within most type IIIa butterfly scales, however, are usually divided into irregular but distinct domains of a few microns diameter (figure 5). By varying the orientation of neighbouring domains, a constant-colour effect is created across the scale and wing through spatial averaging (16, 17). The diffuse angle-independent nature of the structural colour that is provided by type IIIa scales probably helps in defense from predation through camouflage.

Figure 5. A single type IIIa scale from Callophrys rubi taken in transmission using an optical microscope. The domaining effect is especially clear in transmission, and a range of colours (complementary to the reflected colours) is seen between domains across the scale. The dark near-parallel horizontal lines running the length of the scale are the ridges.

Type IIIb is rather rarer than type IIIa and is distinctly different. Scales in this category are coloured through Rayleigh scattering from sub-wavelength sized scattering structures. These structures extend from between the ridges on the surface of the scales down into the body of the scale. In Papilio zalmoxis, the scattering structures are in the form of air-filled alveoli approximately 2 m m long and around 220 nm in internal diameter. The simultaneous presence within the alveoli of a UV-absorbing but blue fluorescing pigment enhances the blue colouration (18).


Cuticle-air microstructure plays a central role in a wide range of butterfly wing colouration. It can be responsible for many characteristics, such as narrow-band reflectivity, ultra-high visibility, polarisation effects, ultra-violet reflection, pointillistic colour mixing and specific angle dependencies that are simply not accessible through pigmentation alone. Certain biological purposes of this microstructure are apparent. Ultra-high brightness colour with wide-angle visibility can effect excellent long range intraspecific signalling (19). On the other hand, low brightness and angle independent colour helps with defence from predation. Interpreting reasons for UV reflection and strong polarisation and colour mixing effects is only possible on appreciation of parallel complexities in butterfly visual pathways (20). It has been shown that some lepidopteran eye systems can be based on pentachromatic photoreceptor arrays, some of which exhibit strong polarisation sensitivity (21).

In this overview, three principle categories have been outlined which broadly classify the different wing-scale morphologies of structurally coloured butterflies. (Non-iridescent and androconial scales, which are also derived from the same general scale template, are detailed elsewhere). These classifications enable a comparison between the structurally generated optical properties of different species. Sub-division of the three principle classes provides a more rigorous classification. It facilitates a fundamental understanding of the mechanisms behind butterfly colouration and it contrasts the variations on several structural themes that Nature has used on some of her most colourful creatures.


  1. Ghiradella H., 1991, Light and colour on the wing, Appl. Opt. 30, 3492-3500.

  2. Huxley J., 1976, The coloration of Papilio zalmoxis and P. antimachus and the discovery of Tyndall blue in butterflies, Proc. Roy. Soc. (B), 193, 441-453.

  3. Vukusic P., et al 1999, Quantified interference and diffraction in single Morpho butterfly scales, Proc. Roy. Soc. Lond. (B), 266, 1403-1411.

  4. Herring P. J., 1994, Reflective systems in aquatic animals, Comp. Biochem. Physiol. 109A, 513-546.

  5. Parker A.R., 1998 A unique form of light reflector and the evolution of signalling in Ovalipes (Crustacea: Decapoda: Portunidae) Proc. R. Soc. Lond. B, 265, 861-867.

  6. Parker A.R., McKenzie D.R. and Large M.C.J., 1998, Multilayer reflectors in animals using green and gold beetles as contrasting examples, J. Exp. Biol., 201, 1307-1313.

  7. Liu K., Shigley J.E., and Hurwit K.N., 1999, Iridescence colour of a shell of the mollusk Pinctada Margaritifera caused by diffraction, Optics Express, 4, 177-182.

  8. Land M.F., 1972, The physics and biology of animal reflectors, Progr. Biophys. Molec. Biol. 24, 75-106,

  9. Nijhout H.F., 1991, The Development and Evolution of Butterfly Wing Patterns, Smithsonian Institute Press, Washington.

  10. Ghiradella H., 1998, Hairs, Bristles and Scales, Microsc. Anat. Invert. 11A, 257-287.

  11. Vukusic P., Sambles J.R., and Wootton R.J., in preparation.

  12. Lawrence C.R., Vukusic P., and Sambles J.R., in preparation.

  13. Huxley J., 1975 The basis of structural colour variation in two species of Papilio, J. Ent., (A), 50, 9-22.

  14. Vukusic P., Sambles J.R., and Lawrence C.R., 2000, accepted for publication in Nature.

  15. Sanders J.V., 1968, Diffraction of light by Opals, Acta Crst. A24, 427-434.

  16. Morris R.B., 1975 Iridescence from diffraction structures in the wing scales of Callophrys rubi, the Green Hairstreak. J. Ent., (A), 49, 149-154.

  17. R. Vane-Wright, private communication.

  18. Huxley J., 1976, The colouration of Papilio zalmoxis and P. antimachus, and the discovery of Tyndall blue in butterflies, Proc. R. Soc. Lond. (B), 193, 441-453.

  19. Silberglied R. E., 1984 Visual communication and sexual selection among butterflies, p207-223 in The biology of butterflies, Symposium of the Royal Society of London, no. 11, Academic Press, London, eds. Vane-Wright R.I. & P.E. Ackery.

  20. Swihart S. L., 1972, Modelling the butterfly visual pathway, J. Insect Physiol., 18, 1915-1928.

  21. Kelber A., 1999, Why false colours are seen by butterflies, Nature, 402, 251.

Home > Research > Electro-magnetic materials > Iridescence in butterfly wings > Classification of butterfly scales

Help for Netscape 4 users