Berthier - Iridescences-The Physical Colors of Insects
Berthier - Iridescences-The Physical Colors of Insects
Berthier - Iridescences-The Physical Colors of Insects
Serge Berthier
Iridescences
The Physical Colors of Insects
Serge Berthier
Institut des Nano-Sciences de Paris
UMR CNRS - Universite Pierre et Marie Curie
Universite Denis Diderot
4 Place Jussieu
Paris, 75252
France
Translator
Capucine Lafait
7 Rue Christian Dewet
Paris, 75012
France
C 2007 Springer Science+Business Media, LLC
All rights reserved. This work may not be translated or copied in whole or in part without the written permission of
the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief
excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and
retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter
developed is forbidden.
The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identied
as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights.
9 8 7 6 5 4 3 2 1
springer.com
About the Author
Serge Berthier teaches physics at the Denis Diderot-Paris 7 University and researches biologic
structures, colors and biomimetism at the Institut des NanoSciences de Paris (Pierre and Marie
Curie-Paris 6 University and CNRS). He teaches solid state optics in the post-graduate degree
Optics and Material and the post-graduate research degree Optics and Photonic for the Denis
Diderot University. He also teaches electromagnetism and laser physics to undergraduates at the
Paris-Jussieu Institut of Technology.
v
Foreword
Here is a book to recommend for various reasons. First of all, let us note that the rst chapter
introduces the reader into a reection on the physical explanation of an observation. The origin
of buttery colors has interested a great physics experimenter, A.A. Michelson, Nobel Prize in
1907, thought it proceeded from selective reection over buttery wings and defended this idea
during his whole life. The book he wrote in 1927 at the age of 75 and which contains all of
the conclusions of his work, ends with a chapter entitled metallic colors of birds and insects.
At the same time, another great physicist, Lord Raleigh, claimed that observed colors were
caused by light interferences. How is it possible to defend two different explanations for similar
observations? Simply because the knowledge of the objects observed was inappropriate. Here is a
simple case which shows how a scientic explanation can evolve with the knowledge of the object
observed.
S. Berthiers book beautifully shows that it is possible to teach physics in relation with biology,
which makes the latter certainly more attracting and lively. One is here offered an optics course
through a treatise on Lepidoptera and Coleopteron colors. This shows that it is possible to teach
physics in a painless way.
Lastly, the conclusion underlines that the study of buttery iridescence entice researchers
to nd ways to apply the same principles to textiles. It also highlights that Butteries and
Cincidelidaes properties could be used to ensure the identication and protection of bank
notes.
As founder of the laboratory in which S. Berthier works, I am very happy and proud to present
this book. Its author has a gift for pedagogy and this book is fascinating at all levels. It should
interest young people, as young as high school seniors, but also every people who likes to learn
and understand nature further. This book succeeds in being concise and clear, but also in making
one marvel. It is also a step forward as concerns pluridisciplinarity. I wish that it is successful, as
it deserves, and I congratulate the author.
Florin Abeles
Professor Emeritus of Pierre et Marie University, Paris.
vi
Acknowledgments
Part of this book is a new, improved, and substantially modied version of the rst book, Buttery
Colors or the Imperative Beauty. I wont mention again all the people who contributed to the rst
book. They will recognize their contribution in the present work that made up Imperative Beauty
and was the basis of Iridescence. I would like to thank them once again, the pupils, students,
and teachers who gave a reason for being to these publications. I thank my colleague and friend
Michel Perreau for his uncompromising correcting of the rst book and making pertinent remarks
about the present book, which were very useful to me.
Other people took over, furthering and widening the scope of the rst exploratory work.
A schoolboy also whose stay was too short: Antoine Baillet. Thank you, my dear Antoine!
Spadework, exalting at rst but also quite laborious, was performed joyfully and contentedly by
Aurelie Tournie and Marius Knoch, students at the Institut Universitaire de technologie (IUT)
Paris-Jussieu, and Mathieu Gueguen at IUT of Annecy. They made a great number of the most
beautiful of the photonic microscopy photos presented here. I thank them for their patience and
congratulate them for succeeding in conciliating scientic rigor and artistic sensitivity.
Postgraduate students focused more on applications. Passing from quality to quantity does
not always prove uplifting. If their conclusions have only a small place in the present book, they
still made the latter possible in terms of logistics. Thanks to Jean Druille and Loic Ledernez,
students of the Optics and Materials Master at the University Denis Diderot in Paris.
We aimed at widening the scope of the study of iridescence beyond physics and butteries
in the present publication, which is beyond our qualications. That is why we relied on the
work of specialists, especially Pr. Henri Descimon, of Marseille University, on pigments. He also
provided us with a certain number of photos illustrating his conclusions. I want to respectfully
thank him.
I would also like to thank my friend Emanuel Fritsch of the Institut des Materiaux de Nantes
for photos of opals, as well as Philippe Lalanne of the Institut dOptique dOrsay, my colleague
of the Optics and Photonic post-graduate degree, for the calculations of electromagnetic elds
diffracted by the Morphos.
These research works were performed at the color and biomimetism group of the Labora-
toire dOptique des SolidesUniversite Pierre et Marie Curie, Paris. I thank all the staff of the
laboratory, beginning with its director, Jacques Lafait, for his unbreakable support. Some people
directly contributed to this work, like Claude Sella, who suggested I should use ionic thinning
on elytrons, thus allowing me to distinguish in relief chitin sticks. These elytrons had rst un-
dergone optical polishing in our crystallography workshop in Michelle Jacquets expert hands,
who I friendlily thank. A special thanks to Eric Charron, who performed spectogoniometric
measurements and established Morphos diffraction maps, as well as to those who patiently and
kindly tried to pacify my relationship with computers and get me out of the incredible messes I
get myself in: Claude Naud and Gerard Vuye.
vii
viii Acknowledgments
Most of the scanning electron microscopy photos were taken by Stephane Borensztein at the
Laboratoire de Physique des Liquides et Electrochimie at the University Pierre and Marie Curie.
Transmission electronic or photonic microscopy sections of scales and alary membranes were
made at the Laboratoire de Biologie Marine at the same university. Thanks to Jean-Pierre Lechair
and Ghislaine Frebourg for welcoming me and for their beautiful work.
I was about to forget my friend and ofcial supplier of butteries and other insects, Claude
Nature. You can go to his shop 6, rue des Chantiers in the 5th in Paris to get a good dose of marvel
and good spirits.
Lastly, authors are unbearable persons. Thanks to my children, Valerian and Juliette, for putting
up with my unavailability and sudden changes of moods with so much kindness and humor.
Annie Fontaine had her share, and in addition, she patiently and rigorouslyand tactfully,
authors being touchy personscorrected proofs. Thank you for the warmth and tenderness you
gave me.
Contents
1 Iridescence ...................................................................................................... 1
About the Book................................................................................................. 5
2 Why Colors...................................................................................................... 7
Variations in Colors ........................................................................................... 9
Colors: Another Classication ............................................................................. 10
Mimicry........................................................................................................... 10
ix
x Contents
Conclusion........................................................................................................... 151
Subject Index ....................................................................................................... 157
General Index....................................................................................................... 158
1
Iridescence
Solid state physics is traditionally classied species, mostly tropical. Coleoptera and even
among hard sciences, in the literal sense by certain birds can actually surpass them as far as
students and in the gurative sense by the this aspect is concerned. It is iridescence, an im-
Academy. To introduce the study of optical propriate term in French, yet a last allusion to
properties of buttery wings is, at best, mere the Ancients and the messenger of the Olympic
wandering and at worst, provocation. Since I gods, the winged goddess Iris draped in her
have started this research, I have been told so rainbow. From the very end of the 19th cen-
many times, though always politely, that in ac- tury, iridescence started to intrigue scholars:
cordance with the current symbolism of butter- biologists of course, who gave the name of the
ies, I was itting around. Yet, butteries have goddess to a great number of species present-
not always stood for inconsistency, futility and ing the qualitywhich we tried to illustrate in
banter! This view appeared as late as in the 17th this bookbut also physicists. If artists would
century and applies to diurnal butteries only. then stick to a frivolous symbolism, things were
Moths are not luckier though. In his introduc- much less easy on the other side! On one side
tion to devotion life (1608), Saint-Francois de the poet:
Sales, a connoisseur, immortalizes the symbol-
ism of moths: Natre avec le printemps, mourir avec les roses,
Sur laile du zephyr nager dans un ciel pur
As the small buttery perceiving the candlelight goes Balance sur le sein des eurs a peine ecloses,
towards it in a peculiar manner to see if it is as soft as it is Senivrer de parfums, de lumiere et dazur,
beautiful, and hurried by his fantasy, never stops unless Secouant, jeune encore, la poudre de ses ailes,
it gets lost on its rst attempt: so the young heart goes for Senvoler comme un soufe aux voutes eternelles;
voluptuous ames. Voila du papillon le destin enchante!
Il ressemble au desir qui jamais ne se pose
The Ancients had a completely different Et sans se satisfaire, efeurant toute chose
view, which is more relevant to the purpose Retourne enn au ciel chercher la volupte.
of this book: The buttery embodies immor- Lamartine, Nouvelles meditations poetiques
tality of the human being. In Greek, the word
psyche designates both the human soul and To be borne with the spring, and die with roses,
the buttery. Metamorphosis, the mysterious On zephyr wings to swim in a clear sky
and obscure passage from earth to ether, offers Rocked in the hearts of blooming owers,
To get drunk on fragrances, on light and azure,
a simple symbolism to the rst Christians who
Flapping the dust out of its new borne wings,
adopted it. To y like a breath to the eternal skies,
But let us come back to earth for a mo- Here is the enchanted fate of the buttery!
ment, and return temporarily to physics. One It is like desire which never rests,
of the most fascinating aspects of buttery col- And never contented, always eeting
ors, in addition to their innite diversity, is cer- At last goes back to the sky to nd delight.
tainly the metallic and changing aspect of some Lamartine, New poetic meditations
1
2 Chapter 1 Iridescence
And we must admit that the controversy Figure 1.3. Scales of the dorsal side of Morpho
that opposed two physicists who were among godarti. Picture obtained by transmission photon
microscopy. Each scale is about 200 m long by
the greatest of the time could puzzle people 50 m wide. One can distinguish the main longitu-
and make them cautious. Let us start with dinal striae network and, in places, the transversal
Michelson himself, the rst American to be counter-striae network.
Iridescence 3
remaining classes of colors: interferences, diffraction . . . colors rules out any kind of visualization. The
(Phil. Mag. 21 (1911) 554) problem can only be addressed through optical
measurements of the whole wing as it enables
From the other side of the Atlantic Ocean, an-
us to track down the nature and structure of
other famous name, Lord Raleigh (4th baron),
its components. This faction is mainly com-
president of the Royal Institution and professor
posed of physicists, like B. Walter, and Bie-
at the Imperial College of London, and advo-
dermann, a biologist who will later side with
cate of the interferential approach, replied:
Raleigh. Michelson based his theory on mea-
It is singular that the explanation of some of the most surements of polarization of light reected by
striking and beautiful of optical phenomena should be still the wing under various angles of incidence (el-
matters of controversy! (Phil. Mag. 37 (1919) 98) lipsometric measurements, as one would call
them nowadays) very similar to those obtained
In order to understand how great physicists
on thin lms of aniline, or magenta. This was a
can have quarreled over such a subject, let us re-
forceful argument, which has yet to be turned
member that at the turn of the 20th century, the
back.
only means of observation available was tradi-
In the other faction were the defenders of
tional photonic microscopy, which was indeed
structural origin. In this faction, Lord Raleigh
unable to resolve structures capable of creat-
and numerous biologists attempted to make
ing such colors, whatever they were. The im-
out this structure. It is to them that we owe ex-
age it could give of a scale resembled the one
tremely precise microscopic observations and
depicted above, and originated many a mis-
outstanding drawings of the scales of a great
take. One can clearly distinguish longitudinal
number of butteries and of sections that
striaeand even a square pattern sometimes
they did thanks to rather rudimentary mi-
which immediately reminds one of a grating.
crotomes. The weightier argument in favor
Hence the hypotheses of diffraction that were
of the thin lm hypothesis as defended by
made at a certain time. Two types of experi-
Lord Raleighwhich we will fully illustrate
mental approaches were then considered and
in this bookis the alteration or even the to-
two factions emerged opposing the different
tal disappearance of colors when scales are
protagonists of the controversy.
immersed in a liquid with an adequate opti-
One faction was headed by Michelson and
cal index. This implies that the structure re-
included those who think that colors result
sponsible for colors remains open, enabling
from absorption and selective reection of a
the liquid to penetrate it. This was not eas-
metallic type. In almost all of their papers, they
ily conceivable, but it excluded the hypothe-
refer to thin lms of aniline or cianinne The
sis of selective reectionwhich, on the con-
rst consequence of which is that it is not nec-
trary, would imply even brighter colors in this
essary to carry a minute observation of scales,
case. It is quite hard to understand how physi-
since the atomic or molecular origin of the
cists like Michelson and Walter could have ne-
glected this question. Another physical source
of colors, among which skystudied by Lord
Raleigh the fatheris a familiar example is the
selective scattering by small particles in numer-
ous living beings (feathers of certain birds,
muzzle of mandrills, human iris, and so on).
Tyndall observed this as early as the late 19th
century (1873), and so did Madoul in 1903.
When these particles are small enough, they
Figure 1.4. Structures and pigments share the visi- only scatter the shortest wavelengthspurple
ble spectrum for the production of colors among but-
teries. Green, in the middle of the spectrum, often and bluethe longest being transmitted and
results from combined effects. This gives the process eventually absorbed by deep layers. This is
a large exibility and allows a cheap adaptability. the case of many Lycaenidae, one of the rare
4 Chapter 1 Iridescence
examples of blue butteries living under our (the majority of semi-precious stones) or inter
tempered climate. When larger, the particles or intra-band transitions of crystalline material
scatter longer wavelengths and we obtain the (for example, metals and semiconductors). If
white of the Pieridae. effects like camouage are obviously useful,
Those physical phenomena generally cause we can wonder why a buttery develops very
cold colors, or high energy colors, ranging loud motifs or colors. Butteries belong to the
from near ultra-violet to blue-green, while pig- class of insects, which entails two important
ments usually create hot colors, from red to consequences. First, because of their position
yellow. Indeed, blue pigments are rare in the in the food chain, they undergo great selective
animal world: there is a carotenod in the shells pressure. Second, thanks to their short life
of some crustaceans, like lobster, while haemo- span and high reproductive rate, they can
cyanin is quite widespread among mollusks, rapidly evolve. Colors are, of course, only
crustaceans and arachnids. As we will see in the a small element of the gear that butteries
chapter on pigments, the scarcity of blue pig- develop throughout their ght for life. One
ments in the animal world can easily be under- still remains fascinated by the ingenuity that
stood. Basic organic molecules that compose they show in this domain, and even more in
blue pigments naturally absorb in the ultra- the extraordinary saving of means to develop
violet, the violet, and the blue, making them their various tricks. Indeed, if through our
appear as complementary colors: red, orange, scientic approach, we distinguished the main
and yellow. However, there are no clear reasons coloration processes, no buttery actually uses
why structural colors remain conned to vio- one to the exclusion of another! The effects are
let and blue. We are forced to notice that in this almost always combined in order to obtain,
case, exceptions still exist, but they are very few. with only a few basic elements, the innite
Nevertheless, this splitting of the visible spec- palette of colors that one can observe. Why
trum, in addition to the fact that it offers great develop a green pigment when it is possible
saving of means, enablesas we will seethe to obtain the color by mixing an interferential
creation of composed colors, of mixed origin, blue or a blue of scattering with a pigmentary
one of the compounds of which (physical) is yellow, a very common melanin? In the same
iridescent and the other (pigmentary) is not. way, Morphos, the undisputed champions of
This results in remarkable and subtle colored interferential colors, distinguish themselves by
variations. combining pigments that form a background
Even if it is long, difcult, and incomplete, more or less opaque and accentuating or, on
the study of pigments has apparently been an the contrary, diminishing the their vividness,
easier task for the scientistand not for the but- sometimes even annihilating it by curving its
tery. The chemical analysis requirements are wings, in order to obtain the dull blue of a mere
decreasing but still involve a substantial quan- Argus?
tity of material. The total of the scales of a big Butteries are often considered to be the
Pieridae does not weigh more than one to two standard bearers of insects, decried crea-
milligrams, from which the pigment only rep- tures that sometimes provoke brutal reac-
resents a small percentage. So, not surprisingly, tions among us. Butteries, at least diurnal
it required some 215,000 butteries in 1925 to species, are exempted because of their evanes-
extract . . . 39 grams of pigment. cent beauty. The color and the texture of their
We have just briey exposed the different wings, which remind us of the most precious
means used by butteries to adorn themselves fabric, play an important role in our infatu-
with such wonderful colors, and conversely, to ation. It is actually to answer the questions
make themselves invisible to our eyes. So but- of a renowned textile manufacturer that this
teries actually use all the processes of color research project was launched at the Labo-
creation that are available to living beings. ratoire dOptique des Solides of the Univer-
They are barred from some other sources, es- site Pierre and Marie Curie in Paris. How can
sentially colors created by the crystalline eld one manufacture, without using any pigment,
About the Book 5
a colored ber to create fabric and garments cles, wings, and structures of butteries and
with the same iridescence as butteries? We Coleoptera. If, once hatched, the imago keeps
had just entered a fascinating and extraordi- its colors and motifs forever, the latter under-
narily complex world. Yet, the richness of the goes changes at some crucial stages of the insect
explanations was as fascinating and other po- life. These critical stages will be described in
tential applications were found, which were those chapters. I focused on the scales and cuti-
even more astonishing. I will mention these cles of the insects where colors primarily dwell.
biomimetic applications of butteries in the In this part, I presentedquite synthetically
correspondent chapters, which will lead us given their diversity and with a great num-
quite far from the little itting buttery, so dear ber of illustrationsthe arrangements, shapes,
to Lamartine. and ne structures of the scales. The colors of
insects are changing in many ways. They are
iridescent of course, in that they change accord-
About the Book ing to the point of view or lighting: that is the
subject of the present book. But they can also
I tried to write this book so that it could be change under natural or articial constraints.
read at different levels or at various speeds to Even if this is never the case with butteries
enable everyone to get something out of it. This and hardly ever the case for Coleoptera, I found
book is for numerous audiences, including the it important to shortly mention the structures
following: students who often tackle in a disor- that enable this changing in colors. The study
ganized manner the various subjects gathered of material known as X-chromes, which can
here; physicists or engineers striving for origi- change colors under a constraint X, represents a
nal structures and new applications; and biol- very active eld for civil and military research.
ogists to whom, as I am not a biologist, I have I also present several structural principles and
no pretense of teaching anything but to whom some simple experiments to determine the ori-
I would like to remind or to put together some gin of colors in insects.
notions of physics that may be somehow dis-
tant. This book is also for the inquiring mind or
the artist who would like to go a little beyond
fascination. This book falls into three parts that
are not necessarily related. The main theme
coloris extremely rich and complex as it re-
lates to a variety of elds.
When one has seen a buttery eaten by a
bird and another buttery remain perfectly
invisible to the brushing wings of the same
bird, the reason for being of those colors, so
beautiful and loud at the same time, cannot
but appeal to us. That is why I decided to start
this book by trying to answer this question.
Even before describing the orders that concern
us hereLepidoptera and ColeopteraI want
to address this questionbecause the question
goes far beyond the classication of insects. In
this part, I evoke color as part of the defen-
sive gear of insects and generally present the
various types of mimicry principles. The next Figure 1.5. Iris, gods messenger. Spreading her
scarf, she makes appear the rainbows (detail, Louvre
two chapters, structured along the same line, Museum). Inset: Aparatura iris, the changing Great
mention a certain number of elements, which Mars. One of the rare iridescent blue butteries in
we all know more or less about: the life cy- our countries.
6 Chapter 1 Iridescence
Whatever its origin, color results from an in- tion and modeling might sometimes seem too
teraction between light and matter. This inter- detailed. If the results only concern specialists,
action, addressed in Chapter 6, is the part that it will still provide the inquiring mind with an
is the most distant from the view of the little idea of the secrets of nature and the extent of
buttery. In addition to an inevitably super- knowledge available, partly unexplored.
cial presentation of notions of solid state optics Still in this part on electromagnetic proper-
and index of material, I draw a panorama of the ties of insects, a chapter is dedicated to the
effects of polarization of light and of their treat- radiation qualities of wings. Here, we reach
ments. If insects are vividly colored animals, the limits of the study since radiations are
they also behave as polarizers. Polarization is not comprised in the colored spectrum but in
a chromatic character that we are not sensitive the infrared. Color is nevertheless related to
to and can originate interesting discoveries and this problem as it directly affects the absorp-
applications. This chapter is certainly the most tion of solar energy by dark butteries, or its
complex of the book and if the reader avoids reection towards the body by light-colored
it at his rst reading, he will still have a good butteries.
understanding of the book. Such a reader will Lastly, as a counterpart of the rst chapter on
be able to come back to it afterwards if cer- the reason of colors and after this long devel-
tain exposed phenomena stimulates his curios- opment on their physical and chemical origins,
ity enough that the equations dont scare him the book ends with a chapter dealing with col-
anymore. orimetry: how to name a color, measure it, and
The gratication for such an effort can be how humans and butteries perceive it.
found in the third main part, which describes This book will have achieved its purpose if it
the various types of colors produced by the can stimulate the readers desire to understand
insect, phenomenon by phenomenon. Each of the reason of being of those colors and how they
them is illustrated at all different scales by one are produced. The reason why is a fascinating
or several type-species that uses it primarily. subject. As far as the how is concerned, one
For some of the species and for the require- should just imagine that an equation can be
ments of certain specic studies, the descrip- beautiful too.
2
Why Colors
Very few species in the animal world have such which the best trick is total immobility, which
diversied colored motifsas much in the pat- drives us back to the rst imperative.
tern as in the colors themselvesas butteries. In order to comply with the three impera-
These remarkable variations are the mere coun- tives that can be contradictory, butteries have
terpart of the innite diversity of biotypes. Bio- developed an amazing palette of processes,
types in which adult butteries must within a among which color, or more generally optics,
short period of time reproduce, feed, and eat play an essential and sometimes primary part.
without being eaten. This equates to three im- As we will see later on, pigments spring from
peratives: to move, to recognize, and to escape preexisting metabolic compounds. They are of-
predators. ten toxic and stocked we are tempted to say
for lack of anything betterin wings, which
To Move are dry organs where toxic compounds cannot
Imagos feed mostly on nectar, a mix of water cause any harm.
and simple sugars that specic owers pro- Coloration, and thus the pigmentary func-
duce, which butteries must recognize and ap- tion, is a secondary consequence of the proper-
proach. They must y to feed but also to seek ties of those compounds. Yet, it still plays an im-
partners and reproduce. portant part in the vital processes listed above.
To Recognize All of them are closely linked, even when we
Let us imagine a triangle of senses, the three try to study them separately.
angles that represent of the main sensorial stim- Butteries have to y in order to feed and
uli: olfactoryl, visual, and auditory. On this tri- to reproduce. To reproduce they must make
angle, butteries would, contrary to the ma- themselves recognizable, hence make them-
jority of their predators, be situated on the selves visible. These two vital activities make
chemical/ visual base. Butteries are known them easy targets for predators. What role can
to be quite sensitive to odors and pheromones. coloration play in the strategies designed to
But the chemical messages conceal numerous conciliate the two contradictory activities? This
disadvantages: Their dynamic and variability role obviously depends on the ecologic context
are limited and they depend on uncertain exte- of butteries and predators, which must trap
rior conditions such as wind. Optical recogni- the buttery next to one of the three angles.
tion is therefore vital and very well developed.
To Display
To Escape Predators A great number of butteries are not palatable,
Lepidoptera have no system of active defense either because of their repelling taste or because
but their predators are many and fearsome: they represent a danger for predators (toxicity).
They are mostly birds, reptiles (chameleons, Thus, it is in the interest of these species to show
lizards, and the like), batrachians, spiders, and it and deliver a clear message to predators. That
other insects (for example, mantids) against is why warning colors can be loud and intense.
7
8 Chapter 2 Why Colors
This widely accepted interpretation corre- isokinemy, that is the same mobility as their
sponds to the decline of the dark form noticed models. The objects that are the more often
over the last decades after anti-pollution mea- imitated are leaves and bark but also include
sures were launched. lichens and bird excrements. We wont address
the genesis of mimicry and camouage; sev-
eral theories were offered that involve natu-
Colors: Another Classication ral selection in the origin, dynamic, and sta-
bility of the phenomenon. These questions are
As regards selection, we can try to classify col- still unanswered today and we will only briey
ors not according to their physical character- set out the modalities of the main types of
istics or physico-chemical origins, but to their mimicry.
role in the strategy for survival and develop-
ment of a species. We can, as the English nat-
uralist Alfred Russel Wallace did in the late Mimicry
19th century, distinguish colors between those
designed to warn and those for camouage. As is often the case, the fathers of important
The latter allow the insect to become invisible discoveries tend to remain unknown from the
in its environment for defensive or offensive public, whereas their work comes at the right
purposes. As we have seen, there is a price time to support great theories with a more uni-
to pay, which is a decrease in mobility. On versal impact. Henry Walter Bates, Fritz Muller,
the contrary, warning colors, which are very and on another hand, Charles Darwin, illus-
loud a priori, are related to a real danger or a trate this.
faked one from the viewpoint of predators. A H.W. Bates, an English naturalist and ex-
few years later, Sir Edward B. Pulton offered plorer, was born in Leicester in 1825 and
another classication by distinguishing colors died in London in 1892. He explored the
that deceive from those that dont. The latter Amazon valley for 11 years and came back with
warn against a real danger: toxicity, venom,
spine, repelling taste . . . These colorations are
called sematic colorations.
Deceiving colors, called apetetic, include all
hiding or cryptic colors and warning col-
ors faking a danger or pseudo-sematic. Each
branch consists of other ramication depend-
ing on whether the insect carrying colors is a
prey (procryptic) or a predator (anti-cryptic),
whether it tries to repel (pseudoaposematic),
or, on the contrary, to attract (pseudepise-
matic). In this latter group, which is actually
quite rare within the buttery population, the
widespread cases of auto-mimicry, in which
the motifs of a nonvital body part (like poste-
rior wings) fake those of the vital part (the head
or anterior wings): Those are called parase-
matic colors.
While sematic colored butteries, which
have nothing to hide, have a wide range
of colors and patterns, apathetic must re-
sort to isochromy with their model or with
a widespread element of their environment Figure 2.5. W. Bates, lithography by W. Purkiss
(homotypy) and to isomorphy, and even (1880). C. Hulton Picture Library.
Mimicry 11
or it is warned against a real danger by means of would annihilate the warning effect of sematic
a minimalist signal, like a color, common to dif- colorations of the latter. In this way, Batesian
ferent uneatable species (Mullerian mimicry). mimicry is similar to some parasitism from
Here there is no deceiving in the proper sense, which a model can escape by changing its
so the predator is described as an indifferent markings. This can lead to a parallel evolution
dupe. The two strategies require a geographi- of the mimic and the model, the latter becom-
cal concordance of the various protagonists. ing more and more different from the former,
which, put in danger, always tries to resem-
ble it.
Batesian Mimicry
This is certainly one of the most remarkable
Mullerian Mimicry
protective strategies of Lepidoptera. Here, the
mimic, an eatable buttery, imitates the mark- Species that are truly protected because they
ings and ying manners of an uneatable in- are uneatable or dangerous and present se-
sect for whatever reason: toxicity, repelling matic colors are not ipso facto safe from preda-
taste, or venom. This kind of mimicry can be tors. The danger here occurs during the learn-
both intra or inter order/group. Swamps ing stage of the predator, which associates an
Hymenopteraare often imitated by Lepi- annoyance to a warning signal only after exper-
doptera. If it guarantees a relative immunity imenting it many times. This is often deadly for
to the mimic, Batesian mimicry puts the model a buttery, yet the risk gets smaller according
in danger and its efciency depends on the ra- to the clarity and non-ambiguity of the mes-
tio between the mimic population and the imi- sage. The latter must be strong. As sematic
tated one. Predators learn their lessons through colorations are loud, they should not be too
successive successes (eatable insects) and fail- many. Thus, protected species should always
ures (uneatable insects); if the proportion of present the same markings to predators. In-
the former outnumbered that of the latter, it deed, they reduce and share the chances of
Our present purpose is not to write a book on We can dene other big groups that are re-
butteries, but rather a book on their colors. It is lated to the previous ones and usually com-
still useful to begin by situating butteries into monly known as micro and macro Lepidoptera.
classications and review some general facts The latter group is composed of the most devel-
concerning their anatomy, development, and oped diurnal and nocturnal butteries, while
the structure of their wings. former group includes the less developed and
Among insects, a group that already out- usually small nocturnal butteries.
weighs the animal world by its remarkable
diversity, ve orders outclass all of the oth-
ers by the number of species composing Outlining a Classication
them. Those are Coleoptera (300,000 species),
Hymenoptera (250,000 species), Diptera and Distinctions are sometimes more marked in
Hemiptera (150,000 to 200,000 species), and English. Indeed, the word butteries comprises
Lepidoptera with about 170,000 species. It is all of Rhopalocers, the real diurnal butter-
mere utopia to try to establish a comprehen- ies, and moths, the Heterocera, mostly noc-
sive classication of such a huge number of turnal butteries with feathery or threadlike
species, all the more since only 5 to 20 percent antennas. These terms designate some use-
depending on ordersof the total have of- ful categories, the limits of which still remain
cially been described and studied. blurry. However, they are quite acceptable
The order of Lepidoptera comprises diur- for us.
nal and nocturnal butteries. The word derives Heteronera are subdivided into a large num-
from the Greek term Lepidos (scales) and Pteris ber of superfamilies, which will not be de-
(wings), and therefore means scaled wings or scribed here. Indeed, the majority of butter-
wings covered with scales. This is one of their ies presented in our study come from one
main characteristics, and yet some of them only superfamily only, the Papilionoidea, consisting
partially present scales while others are totally of 14 families, with one well-known excep-
deprived of them. tion, the Uranoidea, illustrated by the famous
The large majority of nocturnal butteries Uraniidae. It is impossible to describe in a few
and the total of the diurnal belong to the in- lines insects as diverse as butteries, which are
fraorder of Heteroneura, which means that adapted to almost all environments, and the
they present different veins on their anterior size of which varies from 2 mm for the mi-
and posterior wings (Figure 3.9). Homoneura nuscule Nepticulides to more than 20 cm for
consists of some primitive butteries present- the big saturnides. We will only briey mention
ing the same veins on both wings. We will their development and characteristics. Butter-
hardly mention them, as they have little or no ies appeared in the early Jurassic, 200 million
coloring. years ago, and despite their apparent fragility,
14
Summary Description of Butteries 15
Cremaster
Wings
Antennae
Eyes
Figure 3.1. Schematic representation of a caterpillar
highlighting the main anatomic partsArcherontia
atroposthe deaths-head moth. A caterpillar can Figure 3.3. Schematic representation of a Melitea
look quite different from one shedding to the phoeba chrysalisJapan. The Chrysalis is the rst
other. More exposed than the imago because they specialized stage of adulthood including all the fu-
move relatively little, caterpillars possess a sub- ture organs. Compound eyes are perceivable, while
stantial defense arsenal, mainly camouage and antennae, proboscis, wings and legs are along the
mimicry/mimetism like the imago, but also chemi- body, distinguishable in relief. The external cuticle
cal means. can be highly colored.
16 Chapter 3 Lepidoptera Description and Scales of Observation
(a)
Imago
The imago, the last stage in the develop-
ment of the buttery, essentially concerns
reproduction. The imago is mobile and pro- Figure 3.11. Series of dislocations on the ventral side
vided with all the organs necessary to recog- of a Charaxes jasius anterior wing.
20 Chapter 3 Lepidoptera Description and Scales of Observation
Figure 3.13. Morpho godarti ocelli, ventral side. On the left image, the ocellus is spherical and its concentric
circles present the same thickness. On the right image, the ocellus, in the neighboring cell, starts elongating,
the circle thickness increasing towards the wing apexlocated on the right of the images. Blue scales,
which are almost absent in the spherical ocellus, cover half of the central motif in the elongating ocellus.
Generally conned in an alary cell, ocelli can sometimes substantially expand onto the neighboring cells
(Caligo idomeneus).
22 Chapter 3 Lepidoptera Description and Scales of Observation
the same, we will study it in the opposite direc- remarkable surface structure that scarcely af-
tion, in ascending order. Zooming out, color is fects coloration, but that slightly diffuses light
created at the bottom of the scale and is modi- and gives a pale, velvety aspect to these areas.
ed at each rung to eventually obtain the nal One can roughly imagine the wings of an
macroscopic effect. ideal Lepidoptera as two triangles, the an-
gles of which are more or less rounded off. The
angles, sides, and main concentric areas have
Macroscopic Level: Structure of the Wings been designated with different names, which
Lepidoptera possess four membranous
wings covered with relatively few veins in
comparison to that of other insects. The wing
venation characterizes species and thus allows
one to classify species.
The wing consists of one double membrane,
the two of which developed independently and
united at the nal formation stage of the wing.
That is why the colors, motifs, and implanta-
tion of the scales of the ventral and dorsal sides
of the wing are totally independent and often
highly contrasted. Already developed in the
chrysalis, the double membrane is covered by
tracheas. During their development, the two
membranes join each other and at the same
time, are sclerotized along tracheas forming the
veins, which continue to transport the nervous,
gas, and haemolymph messages.
These membranes are the support of the
scales, the surface on which colors mainly
reside in Lepidoptera. Diffuse pigments can
nevertheless be found on membranes, like
that found with Graphium (see Chapter 11). In Figure 3.15. Schematic representation of a Lepi-
this specic case, the membrane consists of a dopteron scale.
Description of the Wings 23
Tillards system, which is more adapted to des- joining the radial and the cubital, and from
ignate triangles than the Anglo-Saxon numer- which proceed the medial ones: It is the dis-
ation. The majority of secondary veins actually cal crossvein. But of course, scarce are the but-
proceed from two main big veins, known as teries that match this diagram, even without
radial and cubital. Their subdivisions are num- counting the supernumerary, merged, or sub-
bered from the rear to the back. The previous divided veins . . .
ones are the subcostal, and those between the The areas delimited by the pattern of veins
two medial and the following ones are the ab- and the edge of the wings, known as alary cells,
dominal. The main veins are all oriented from are more important to us. They were named af-
the base towards the periphery of the wing, ex- ter their inferior vein, except for the triangular
cepted for one transversal vein, the only one area delimited by the radial, cubital, and discal
even sometimes on the whole wing, and it along the median axis of alary cells, depends
proves quite difcult to experimentally deter- on a specic property of the cells of the motif
mine its biogenetic origin. This is not the case center. Known as the focus, it has been possi-
where ocelli are concerned, which consist of rel- ble to experimentally shift this group of cells
atively small motifs that are always precisely leading to the formation of an ocellus on the
situated. place of the shift, or to remove it, which brought
An ocellus is characterized by a specic pig- about the complete disappearance of the ocel-
mentation and an arrangement of the scales lus. These experiments also establish a rather
contrasting with its environment. The ocellus detailed chronology of the development of the
itself generally includes several colors, yet it ocellus and show that the motif is denite about
does not affect the morphology of its scales. It three days after the wing has been totally pig-
is an established fact now that the arrangement mented, which occurs within the 24 hours be-
of ocelli, which are always placed precisely fore hatching.
Description of the Wings 27
The foci are thus a source of information substance, an inhibitory, which allows the syn-
indicating the place where the ocelli can de- thesis of the melanin only when the concen-
velop. The geometry of the motifs of the ocel- tration in the activator is superior to that of
lus can be explained by a double process of the melanin. The fact that the nonspherical
activator-inhibitory kind; this is the theory of or composed ocelli remain symmetrical in re-
the two gradients. lation to the median line of the alary cell
The focus diffuses a substance, an activator implies that the foci (or focus) of this sec-
the rst gradientleading to the synthesis ond substance is situated at the base of the
of a dark melanin after its concentration has wings and that the inhibitory diffuses mostly
reached a critical limit. Under this limit, the towards the peripherythis is the second
scales synthesize another pigment, a yellow gradient.
papiliochromis for instance. This limit may The respective shapes of the two gradients
be determined by the presence of a second allow us to reproduce most of the motifs thanks
28 Chapter 3 Lepidoptera Description and Scales of Observation
to few parameters. An isotropic scattering from here could result from the fusion of peripheral
the foci would thus create circular motifs if the rings of large ocelli.
second gradient were absent (which entails a Gradients of activator and inhibitor concen-
uniform concentration on the whole surface of tration seem to be the key to the system of ar-
the ocellus) and elliptic motifs if the gradient is rangement of the colored motifs of buttery
strong with the main axis towards it. wings. However, the reality is obviously more
It is quite possible, but not yet established, complex and may imply more than two com-
that the motifs of the symmetric systems are ponents (inhibitor of inhibitor) and saturation
also related to rows of foci running across the phenomena of one component with the emer-
wings from the edges to the interior and ab- gence of a limit.
dominal edges. As a plane wave can be split up Finally, the detailed chronology of chemi-
into overlapped spherical waves, straight lines cal reactions turning tyrosine into melanin or
Figure 3.26. Iridescence effects can be substantially weakened by the scales geometry and arrangement. The
most remarkable effect results from at scales that are little or not at all covered by cover scales, like in Morpho
menelaus or Morpho cypris. In Morpho anaxibia, on the right, structural scales are slightly convexhardly
affecting interferential effectsbut are completely covered by highly scattering cover scales. Iridescence
remains, yet the buttery looks dull, compared to the bright Urania Chrysiridia madagascariensis, the structural
scales of which are highly curved but not coveredon the left.
papiliochromis represents a temporal indica- tennas, where they play sensorial, respiratory,
tor that could explain many details of the poly- and odoriferous roles. To make our description
chromatic structure in a given motif. more simple and before describing scales in de-
tail, we would like to distinguish pigmentary
Scales: Structures and Morphologies scales and the colors they produce from the
structural ones, the color of which primarily
Nomenclature Attempt
originates from physical effects related to the
If we focus on the scales covering wings, one ne structure of the membranes. The caterpillar
should remember that the phaneres recover the already shows embryonic wings as tiny spots.
whole of the body, including legs, head, and an- The imaginal discs split up into vesicles during
the successive shedding. At last, during the On the surface of the scales, an epicuticular
nymphal shedding, the latter evolve to form the skeleton will emerge that will produce the vari-
pterotecae. Onto the folds of these future wings ous structural systems generating physical col-
wrinkled up in the chrysalis, some cells differ- ors. In the same time, pigments will be stocked
entiate turning into trichogenous cells on one into the epicuticale system or outside of it as
hand and tormogenous on the other hand. The ovoid granules.
former will produce attened and altered hair, The general morphology of the scales is as
scales, and the latter will create scales peduncle. fascinatingly diverse as their roles are. We
propose to give an outline of the former striae that are visible through an optical mi-
which has never been completedin order to croscope, and on other scales, a secondary pat-
establish the nomenclature. tern of transversal striae. Other, even smaller
The scales of Lepidoptera usually consist of structures can be perceived that are shut in
two chitineou layers that form the limb. Facing those patterns. The lamellae composing the
the alary membrane, the inferior side is gen- two sides of a scale are joined inside by vertical
erally smooth or slightly undulating, whereas trabeculae.
the superior side, which is turned outward, is As we will see, scales are implanted in a
thicker and shows a great diversity of struc- more or less regular manner into the cuticle
tures. Almost all of the scales present a regular of the alary membrane, which is curved as a
pattern of more or less spaced out longitudinal cupule with angular edges, the axis of which
32 Chapter 3 Lepidoptera Description and Scales of Observation
inclines towards the apex of the wing. Scales Peripheral scales are usually quite long,
present a great variety of shapes as shown sometimes pilliform, but they can also be
by the illustrations in this book: at, concave, haphazardly scattered throughout the wing
undulating, threadlike, pilliform . . . These var- among the standard scales. The end of the limb
ious shapes inuence colors and must be con- can present extremely various shapes: curved,
sidered when analyzing the colorimetry of dense, or even lamentous.
the wing. Within a species, the shapes of
scales greatly depend on their central or pe- Pigmentary or Structural
ripheral implantation on the wing, and also Scales are the center of colors among butter-
sometimes on the implantation of one single ies. We will rst make a distinction concern-
cell. ing the origin of these colors.
Figure 3.33. SEM image of a Morpho menelaus ventral side pigmentary scale. The key to Lepidopterons
colors is concealed in this extraordinary and mysterious architecture. The various elements composing the
thin structure of scales can all be seen here; keystone supporting striae, lamellae partially overlapping one
another along striae and held by a very regular network of microtrabeculae. All scales, whether pigmentary
or/and structural, use these elements to differentiate. Here, the pigment can be found in a diffuse state in
the alary membrane scales, but it can also be found in others as small granules hanging from keystones.
The lamellae, little developed here, dont take part in color. Color actually proceeds from the lamellae that
are excessively long among certain structural scalesproducing real multilayers, sometimes up to a dozen!
Other structures can be contained in the scale thickness, thus not visible by using SEM.
A certain amount of scales are provided with sponds to about a quarter of the visible wave-
more or less complex and regular structures lengths of the spectrum and leads to colored
on their external membrane or into the mem- optical effects. These scales are designated as
brane itself. The standard dimension of those structural. Others, which dont tend to show
structures is comprised between a hundred such structures, have stocked a certain quantity
and several hundred nanometers, which corre- of pigments, either in the constitutive material
Figure 3.34. Scheme representing the arrangement of lamellae in striae. Lamellae are in equal number
anywhere on the stria and form, with the scale superior membrane, an angle that can vary by around 10
according to species.
34 Chapter 3 Lepidoptera Description and Scales of Observation
(a) (b)
(c) (d)
Figure 3.35. Diverse structures of striae. SEM view of Morpho godarti structuralabove, on the leftand
pigmentary striaeabove, right. Structural striae here consist of ve overlapping lamellae kept wide apart
from the membrane by a system of long arches. Lamellae, their overall covered surface attened, are almost
cylindrical at their ends. They are linked to one another by a thin network of vertical ridges around a hundred
nanometers apart. Pigmentary striae actually present the same structure, yet lamellae, very short, do not
overlap. The widening main arches are present all along the inter-striae space, forming small compartments,
in which granules of pigment can be found among some species. Reinforcement pieces are well developed.
Below, Archeoprepona striae: pigmentary on the right and structural on the left. The general arrangement
is roughly similar to that of Morphos, but the structural scales lamellae, which are cover scales here, are
peculiarly distorted.
itself or externally as granulesthese scales tary scales are often lobed and overlap on a
are pigmentary. The distinction is not always signicant part of their surface.
so denite though, and numerous structural Among scales with a well-established role,
scales can also include diffused pigments in we can mention the beautiful androconia,
their structures. a characteristic of Lepidoptera. Aphrodisiac,
Structural scales generally show simple these scales are often very long, sometimes
geometrical shapes that can efciently re- ending in various capitate laments or plumes.
cover without overlapping, whereas pigmen- They are hollow and relate to a glandular cell
Description of the Wings 35
(a) (b)
(c) (d)
(e)
Figure 3.36. Various inter-striae spacings. From top to bottom and left to right, two views of Polyommatus
icarus, and Papilio machaon, Morpho cypris, and Catopsilia orella.
36 Chapter 3 Lepidoptera Description and Scales of Observation
the scented secretion of which goes through the covering the latter by overlapping each other
cuticle and spreads its fragrance around the in- a little bit. Their shape is usually simple with
sect. Androconia can be either grouped in var- regular or slightly dense edges.
ious spots on the wing, or scattered all over the As concerns physical colors, scales are often
surface between the cover scales. arranged in two layers: the under layer scales,
known as ground scales and the outer ones, or
Arrangement of the Scales cover scales on top, which are always very dif-
There is no characteristic arrangement of scales ferent in terms of shapes and structures. De-
on the membrane. If in primitive groups scales pending on species, structural scales can ei-
are spread about without any apparent logic, ther cover (Archeoprepona, Uranidae . . . ) or line
they canespecially among the developed and at the bottom (Morpho), but in the last case,
vividly colored species that concern usshow the covering scale must be transparent or at-
a quasi crystalline-like arrangement. As we will rophied so that light can reach the structural
see in our physical study of the optical proper- scales. In the rst case, the ground scales are
ties of wings, the alary membrane only takes often highly pigmented and form a dark screen
part in the infrared properties of the wing, and against which other scales stand out. Ground
not in the visible coloration. Whatever their and cover scales emerge alternatively along the
originphysical or pigmentarycolors arise same line, the ones covering the others, ground
in the scales. An isolated scale includes all the ones spreading almost over the whole surface
effects of the whole wing. The way it is im- without overlapping.
planted and arranged among the other scales However, we will mention certain species
may greatly inuence the nal aspect. of Morpho among which this distinction is not
With the exception of scattering, colors of a obvious anymore. While presenting structural
physical origin imply a great regularity of the colors, cover scales eventually slip between the
structure, at the microscopic level as well as at ground scales. Covered with only one layer of
the macroscopic, and the effect could be under- iridescent scales, the buttery is radiant. The
mined, or even annihilated if scales were dis- two types of scales have almost similar shapes.
organized. That is why the structures, shapes, However, if cover scales are composed of struc-
and implantation of iridescent scales show a tures, they still arent pigmentary. When im-
rather strict order. They tend to be parallel mersed in a liquid of index, they disappear
to the membrane or slightly inclined, evenly completely and only ground scales remain.
Description of the Wings 37
Figure 3.38. Several internal structures. From top to bottom and left to right;
(a) Papilio blumeia cover scale on a striae
(b) The same butteryground scales
(c) Ornithoptera priamus poseidona cover scale on top of two ground scales
(d) The sameanother type of cover scales.
hundred microns long. Either pigmentary or striae, in most cases parallel to their main
structural, almost all of them present on their axismore rarely oblique or as a spiralmore
superior side a pattern of longitudinal striae or less thick, spaced, and easily distinguish-
and also some other transversal structures, able thanks to an optical microscope. Over the
which we will describe in detail later. Here largest surface of the scales, striae tend to be
is the source of colors, which can still be parallel to each other, crossing the scale from
greatly affected by the general shape of the one side to the other and regularly spaced,
scales. The pigmentary and physical effects especially on the scales producing physical
are clearly distinguishable on at or slightly colors. On the contrary, they can be totally
curved scales. Structures are always situated disorganized on some specialized scales like
on the superior face, while potential pigments androconia.
are found on the inferior face or among the The distance separating each main striae
ground scales. The whole structural surface is varies from 500 nm on certain structural
exposed to radiance. But numerous scales in- scales, which corresponds to 2000 striae per
clude more complex sections (concave, con- mm, to more than 2 m on pigmentary
vex, undulating, or twisted), which can lead scales.
to combined effects that resemble iridescence, The thickness of striae also varies substan-
without being iridescence stricto sensu. This is tially. The highest ones can be as high as a
the case of certain concave structural scales, micron (among Morphos for instance), but less
the apex of which is directed towards the ex- than the hundred of nanometers. All striae
terior of the wing, thus revealing the infe- share a rather simple structure, which each
rior pigmented face of the scales, when ob- species will modify at will by developing or
served from the apex towards the base. In inhibiting certain elements, sometimes exces-
this direction, they look matte and brown, sively.
whereas they turn iridescent in the opposite di- Striae are composed of more or less long
rection. According to this principle, scales pre- lamellae, arising at regular intervals on the
senting several undulations show from a dis- superior membrane of the scale and overlap-
tance successive bright stripes corresponding ping like tiles. Lamellae are thus all inclined
to the crests of the waves and dark stripes cor- on the scale according to an angle of about
responding to the troughs, and thus create a 10 . The longer the lamellae are, the more
changing combination thanks to the incidence they are to overlap. Hence, there are some
of viewpoint. However, this is more the result cases where as much as 12 lamellae overlap.
of effects of geometry and shade than strictly Their length is also remarkably constant and
iridescence. almost equal to a whole multiple of the in-
The morphology of the scales has a great in- terval between their bases, so that the num-
uence on the color of the wing, as well as its ber of layers is constant on every point of the
intensity, purity, and on optical effects in gen- stria.
eral. And yet, the diversity of geometries is
such that it is impossible even to mention them.
The Interstriae Spacing
We will study in detail a few cases to convey
this diversity. The diversity is as great concerning interstriae
spacing. The latter plays a signicant role espe-
cially among butteries that diffuse light since
Fine Structures; Striae and
it is there that structures take place. We will
Interstriae Spacing
describe them fully in the chapter dedicated
We have already mentioned the complex sys- to scattering and will now only give an idea
tem of striae present on the superior surface of of their diversity. Let us mention that this area
the scales. This characteristic is quasi general in the membrane often shows pores, which al-
among Lepidoptera. Apart from a few rare low the gas exchanges with the interior of the
specic scales, all of them include longitudinal scales.
Description of the Wings 39
Coleoptera represents an extraordinary order caterpillars, have legs and pseudopods, while
from every point of view and we will study it oligopod ones are not provided with pseu-
in detail. It is indeed the most important or- dopods, and apodes larvae at least, which
der in the animal world, including the biggest are even completely deprived of any locomo-
insects but especially the most colorful ones. tive organs. Most of Coleoptera have oligo-
Coleoptera appeared before Lepidoptera some pod (carabidae, scarabaeidae) or apod (scolyti-
280 million years ago. They are characterized dae) larvae. These larvae undergo successive
by sclerotic anterior wings: the elytrons. The metamorphoses before reaching the nymphal
latter are not involved in ying but form an ex- stage.
tremely strong protective case (koleos in Greek) If numerous Coleoptera are carnivorous
after which the group was named. and therefore possess strong mandibles, many
have also developed some chemical protec-
tion, which make them repellent in terms
Coleoptera of taste or often even toxic. Like butter-
ies, Coleoptera use Batesian and Mullerian
As indicated by their name, Coleoptera are mimicry, in which their extraordinary colors
characterized by elytrons, anterior sclerotic play a critical part, of course. We wont elab-
wings that recover and protect the abdomen. orate on the subject.
Posterior wings are the only ones used for y- These colors generally recover the whole
ing; they are membranous and folded both lon- body, including legs, antennas . . . and they are
gitudinally and transversally under elytrons, particularly visible among elytrons. Among
which remain folded or half-open during Coleoptera, the cuticle is an incredibly com-
ying. plex material that we will study later. Let us
Among many Coleoptera that lost the ca- notice for now that, contrary to butteries and
pacity of ying, posterior wings can be partly even concerning physical colors, Coleoptera
or totally atrophied. The order is divided into can present optical structures in their scales,
these suborders: archostemata, myxophaga, inside of the wing itself and sometimes in both
and adephaga, composed of the carnivo- at the same time. This must be the price for the
rous Coleoptera including cicindelidae and most beautiful colors existing in nature.
polyphaga. They can be found in almost ev-
ery environment, including running and sea
The Cuticle of Coleoptera
waters.
The great variety of insect larvae is usu- In order to address this second place where
ally classied according to functional rather color can be found that is the carapace
than phylogenetic criteria; hence the tradi- of Coleoptera, we must describe this ex-
tional distinction between polypod, oligopod, traordinary structure covering arthropods: the
and apod larvae. The polypod larvae, like the cuticle.
40
Coleoptera 41
(a) (b)
into many sub-layers that can individually play and oriented parallel to the epicuticle, forming
a part in the coloration of the elytron thanks to layers that produce lamellae by superposing.
their structures. It is this system of lamellaes that originates the
We wont elaborate on this subdivision, the numerous optical phenomena resulting in the
classication of which hasnt been established physical colors of Coleoptera.
yet, but we will study one of the components At this point, helicoid arrangement can oc-
that play an essential role in the elytron color: cur. This is a most peculiar and characteris-
chitin. tic phenomenon providing arthropods with
The chitin molecule is a polysaccharide unique properties of circular polarization.
with a structure similar to vegetal cellulose.
It is a long chain composed of monomers of Helicoidal Structure
N-acetylglucosamine. About twenty in total, In the exocuticle, layers are parallel to each
these molecules form two or three rows, thus other, whereas each microbril turns following
creating almost crystalline sticks: the microb- a relatively constant angle, which results in a
rils. The latter are coated with a proteinic matrix helicoid structure similar to that of some liquid
Epicuticle
Exocuticle
Cement
Waxe
Procuticle
Superficial layer
Cuticle
Outer epicuticle
Endocuticle
Inner epicuticle
Exocuticle
Figure 4.6. The main subdi-
visions of arthropodes cuti-
cle. Divided between endo-
cuticle and exocuticle, the
procuticle lies on the epider-
Epidermis mis cells and is covered by a
very ne epicuticle. Numer-
ous ducts and canaliculi run
across the whole.
Coleoptera 43
crystals in the cholesteric phase. Yves Bouli- The analogy with the cholesteric phase of liq-
gand is the rst who demonstrated it concern- uid crystals may be fortuitous, yet the formal-
ing seashells and later, Neville demonstrated it ism elaborated by Fridel will allow us to briey
concerning insects. The oblique section of such study the phenomenon.
a structure shows a puzzling arrangement of This phase appears with molecules known as
arches, a result of this helicoid piling accord- chirals, which means that their image cannot be
ing to Bouligand. superposed in a mirror. This is the case of every
One can also notice it on a section of the tu- organic molecule including at least one asym-
bercles or concavities in the elytron. The lay- metrical carbon with four different groups.
ers thus look like concentric circular bands The main physical result of this characteristic,
in which brilles are all parallel to each called chirality, is an optical activity, which en-
other. tails that the direction of the light polarization
(a) (b)
(c)
Figure 4.11. Views of the surface structures of elytrons at different scales. Millimetric alveolae in a cassida:
below. In the center, the slightly depolished surface of a Scarabeide Eudicella graf. Left, Plusiotis cupreomaig-
inata smooth surface.
46 Chapter 4 Coleoptera Description and Observation Scales
(a) (b)
Figure 4.15. Various arrangements of Coleopteron scales. Above, sh-type arrangement: Hoplia cerulea C.
Scales overlap to a large extent. Below, scales are more scattered, like pebbles: Curculionidae.
48 Chapter 4 Coleoptera Description and Observation Scales
is covered with scales like Lepidoptera. Their cuticle. However, there are cases presenting a
structure is nevertheless quite different since sh type arrangement in which thin and at
we always nd the same internal crystalline scales overlap.
organization, which is rare among butteries. With a few exceptions of the latter kind, the
There are numerous families of Coleoptera pro- scales of Coleoptera are thick, hard, circular,
vided with colored scales. The principal one is and convex. They are usually attened against
the Curculionidae, which includes our dear and the elytron, making it difcult to perceive the
rather dull weevils, but also the Scarabeidae peduncle. Their external surface is the more
and especially cetonia, well known for their ex- often loosely folded. However, it can be cov-
traordinary metallic shine. ered with tiny nipples resembling those of the
Contrary to Lepidoptera, scales are gener- corneal lens of ommatidium (compound eyes
ally arranged without apparent order and often of many butteries), where they can play the
scattered, discovering most of the underlying same role of antireecting layer.
5
Changing Colors: Structures
or Pigments?
49
50 Chapter 5 Changing Colors: Structures or Pigments?
Let us begin with basic remarks: Pigmentary colors or colors proceeding from
selective scattering hardly vary, whatever
Blue pigments are rare in the animal world.
the angle of observation or illumination, ex-
Most of them are metallo-protein pigments
cept when angles are very high and for
characterized by the presence of one or sev-
highly polarized light. The corresponding
eral metals on various protein molecules,
aspect is matte, even dull.
the most common of which is haemocyanin,
which derives from copper. It is usually These experimental techniquesmodica-
found among mollusks, crustaceans, and tion of the index or of thicknesspermit us
some arachnids in solution coloring plasma in many cases to distinguish between struc-
blue. tural and pigmentary origin of colors. They
On the contrary, red, yellow and brown pig- always lead to the alteration or disappear-
ments are widespread, especially among ance of colorsa real opportunity for nature
butteries, either as a product of degradation and the opportunity for us to mention all the
(such as melanins, widespread in the animal types of color alterations, of chroma, observed
world), or as tetrapyrolique pigments (like among insects. Let us only note that color al-
hemoglobin) from which derive the vivid terations are quite rare among the latter. There
reds of numerous vanessas (Inachis io, Aglas are physiological modications, rather quick
urticae . . . ). and generally reversible, and morphological
changes generally irreversible resulting in a
If we now examine the wings from the point
new pigment. We will focus on the former,
of view of the variations in colors related to
which are obtained through the modication
the angle of incidence, we can establish a new
of the cuticle structure or the migration of
selection:
pigments.
Interferential or diffractive colors change sub- We will call the devices leading to color vari-
stantially depending on the angle of observa- ation with the generic term X-chrome, X desig-
tion or illumination. This is the phenomenon nating the factor causing the variation, thermo
of iridescence, one of the most spectacu- for temperature, gonio for the angle, and hy-
lar among tropical butteries like Morphos, gro for humidity . . . Many of those phenom-
Archeoprepona, and Urania, but also, for in- ena dont involve color strictly speaking but
stance, among the little common forester rather brightness, that is why we will briey
(Procris Statice) in temperate regions. The mention them. Goniochromic structures on the
aspect of iridescence is here bright and contrary are the very focus of the present
metallic. book.
54 Chapter 5 Changing Colors: Structures or Pigments?
Changing Colors
Goniochromic Structures
They strictly concern iridescence. Indeed, color
changes according to the viewpoint or angle
(gonio in Greek) of light, which means that color
is physical here, except for the very specic
56
LightMatter Interaction, Polarization, and Optical Index 57
Denitions Relations
Linear Polarization
Let us rst note that when a wave, even non- senkrecht) and the other Ep being parallel to the
polarized, falls on a surface, specic directions plane (parallel, in German).
show up ipso facto. If one calls plane of inci- The two elds dene two perpendicularly
dence the plane dened by emergent and in- polarized waves, which, if they follow Snell-
cident rays, whatever the direction of the in- Descartes laws establishing their propagation
cident eld, it is always possible to express it directions, can interact with matter in differ-
as two vectors: one Es being perpendicular to ent ways. One of the most noticeable con-
the plane of incidence(from the German term sequences is the angle dependence of their
reected intensity (Figure 6.3), which is often Coleoptera, that we should present: circular, or
observed among butteries. more generally, elliptic polarization.
Although that of the s wave, with its electric The vector nature of the electromagnetic eld
eld remaining in the layer plane, whatever the entails a matrix treatment of this kind of prob-
incidence, hardly depends on the latter, that lems. We wont try to study this matrix formal-
of the second wave p presents a surprising ism here. We will only mention it through a
angle variation. Under a particular incidence, descriptive and intuitive approach, which will
Brewster incidence, the p wave reectivity is allow us to give a good overview of the differ-
equal to zero! If one illuminates a diopter under ent phenomena observed among insects. When
the B Brewster incidence, the reected light one studies a single vector, like the E eld for
is strictly s polarized. A useful consequence, instance, it is common and convenient to ori-
which is well known among photographers, is ent one of the reference axes parallel to the
that under this incidence, it is possible to sup- vector. Indeed, there is no point in consider-
press the reections of a surface. Another con- ing two componentswhich are not equal to
sequence that directly concerns us is that under zerowhen one is enough.
the same conditions, the light reected by some However, we wont proceed this way. Let
butteries can be highly polarized. It is well us consider a reference in which the vector E
known that contrary to humans, many insects has two components: Ex and Ey (see Figure
are sensitive to light polarization. Their chro- 6.4). In addition, if this eld oscillates with
matic messages are thus enriched with a new a given frequency and moves in a direc-
aspect! The Brewster angle only depends on the tion k, with a speed c, so does each of its
indices of the materials bathing the diopter: components, which thus behave like two inde-
pendent waves and form the two polarization
tgB = n/n0 (6-1) states of the electromagnetic wave. In any wave
Brewster angles of biological materials, with planea plane perpendicular to the propaga-
indices close to n = 1,5, put in air (n0 = 1) range tion directionthe E vector extremity moves
between 50 and 60 . forward and back along a straight line segment.
This proves that the plane wave is indeed po-
Circular Polarization larized linearly.
It is well known that in vacuum, light speed
Linear polarization as mentioned above is a is c 3 108 ms1 . Yet, in any other material,
mere particular state of light, which is im- this speed diminishes proportionally to the op-
portant for us since it causes interesting opti- tical index value of the material:
cal phenomena in butteries. However, there
is another state of light generated by some v = c/n. (6-2)
LightMatter Interaction, Polarization, and Optical Index 59
r The eld can nally tend to orienton inary part characterizes the absorption of the
averagemolecules containing a permanent material.
electric dipole in a given direction. = 1 + i2 ,
(6-5)
In any case, there is a displacement of n = n ik.
charges. In electromagnetism, this displace- We wont elaborate on the calculation of the
ment is called D. As long as the applied eld dielectric function, which doesnt concern us
is not too stronglike with sunlight and tradi- directly. A simple method consists in consid-
tional electric lampsthe displacement is pro- ering the charges in motion as small masses
portional to the electric eld: carrying electric charges linked to the nucleus
or to neighboring charges by springs: This is
D = E. (6-3)
the mechanistic model, basic but allowing us
The proportionality constant is a character- to understand most of phenomena. For cer-
istic of the material, or rather of its response to tain specic frequencies, 0 , this oscillator res-
the action of light. Since it depends on the wave onates and the absorption is maximal. The
frequency, it is called dielectric function of model leads to the following expression of the
the material. By denition, the optical index is dielectric function:
the square root of : Ne2 /m
() = n2 () = 0 + 2 ,
n() = () (6-4) 0 2 i
() = 1 () + i2 () and n() = n()ik().
This mere denition doesnt tell us anything
about the aspect, the value, or the frequency (6-6)
dependence of this index . . . Let us now con- N is the number of oscillators, i.e., the num-
sider the latter, with the least equations pos- ber of molecules or polarized atoms per unit
sible. It is important to note rst that if the volume, and is a damping term, both param-
displacement D (i.e., the effect), is indeed pro- eters characterizing each kind of atoms. As il-
portional to the applied eld E (i.e., the action), lustrated by Figure 6.7, the 2 imaginary part
they are not systematically in phase; a eld can variation, directly linked to the optical absorp-
vary very quickly, while a charge can be slowed tion, looks like resonance peaks centered on
down in its motion. the various 0 frequencies. As concerns the
This phase difference is expressed by making 1 real part, it tends towards the static dielec-
become a complex quantity with a real part, tric constant (0) of the material at low fre-
1 , and an imaginary one, 2 , which directly ac- quency ( 0 ), whereas at high frequency,
counts for the energy provided by the wave in it tends towards that of vacuum by decreasing.
order to move charges, i.e., the energy absorbed This gradual decrease of the real part of the di-
by the material. Like the dielectric function, the electric function of insulators is important and
index n also becomes complex and its imag- worth mentioning. One can show that the real
62 Chapter 6 Physical Colors, Chemical Colors Basics of Solid State Optics
and imaginary parts of the dielectric function dielectric function, but that, on the contrary,
are not independent, and that they are related every integral has a positive value at low fre-
to each other by integral relations: Kramers- quency ( n ) and their different contribu-
Kronig relations. tions n are additive and approximately equal
to:
2 ( )
1 () 0 = 2 2 d ,
2
2 2 ()
0
(6-7) 1 n d . (6-9)
[1 ( )0 ]
2 () = 2 2 d
. picn
2
0
In other words, each absorption peak con-
I decided to mention these relations here be- tributes to 1 at low frequency, and so does
cause they have an important physical mean- the index real part n, which is always larger
ing. First, with one componentgenerally 2 , in the red side of an absorption peak (low fre-
which can be more easily determined through quency) than on the blue side (high frequency).
experimentone can calculate the other one. This is nothing than the mathematic expres-
Second, these relations express the index dis- sion of a well known phenomenon: The larger
symmetry on both sides of a resonance, and the mass, the more difcult to make it oscil-
therefore selective reection, i.e., surface color, late at high frequency. In the case concern-
that was so dear to Michelson. ing us, when frequency is gradually increased,
As has been seen, the 2 imaginary part of the molecules, atoms, and electrons stop vibrating
dielectric function looks like peakspossibly successively. At each step, the dielectric func-
severalcentered on a n frequency, and be- tion decreases to nally reach the value of the
tween these peaks, 2 is almost equal to zero dielectric function of a material in which noth-
(6.7). The integral resulting in 1 as a function ing vibrates: vacuum (0 ).
of 2 , which is a sum on all frequencies can Let us nally note that the index varies sig-
thus be expressed as a series of integrals over nicantly only when it is close to resonance
the only ranges where 2 is not equal to zero, frequencies. Outside of these high absorption
i.e., on the various absorption peaks: areas, within transparency windows, the index
2 2 ( ) real part is not strictly constant and slightly
1 () = 0 + 2 2
d . varies with wavelength. Rather than using ex-
n
picn pression (6-6)which still remains valida
(6-8) better account of the variation is given by an
One thus notes that at high frequency ( empirical expression. This polynomial expres-
n ), all of these integrals are equal to zero sion as a function of , or , is easier to use. Its
and 1 tends towards 0 , demonstrating that coefcients are determined by tting the exper-
the various absorptions dont take part in the imental variations of the index of each material.
index values. One can thus show that the vol- quently the dipole value depend to a great ex-
ume average of the dielectric functions rep- tent on the inclusion geometry. One thus intro-
resents a good approximation of expression duces a shape factor, called depolarizing factor
(6-12). This is the linear theory of mixing, which A, which modulates the intensity of the dipole
we will be able to apply to butteries in most generated by the eld. It is generally difcult to
cases: calculate this coefcient. It leads to integral ex-
pressions with no analytic solutions, which can
e = pi + (1 p)m . (6-13)
be determined only numerically thanks to aba-
Maxwell Garnetts research focused on the cus. In the more common case of spheroids and
optical properties of colloidal suspensions of other simple geometries, these elliptic integrals
metals and on the colors of stained-glass win- are expressed by using basic functions de-
dows obtained from metal inclusions in glass. pending on the inclusion eccentricity. We are
It is to this phenomenon, among others, we only concerned with their graphic represen-
owe the extraordinary blues of Medieval Age tations and more especially the extreme cases
stained-glass that illuminate our cathedrals. In of long innitely inclusions, like chitin sticks,
this specic case, the quantity of metal inserted or extremely at inclusions, like stria lamellae.
is extremely small and metal atoms, lost in a These shapes are the closest to that observed in
dielectric ocean, join together and form small our biological structures. The coefcient varies
aggregatesseveral tens of nanometers big from the value A = 1 for a at inclusion to A =
that are somewhat spherical at this stage. In 0 for a very elongated inclusion when the eld
order to use this approach in the framework is oriented in the direction of the revolution
of the biological structures concerning us, one axis. In the perpendicular direction, it is equal
must consider nonspherical inclusions. This to half of (1-A), i.e., 0 in the rst case and 1/2 in
development of the model was started more the second. Those are the only values needed
than half a century later. When an inclusion is to model optical structures of insects.
subject to an electric eld, a displacement of A mean eld calculation according to
charges systematically occurs. This results in a Maxwell Garnett results in the following
charge accumulation on the periphery of the in- tensor expression of the dielectric function,
clusion, which thus behaves, as a rst approx- established by Cohen et al. in 1973:
imation, like an oscillating dipole if the eld is
i A + p(1 A + m 1 p (1 A)
periodic, which is the case in optics. One can e = m ,
intuitively understand that the distribution of i A 1 p + m 1 A(1 p)
charges on the inclusion periphery and conse- (6-14)
This tensor expression reduces to the where is the wave function of the particle and
Maxwell Garnetts scalar expression when A = E its energy. We wont try to solve this equation,
1/3, which is depolarizing factor of a sphere. It which is actually not so easy in the present case.
is important to note that for a given structure, e We will rather keep in mind that its solutions
wouldnt have the same value for two different are discontinuous; propagation cannot occur
orientations of the electric eld. This is partic- with any energy. Allowed energies are grouped
ularly true in Morphos, when the structure is in bands separated by energy areasforbidden
illuminated with Transverse Electric or Trans- areasthat the particle cannot experience. The
verse Magnetic polarized light. This anisotropy potential periodicity causes a partial quanti-
of the material index results in both colors and cation of energies.
polarization effects observed in Morphos. Let us now come back to optics, and more
precisely to buttery structures. The particle is
now a photon and its associated wave is the
Structural Origin of Colors; What electromagnetic wave. What happens when the
Kind of Logic for Structures? particle tries to propagate within this medium
presenting a periodic alternation of index, the
The Ordered Phase; Photonic Crystals equivalent of the electron electric potential?
Before Time Strictly the same thing happens! The equation
to solve is now Helmholtzs:
In the traditional sense of the term, the crys-
tal phase characterizes a periodic organization n2 2
2 = 2 , (6-16)
of atoms both at short and long distances. In c
any place on the crystal, one can observe that which is expressed in the exact same way
atoms are distributed in the same way. This as Schrodingers equation (the frequency
strict periodicity has surprising consequences replacing the E energy) and leads to the same
on particles trying to move within the crystal. type of solutions: successive bands of per-
In order to solve such a problem, quantum me- mitted frequencies alternating with bands of
chanic starts by strictly modeling the shape of forbidden frequencies. The index periodicity
the potential subjecting the particlegenerally causes a partial quantication of the frequency
an electronwhen the latter comes near atoms. range.
This approach substantially simplies calcu- The term photonic crystal proceeds from
lations, without losing any physical meaning, this analogy between electrical phenomena in
and gets us closer to the optical phenomenon a crystal and optic phenomenaalso called
that are interested in here. photonicin a structured medium.
One then has to solve Schrodingers famous It is in this context that we will describe the
equation, which in this case is expressed as fol- structures of the wing and the associated phe-
lows: nomena. From the point of view of geometry
indeed, one can classify photonic crystals ac-
2m cording to the number of dimensions in which
2 = - 2 (E V), (6-15) periodicity develops.
h
66 Chapter 6 Physical Colors, Chemical Colors Basics of Solid State Optics
p
la
overlap and cannot interfere.
er
Ov
interference phenomena encountered in a thin ing the two examples presented in Figure 6.12,
layer. disorder can proceed from two different ways.
After studying ordered structures, or crys- The method that may be the more intuitive con-
talline phases, let us now see what happens sists in arranging elements at random (Figure
when this beautiful arrangement deteriorates, 6.13 (b)). One can easily understand that there
i.e., when it gets in disorder and when the phase is no phase relation between the two waves
becomes amorphous. diffracted by each of the elements. In fact, such
a structure doesnt lead to any diffracting phe-
Amorphous Phase nomenon.
One can also, as shown on Figure 6.13 (a),
All the phenomena that have been mentioned
randomly remove elements from an ordered
(interferences, diffraction) occur because there
structure. There is no order over a short or long
is a relation between the phases of the two
distance anymore. It is indeed an amorphous
waves diffracted by two neighboring objects,
phase, although all the distances between
and because the periodicity of the two ob-
elements remain a multiple of the initial crystal
jects enables this relation to spread to all the
period. In a given direction, diffracted waves
waves, thus leading to a global effect that can
can thus very well be in phase, actually out of
be observed. Were the periodicity lost, the ef-
phase of n wavelengths the one to the others,
fect would vanish. Phase relations would still
with n varying randomly. Consequently, noth-
occur here and there, but they would not pro-
ing prevents interference phenomena from tak-
duce any global phenomenon. This is amor-
ing place. Yet, this hardly occurs with natural
phous phase. Light is scattered. As the beau-
light. The reason is quite simple: waves are not
tiful arrangement has been lost, only the sizes
innite but truncated. They have a beginning
of the scattering objects and the compactness
and an end. One thus speaks of trains of waves,
of the whole play a part. This is this type of
and the nite length of these trains is called .
structure that produces most of the blue and
If the distance between two objects is superior
white colors found among butteries in our re-
to this length, then the two resulting waves
gions. The very ne structures of Argus pref-
even if they were in phasedont superpose
erentially scatter blue, whereas the structures
and therefore cannot interfere. We will call co-
of bigger ovoid grains of Pieridae scatter all
herence length, this distance beyond which two
wavelengths, hence their whitish color.
waves in phase dont interfere anymore. Let us
only remember that a periodic structure is not
An Important Parameter: Coherence Length
necessarily sufcient to produce constructive
Disorder can be seen as quite an intuitive no- optical phenomena and that the latter can only
tion, but it is actually very complex and we proceed from ranges that are smaller than the
wont tackle it. Let us only note that consider- coherence length.
7
1-Dimensional Structures:
Interferences
68
Theory Recalls 69
Ep r
Es
i r i
ni
n I K
e n
r
t J
t
Figure 7.1. A diopter and its normal form an orthog- Figure 7.2. The optical path of the rst rays reected
onal coordinate system in which any wave can be and transmitted by a thin layer. The intensities of
split into two waves vibrating respectively in the the rst two reected rays are almost similar. The
plane of incidence (Ep ) and in a plane parallel to the intensities of all the transmitted rays, excepted the
layer (Es ). rst one, are negligible.
transmitted intensities with the incident inten- direction than in its amplitude falling on the
sity, which can be directly obtained from spec- layer inferior side.
trometric measurements: The Thin Layer
Re {n} 2 The wave reected on the rst interface is ob-
R = |r|2 and T = |t| (7-3) viously not affected by the presence of the sec-
ni
ond one, but the transmitted wave now plays
In the case of butteries in a weakly or non- the role of an incident wave with a different an-
absorbing medium like air, which we will often gle t . It here splits, following the same laws at
encounter, the reection and transmission fac- the superior interface into a transmitted wave
tors become: of high amplitude and a reected wave of low
n 1 2 2 2 amplitude. The latter, back on the upper in-
R= and T = n . (7-4)
n + 1 n + 1
terface, splits again into a transmitted wave
and a reected wave. Although this play of
Before going further, let us note that indices come and go between the two faces of the layer
of the most widespread nonabsorbing natu- goes on innitely, producing an innite num-
ral/biological materials (for example, water ber of transmitted rays, it is not necessary to
and chitin) are not very highapproximately analyze it further. By following point by point
4% in the mentioned casesand that most of the trajectory of the rst transmitted wave and
the incident wave is transmitted. It is therefore the evolution of its amplitude, one can easily
a wave that is more affected in its propagation observe that the rst two reected rays have
approximately the same low intensity, whereas All basic physic phenomena are contained
that of the following reected rays are insignif- in the treatment of the unique layer and to
icant. In the same way, only the intensity of the mathematically analyze the multilayer would
rst transmitted ray is important; interference not provide us with much more information. It
phenomena proceeding from the reection on only deprives us of the pleasure to present my
thin layers can be considered as phenomena teachers, F. Abeles, delicate matrix treatment,
of two waves with almost similar intensities, in which every layer is represented by a matrix
and thus of maximum contrast. These two rays, linking the elds at the entry and at the exit,
generated by the same incident ray, are coher- the effect of the multilayer being obtained by
ent and interfere. In a given direction, these in- multiplying matrices.
terferences are constructive only for a given
wavelength, which indeed gives the wanted Summary
iridescent effect in the visible. The main results, as derived from the study
The period difference between the two rst of the unique layer or multilayer are: (a) there
reected rays is: is no absorptionthe multilayer interferen-
tial mirror is not a colored lter in the lit-
= 2ne cos t + . (7-5) eral meaning of the wordand nonreected
2
wavelengths are transmitted. The color of the
The latter are thus out of phasedestructive
transmitted light is thus the complementary
interferencesif d is equal to an odd number
of that of the reected light. (b) As shown
of half wavelengths:
by the equation (7-8), the wavelength re-
ected by a multilayer mirror varies linearly
(2k + 1) = 2ne cos t + (7-6)
2 2
For a given conguration, there is a minimal
reection for each wavelength verifying
N = 10 N=5 nh = 1.56
0.8 nl = 1.00
0.6 N=2
0.4
(a) (b)
(c)
Figure 7.7. Some examples of interferential colors. (a) a solid-solid multilayeran Aliotis Iris nacre. (b) a
liquid-liquid multilayer; a thin layer of petrol over a water lm. (c) a solid-gas structure quite similar to
that of Butterieshummingbird feathers.
Type Butteries with Convex after them: Morpho type and Urania type scales.
Structures Indeed, they are big butteries presenting areas
of rather homogeneous colors and are well-
Chrysiridia madagascariensis, Urania leilus, suited to routine spectroscopic measurements.
and Prepona We will comply with this convenient tradi-
tion. Yet, facility here is only apparent since
Interferences and diffraction are the two physi- at the microscopic scale, the two families do
cal phenomena leading to the most spectacular present two specic types of scales. Yet con-
iridescent effects. They occur among miscella- cerning Uraniidae, their scales are quite de-
neous buttery species throughout the world, formed, which makes it difcult to get a gen-
yet hardly with the same brightness as the large eral view, and as concerns Morphidae, the effects
neotropical family of Morphidae and the more produced at the smallest scale are then modu-
smaller one of Uraniidae represented by the lated.
Madagascan species Chrysiridia madagascarien- We will start our study of iridescence
sis and South-African Urania leilus. as produced by convex structures of Urani-
For practical reasons, the study of colored idae (mostly Chrysiridia madagascariensis) and
effects has always been carried on these two Charaxinae (Prepona), which provide a better ba-
families. Certain scales have even been named sis for the interpretation at a small scale. We
Type Butteries with Convex Structures 73
(a) (b)
Figure 7.11. The two scales types of Urania leilusin arrangement (right) and isolated (left). Ground scales
form a black screen under cover structural scales that present an interferential color variation from the crest
to the root.
Type Butteries with Convex Structures 75
(a) (b)
Figure 7.13. SEM views of the dorsal side of Chrysiridia madagascariensis, (a), and of Urania leilus, (b). They
show the general arrangement of structural scales and their signicant convexity. One can distinguish
pigmentary ground scales under the indented apex.
absorbing screen on which the second layer of two types of scales present not very distinct
structural scales is placed. The latter, strongly longitudinal striae, with a larger spacing on
convex, slightly pigmented, and their sides ad- the structuralabout 4 m apartthan on the
joining, form long rolls resembling waves, on pigmentary ones (1 to 2 m). There are al-
the crests of which the light is reected, thus most no counter-striae on the structural scales,
drawing long colored bands transversal to the while pigmentary scales present some that are
wing (Figure 7.18.). An enlargement with scan- very distinct and regular with spaces from 230
ning electron microscopy enables us to under- to 300 nm. It is this quite condensed alveo-
line the general shape of the scales as well as lar structure, which is responsible along with
their supercial structures (Figure 7.12). The melanin for the strong visible absorption of the
ground scales, that we will further describe in croscopy (TEM) reveals the superior mem-
the study of the thermodynamic properties of brane consisting of 13 to 15 overlapping al-
butteries. ternate layers of air and chitin, 90 nm thick
Let us now observe the striae of the struc- on average, and supported by microtrabeculae.
tural scales. They show a traditional structure TEM also reveals the external layer presents
composed of short lamellaes that dont over- many pores. The penetration speed of the
lap, the length of which is almost equal to various index liquids in the scales suggests
their spacing. Given their structure and spac- that the deeper layers should also include
ing, they cannot be the source of the strong pores.
reection that one can observe. One must get Between two striae (that is, on distances
into the scales in order to discover the se- about 4 m), the structure is rather at and is
cret of the Urania. Transmission electron mi- easily modeled by a symmetric multilayer of
Figure 7.16. Cross section of a Chrysiridia madagascariensis structural scale cut almost through the middle.
The spacings between two striae, slightly concave, can be likened to planes tilted of about 10 in relation to
each others.
Colored Effects 77
(a) (b)
(c) (d)
Figure 7.20. Structural organization of PapilioPapilio ulysse. (a) cover structural scales. (b) a striated net-
work and undulations forming a counter-striae network. Below, a section performed perpendicularly to
striae next to the center of a basin and showing the (c): SEM, (d): TEM.
polarization effects are minor because of the proceeds from a pointillist effect of the out-
low incidence of the wave. Even further from line, blue framing green at an invisible scale.
the center, the result is totally different. The in- Interferential colors generated by curved struc-
cidence increases, getting closer to the Brew- tures always entail polarization effects by re-
ster angle (Figure 7.22). Polarization effects ex- ection. In the case interesting us here, these
pand and the blue becomes more intense. A effects also occur at the scale of the thin struc-
double reection occurs, making light emerge tures of scales (basins), not always perceivable
in the incident direction. There is no irides- at the buttery scale. The reason why is quite
cence anymore but a mixing of colors with the simple, as illustrated in Figure 7.24.
wavelength coming from the center. Here is a When a nonpolarized wavelength falls onto
peculiar case of combination of structural col- a basin, under normal incidence for instance,
ors. The nal color, with a green predominant, we know that in the areas that are inclined
80 Chapter 7 1-Dimensional Structures: Interferences
Figure 7.21. Iridescent effects on Papilio illuminated under grazing incidence on the right hand-side, and
under normal incidence on the left hand-side. From top to bottom: Papilio ulysse, papilio paris, and papilio
blumei.
Figure 7.22. Reection in a concave multilayer. Under low incidence (a) the center is illuminated under near-
normal incidence and produces a generally yellow-brown nonpolarized color by interference. Incidence is
a bit higher on margins, the color approaches green, goes out under an oblique angle, and remains weakly
polarized. Closer to the basin (b) incidence is close to the Brewster angle, itself close to 45 . The reected
light is blue, highly polarized, and goes out after a double reection, in the direction of incidence. (c) Aspect
of a basin illuminated under normal incidence.
Papilionidae: Solid/Gas Structures 81
according to the Brewster angle B , only the to it. When the double reection occurs with
s component is reected. If the angle is not ex- light deviating in the incident direction, there
actly equal to B , a small part of the p wave will are two successive reections both increasing
also be reected. Within an angle of 10 around the proportion of wave s over wave p, which
B , the p proportion doesnt exceed 10%. This results in a highly polarized ray, even if the an-
case is particularly interesting, because if the gle is not optimal.
Brewster angle is not exactly equal to /4 However, such a device doesnt create a
more around 55 it remains still very close polarization effect at the scale of the whole
(a) (b)
buttery, or only indistinctly in some cases. This alveolar system includes small protuber-
Basins present a relatively central symmetry, ances surrounded by some ten alveoles and
which makes that wave s that is reected by presenting a color that can be quite different
the two opposite edges of the basinlet us say from the rest, sometimes violet-blue.
the left and right edgesis a wave p from the The cuticle has a traditional structure, bril-
viewpoint of the two perpendicular edges (top lated in the exocuticle and multilayered in the
and bottom). At the scale of the scales or of epicuticle. There are up to 12 stacked solid lay-
the whole wing, the reected ray includes as ers of 150 to 200 nm. They appear slightly con-
many s polarized waves as p polarized waves. trasted under the electron microscope, hence
The Papilio Ulysse is a good illustration. A dis- a slight contrast in the optical index, probably
crepancy can only occur when certain edges minor yet sufcient to create the resulting col-
are constantly longer than their perpendicular ored effects.
counterparts, like Papilio paris for instance, in All the optical phenomena described in Pa-
which the blue chromatic component is slightly pilios can be found here, especially polarizing
polarized after the double reection. effects in alveoles. However, the latter dont
show any lengthening in a denite direction,
that is why there is no polarization including
Cicindelidae: Solid/Solid Structures the whole insect.
(a) (b)
(c)
Figure 7.28. SEM images of volume and surface structure of Cincidela campestris. (a) Surface alve-
olae and protuberances. (b) A section of the cuticle showing the exocuticle chitineous brillae
and the nonbrillated epicuticle. (c) The multilayer hemispheric structure of the epicuticle in an
alveola.
that generate the peculiar polarization effects, This case is an anisotropic solid/solid mul-
as we have already described in the previous tilayer structure with a director rotating from
chapters. one layer to the other, creating a periodic gra-
dient of index, though quite weak, in the epi-
cuticle. This little difference in index is com-
pensated by the large number of layers leading
to the famous bright glance. It is important to
note that there is no alternation of materials
with low and high indices in the multilayer.
Yet, the latter results from the rotation in its
plane of a layer composed of a single mate-
rial with two different indices in perpendicular
directions. Interference occurs between layers
with the same direction. It is associated with
Figure 7.29. The Rose chafer Cetonia aurata in the an optical activity, which can be demonstrated
ying position. by using a /4 slide introducing a delay of
84 Chapter 7 1-Dimensional Structures: Interferences
130 to 135 nmso, a /4 slide for a wavelength The Scarabeidae family includes other mem-
= 550 nm corresponding to the green color of bers with elytrons covered with scales. The
the Rose chafer for instancefollowed by a lin- arrangement, morphologies, and structures of
ear polarizer. these scales differ greatly from one species to
All the measurements suggest that polariza- another. We have already mentioned in the
tion is counterclockwise on the two elytrons third chapter the singular arrangement of Ho-
and that the axial symmetry of the insect is plia cerulea scales, resembling that of a sh, but
thus not respected. The origin of the layer ro- also as singular the scales of Hoplia argentea that
tation remains obscure. It could be the result are covered with thin radicles.
of the daily alternation of illumination dur- In the absolute, treating such a problem is
ing the nymphal stage. It can be suppressed tedious, yet it is somewhat simplied in the
by keeping the insect exposed to a constant present case, in which the coefcient of phase
illumination. delay = (ne n0 )k0 is large compared to the
(a) (b)
Figure 7.31. Cross section of Hoplia scales. On the left, a female Hoplia argentea scale and Hoplia cerulea
on the right. Layers are well visible and set equally apart, but their individual structures are quite
disorganized.
Cetonia: Circular Polarization 85
rotation coefcient of the director given by angle and the one remaining parallel to the
= Z. (ne and n0 are the extraordinary and quick axis (Ex on the gure) is ahead with a
ordinary indices of the material, k0 the wave phase difference of .
vector, and Z the traveled distance measured
from the surface.)
One thus demonstrates that the Jones matrix
of the system giving the polarization state of
the wave after a travel on a distance d can be
expressed as follows:
d
exp j 0
2
T = R(d) ,
d A microscopic observation reveals in the
0 exp j scales a multilayer solid/gas-type structure
2
without any specic orientation of brilles. Ho-
i.e., like the product of an optical activity and plia argentea radicles, whether male or female,
a dephasor. 300 nm in diameter and 1 to 2-m long, scatter
After a travel on a distance d in such a ma- the reected light and weaken substantially the
terial, the two components have rotated of an elytron brightness.
8
2-Dimensional Structures:
Interferences and Diffraction
86
Theory Recalls 87
Figure 8.5. (a) Selective diffraction. Long wavelengths are either transmitted or absorbed. Only short
ones can be found in each order, including in order 0. (b) High dispersion. Under normal incidence, a
2000 groves/mm grating diffracts only the wavelengths shorter than 500 nmblue-greenin the rst
order, while they are all present in the 0 order.
30 species and many sub-species, all present- a priori. It is also a Morphalidae widespread in
ing a strong sexual dimorphism. That is why it the Amazonian basin and even in Bolivia. The
is not an easy task to choose a species for our male is entirely blue on its dorsal side, except-
study. Our predecessors tended to choose one ing the abdominal and costal edges, fringed
more because of some opportunity rather than with black. It is cryptic on its ventral side
based on rigorous scientic reasons. with three ocelli on posterior wings and two
All male Morphos show marked iridescent ef- on the anterior, all little developed. As has al-
fects with a predominant blue. Some are very ready been shown in Chapter 5, these colors
dark, of a deep blue, while others are almost vary with the exterior medium index, demon-
white with only a slightly bluish touch. Many strating without any ambiguity their structural
present marked bright colors, whereas others
have a matte surface that seems to extensively
diffuse light. One could thus classify them very
basically in a plane according to two criteria:
brightness being the rst coordinate and rang-
ing from 1 for a mirror to 0 for a perfectly
scattering surface, while the second coordinate
representing spectral purity ranging from 1 for
a monochromatic color to 0 for white. This is
merely a subjective classication, which isnt
based on any measurements. We have chosen
our standard Morpho (Morpho menelaus) among
the species classied in the center of this plane,
eventually selecting the more common ones in
order to study them morphologically and opti-
cally, and to widen our study with some other
species of the plane by focusing more on dif-
ferences.
Figure 8.7. Reection factor of a Morpho menelaus whole wing for different negative incidences, (from the
base to the apex). Angles have been measured from the normal to the wing plane. One must add 10.5 in
order to obtain incidence on the interferential structure.
origin. Immersed into a medium with a rather tion, amplitudes differ, as can be seen on the
high index, close to that of chitin, and when col- chromatic lines obtained from spectral mea-
ors have totally vanished, the buttery looks surements and which, contrary to an isotropic
extremely dark, revealing concentrated pig- thin layer, do not overlap. This observation
ments on either face. questioning Michelsons interferential hypoth-
Iridescence is particularly marked, as shown esis could, on the contrary, conrm it. As has
by the series of pictures in Figure 8.8. The been seen in the last chapter, the reection min-
evolution and characteristics of colors will be ima of an electromagnetic wavelength on a thin
studied further, but let us now rst focus layer are given by:
on the aspect of the buttery under various
kmin = 2ne cos t
illuminations. What is the most remarkable
initially is the important dissymmetry of col- This equation can be rewritten by involving,
ors in the wings when illuminated under the not the t refracting angle, which we dont a
same angle from the right or the left. This sug- priori know, but the i incident angle. Snells
gests that reecting structures on the wings law allows to modify this variable and results
are inclined. An obvious fact that Michel- in the following apparently more complex ex-
son missed, yet he had then to assume that pression:
the structure existed. Visible and near-infrared
spectrometric measurements, performed on k2 2
sin2 i = + n2 (8-4)
the big homogeneous areas of cells M3, Cu1, 4e2 m
Cu2, and discoidal, conrm this initial visual Nevertheless, under this form, one notes that
impression. sin2 i varies in a linear way with the square of
Indeed, through measurements performed the wavelength m. The slope of this straight
from the apex towards the base or the re- line gives the thickness of the layer and the ab-
verse, one can observe a shifting in the spec- scissa at the origin its index. In pre-computer
tral structures towards blue when the incident times, this simple graphic method allowed us
angle increases, with maxima and minima in to determine indices and thickness layer. Sub-
various places for a given incidence. In addi- jected to spectral measurements performed on
Type Butteries with 2-Dimensional Structures 91
Figure 8.8. Iridescence effects on Morpho godarti exposed to collimated illumination. One can note that
colors vary quite substantially depending on the illumination directionfrom base to apex or the
opposite.
M. menelaus, one notes that, like Michelson, The dissymmetrical colors made us conclude
this linear variation isnt respected, and by far. that the structure was inclined. It is therefore
Experimental points form large arcs of circle: necessary to correct every incidence angle of a
concave for positive incidences and convex for constant angle . Once done, all the points of
negative ones. the two arcs of circle converge towards their
92 Chapter 8 2-Dimensional Structures: Interferences and Diffraction
common chord, giving thus the index, thick- tivity, as has been seen in higher proportions
ness, and angle of the reecting structure, here among C. madagascariensis.
about 10 . Measurements performed in the vis-
ible and the near infrared on zones of complete
wings or on wings deprived of the scales of the Structures of Wings and Scales
dorsal side, and then of the ventral, allow one
to underline the inuence on the total reec- Let us discover this structure by zooming in
tion of each layer. Dorsal scales are responsi- by six orders of magnitude. By using optical
ble for the visible structure of spectra and thus microscopy, one observes that scales are here
color, whereas the alary membrane and pig- again arranged in two layers, yet structural
mentary scales are in charge of infrared reec- ones are in the bottom. Cover scales are thus
Reflectivity
0.8
0.4
deprived of pigment so that light can come and disappear completely and one can note that
play on the structures of the ground scales that ground scales, if they have lost their colors, still
are partly covered. Immersed into a liquid of serve as a dark screen. From different origins
adequate optical index (Figure 3.22), the latter among C. madagascariensis, the two structural
and pigmentary effects here on the contrary
proceed from the same scales. The inuence
of cover scales is minor and it is possible to
observe them by using reectance optical mi-
croscopy only under grazing incidence, where
a slight scattering effects makes them appear
whitish by diminishing the brightness of cov-
ered areas.
Scanning electron microscopy reveals the
same dense arrangement of structural scales
that form a regular background all over the
wing almost without overlapping. Structural
Figure 8.12. Optical microscope view in reection scales are at, rectangular (80 m 200 m)
of Morpho menelaus dorsal scales. Long and narrow, and slightly rounded and fringed at the apex.
cover scales are almost invisible in transmission or The cupule axis from which scales issue, is
in reection under normal incidence. Immersed in strongly inclined on the wing plane, thus form-
an index liquidFigure 3.22structural scales look ing a layer parallel to the alary membrane.
brown. They contain pigments and take part along
with ventral scales, in the formation of an opaque Striae are also dense (1600 to 1800 per mm)
screen that absorbs the wavelengths that are not re- and seem very deepsome even tend to join
ected by the structure. in places. Cover scales are slightly longer and
94 Chapter 8 2-Dimensional Structures: Interferences and Diffraction
(a)
(b)
composed of 1600 lines per mm on average, ite materials and the denition of the index of
thus only including three orders (1, 0, +1) the components requires a rather complicated
in which, as has been mentioned, the smaller modeling, which wont be fully presented
wavelength are less deviated than larger ones. here, as well as a theory of effective medium.
More generally, and if striae are very dense Maxwell Garnetts theory as mentioned in
(d e), the whole behaves like a quasi- Chapter 6, with a depolarizing coefcient A
continuous multilayer generating thin layer in- corresponding to the elongated shape of inclu-
terferences. And when one exposes the insect sions, is well adapted to the present problem.
to a diffuse light (falling under all incidences), It leads to an anisotropic solid phase index
more short wavelengths will be observed un- chitin and proteinwhich means that it shows
der a high angle of incidence than large ones. a different value depending on the direction,
The two phenomena are antagonist and unless that of striae, or the perpendicular one. It is in-
one is predominant, this results in color mixing teresting to note that the two index curves inter-
and thus to a reduced spectral purity. sect in the 450 nm500 nm spectral range; that
Despite the good atness of the Morpho is, the blue-green part of the spectrum that cor-
menelaus scales, their orientations prove too an- responds to the maximal reection of Morphos.
archical and the wing distortions are such that This certainly explains why researchers have
it is illusory to expect to observe the diffraction so far missed this anisotropy, as they used to
orders from the entire wing. It then becomes dene indices through immersion techniques.
possible to represent the wing of Morphos as a As has already been described, the method con-
stack of alternated layers of low and high in- sists in dipping the wing into diverse index liq-
dices. High-index layers that contain lamellae uids and seeking the perfect match revealed
mostly consisting of chitin with a few air, while by the disappearance of colors. This merely
low-index ones are quasi-symmetrical. These matches the index with the wavelength of the
layers are thus basically composed of compos- color. In the present case, it is isotropic.
Figure 8.19. The three stages in the modeling of Morpho structures by using an effective medium theory.
(a) Denition of a schematized structure model. (b) Determination of effective heterogeneous lms. (c) Ho-
mogenizing of lms by using the effective medium theory.
Figure 8.20. Modeling of the reection of Morpho menelaus wing under normal incidence by a stack of
14 alternated air/chitin layers. Thicknesses are in fairly good agreement with those deduced from TEM
observations.
Cuticular index
Transverse Electric
Transverse Magnetic
1.5
1
Figure 8.21. Opposite. Variations of the
400 450 500 550 600 650 700 750
anisotropic index of the cuticle scales de-
duced from the model (see Figure 8.19). Wavelength (nm)
98 Chapter 8 2-Dimensional Structures: Interferences and Diffraction
Figure 8.24. The basic structure of iridescent butteries. First, a scattering layer formed by cover scales.
Its efciency depends on the covering ratio ranging from 0 in M. cypris to 80% among M. godarti. Urania
are deprived of scattering layer. Second, a reecting layer formed by structural scales, which are cover
scales among Urania and ground scales among Morphos. Its efciency depends on the number of layers, the
spacing between striae, and the scale shape. Finally, an absorbing layer formed by ground scales, the alary
membrane, and the ventral side. Its absorption factor depends on the pigment concentration but also on
the structure of pigmentary scales.
(a)
spaces separating the latter are slightly wider the two membranes of the wing develop in-
(1200 striae per mm on average instead of 1800 dependently at the nymphal stage, so that the
among M. menelaus). The multilayer effect is motifs on the two faces of scales of different
diminished and more intertwined with the an- structures do not show any similarity. It is the
tagonist effect of the grating. Iridescence is less case of M. cypris that presents, like most of
marked, and more due to variations of reected the other members of the genus, cryptic col-
intensity than to real color changes. ors ranging from brown to black on the ven-
Let us now mention brightness variations. tral side, precisely excluding the dorsal white
Among the numerous very bright Morphos, in- spots. This perfect correspondence doesnt pro-
cluding M. aega as studied by Michelson, we ceed from the simultaneous development of
selected Morphos cypris. This buttery is char- the faces, but from the relation of cause and
acterized by long white streaks on the medial effect.
areas of its wings. These streaks rigorously Concerning the brightness of M. cypris, high
conrm the conclusions of our previous study magnications are not necessary to nd its
and we wont elaborate on them. This butter- cause. Photonic microscopy instantaneously
y highlights particularly well the inuence offers it. Cover scales are completely atrophied
of the absorbing sub-layer. As has been seen, and expose the total surface of structural scales
102 Chapter 8 2-Dimensional Structures: Interferences and Diffraction
R
I
Ordre R1
0.3
TE
R1
0.2 R-1
R0
TM 30 20 10 0 10 20 30
0.1 Facteur de reflexion (%)
R
Ordre 0 (2)
TE
TM
400 500 600 (nm)
Ordre R-1 I
0.2 R1
TE
R-1
0.1 R0
30 20 10 0 10 20 30
Facteur de reflexion (%)
TM
400 500 600 (nm)
Figure 8.33. Reection factor. The same effect as the terferences or diffraction, wavelengths devi-
previous, yet it is less distinct among Morpho godarti ated forward are selected. Here is the origin
asarpaTE on the right, TM, left. of color and iridescence.
(a)
Figure 8.38. (a) and (b) The various Morpho species examined. The genre includes some 30 species, most of
which are iridescent. Yet, the micro structural study of the latter is incomplete.
(a) (b)
Figure 8.39. Colias crocea in the visible (a) and in ultraviolet (b). The wings of the female uniformly absorb
in the ultraviolet spectrum, whereas the male posterior wings reect it partially.
Figure 8.42. SEM image of Gonepterix cleopatra Anterior wings. (a) General view showing the two scale
types, very distinct here. (b) Multilayer structure of cover scales. They are responsible for iridescence in
ultraviolet. (c) Ground scales contain numerous pterinosomes and generate the orange-yellow color in the
insect.
108 Chapter 8 2-Dimensional Structures: Interferences and Diffraction
Figure 8.43. The same arrangement among females (a) yet structures are identical in both types of scales
(b) and (c). Lamellae are short and do not overlap and both scales contain pterinosomes.
wave, making it turn green. This corresponds light, the wavelength is diffracted on one side
to theoretical assumptions, yet it goes further. and in TM polarized light on the other side.
In fact, this diffracting structure polarizes light, Although it is accepted that Morphos distin-
even (and especially) under near-normal inci- guish the two polarization states, the advan-
dence, which is quite singular! In other words, tage of the phenomenon for them remains un-
diffraction orders are polarized in very dif- clear, so is its origin. One can at least argue
ferent ways when the structure is exposed to that it is due to the arrangement in staggered
a natural non-polarized light. This is clearly rows of the lamellae on both sides of the striae
revealed by the calculated spectra. The order vertical axis, since the effect vanishes on a sym-
R1 is mostly Transverse Magnetic (TM) polar- metrical structure.
ized, whereas the opposite order R1 is Trans-
verse Electric (TE) polarized. As for the spec-
ular order R0 , not only is it little polarized, Mysterious Adonis
but it also shows a very low intensity. The cal- Every family has its lame duck. I would like to
culations and maps of diffracted elds offer complete this presentation by mentioning the
a different vision. Exposed to a polarized TE particular case of the beautiful Morphos adonis.
Polarization Effects 109
(a)
Figure 8.45. SEM image of posterior wing scales of
female Colias crocea. The structure is similar to that
of male posterior wings. Lamellae hardly overlap
and numerous granules of pigment can be found
between striae, like in the male.
Figure 8.46. A few examples of butteries showing iridescence in ultraviolet. They are all widespread in
France, excepting Phoebis rurina from Brazil.
visible, which makes these butteries particu- rior, posterior, or even both. On the other hand,
larly interesting, yet complex. To be able to dis- it insures a certain distinction between mimetic
play characteristics that predators cannot see species, which could otherwise be taken in by
presents at least two advantages. On the one each other, hence hindering reproduction.
hand, it results in a better sexual recognition This is true concerning both Batesian and
between partners of a same species showing Mullerian mimicry and can affect males as well
the same colors and motifs, whether cryptic as females.
or mimetic, in the visible. The differences be- Spectrometric measurements performed on
tween males and females are generally found the anterior and posterior wings of female
on large areas of wings or whole wings, ante- and male Colias crocea underline the different
Polarization Effects 111
112
Theory Recalls 113
(a) (b)
Figure 9.3. Polarization effects observed on Curculionidae scales, under natural light (a), and polarized light
(b).
114 Chapter 9 3-Dimensional Structures: Crystalline Diffraction
(b)
Polarization Effects
The rigorous treatment of the polarization of
wavelengths refracted by a grating or a crystal
(c)
(a) (b)
Figure 9.7. A natural Australian opal and its crystalline structure, left. It is less ordered than in Curculionidae
scales. Photo and collection E. Fritsch and A. Barreau, Institut des Materiaux, Universite de Nantes.
is generally intricate, and wont be tackled here. polarized than others and can be suppressed by
One can convey an idea of it, in terms of qual- using a polarizer. Chromatic effects are some-
ity at least, by keeping a Bragg interferential times quite impressive locally. And yet it is
approach and reminding one that on a diopter, more often difcult to distinguish them at a
the s and p components of a wavelength are macroscopic scale, since natural light is little
reected differently. In a similar way, when a polarized and the scales orientations are quite
crystal is illuminated more or less diffusely, the anarchical.
color observed in a given direction results from
the colors created through interferences of re-
ected wavelengths under various incidences, Insects and Photonic Crystals
and thus differently polarized. Some chromatic
componentsas reected under an incidence The formalism that Bragg developed in or-
close to that of Brewster angleare thus more der to analyze the diffraction of X-rays by a
crystal dates back to the early 20th century,
whereas passion for photonic crystals is a very
recent phenomenon. One should be wary of
too-simple transpositions and make sure that
basic hypotheses remain unchanged. Here, the
initial hypothesis is that structures are actually
periodic on rather long distances and that the
period of the grating and the incident wave-
length present approximately the same size. As
the following illustrations will show, these con-
ditions are met at the scale of the insect scales,
but not at the scale of the insect itself. Although
colors are beautiful locally, the insects tend to
look quite dull.
A good illustration of this is provided by
certain members of the Curculionidae big fam-
Figure 9.8. Two Curculionidae. Eupholus sulcicollis
from New Guinea, on the left, and Eupholus humer- ily, as represented by weevils in our regions.
alis from Papua, right. Like the rest of the family, They can actually show quite nice blue or
they are entirely covered with scales. green colors in other regions. The whole body,
116 Chapter 9 3-Dimensional Structures: Crystalline Diffraction
including legs, head, antenna . . . , is covered by undulations and they dont take part in the
thick and hard scales that seem to be scattered coloration of the insect. When they are me-
in an anarchical way. Two types of scales can chanically broken after freezing, they reveal a
be distinguished: on the one hand, rather at quasi-perfect tetrahedral crystalline structure,
scales plating the elytrons surface and on the actually very close to that of opals, and often
other hand, rare ones that are pilliform and leading to the same iridescent effects. Though
bristled. The former present various shapes- rarer, similar structures were found among cer-
spherical, more or less rectangular or oblong tain butteries such as Parides seostris by Pete
and can also include an indistinct surface Vukusic and Roy Samble from Exeter Univer-
structure. They seem to be striae rather than sity in England.
10
Amorphous Structures: Scattering
117
118 Chapter 10 Amorphous Structures: Scattering
(a) (b)
Figure 10.2. Single scattering (a). Particles are far away from one another and the thickness of the scattering
medium is weak; each particle receives only the main incident beam. Multiple scattering (b). The particle
concentration is higher. Their response is still that of a sphere but they are also subjected to the eld scattered
by neighboring inclusions.
Theoretical Recalls 119
In nature, scattering particles are rarely de- wavelength, they are as follows:
prived of pigments. A widespread case is 9 |a1 |2
i// = cos ,
2
melanin, which often appear as granules, like 2
4k r 2 1
i= i// + i .
in the iris in the eye, for instance. As will 9 |a1 |2
2
be shown, melanin is a pigment ranging i = ,
2
4k r 2
from black-brown to yellow, i.e., absorbing the (10-4)
smallest wavelengths, and mostly the blue. If where k is the wave vector, r and are the
the size of granules is such that it produces polar coordinates of the measurement point
scattering, the scattered spectrum as well as in a location, the center of which is occupied
the transmitted one are here deprived of any by the scattering particle, and a1 is a scatter-
blue wavelength. Usually containing a lot of ing coefcient depending on the size parameter
blue wavelengths, the former thus turns more and the particle index. Let us only note with-
green. The second one, which was already de- out elaborating that when the incident wave
prived of blue wavelengths, is not altered much is polarized perpendicularly to the scattering
and remains red. This combination of effects plane, scattering is isotropic, but that it varies
gives birth to many green colors, such as that like the square of the cosine of the scatter-
of lizards, grasshoppers, and many others. ing angle when polarized perpendicular. This
Polarization Effects
Polarization effects generated by these scatter-
ing phenomena are important in many regards.
They can be found in both the light scattered by
insects, which gives them their colors, and also
light that sometimes illuminates themsunlight,
for instance. The directions of polarization of a
scattered wave are marked within the scatter-
ing plane formed by the direction of the inci-
dent ray and that of observation.
My calculations that have only been briey
mentioned allow one to determine the scat-
tered intensities in a given direction for a polar- Figure 10.14. The common Pieridae, Pieris brasicae,
ization state of the incident wave. For a given female.
Type Butteries with Scattering Structures 123
implies that even when the incident wave is not scattering that can both substantially split col-
polarized, part of the scattered wave is polar- ors, scattering is relatively little selective. This
ized and always perpendicularly to the scatter- is easily understood when one remembers that
ing plane. The P polarization ratio, dened as a particle of a given size will highly diffuse the
the following product: radiations of wavelengths that are shorter to
i i// itself, whereas it wont diffuse much that of
P= , (10-5) longer wavelengths. Consequently, scattering
i + i//
by itself cannot produce a color situated at the
becomes, according to the previous equations: center of the visible spectrum or at its red end.
Very small particles can indeed diffuse blue
1 cos2
P= , (10-6) wavelengths but hardly others. Nevertheless,
1 + cos2 when one increases their sizes, they wont dif-
which demonstrates that at right angle of the fuse just a greater wavelength, but more and
incident direction, the scattered wave is even more wavelengths, from green to yellow . . . the
totally polarized. This effect is similar to the superposition of which leads to an increas-
total polarization obtained in reection at the ingly pure white. This process implies a mixing
Brewster angle. The scattering diagrams of rather than selection of colors, and the resulting
waves polarized perpendicularly and parallel colors are not very pure. More marked colors
to the scattering plane are shown on Figure will appear, provided scattering be combined
10.13, as well as their result when the incident with a more selective phenomenon, which is
wave is not polarized. pigmentary absorption.
We will illustrate this phenomenon with two
Type Butteries with Scattering species that live in our temperate regions and
that are situated at the two ends of the spec-
Structures
trum, as has just been mentioned: the male
Common blue Polyommatus iracus and Celas-
Pieridae and Lycenidae
trina argiolus on the one hand and the white
White or Blue Pieridae Pieris brasicae on the other hand. As
Scattering as a source of colors is widespread has been already said, those are the only colors
in nature. Although it may seem quite labo- that can be obtained through scattering. But the
rious, the presentation we have just made of two species also happen to illustrate the two
the phenomenon enables to convey its intri- possible ways of scattering: a high index par-
cacy. After this long chapter on thin layer in- ticle immersed in air or, on the contrary, an air
terferences and the fact that theoretical pre- inclusion in a material with a higher index.
dictions and experimental conclusions corre-
spond, the present chapter may appear quite
short and disappointing. Yet, it will be justied
by the following study of the wingaccording
to the same zooming in approach. Whether it is
from the microscopic point of view, from that
of each scattering particle, or the macroscopic
and static point of view of their distribution
on wings and their interactions, formalism is
much more complex than what has been seen
so far. As cross-sections observed by electron
microscopy will show so wonderfully, one is
here confronted with an extraordinary accu-
mulation of mathematical and calculation com- Figure 10.15. Diffuse reection factor under normal
incidence of a Pieris brasicae anterior wing. Reec-
plexities, these butteries thus proving impos- tion is strong throughout the visible spectrum.
sible to model so far. Let us come back from UV and part of the blues are absorbed, hence the
the beginning. Contrary to interferences and slightly yellow color of the wing.
124 Chapter 10 Amorphous Structures: Scattering
(a)
(b)
(c)
Figure 10.16. (a) SEM image of the extremity of a Pieris brasicae scattering scale. (b) view of the center of the
scale approximately showing a non-covered area and the neighboring scales. In covered areas, striae are
denser, counter-striae are thicker, and inclusions are fewer. Pieride scales also present traces of leucopterine,
mostly on the ventral side of posterior wings, which gives them a slightly yellow color. Pterinosomes: (c).
the visible spectrum, yet the modeling of which There is no inclusion here. The bottom of com-
remains to be established. partments is lined with an air spongy structure.
It is difcult to evaluate the structure thickness
by using Scanning Electron Microscope (SEM)
Blue images. Yet one can roughly estimate the aver-
The common blue appears like the negative of age size of pores, here below 100 nm. The ef-
Pieris brasicae. Most of the scales, with a more ciency of scattering diminishes very substan-
geometrical shape, prove quite disorganized, tially for long wavelengths, which makes short
while some very long threadlike scales are scat- blue and violet wavelengths become more visi-
tered on the whole. Spacing between striae is ble. The intensity remains weak and the purity
well marked, lamellae are short and not cov- much lower than that of Morphos obtained by
ered, and counter-striae are well developed. interference.
If the present book is mostly dedicated to phys- energy to the latterand thus disappearif
ical colors, they are still a minority in nature. the energy corresponds to the difference of en-
Most of animal colors derive from a pigment, ergy between two levels of the atom. After a
and although these substances dont produce photon is absorbed, the atom is in an unset-
iridescence or very little in the usual observa- tled stateknown as excitedand goes back
tion conditions, they still have their place in to its fundamental state. Several processes al-
a book dedicated to buttery colors. We will low this. Symmetric to the previous, the rst
mention them along two distinct lines, rst the one, is radiative relaxation. The atom goes back
physicochemical aspect, which will include an to its fundamental state by emitting its excess
overview of some important and distinct fam- energy under the form of a photon. An exte-
ilies of pigments as well as the physical pro- rior observer wont notice anything: A photon
cess causing color: selective absorption. Since enters and another, identical to the rst, goes
buttery colors are observed in reection, this out. The incident light is not modied and no
aspect of the problem, generally less known, color appears. However, such relaxation can be
will be further studied. Secondly, we will tackle non-radiative. Surplus energy can, for instance,
the physiological aspect of pigmentation, as make a crystalline grating or its neighboring
well as its origin and from the point of view molecules vibrate, thus leading to an increase
of evolution. in the temperature of the object. Yet, there is no
emitted light anymore. The incident photon is
Selective Absorption and Colors indeed absorbed during such a process. Yet, for
of Pigments the observer, this or these photons have disap-
peared from the incident light spectrum, and
Absorption is a transfer of energy between the the color of the body somehow corresponds to
incident electromagnetic wave and the atoms the complementary of the absorbed color.
or molecules composing the illuminated object, In order for such a phenomenon of coloration
or more precisely, the electrons of these atoms to occur, the difference of energy between the
that are distributed on various orbitals, each two levels must correspond to a frequency that
characterized by one energy level. is perceivable by the human eye. For us, this
Transitions between the different levels pro- corresponds to wavelengths ranging from 380
ceed from absorption or emission of the corre- to 680 nm. It is generally the case for isolated
sponding energy. In the case that concerns us atoms in a gas, but not atoms that are linked
here, the energy that is absorbed comes from to each other. In this latter case, the absorp-
the incident light, which will thus be deprived tion bands of electrons forming covalent bonds
of certain wavelengths and its color will be con- are usually situated in the ultraviolet. Pigments
sequently changed if the latter are situated in and dyes belong to a very singular category of
the visible part of the spectrum. When a pho- substances, in which the distribution of elec-
ton interacts with an atom, it can transfer its trons is such that absorption bands are brought
127
128 Chapter 11 Pigments and Pigmentary Colors
Melanins
Melanins form a large family of miscellaneous
pigments, widespread in both vegetal and
animal realms. They certainly represent the
most common, yet most complex, pigments. Figure 11.8. Albinism occurs when melanin is not
synthesized.
They are polymers, the biological precursor Here, Pieris brasicae. Above, the fellow individual
of which is tyrosine. Their colors range from is normally pigmented, white areas are colored by
yellow to brown and black. In the animal pterines and the apex by melanins. The individual il-
world, there are two distinct kinds: phaeome- lustrated below, presents normally synthesized pter-
lanins (yellow to reddish) and eumelanins ines, but the apex scales are depigmented due to the
incapacity of synthesizing pterines. (11.7 and 11.8
Photo and collection H. Descimon)
Leucopterine
Isoxanthopterine
Xanthopterine
3
Sepiapterine
Erythropterine
(a) (b)
Ommines are little soluble in water and it is dopa) for the simplest ones. The structures
quite difcult to study them from a chemi- of other more complex ones are little known
cal point of view. They are mostly found in and only the structure of papiliochromic II has
eyes. Other pigments, red and highly insoluble, been discovered. Their coloration ranges from
that can be observed in Nymphalides (Precis light yellow to red, and as their names suggest,
coenia), prove to belong to the same chemical they are present in Papilionides. A polymor-
family since they derive from tryptophane. phism determined by one gene and opposing
It was established through blending it with a light yellow pigment to an orangey one has
tryptophane14 C. been researched among Zerynthia. The Greek
species, Z. polyxena, thus includes an orangey
type, Ochracea, present in both males and fe-
Papiliochromes
males. Among Z. rumina, it is the canteneri type
They are close to ommochromics and de- characteristic of Southern Spain and Maghreb,
rive from cyclic amino-acids, tryptophane which is dominant and tends to only occur in
(via cynurenine) and from phenylalanine (via females.
Selective Absorption and Colors of Pigments 133
(a) (b)
(a) (b)
Figure 11.13. Flavonoids among various butteries from different genus. Lysandra coridon, Leptidea sinapis,
Melanargia galathea. Above, normal aspect. Below, insects immersed in ammoniac vaporscharacteristic
reaction of avonoids.
134 Chapter 11 Pigments and Pigmentary Colors
Papiliochrome 2
135
136 Chapter 12 Thermoregulation and Spectral Selectivity
Energetic Balance
A Few Denitions
Figure 12.3. Photothermal collector losses. Optical
Endothermal sources of heat aside, as concerns losses are due to light being reected on wings. They
are very small in black butteries. Losses proceed-
the Q energetic balance of a solar collector, solar
ing from convection and conduction are small when
energy incident on the whole collector surface the insect is at rest, but can be substantial yet use-
represents the credit, whereas the debit is the ful during ight in order to get rid of surplus heat.
convective, radiative or optical loss: reected Losses due to radiation are predominant.
Energetic Balance 137
50C
0.5
40C
0.2 0.5 1 2 5 10 20 50
Efficiency
e increases
decreases
e increases
T increases
Efficiency
Figure 12.8. Reection, transmission, and absorp- increases
tion factors of Archaeoprepona black areasCu1 and
Cu2 baseshowing the strong spectral selectivity
of wings near the thorax and the combined action Figure 12.10. Temperature stabilization cycle of a
of reection and transmission that weakens the ab- buttery black wing. When temperature increases,
sorption in the infrared. the captor efciency decreases and the way round.
Energetic Balance 141
thus long wavelengths and the absorption peak the second case. The radiated energy is here
lies outside of the emission spectrum. In this proportional to the surface common to the ab-
case, the emissivity is lower, the radiative ef- sorption peak and the emission spectrum. At
ciency increases, and the temperature raises. 30 C, when the peak does not take part in the
The real efciency of this stabilization cycle is thermal emissivity at all, radiative losses reach
not easy to precisely determine when one does 275 W.m2 , but go over 500 W.m2 at 50 tem-
not know the temperatures that wings reach perature. The temperature of a buttery warm-
when exposed to the sun. One can still evalu- ing itself in the sun remains quite inferior to
ate the emissivity for the extreme cases of wing these extremes, yet the previous rough calcu-
temperature; for instance, 50 and 30 C. In the lations demonstrate that this auto-stabilizing
rst case, the peak totally overlaps the emission system can play a substantial part in the insect
spectrum, while it is outside of the spectrum in thermodynamic.
13
Vision and Colorimetry
142
Sources and Illuminants 143
Figure 13.2. The physical components of a visual stimulusarbitrary units. Light perception depends on
the spectral composition of the source, (a) on the reectanceor transmittance for transparent objectsof
the observed object, (b) and on the sensitivity of the various eye receptors to different wavelengths (c). The
corresponding brain representation is more complex to quantify, for instance, as coordinates within a color
spaceand is the goal of colorimetry.
144 Chapter 13 Vision and Colorimetry
Colorimetry Notions
The Perception of Colors by Humans
Let us now study the aspect of the color percep-
tion process that is certainly the most complex;
the eye response V() and its interpretation by
the brain. Figure 13.4. Monochromatic solar illumination in
relation to the wavelength. Outside of the atmo-
The human eye contains two kinds of re-
sphere, the solar spectrum can be modeled as the
ceptors with very different properties. Cones, emission spectrum of a black body at 5800 K. At
which possess a high resolution yet weak sen- ground level, and according to the crossed air mass,
sitivity resolution power/capacity, are mainly it is slightly shifted towards red and presents nu-
concentrated in the retina central area: the merous absorption bands by the atmosphere com-
ponents. For an air mass of 4, spectrum intensity
fovea centralis. They are both photometric and
is substantially weakened and it is shifted towards
chromatic receptors allowing photopic vision red.
and colored diurnal vision. Mowing away from
the retina center, one then encounters rods
that are very dense at the periphery. Rods al- grey levels. The sensitivities of the two col-
low nocturnal scotopic vision. Those photo- lectors differ by more than three orders of
metric receptors are highly sensitive, but do magnitude (and their spectra are shifted); the
not perceive colors and only give an image in maximal sensitivity of cones peaks around
550 nm approximately(yellow-green), while
that of rods peaks around 500 nm (at the limit
between green and blue). Since the latter dont
take part in color perception, the denition of
the reference observer involves the response
curves of the various cones only. There are
three types of cones, each containing a pigment
presenting a maximal sensitivity in a specic
part of the spectrum: long (L), medium (M),
and short wavelengths (S). Sensitivity maxima
Figure 13.3. A color corresponds to a dominant are situated respectively at 560 nm in the red,
wavelength: 1 or 2 . This does not necessarily en- 540 nm at the yellow-green limit, and at 420 nm
tail that the wavelength is really present within the in the blue. These cones are sometimes des-
received spectrum, since two wavelengths can com- ignated by their predominant color: red (R),
bine and thus produce the effect of a third one (a). green (G), blue (B). These different types of
For a given dominant 0 wavelength, light intensity
can be more or less strong (A and B) and color more cones are not distributed evenly on the retina,
or less pure (C) according to the spectrum height especially the blue ones that do not represent
and width respectively (b). more than 2% of the fovea cones.
Colorimetry Notions 145
with:
r+v+b=1 (13-4)
Figure 13.11. SEM general image and details of Parnassius apollo compound eye.
intermediary between that of air (n0 = 1) and that could reveal the presence of the insect to
of the corneal lens (n 1,5). This index, ap- predators. These devices, known as moth eyes
proximating 1,2, can be evaluated thanks to structures, are also used in the space industry.
Maxwell Garnetts theory or through its ap- How images are formed through such an op-
proximation to neighboring indices. This effec- tical device is complex and varies according
tive layer behaves like an anti-reecting coat- to species. The apposition eyes, in each com-
ing, both allowing to reduce optical loss at the pound of which rays enter and give an inde-
ommatidium entrance and to avoid reections pendent signal, are distinguished from the su-
perposition eyes of generally nocturnal species,
in which rays can be combined on the same re-
ceptor. These eyes do not allow any focalization
but are highly sensitive to motion. A fact that
concerns us more is that butteries can perceive
colors well and that their sensitivity spectrum
is shifted towards ultraviolet as compared to
Figure 13.12. Comparison between the photorecep-
tors sensitivity spectra of humans, Lepidopterons, ours. They can therefore perceive on their fel-
and Birds. The buttery spectrum goes far beyond low creatures motifs that are invisible to preda-
that of its main predators in ultraviolet. tors.
Buttery Vision 149
point and Arago pointthat are not always vis- prehensive, it is accepted that a great number
ible. of them can perceive polarization states. This
This has two important consequences. First, perception is based on the variations in absorp-
when the sky is clear, diffuse light falling on tion of the electric eld with its orientation. In-
an object or an insect is polarized mostly per- deed, it has been discovered that the absorption
pendicularly to the scattering plane. The po- of some materials depends on the polarization
larization rate can reach about 70% at 90 from states of the incident light, either rectilinear or
the sun direction. According to the orientation circular. Known as dichroism, the phenomenon
of the insect or of the structure in relation to is caused by the shape of absorbing molecules,
this plane, polarization effects by reection on which are either elongated in a single direction
wings can be amplied, thus increasing the or in spiral. A linear dichroism has indeed been
contrasts. discovered in the visual cells of rhabdomes in
This also entails that a measurement of sky some insects, and orientation maps have even
polarization can reveal the direction of the sun, been established. It has thus been proved
even if is hidden. It has been established that even if their meaning remains obscure
bees actually use this property to nd ones that all polarization effects generated by the
bearings, and perhaps also by some big mi- complex structures previously mentioned
grants like Danaus plexippus. can be perceived by most insects and there-
(2) Even if the study of insect sensitivity to fore represent one more vehicle for visual
light polarization is still far from being com- communication.
Conclusion
Since its creation, the laboratoire doptique des Let us rst describe the traditional struc-
Solides, a CNRS-Pierre et Marie Curie Univer- ture of a fabric. Its basic component is ber
sity unnite mixte, has focused on the study synthetic in the present case. It is a long ber,
of optical properties of inorganic solids, rst usually about 10 nanometers in diameter, com-
on metals, and later on semiconductors, di- posed of various more or less long polymer
electrics, and inhomogeneous materials. Noth- chains, the nature and arrangement of which
ing to do with butteries indeed. And yet, determine the mechanical properties of the
the further we study wings, the more simi- ber. It is on this ber that the optical struc-
larities we nd. Or rather, we realized how ture generating iridescent colors are produced.
much butteries were ahead of us concern- These bers are then gathered in a varying
ing optical phenomena. They in fact proved to numberusually 60and form the thread that
conceal all that concerned us: dielectric multi- is woven or knitted to create fabric. During
layer, scatterers, selective materials, disordered the various stages, one can notice that the ba-
mediaand that only concerns wings: all sub- sic element supporting the very irregular op-
jects that we were studying. In this way, but- tical structure is subjected to numerous dis-
tery wings became quite naturally part of our tortions at extremely different scales. The ber
research. Materials are new, yet structures are curving rays, for instance, can be measured in
traditional. Our research resulted from a pas- tens of micrometers when twisted into a thread,
sion dating back to my childhood and a denite and some millimeters once woven. Finally, the
industrial request. creases in the nished garment present distor-
Within the framework of a prospective re- tions measured in centimeters.
search on the fabric of the future, we were The main difculty is then to model, or at
asked about the origin of iridescence among least evaluate, the evolution of the optical prop-
butteries and its possible application to tex- erties at these various scales. Physical phe-
tiles. The answer to the rst question is the topic nomenon involving elements occurring at dif-
of this book. It didnt reveal more complex, yet ferent scales are among the most complex.
it proved to be more diverse than expected. In fact, it is to them that one is confronted
As concerns the application to bers, and even when one wants to evaluate the visual effects
garments, represents a real challenge, even in- of a garment, the iridescent colors of which
dependently from any technical creation. It is generated at the smallest scale and imposing a
worth mentioning the method. As the intro- strict orderare modulated by local entropies
duction was dedicated to the buttery, so the of increasing scales. The solution we chose is
conclusion has to be. One thus starts by den- directly inspired by Urania scales. It consists in
ing precisely the optical properties of a plane distributing over the ber as it is being woven,
multilayer. It is then rather easy to optimize the a certain number of thickness and index lay-
system, based on the materials available for tex- ers corresponding to a given color. Calculations
tiles, in order to obtain a greater iridescence. show that this number actually does not need
151
152 Conclusion
was started prudently, yet pupils quickly took blue wing (it immediately turns green, then
an interest in itfrom middle school interns, blue again when dry). The same experiment
who thought it was impossible to work seri- performed on a blue Argus results in the disap-
ously with this material, to high school sci- pearance of blue without the green stage! The
ence juniors and seniors, who discovered (bet- three main sources of colors in the living world
ter late than never) a totally different aspect of have thus been experimented. Then, a group
physics. They would thus attend conferences observes and carries out the same experiments
on butteries, sometimes several hours long, by using an optical microscope, modifying in-
without showing impatience, and I would even cidence, while another group researches the
say with a certain enthusiasm . . . and at the various origins and properties of light. These
end, they would realize they had actually classes went out, to discover in museums how
studied physics for four hours. The interdis- it works among birds, shells, stones, and so
ciplinary study of butteries proved a success on. When they realized that iridescence could
in high school, and a great way of introduc- not be generated by the striae they had ob-
ing physics. We wished these bridges between served with the optical microscope and that
different subjects could be implemented in they had to change the scale once again, we or-
universitiesthis modest rst experiment in ganized several scanning electron microscopy
high school being quite encouraging. This is sessions at the university. The resulting pic-
in fact worth describing it. Launched between tures were then studied in high school: to have
biology and physics via butteries, the ex- a class count the diffusing particles of a Cab-
periment could very well be applied to any bage Pieride is worth any image-processing
other organism that is a little popular accord- software . . .
ing to meshells, beetles, and, during spring In the end, these pupils who tend to con-
months, grasshoppers, maybugs, or owers sider physics as one of the most boring subjects
and between any subjects: thermodynamic, discovered physics and basic optics notions
mechanics, chemistry, mathematics, and even via a more appealing subject like biology. No-
arts or music. I had the good fortune to meet tions that they will be less reluctant to study
a biology and a physics teacher who agreed to further in class. For instance, they saw a
adapt classes and dedicate some time to imple- wavelength and evaluated its length for the
ment the project together. It entailed changing rst time. Better still, the experiment showed
the traditional schedule of one-hour physics them the advantages of interdisciplinary stud-
class and a break followed by a one-hour bi- ies, how interfaces enrich one another, but also
ology class. So, we asked pupils to resolve a how jargons could represent a barrier between
complex interdisciplinary problem: How and two communities. May they persevere in this
why do butteries present such colors? This re- path, and always remember this new butter-
quires quite simple experiments that anybody y effect. And may this book guide them. I
can carry out: a piece of red wing is dipped into would be very proud. It would be my best
acetone (nothing happens), a piece of Morpho achievement.
Subject Index
157
General Index
158
General Index 159