Near-Infrared (NIR)-Reflectance in Insects

7/28/2019 Near-Infrared (NIR)-Reflectance in Insects

1/33

Near-Infrared (NIR)-Reflectance of Insects 183

Entomologie heute 24 (2012)

Entomologie heute 24 (2012): 183-215

Near-Infrared (NIR)-Reflectance in Insects

Phenetic Studies of 181 Species

Infrarot (NIR)-Reflexion bei Insekten phnetische Untersuchungen an 181 Arten

MICHAEL MIELEWCZIK, FRANKLIEBISCH, ACHIM WALTER & HARTMUT GREVEN

Summary:We tested a camera system which allows to roughly estimate the amount of reflectance prop-erties in the near infrared (NIR; ca. 700-1000 nm). The effectiveness of the system was studied by tak-ing photos of 165 insect species including some subspecies from museum collections (105 Coleoptera,11 Hemiptera (Pentatomidae), 12 Hymenoptera, 10 Lepidoptera, 9 Mantodea, 4 Odonata, 13 Orthoptera,1 Phasmatodea) and 16 living insect species (1 Lepidoptera, 3 Mantodea, 4 Orthoptera, 8 Phasmato-dea), from which four are exemplarily pictured herein. The system is based on a modified standardconsumer DSLR camera (Canon Rebel XSi), which was altered for two-channel colour infraredphotography. The camera is especially sensitive in the spectral range of 700-800 nm, which is well-suited to visualize small scale spectral differences in the steep of increase in reflectance in this range,as it could be seen in some species. Several of the investigated species show at least a partial infraredreflectance. NIR-reflectance is especially pronounced in specimens of an overall white, red, orangeand yellow colouration, but was also found in numerous green insects (e.g. the leaf katydidsAncylechafenestrataand Stipnochlora coulonianaand the walking leafPhyllium celebicum). In contrast, other green

wings, as for example the metallic green wings of the butterflyTroides priamusor the metallic greenelytra of several jewel beetles such as Chrysaspis aurovittata, do not reflect NIR-radiation.

Keywords:infrared photography, NDVI, infrared reflectance, cuticle, insect, phenotyping

Zusammenfassung:Wir stellen ein Kamerasystem vor, mit dem man in erster Annherung dieReflexionseigenschaften von Insekten im Nah-Infrarotbereich (NIR; etwa 700-1000 nm) bestim-men kann und haben dies anhand von entsprechenden Aufnahmen von 165 Insektenarten ausSammlungen (105 Coleoptera, 11 Hemiptera (Pentatomidae), 12 Hymenoptera, 10 Lepidoptera,9 Mantodea, 4 Odonata, 13 Orthoptera, 1 Phasmatodea) und 16 lebenden Insektenarten (1 Lepido-

ptera, 3 Mantodea, 4 Orthoptera, 8 Phasmatodea) geprft, von denen wir vier Arten hier exemplarischabbilden. Das System beruht auf einer modifizierten Standard-Digitalen Spiegelreflexkamera (CanonRebel XSi), welche fr den Einsatz als Zweikanal-Farbinfrarotkamera umgebaut wurde. Die Kameraist besonders empfindlich im spektralen Bereich zwischen 700 und 800 nm, da in diesem Bereich oftkleinere Unterschiede der Steigung im Anstieg in der Reflexion bei verschiedenen Arten unterschiedenwerden knnen. Von den untersuchten Insektenarten zeigen einige zumindest partiell eine NIR-Reflexion. Diese ist besonders deutlich bei weien, roten, orangefarbenen, gelben und besonders aus-geprgt bei einer Reihe von grnen Insektenarten (z. B. bei den BlattheuschreckenAncylecha fenestrataundStipnochlora coulianasowie dem Wandelnden BlattPhyllium celebicum). Im Gegensatz dazu zeigen dieFlgel anderer grner Insektenarten, wie die metallisch grnen Elytren einer Reihe von Prachtkfern,z. B. Chrysaspis aurovittata, oder die leuchtend grnen Flgel des Schwalbenschwanzes Troides priamus

keine NIR-Reflexion.

Schlsselwrter: Infrarot-Fotografie, NDVI, Infrarotreflexion, Kutikula, NIR, Insekten, Phno-typisierung


2/33

184 MICHAEL MIELEWCZIK, FRANKLIEBISCH, ACHIM WALTER, & HARTMUT GREVEN

1. Introduction

Insects are variously coloured and thecolouration has variegated functions suchas advertisement, warning, crypsis etc. (forreview see COTT 1940; FOX1979; CHAPMAN1998; LUNAU 2011).Typically, the coloura-tion is produced by physical means (e.g.diffraction, interference and scattering bynanostructures) and/or by pigments thatabsorb or emit light, and both, nanostruc-tures and pigments, have been thoroughlystudied in insects (for review see CROMARTIE1959; FOX1976; VIGNERON & SIMONIS 2010).

However, investigations concerning insectsprimarily cover colouration patterns andtheir spectral reflectance characteristics inthe range from 400 to 700 nm, which is vis-ible for humans and many other vertebrates(visible spectrum of light = VIS), and theshort-wave spectrum (ultraviolet/UV, 300-400 nm). Perceiving these patterns enableinsects to recognize mating partners, hosts,food etc. (seeMENZEL 1979; KELBER et al.2003). Studies of infrared reflectance ofthe insect integument, i.e. between 700 and1400 nm (near infrared = NIR), are rare,even though initial studies of insects werealready performed by some of the earlypioneers in infrared photography (TRISTAN& MICHAUD 1916). COTT (1940) was oneof the first authors, who paid attention toinfrared photography to study different

green animals that may greatly differ in theirabsorption of infrared light. He presentedinfrared photographs showing a tettigoniidgrasshopper that clearly contrasted againstthe background of green leaves, whereasthe green caterpillar of Smerinthus ocellatusmerges with the surrounding vegetation (seeCOTT 1940, plates 3 & 5).In the present note we revive these studies

by screening NIR-refl

ectance of a largernumber of dead and living insects. In de-tail we 1) report on some basic principlesof the red-edge based NDVI-photogra-phy we used for this phenetic screening,

2) present the results obtained from al-together 181 species including some subspe-cies (some of them are shown in NDVI-photographs), and 3) shortly touch thenature of infrared reflectance and itspossible adaptive value. Generally, our studyindicates that NIR-reflectance of insectinteguments seems to be more widespreadthan the existing literature suggests.

2. Material and methods

2.1. Animals

We examined 165 species from the collectionof the Aquazoo/Lbbecke-Museum Ds-seldorf and the Entomological Collection ofthe ETH Zrich, i. e. 105 Coleoptera, 11 He-miptera (Pentatomidae), 12 Hymenoptera,10 Lepidoptera, 9 Mantodea, 4 Odonata,13 Orthoptera, 1 Phasmatodea (Tab. 1; seeappendix) and 16 living insect species, i.e.1 caterpillar (Lepidoptera), 3 Mantodea, 4Orthoptera, 8 Phasmatodea) (Tab. 2; seeappendix). We included a large numberof green specimens, as it is known from

vertebrates that even though specimenscan show a comparable green colouration,reflectance in the NIR-might be completelydifferent (SCHWALM et al. 1977; DODD 1981;KREMPELS 1989; EMERSON et al. 1990). Fromthe museum specimens colour photos andinfrared images were taken under the same

conditions, the latter with a modified DSLR(digital single-lens reflex) camera (see 2.2.).Colour images were acquired with a CanonEOS 550D and 600D camera. Images ofspecimens from terraria were taken underthe local lighting conditions. Green leaves

were used as references to quantify theamount of infrared reflectance. All images

were acquired in highest quality as JPEG or

Canon Raw imagefi

les.Generally we adopted the scientific namesof the insects directly from the collections.In a few cases we changed the species epi-thethon or the genus name.


3/33



2.2. Digital infrared and red edge pho-tography

For image capture a modified handheld 12.2megapixel DSLR camera (Canon Rebel XSi,modified by LDP LLD, Carlsted, USA, www.maxmax.com) was used. Such cameras aresuccessfully used in monitoring the coverof vegetation (e.g. BOKHORST et al. 2012)by calculating the normalized difference

vegetation index (NDVI) as described be-low. The camera is therefore advertised asNDVI-camera, a term we also use in thisstudy. Here we apply the same camera to

differentiate NIR-reflecting and NIR-non-reflecting insects and their surroundingsurfaces. The modified camera detects lightin two wave length bands in the range ofblue light (B, 370 to 480 nm) and in therange of the so called red edge (RE). Thered edge lies between red and near infrared(NIR) and is characterized by a wave lengthranging from 675 to 775 nm.

The 2-band model offers the possibilityto differentiate between NIR-reflectingand NIR-non-reflecting surfaces alreadyby viewing the images. The former appearstrongly red, whereas the latter appear(dark) blue. Discrimination of NIR-reflect-ing and non-reflecting surfaces can be easilyperformed using the 2-band images withoutnormalization (see below), as human visionitself can differentiate blue and red. For

an initial test of the camera we used twogreen vertebrate species (Morelia viridisanda green Furcifer pardalis), which are known todiffer in their infrared reflectivity marked-ly (DODD 1981; KREMPELS 1989; GEHRING& WITTE 2007) (Fig. 1). Two examplesfor colour rendering of different naturalmaterial provided by the modified twochannel NDVI-camera are given in figure

2. Differences are quite clear; therefore wethink that normalization procedures for theimages are not necessary in many cases.However, for quantitative analyses andcomparison of insect images taken under

different light intensities, the calculation ofthe NDVI-might be beneficial.

NDVI-= (RE-B)/(RE+B)

Using the equation, we tested the hypothesisthat the flash integrated in the camera anda white balance (white card) are useful tostandardize image quality and small scaledifferences under different light regimes(Fig. 3 A). We selected areas on images oftypical background vegetation such as grass(examples can be seen in Fig. 2 A, B).

To evaluate the image quality and identify anoptimal image procedure for image acquisi-

tion we therefore compared NDVI-resultsunder sunny and shadowy conditions (Fig.3 B). Based on those findings, which clearlyshow that good illumination conditions arepreferable, we used the internal camera flashfor all photos. Images of the insect collections

were acquired with an additional illumina-tion provided by a 500 W halogen lamp toprovide sufficient light in the NIR-range ofthe spectrum. All figures were prepared using

Adobe Photoshop CS5 and Corel Draw X4.Box plots in figure 3 were created and basedon statistical analysis performed with R 2.15.1(R-DEVELOPMENT-CORE-TEAM, 2008).

2.3. Abbreviations

B = blue light; NDVI-= normalized differ-ence vegetation index; NIR-= near infrared

(700-1400 nm); RE = red edge (~700-800nm); VIS = visible spectrum (400-700 nm);CCD = charged coupled device; CMOS =complementary metal-oxide semiconductor;DSLR = digital single-lens reflex.

3. Results

3.1. Optimizing image acquisition

The NDVI-camera is not only highly effi-cient in analysing differences of vegetationreflectance, but also to differentiate highfrom low reflectance items (Fig. 3 A, B).


4/33


Fig. 1:The NDVI-camera clearly shows reflectance differences in the NIR. Contrary to the green

tree python Morelia viridis(A), the green panther chameleon Furcifer pardalis(B) shows a high NIR-reflectance, which is comparable to that of green coloured leaves (C).Abb. 1:Die NDVI-Kamera zeigt deutlich Reflexionsunterschiede im Infrarotbereich. Im Gegensatzzum grnen Baumpython Morelia viridis(A) zeigt das grne Pantherchamleon Furciger pardalis(B)eine starke Reflexion, hnlich der Reflexion der grnen Bltter (C).


5/33



Fig. 2:Colour rendering of the infrared sensitive NDVI-camera.ALandscape with different materials.Vegetation appears pink to light red (high NIR-reflectance), stones and sand are violet to blue (lowNIR-reflectance). B Depending on shadow, illumination, kind of leaves and light quality, vegetationmay show small variations, but the high reflectance in the NIR is always obvious. However, othercolours than green, e.g. the yellow fur of a tiger (Panthera tigris), reveal a similar NIR-reflectance asthe vegetation.Abb. 2: Farbrendering der infrarotsensitiven NDVI-Kamera. A Landschaft mit verschiedenenMaterialien. Die Vegetation erscheint pink oder rot (hohe NIR-Reflexion), Steine und Sand sind

violett bis blau (niedrige NIR-Reflexion). B In Abhngigkeit von Schatten, Lichtintensitt, Blattbe-schaffenheit und Lichtqualitt kann die Farbdarstellung der Vegetation variieren, jedoch ist die starkeReflexion im NIR klar erkennbar. Auch andere sichtbare Farben als Grn knnen eine hnlich starkeInfrarotreflexion aufweisen, z.B. weist das orangerote Fell des Tigers (Panthera tigris) eine hnlicheNIR-Reflexion auf wie die Pflanzen im Schatten.


6/33


Using a white card and the integratedcamera-flash clearly shows that undershadowy conditions NDVI-values of dif-ferent background vegetation increase and

variation decreases (Fig. 3 A). Further, it isshown that under strong light regimes thedifferences between samples are more pro-nounced (Fig. 3 B). Images of specimens

were thus all acquired using automaticwhite balance and the internal compulsoryflash of the camera.Even though limited to a two colour mo-del, the acquired images share a coloura-tion scheme, which is highly comparable

to those of film-based, classic 3-colour-infrared photography (for comparisonssee GIBSON 1978).

3.2. NIR-reflectance in insects

Generally, our NIR-images reveal a remark-able variation not only concerning theoccurrence of NIR-reflectance, but alsoconcerning the reflectance patterns (for anoverview see Tabs. 1, 2; Figs. 4-10). Only fordescription we herein use five categories tocharacterize reflectance: no, very weak, weak,medium, strong (see Fig. 4).

3.2 1. Coleoptera (museum specimens)

Black beetle species such asCarabus scabrosustypically show no NIR-reflectance. In con-trast, most dark brownish to red brownishspecimens show a weak to medium reflec-tance. Such beetles were found in nearly allgroups of brownish coleopteran investi-gated, e.g. in the Buprestidae, Dynastidae,Elateridae and Scarabidae (see Tab. 1; Fig.5). Beetles, which possess orange, yellowor red colouration or markings, typicallypossess a higher NIR-reflectance than dark-brown specimens. In some cases (e. g. in

Eudicella smithior Sternotomis bohndorfii) theorange legs or antennas showed a mediumNIR-reflectance. The highest phenetic plas-ticity was found in green-coloured beetles.Most metallic green beetles, as for exampleSternocera sternicornis,Sternotomis bohndorfiiandStephanorrhina guttata, do not show a markedNIR-reflectance, in contrast to their orangepatterns. There was also a small numberof green beetles (for example Smaragdesthesafricanaand Caelorrhina superba), in which amedium reflectance could be found. In C.superiorfive of the examined specimens didnot show any NIR-reflectance, while two

Fig. 3:A NDVI-values of grass vegetation under shadowy (left) and sunny (right) conditions asaffected by the corrective treatments (norm = without corrective treatment, wb = white card usedfor image capture, wb+flash = white card and internal flash used for image capture). The NDVI-values increase, when a white card and a white card plus internalflash of the camera are used under

shadowy conditions. Under sunny conditions NDVI-values are highest and corrective measures haveno effect. In the boxplots (n = 9 replicates) the horizontal line represents the median, the bottom andthe top of the box show the 25th and 75th percentile, respectively, and whiskers represent either themaximum value or 1.5 times the interquartile range, which ever is the smaller. Points higher than 1.5times are shown as single points (potential outlier). BDifferences of various background vegetationsamples under shadowy (top) and sunny (bottom) conditions (n = 3). A better light regime reducesthe variance and allows a better discrimination of the different surfaces.Abb. 3: ANDVI-Werte von Grasvegetation unter schattigen (links) und sonnigen (rechts) Umwelt-bedingungen in Abhngigkeit von Korrekturen whrend der Aufnahmen (norm = ohne Korrektur,wb = Weiabgleich mit weier Karte whrend der Bildaufnahme, wb+flash = Weiabgleich mit

gleichzeitig verwendetem Blitz). Durch den Einsatz der weien Karte und der weien Karte mitBlitz steigt der NDVI-Wert unter schattigen Bedingungen. Unter sonnigen Bedingungen sind dieNDVI-Werte am grten und die zustzlichen Korrekturmanahmen haben keinen Effekt. In denBoxplots (n = 9 Replikate) reprsentiert die horizontale Linie den Median und das untere und obereEnde der Box jeweils den 25- und 75-Perzentil. Die Whisker zeigen entweder den Maximalwert


7/33



oder den 1.5-Interquartilbereich. Einzelpunkte zeigen Werte auerhalb dieses Bereichs (potenzielleAusreier). B Unterschiede zwischen verschiedenen Vegetationsbeispielen bei schattigen (oben)und sonnigen (unten) Bedingungen (n = 3). Bessere Lichtverhltnisse reduzieren die Varianz undmachen verschiedene Oberflchen besser voneinander unterscheidbar.

norm wb wb +flash

corrective treatment

norm wb wb +flash

corrective treatment

shadow sun

shadow

sun

sample


8/33


Fig. 4:Different Pentatomidae species in a showcase. Overview (A) and NIR (B-F) and conven-tional color photos (G-K) of selected species, which either do not reflect NIR (B) or reflect (C)very weak, (D) weak , (E) with medium intensity and (F) high. (B+G) Trochycocoris rotundatus, (C+H)Dolycoris baccarum, (D+I) Chlorochroa juniperina, (E+J) Palomena viridissimaand (F+K) Nezara viridula.

Abb. 4:Verschiedene Pentatomiden in einem Schaukasten. bersicht (A) sowie NIR- (B-F) undFarbaufnahmen (G-K) einzelner Arten, die entweder (B) kein NIR reflektieren oder (C) sehrschwach, (D) schwach, (E) mittelstark und (F) stark reflektieren. (B+G) Trochycocoris rotundatus,(C+H) Dolycoris baccarum, (D+I) Chlorochroa juniperina, (E+J) Palomena viridissima, (F+K) Nezaraviridula.


9/33



showed either a weak or medium to strongreflectance (Fig. 6). The latter specimensshowed a slight reddish tint in their greencolouration when viewed in the visiblespectrum.

3.2.2. Hemiptera (Pentatomidae, mu-seum specimens)

Black specimens, e.g. Trochicocoris rotundatus,did not show any NIR-reflectance in theblack parts of their body (see Fig 4, Tab. 1).In contrast, in green and brown specimensof Chlorochroa juniperina, Dolycoris baccarum,

Palomena prasinaand Palomena viridissimava-rious shades of NIR-reflectance were found(Fig. 4 A, C-E). Nezara viridulashowed anexemplarily high NIR-reflectance.

3.2.3. Hymenoptera (museum speci-mens)

NIR-reflectance of the black body parts waslow in all Hymenoptera species investigated(Tab. 1), whereas the yellow striped mark-ings of different wasps showed a mediumreflectance. The transparent wings appearingyellowish in the visible spectrum showed ahigher reflectance.

3.2.4. Lepidoptera (museum and livingspecimens)

Dark patterns of caterpillars from muse-um specimens generally showed a reducedor no NIR-reflectance. The green cater-pillars ofEndromis versicolouraand Saturniapavoniarevealed a medium reflectance. Ahigh NIR-reflectance was found in livinggreen caterpillars of the Atlas moth A tta-cus atlas(Fig. 7). Cocoons of this speciesshowed a weak to medium reflectance

comparable to that of bark. Brown wingparts of adult butterflies such as Preponachromustypically show a weak to mediumreflectance. The light blue markings ofthis species and the Paris peacockPapilio

parisshowed a somewhat higher reflec-tance, compared to the rest of the wings. Ahigher reflectance was also found in yellowor orange markings ofPapilio euchenorandTeinopalpus imperialis. The green iridescentareas on the wings of T. imperialisand P.parisshowed a very weak NIR-reflectance,if any.

3.2.5. Odonata (museum specimen)

In contrast to the yellow spots on the abdo-men of a female L ibellula depressa, the darkbody parts of Odonata species did not show

any NIR-reflectance (see Tab. 1). Whitishtransparent exuviae of various unclassifiedspecies generally showed a high reflectance(data not shown).

3.2.6. Mantodea, Orthoptera andPhasmatodea (museum and livingspecimens)

A green living specimen ofMantis religiosashowed an extremely high NIR-reflectance,comparable to that of green leaves (Tab. 2).

The brownish wandering violin mantisGon-gylus gongylodesshowed only a weak to me-dium reflectance, similar to the twigs in itsterrarium. Similar observations were madeon several other green and brown speciesof Orthoptera and Phasmatodea (Tab. 2).Green leaf katydids such as Ancylecha fe-

nestrataand Sti lpnochlorchis coulonianashoweda strong NIR-reflectance comparable tothat of green leaves (Fig. 8). Yellow or greencoloured body parts of Schistocerca gregariashowed a strong reflectance. This alsoapplies to juvenile stages that, however,possess large black, non-reflecting mar-kings. Very young nymphs are completelyblack and easy to discriminate from back-

ground vegetation in the NIR-photos. Wealso observed a slight sexual dimorphismin this species regarding NIR-reflectance(Fig. 9). A very high NIR-reflectance

was also observed in green Phasmato-


10/33


Fig. 5:Specimens of different beetle species shown in conventional colour (A,C,E) and NIR-photos(B,D,F). The brown to red brown jewel beetle Sternocera hildebrandti(A+B) shows a medium NIR-reflectance, as the light brown elytra of the stag beetle Dicranocephalus bowringi(E+F). The green

cetoniid Ischiosopha lucivorax(C+D) does not show any NIR-reflectance.Abb. 5: Individuen von verschiedenen Kfern gezeigt in Farb- (A,C,E) und NIR- (B,D,F) Foto-grafie. Die braun bis rotbraunen Prachtkfer Sternocera hildebrandti(A+B) zeigen eine mittlere NIR-Reflexion, wie die hellbraunen Flgeldecken des Hirschkfers, Dicranocephalus bowringi(E+F). Diegrnen Rosenkfer Ischiosopha lucivorax(C+D) zeigen keine NIR-Reflexion.


11/33



Fig. 6:Specimens of the rose chafer Caelorrhina superba(A) differ considerably concerning NIR-reflectance (B). Black beetles such as Carabus scabrosus tauricus(C) generally do not show NIR-reflectance (D).

Abb. 6: Individuen des Rosenkfers Caelorrhina superba(A) unterscheiden sich hinsichtlich ihrerNIR-Reflexion betrchtlich (B). Schwarze Kfer wie Carabus scabrosus tauricus(C) zeigen generellkeine NIR-Reflexion (D).

dea species such as Heteropteryx dilatata,Diapherodes giganteaand Phyllium celebicum.In contrast, brown Phasmatodea as Ex-tatosoma tiaratumappeared darker andblueish in NIR-images similar to the co-louration of twigs and dry brown leaves.

Observations from living specimens gene-rally agree with those of dead museum speci-mens. Yet, it should be noted that colours ofgreen and yellow Mantodea and Orthopteraspecies in the collection tend to fade.

4. Discussion

Techniques of infrared photography are athand now for more than 100 years, beginn-ing with the discovery that vegetation orbetter green leaves possess an extremelyhigh reflectance in the NIR-range of thespectrum and, thus, appear snow-white inmonochrome infrared images (WOOD 1910a, b). This unusual photographic peculiarityis commonly known as the Wood effect,


12/33


Fig. 7:The bluish-green caterpillar of the Atlas mothA ttlacus atlas(A) exhibits a high NIR-reflectance(C), which markedly differs from that of the leaves because of its high blue light reflection. Theseemingly higher NIR-reflectance of the leaves is caused by their low blue light reflectance. Thebrownish cocoon shows a much lower reflectance in both wave bands (B).Abb. 7: Die blau-grne Raupe der Atlasmotte A ttlacus atlas(A) weist eine hohe NIR-Reflexion

auf (C), die sich wegen ihrer starken Blaulichtrefl

exion stark von der der NIR-Refl

exion derBltter unterscheidet. Die scheinbare hhere NIR-Reflexion der Bltter ist durch ihre geringeBlaulichtreflexion bedingt. Die braune Puppe zeigt eine viel geringere Reflexion in beiden Wel-lenlngenbereichen (B).


13/33



Fig. 8:The Malaysian katydid Ancylecha fenestrata(A) shows a strong NIR-reflectance (B).Abb. 8:Die Malaysische Blattschrecke Ancylecha fenestrata(A) zeigt eine starke NIR-Reflexion (B).

a term still widely used by enthusiasticamateurs and professional infrared photo-graphers. Since then infrared photographyhas been frequently used in different fields

of science (e.g. CLARK 1947; GIBSON 1978;VERHOEVEN 2008).More recently film-based infrared pho-tography has been substituted by digital


14/33


Fig. 9:The female of the desert locust Schistocerca gregariaappears slightly darker compared to themale that shows a NIR-reflectance similar to the grass vegetation; the nymphs (on the top) do notreflect any NIR.Abb. 9:Das Weibchen der Wanderheuschrecke Schistocerca gregariaerscheint etwas dunkler als dasMnnchen, das eine sehr starke NIR-Reflexion zeigt; die Nymphen (oben) reflektieren kein NIR.


15/33



Fig. 10:The leaf insect Phyllium celebicum(A) and its NIR-reflectance (B) similarly strong as thesurrounding vegetation.Abb. 10: Das Wandelnde Blatt Phyllium celebicum(A) reflektiert im NIR (B) hnlich stark wie dieumgebende Vegetation.

cameras, with the consequence that co-lour infrared films are no longer available(VERHOEVEN 2008) and that commerciallaboratories have ceased film processing.

In principle, however, all CCD (charge-coupled device) and CMOS (complemen-tary metal-oxide-semiconductor) sensors ofdigital consumer cameras are also sensitive


16/33


to near-infrared up to 1100 nm due tothe intrinsic physical properties of silicon(FREDEMBACH & SSSTRUNK 2008, 2010).

Yet, in front of the CCD sensor a more orless efficient hot-mirror is integrated intothe camera sensor to block infrared light,

which improves the overall image quality.Therefore, any consumer DSLR cameracan be used to acquire NIR-images, ifthe exposure time is sufficiently long andif a filter is used to block the visible light(VERHOEVEN 2008). However, very longexposure times are often not practicableand may reduce image quality, but most

digital consumer cameras can be modifiedinto NIR-sensitive cameras by removingthe internal hot-mirror (used in normalphotography to block NIR-light) and usingdifferent red or infrared band-pass filters,similar to those used in classical infraredphotography (see GIBSON 1978 for filtersused in infrared photography and TETLEY& YOUNG 2007, 2009 for details on infraredcamera modification). Commonly, those fil-ters can be attached either to the front lensof the camera or they are mounted behindthe lens on the sensor during camera modi-fication. As filter usage is not standardizedpresently, digital colour infrared photog-raphy requires digital post-processing, e.g.manual colour adjustments.

The herein presented results show that themodified NDVI-camera used by us is an

efficient and useful alternative to performdigital infrared photography for scientificpurposes (for references on digital infraredphotography see VERHOEVEN 2008). Ad-

vantages are a colour scheme based on twobands, which is not only easy to understand,but also matches partially the false colourrendering of previous colour infrared films(for reference see GIBSON 1978). Concern-

ing applied NIR-photography, the NDVI-camera has several advantages: (1) It is notnecessary to apply any additional filter infront of the lens to block the visible light,as the filter has been integrated back-lens

directly on the CCD sensor by the manufac-turer. In consequence, both the viewfinderand the live view mode are fully functional.(2) There are also no limitations to auto-exposure and sensor cleaning functions.(3) The narrow RE-bandwidth of thecamera limited to the range of about675 to 800 nm. Therefore any problemsrelated to wavelength-dependent shifts inthe camera focus typically occuring in NIR-photography are minimized and even theautofocus can be used. (4) There appearto be no limitations concerning the use of

various lenses.

For general visualization of differences inNIR-reflectance it does not appear to benecessary to normalize the generated images(see equation p. 185), though it can be help-ful whenever quantitative imaging is needed,e. g. in vegetation studies. Under shadowycondition (e. g. cloudy sky, tree canopy, in-door) image quality can be improved by us-ing white cards and internal flashes. That inour study these NDVI-values did not reachNDVI-values obtained in full sun can beexplained by the weakflash of the camera.Using a stronger external flash will probablyincrease these values. We showed that addi-tional light during image capture improvesimage quality for analytical purposes.Infrared photography for image-based phe-notyping of animals has already been pro-posed by DODD (1981) and especially KREM-

PELS (1989). Now digital infrared camerasenormously simplify this method, not onlybecause it is easier to acquire many images ina short time, but also (even more importantto our opinion) because time-consuming andcostlyfilm processing in a darkroom specifi-cally equipped for infrared photography hasbecome obsolete. In addition, recent digitalconsumer cameras have further advantages

very useful for image-based phenotyping, i.e. additional meta-information can be auto-matically integrated into the image duringimage acquisition such as used lens, exposuretime, flash mode etc., which can be accessed


17/33



later on any computer (MIELEWCZIK 2002,2003 a, 2003 b). Also various epithetha canbe attached to the images using additionalmeta-datafields provided by the IPTC imagefile extensions (see for example MIELEWCZIK2007 a), which is helpful in managing hugedatabases with several thousand images. Webelieve it is best to acquire images in RAWand JPEG mode simultaneously. RAWimage files generally provide the highestquality and degree of freedom for post-processing purposes (see for exampleMIELEWCZIK2007 b), while JPEG files aremuch easier to handle in day-to day-use due

to their smaller size.

4.1. Infrared reflectance of insects

The most striking result of our study is thefact that many green insects show a very highNIR-reflectance, which closely resemblesthat of leaves. This was especially found inliving specimens of walking leaves (Phasma-todea) and green Orthoptera such as Phyl-lium celebicumand Stilpnochlora couloniana. Allspecies are known for camouflage and phy-tomimesis (for summary see LUNAU 2011).Spectral similarities between green insects,such as caterpillars, grasshoppers, mantisand walking leaves, and true leaves havebeen known for a long time (BECQUEREL &BRONGNIART 1894; PRZIBAM 1913). However,to our knowledge it has never been reported

for any mantid or orthopteran species thatthis property extends into the NIR-rangeof the spectrum. Spectra of green insectspecies showing a high NIR-reflectance areknown from a caterpillar (Rhodenia fugax:SAITO 2001) and in the metallic green beetlesCalloodes grayanus(PARKER et al. 1998) andCharidotella egregia(VIGNERON et al. 2007).

4.1.1. Museum specimens

We found different intensities of NIR-reflectance in a large number of deadspecimens from the museum collections

that were dried and stored in drawers. Thisshows that these specimens retained thisproperty at least in part. We assume thatNIR-reflectance observed in these speci-mens is largely attributed to the organizationand pigmentation of the cuticle; the under-lying epidermis should be destroyed. Thegeneral lamellate structure for exampleof the elytron cuticle hardly or not to beseen in the exocuticle is conserved (e.g.

VANDE KAMP & GREVEN 2010). The sameapplies for structural colours of the cuticle,e. g. the multilayered reflectors in the cuticleof beetles (SEAGO et al. 2009), which can be

identified even in fossils (PARKER& MCKEN-ZIE 2003; TANAKA et al. 2010; MCNAMARA etal. 2012). Also preserved are dark and brownpatterns caused generally by tanning (elytraare often heavily sclerotized), but also bymelanization, which was seen in many ofthe species examined.However, some pigments appear to fadein course of the time, e.g. in the green andotherwise coloured species of Mantodea,Phasmatodea, Orthoptera and also in lightcoloured caterpillars. Their green and yel-low patterns are caused by the simultaneouspresence of selectively blue (carotenes) andred absorbing pigments, such as chromo-proteins containing non-covalently boundbiliverdin IX as prosthetic group (PRZIBRAM& LEDERER1933; KAWOOYA et al. 1985; OKAY1945; PASSAMA-VUILLAUME & BARBIER1966;

RDIGER 1970; HOLDEN et al. 1987; WIL-LIG 1969). Therefore, we assume that herephoto-oxidation and chemical degradationare responsible for bleaching, but, neverthe-less, some NIR-reflectance is retained.

We think that also some further parametersmay affect NIR-reflectance such as thinnessof the cuticle and their dehydration. Anopposite effect (i.e. higher reflectance) is

known from stacking of leaves due to thereduced transmission (ALLEN & RICHARDSON1968, see also GATES 1980). Further, dryingof the cuticle may lead to another effect, alsoknown from studies of leaves: Scattering co-


18/33


efficients in diffuse reflectance over the VISand NIR-range of the spectrum generallyincrease in dehydrated tissues (ALLEN &RICHARDSON 1968; see GATES 1980). De-hydrated leaves show an increase in reflec-tance over the whole range of the spectrumfrom 400-2500 nm (CARTER1991; RASCHERet al. 2007), although only the increased re-flectance in the range of 1000 to 2500 nm isthought to be directly related to absorptionof water (JACQUEMOUD et al. 1996; RASCHERet al. 2010). Nevertheless, the acquiredNDVI-infrared images of the insects exam-ined herein show areas and patterns of dark

and brown colourations that are retainedin the museum specimens and play a ma-jor role in the overall spectral and spatialabsorption in the RE- and NIR-range ofthe spectrum.Further, our study indicates that insectspecies with a considerably high NIR-reflectance are found throughout the worldand in different taxa. We think that thisphenomenon might be more common inspecies living on leaves. Here, NIR-reflec-tance matches to the green background.Similarly, NIR-reflectance of dark or browninsects may match to their local substrate.Generally, however, the adaptive value ofNIR-reflectance of insects and animals islargely unknown and only a few sugges-tions are given herein: KREMPELS (1989),

who studied infrared reflectance of several

frogs and squamates assumed an adaptivecamouflage of predators and suggestedthat IR-reflectance in coral snakes con-tributes to their aposematic colouration.Interestingly, we found NIR-reflectance inthe green predatoryMantis religiosa, whichin turn may baffle its prey. Brownish orred colouration is accompanied by a weakto medium NIR- reflectance (Buprestidae:

WAGNER1971; and many other dark insectsincluding several phasmids). This fits verywell into the general cryptic protectivestrategies of many stick insects (Phasmato-dae) (for review see BEDFORD 1978).

If these suggestions are right, putativepredators or prey must be able to see inthe NIR-and RE range of the spectrum.Spectral tuning of NIR-reflectance andthe steep increase of reflectance in the REmight be beneficial to baffle predators thatshow a shift toward longer wavelengths intheir spectral sensitivity. Up to date truesensitivity in the NIR,first suggested by earlypioneers in infrared photography (TRISTAN& MICHAUD 1916), has been reported ina limited number of various invertebrateand vertebrate taxa, e.g. Crustacea (Mysisrelicta: LINDSTRM & MEYER-ROCHOW1987),

Coleoptera (Hypera postica: Meyer 1976),Teleostei (Rutilus rutilus: LOEW & LYTHGOE1985; Oreochromis mossambicus: SHCHERBAKOVet al. 2012), Dipnoi (Neoceradotus fosteri:HART 2008), Squamata (Anolis carolinensis:KAWAMURA & YOKOYAMA 1998; PROVENCIO etal. 1992), Mammalia (Mustelus furo: NEWBOLD2007, NEWBOLD & KING 2009). Infrared

vision in birds (VANDERPLANCK1934, WOJ-TUSIAK1949) and turtles (WOJTUSIAK1947)has been doubted (for discussions see forexample MATTHEWS & MATTHEWS 1939, INDEN BOSCH 1987).It is also possible that reflectance in thefar-red and near-infrared has an adaptive

value with respect to interspecific visualinteractions, as it has been suggested forexample in fish and some squamates (PAR-TRIDGE et al. 1989; PROVENCIO et al. 1992).

NIR-reflectance may also affect thermo-regulation as heatload may be increased(for discussion see IN DEN BOSCH 1987).Furthermore, it had been suggested, thatIR-reflectance might be only an inevitableby-product of the colour pattern in the

visible part of the spectrum (INDEN BOSCH1987). For all suggestions convincing ex-perimental evidence is missing.

To our knowledge the structural basis of IR-reflectance is not yet satisfactorily clarified. InLepidoptera pterin is suggested to contributeto the overall reflectance of the longer wave-lengths within the visible spectrum (RUTOWSKI


19/33



et al. 2005) and therefore may also cover theNIR-range. We believe that the very highNIR-reflectance demonstrated in various in-sects or their larvae is based on the structuralproperties of their integument, i.e. the cuticleand epidermis. These properties may lead to abroadband high and diffuse reflectance, span-ning the VIS and NIR-spectrum. The NIR-reflectance is furthermore characterized byabsorption of brownish and dark pigments.

Additional blue and red absorbing pigmentssuch as erythropterin and carotenoids (blueabsorbing) and biliverdin (red absorbing)occurring in both, the cuticle and the epi-

dermis (LENAU & BARFOED 2008), may leadto the green appearance of the specimens. Inaddition, multiple layers with an alternatingpattern of low and high refractive indices, astypical for the arthropod cuticle in general,can define not only visible colours (LAND1966; LENAU & BARFOED 2008), but may alsodefine NIR-reflectance.Especially the various taxa of green metallic-coloured beetles such as Stephanorrhina gut-tataand some butterflies such as Teinopalpusimperialisshow a wide range of variation inthe intensity of NIR-reflectance, which maybe related to the different modes of colourproduction (e.g. scattering, diffraction orinterference effects) (for references see FOX& VEVERS 1960; CHAPMAN 1998; MICHIELSEN& STAVENGA 2008; LENAU & BARFOED 2008).More difficult to explain are interspecific

differences as demonstrated in Caelorrhinasuperba. Here polarizing effects in the nano-structure may affect NIR-reflectance.Our study shows that high levels of NIR-re-flectance are much more widespread amonginsects than to be expected from literature.Especially in green tropical Phasmatodeaand Orthoptera NIR-reflectance appearsto be prevailing.

Acknowledgements

We thank Dr. SILKE STOLL (Aquazoo/Lb-becke-Museum der Stadt Dsseldorf) and

Dr. ANDREAS MLLERand FRANZISKA SCHMID(Institute of Agricultural Sciences, AppliedEntomology Group, ETH Zrich), whomade available to us the insect specimensfrom the collection of the Museum, andSUSANNE TITTMANN (ForschungszentrumGeisenheim) for support in photographyon excursions.

Literature

ALLEN, W.A., & RICHARDSON, A.J. (1968): Interac-tion of light with a plant canopy. Journal ofthe Optical Society of America 58: 1023-1028.

BECQUEREL, H., & BRONGNIART, C. (1894): Lamatire vertechez les Phyllies. Comptesrendus hebdomadaires des sances de lAca-dmie des sciences 118: 1299-1303.

BEDFORD, G.O. (1978): Biology and ecology ofthe phasmatodea. Annual Review of Ento-mology 23: 125-49.

BOKHORST, S., TMMERVIK, H., CALLAGHAN, T.V.,PHOENIX G.K., & BJERKE, J.W. (2012): Veg-etation recovery following extreme winterwarming events in the sub-Arctic estimated

using NDVI-from remote sensing and hand-held passive proximal sensors. Environmentaland Experimental Botany 81: 18-25.

CARTER, G.A. (1991): Primary and secondaryeffects of water content on the spectral reflec-tance of leaves. American Journal of Botany78: 916-924.

CHAPMAN, R.F. (1998): The insects. Structureand function. 3rd edition. University Press;Cambridge.

CLARK, W. (1947): Photography by infrared Itsprinciples and application. 2nd edition. JohnWiley & Sons; London, New York.

COTT, H.B. (1949): Adaptive colouration in ani-mals. Methuen & Co Ltd.; London.

CROMARTIE, R.I.T. (1959): Insect pigments. An-nual Review of Entomology 4: 59-76.

DODD, C.K. JR. (1981): Infrared reflectance inchameleons (Chamaeleonidae) from Kenya.Biotropica 13: 161-164.

EMERSON, S.B., COOPER, T.A., & EHLERINGER, J.R.

(1990): Convergence in reflectance spectraamong treefrogs. Functional Ecology 4: 47-51

FOX, H. M., & VEVERS, G. (1960): The natureof animal colours. Sidgwick and Jackson,London.


20/33


FOX, D.L. (1976): Animal biochromes and struc-tural colours physical, chemical, distributi-onal & physiological features of bodies inthe animal world. 2nd edition. University ofCalifornia Press; Berkeley.

FOX, D.L. (1979): Biochromy Natural coloura-tion of living things. University of CaliforniaPress; Berkeley.

FREDEMBACH, C., & SSSTRUNK, S. (2008): Colour-ing the near infrared. Pp. 176-182 in: Procee-dings of IS & T/SID 16th Colour ImagingConference; Portland.

FREDEMBACH, C., & SSSTRUNK, S. (2010): Auto-matic and accurate shadow detection from(potentially) a single image using near-infraredinformation. IEEE Transactions on PatternAnalysis and Machine Intelligence 165527.

GATES, D. (1980): Biophysical ecology. Springer;New York, Heidelberg, Berlin.

GEHRING, P.-S., & WITTE, K. (2007): Ultravioletreflectance in Malagasy chameleons of thegenus Furcifer(Squamata: Chamaeonidae).Salamandra 43: 43-48.

GIBSON, H.L. (1978): Photography by infrared:Its principle and applications. 3rd edition. JohnWiley & Sons; London, New York.

HADLEY, N.F., SAVILL, A., & SCHULTZ, T.D. (1992):Colouration and its thermal consequences inthe New Zealand tiger beetle Neocicindela per-hispida. Journal of Thermal Biology 17: 55-61.

HART, N.S., BAILES, H.J., VOROBYEV, M., MAR-SHALL, N.J., & COLLIN, S.P. (2008): Visual ecol-ogy of the Australian lungfish (Neoceratodusforsteri). BMC Ecology 8: 1-14.

HOLDEN, H.M., RYPNIEWSKI, W.R., LAW, H.L., &RAYMENT, I. (1987): The molecular structureof insecticyanin from the tobacco hornwormManduca sextaL. at 2.6 resolution. EMBOJournal 6: 1565-1570.

JACQUEMOUD, S., USTIN, S.L., VERBEDOUT, J.,SCHMUCK, G., ANDREOLI, G., & HOSGOOD, B.(1996) Estimating leaf biochemistry using thePROSPECT leaf optical properties model.Remote Sensing of Environment 56: 194-202.

INDEN BOSCH, H.A.J. (1987): ber Missverstnd-nisse bei der Deutung von Infrarotaufnahmenam Beispiel von Chameleo jacksonii. Herpeto-

fauna 51:19-28.KAWAMURA, S., & YOKOYAMA, S. (1998): Func-tional characterization of visual and nonvisualpigments of American chameleon (A noliscarolinensis). Vision Research 38: 37-44.

KAWAOOYA, J.K., KEIM, P.S., LAW, J.H., RILEY, C.T.,RYAN, R.O., & SHAPIRO, J.P. (1985): Why aregreen caterpillars green? ACS SymposiumSeries American Chemical Society 276:511-521.

KELBER, A., VOROBYEV, M., & OSORIO, D. (2003):Animal colour vision: behavioural tests andphysiological concepts. BiologicalReviews ofthe Cambridge Philosophical Society 78: 81-118

KREMPELS, D.M. (1989): Visible light and near-infrared reflectance of amphibians and rep-tiles and the visual system of avian predators(Accipitridae:Buteospp). Dissertation Thesis:University of Miami; Miami.

KUSUDA, J., & MUKAI, J.-I. (1971): A biliproteinfrom the digestive juice ofBombyx moriL. Itspurification and partial structural study of thechromophore. Comparative Biochemistry andPhysiology Part B 39: 317-323.

LENAU, T., & BARFOED, M. (2008): Colours andmetallic sheen in beetle shells A biometricsearch for material structuring principlecausing light interference. Advanced Engi-neering Materials 10: 299-314.

LOEW, E.R. & LYTHGOE, J.N. (1985): The ecologyof colour vision. Endeavour 9 (4): 170-174.

LOWREY, S., DE SILVA, L., HODGKINSON, I., &LEADER, J. (2007): Observation and model-ing of polarized light from scarab beetles.Journal of the Optic Society of America A24: 2418-2425.

LUNAU, K. (2011): Warnen, Tarnen, Tuschen.Mimikry und Nachahmung bei Pflanze, Tierund Mensch. Wissenschaftliche Buchgesell-schaft; Darmstadt.

MATTHEWS, L.H., & MATTHEWS, B.H.C. (1939):Owls and infrared-radiation. Nature 143: 983.

MCNAMARA, M.E., BRIGGS, D.E.G., ORR., P.J.,NOH, H., & CAO, H. (2012): The originalcolours of fossil beetles. Proceedings of theRoyal Society B 279: 1114-1121.

MENZEL, R. (1979): Spectral sensitivity andcolour vision in invertebrates. Pp. 503-580in: AUTRUM, H. (ed.): Handbook of SensoryPhysiology. Vol. VII/6A. Springer; Berlin,Heidelberg, New York.

MEYER, J.R. (1976): Positive phototaxis of adult

alfalfa weevils to visible and near-infrared ra-diation. Annals of the Entomological Societyof America. 69 : 21-25.

LINDSTRM, M., & MEYER-ROCHOW, V.B. (1987):Near infra-red sensitivity of the eye of the


21/33



crustaceanMysis relicta? Biochemical and Biophy-sical Research Communications 147: 747-752.

MICHIELSEN, K., & STAVENGA , D.G. (2008):Gyroid cuticular structures in butterfly wingscales: biological photonic crystals. Journal of

the Royal Society Interface 5: 85-94.MIELEWCZIK, M. (2002): Versteckte Informa-tionen. PC Praxis Foto 2/2002: 113-114.

MIELEWCZIK, M. (2003 a): Eldorado fr Hobby-Fotografen Zusatzinformationen zu Digi-talfotos. Linux User 10/2003: 52

MIELEWCZIK(2003 b): Extended information ondigital photos the perfect picture. LinuxMagazine 11/2003: 34-35.

MIELEWCZIK, M. (2007 a): Exif-Daten Zusatz-Infos nutzen. Chip Linux 1/2007: 54-55.

MIELEWCZIK, M. (2007 b): Raw-Bilder unterLinux. Chip Linux 1/2007: 49-51.

OKAY, S. (1945): Pigmentation of Orthoptera.Nature 155: 635.

NEWBOLD, H.G. (2007): Infra-red vision in ferrets(Mustelo furo). Master Thesis, University ofWaikato; Waikato, New Zealand.

NEWBOLD, H.G., & KING, C.M. (2009): Can a preda-tor see invisible light? Infrared vision in ferrets(Mustela furo). Wildlife Research 36: 309-318.

PARKER, A.R., MCKENZIE, D.R, & LARGE, M.C.J.(1998): Multilayer reflectors in animals usinggreen and gold beetles as contrasting ex-amples. The Journal of Experimental Biology201: 1307-1313.

PARKER, A.R., & MCKENZIE, D. (2003): The causeof 50 million-years-old colour. Proceedings ofthe Royal Society B 270: S151-S153.

PARTRIDGE, J.C., SHAND, J., SRCHER, S.N., LYTHGOE,J.N., & VAN GRONING-LUYBEN, W.A.H.M.(1989): Interspecific variation in the visualpigments of deep-seafishes. Journal of Com-parative Physiology A, 164: 513-529.

PASSAMA-VUILLAUME, M., & BARBIER, M. (1966):Sur la biosynthse de la biliverdine IX parla mante Mantis religiosaet le criquet Locustamigratoria. Comptesrendus hebdomadairesdes sances de lAcadmie des sciences, SrieD 263: 924-925.

PROVENCIO, I., LOEW, E.R., & FOSTER, R.G.(1992): Vitamin A2-based visual pigments in

fully terrestrial vertebrates. Vision Research32: 2201-2208.PRZIBAM, H. (1913): Grne tierische Farbstoffe.

Pflgers Archiv fr die gesamte Physiologie,153: 385-400.

PRZIBRAM, H., & LEDERER, E. (1933): DasTiergrn der Heuschrecken als Mischungaus. Farbstoffen. Anzeiger der Akademieder Wissenschaften Wien, Mathematisch-Naturwissenschaftliche Klasse 70: 163-167.

RASCHER, U., NICHOL, C.J., SMALL, C., & HEN-DRICKS, L. (2007): Monitoring spatio-temporaldynamics of photosynthesis with a portablehyperspectral imaging system. Photogram-metric Engineering & Remote Sensing 73:45-56.

RASCHER, U., DAMM, A., VAN DER LINDEN, S.,OKUJENE, A., PIERUSCHKA, R., SCHICKLING, A.,& HOSTERT, P. (2010): Sensing of photsyn-thetic activity in crops. Pp. 87-99 in: OERKE,E.-C., GERHARDS, R., MENZ, G., & SIKORA,R.A. (eds.): Precision crop protection Thechallenge and use of heterogeneity. Springer;Berlin, Den Haag.

R-DEVELOPMENT-CORE-TEAM (2008): R: A lan-guage and environment for statistical compu-ting. R Foundation for Statistical Computing;Vienna. http://www.R-project.org.

RDIGER, W. (1970): Animal biliproteins. Proceed-ings of the 505th Meeting of the BiochemicalSociety. Biochemical Journal 119 (3): 1P.

RUTOWSKI, R.L., MACEDONIA, J.M., MOREHOUSE,N., & TAYLOR-TAFT, L. (2005): Pterin pig-ments amplify iridescent ultraviolet signal inmales of the orange sulphur butterfly, Coliaseurytheme. Proceedings of the Royal Society B272: 2329-2335.

SAITO, H. (2001): Blue biliprotein as an effectivefactor for cryptic colozration in Rhodiniafugaxlarvae. Journal of Insect Physiology47: 205-212.

SEAGO, A.E., BRADY, P., VIGNERON, J.P., &SCHULTZ, T.D. (2009): Review: Gold bugs andbeyond: a review of iridescence and structuralcolour mechanisms in beetles (Coleoptera).Journal of the Royal Society Interface 6:165-184

SHCHERBAKOV, D., KNRZER, A., HILBIG, R.,HAAS, U., & BLUM, M. (2012): Near-infraredorientation of Mozambique tilapia Oreochromismossambicus. Zoology 115: 233-238.

SCHWALM, P., STARRETT, P., & MCDIARMID, R.

(1977): Infrared reflectance in leaf-sittingneotropical frogs. Science 196: 1225-1227.TANAKA, G., TANIGUCHI, H., MAEDA, H., & NOMURA,

S. (2010): Original structural colour preservedin an ancient leaf beetle. Geology 38: 127-130.


22/33


TETLEY, C., & YOUNG, S. (2007): Digital IRand UV imaging. Part 1: Infrared. Journalof Visual Communication in Medicine 30:162-171.

TETLEY, C., & YOUNG, S. (2008): Digital infrared

and ultraviolet photography using advancedcamera services modified equipment. Journalof Visual Communication in Medicine 32:40-42.

TRISTAN J.F., & MICHAUD G. (1916): Probableperception of light by some animal species.Scientific American January 15: 81, 88-89.

VANDE KAMP, T., & GREVEN, H. (2010): On thearchitecture of beetle elytra. Entomologieheute 22: 191-204.

VERHOEVEN, G. (2008): Imaging in invisible usingmodified digital still cameras for straightfor-ward and low-cost archaeological near-infra-red photography. Journal of ArchaeologicalScience 35: 3087-3100.

VIGNERON, J.P. PASTEELS, J.M., WINDSOR, D.M.,VRTESY, Z., RASSART, M., SELDRUM, T., DU-MONT, J., DEPARIS, O., LOUSSE, V., BIRO, L.P.,ERTZ, D., & WELCH, V. (2007): Switchablereflector in the Panamanian tortoise beetleCharidotella egregia(Chrysomelidae: Cassidi-

nae). Physical Review E 76 031901: 1-9.VANDERPLANK, F.L. (1934): The effect of infra-red waves on tawny owls (Strix aluco). Proceed-ings of the Zoological Society of London104: 505-507.

VIGNERON, J.-P., & SIMONIS P. (2010): Structuralcolours. Advances in Insect Physiology 38:181-218.

WAGNER, G. (1971): Infrarot-Fotografie DerWeg ins Unsichtbare. 2. Auflage. Verlag dieschnen Bcher Dr. Strache KG; Stuttgart.

WELCH, V., LOUSSE, V., DEPARIS, O., PARKER, A.,& & VIGNERON, J. (2007): Orange reflection

from a three-dimensional photonic crystal inthe scales of the weevilPachyrrhynchus congestuspavonius(Curculionidae). Physical Review E75: 41919-1-41919-9.

WILLIG, A. (1969): Die Carotenoide und der

Gallenfarbstoff der Stabheuschrecke, Ca-rausius morosus, und ihre Beteiligung an derEntstehung der Farbmodifikationen. Journalof Insect Physiology 15: 1907-1927.

WOJTUSIAK, R.J. (1947): Investigations on the visionof infra-red in animal. I. Experiments on watertortoises. Bulletin International de lAcademiedes Sciences de Cracovie B 2: 43-61

WOJTUSIAK, R.J. (1949): Polish investigationson homing in birds and their orientation inspace. Proceedings of the Linnean Society ofLondon 160: 99-108.

WOOD, R.W. (1910 a): A new departure in pho-tography. The Century Magazine 79: 565-572.

WOOD, R.W. (1910 b): Photography by invisiblerays. The Photographic Journal 50: 329-338

Dipl. Biol. Michael MielewczikDr. Frank LiebischProf. Dr. Achim WalterETH Zrich

Institute of Agricultural SciencesUniversittstr. 2CH-8092 Zrich/SwitzerlandE-Mail: [email protected]: [email protected]: [email protected]

Prof. Dr. Hartmut GrevenZoologie II der Heinrich-Heine-UniversittUniversittsstr. 1

D-40225 DsseldorfE-Mail: [email protected]


23/33



Appendix

Tab.

1:NIR-reflectanceofinsectsfrommuseumcollectionsasestimatedbyRe

d-E

dge-P

hotography.Numberofexaminedspecimensinbracke

ts

afterthescien

tificname.

Tab.

1:NIR-R

eflexionvonInsektenartenausM

useumssammlungenbasierendaufRed-E

dge-Fotografie.DieA

nzahlderuntersuchtenIndividuen

stehtinKlammernhinterdemwissenschaftlichenNamen.


24/33


Tab.

1:Continued.

Tab.

1:Fortse

tzung.


25/33



Tab.

1:Continued.

Tab.

1:Fortse

tzung.


26/33


Tab.

1:Continued.

Tab.

1:Fortse

tzung.


27/33



Tab.

1:Continued.

Tab.

1:Fortse

tzung.


28/33


Tab.

1:Continued.

Tab.

1:Fortse

tzung.


29/33



Tab.

1:Continued.

Tab.

1:Fortse

tzung.


30/33


Tab.

1:Continued.

Tab.

1:Fortse

tzung.


31/33



Tab.

1:Continued.

Tab.

1:Fortse

tzung.


32/33


Tab.

1:Continued.

Tab.

1:Fortse

tzung.


33/33


b.

2:NIR-r

eflectanceoflivinginsectsasestimatedbyRed-E

dge-Photography.

b.

2:NIR-R

eflexionvonlebendenInsekten

basierendauf

Red-E

dge-Foto

grafie.

Documents

Near-Infrared (NIR)-Reflectance in Insects