Adaptive camouflage: What can be learned from the wetting behaviour of the tropical flatbugs Dysodius lunatus and D. magnus

The neotropical flatbug species Dysodius lunatus and Dysodius magnus show a fascinating camouflage principle. Its appearance renders the animal hardly visible on the bark of trees. However, when getting wet due to rain, bark changes its colour and gets darker. In order to keep the camouflage effect, it seems as if some Dysodius species benefit from their ability to hold a water film on their cuticle and therefore change their optical properties when wetted by water too. This camouflage behaviour requires the insect to have a hydrophilic surface and passive surface structures, which facilitate the liquid spreading. Here we show morphological and chemical characterisations of the surface, especially the cuticular waxes of Dysodius magnus. Scanning electron microscopy revealed that the animal is covered with pillar-like microstructures which in combination with a surprising chemical hydrophilicity of the cuticle waxes, render the bug almost superhydrophilic: Water spreads immediately across the surface. We could theoretically model this behaviour assuming the effect of hemi-wicking (a state in which a droplet sits on a rough surface, partwise imbibing the structure around). Additionally the principle was abstracted and a laser patterned polymer surface, mimicking the structure and contact angle of Dysodius-wax, shows exactly the behaviour of the natural role model – immediate spreading of water and the formation of a thin continuous water film changing optical properties of the surface. B io lo gy O pe n • A dv an ce a rt ic le by guest on November 2, 2017 http://bio.biologists.org/ Downloaded from Introduction Flat bugs (Aradidae), also named bark bugs, comprise a family of insects in the order of Heteroptera (true bugs). In the temperate zone, these animals are often found on or under loose bark of dead trees; while many tropical species are found in leaf litter or on fallen twigs and branches or they dwell on the bark. Most if not all members of the family are supposed to be mycophagous. Flat bugs are relatives of the more familiar shield and tree bugs (Pentatomidae/Scutelleridae) (Schuh & Slater, 1995). They typically rely on their excellent camouflage, i.e. their colouration, often mottled brown/black/grey/yellow, makes them quite difficult to detect for predators like birds. This was for example investigated by Johansen et al. (2010). This work investigates flat bug species Dysodius lunatus (Fabricius, 1794, Figure 1A) and Dysodius magnus (Figure 1B) (Heiss, 1990), which live on the outside of the bark of trees in the neotropics where they feed on fungi (Panizzi & Grazia, 2015). Their cryptic nature should be apparent from Figure 1C where it is difficult to make out where the bark ends and the animal begins. However, these flatbugs face a problem: Bark changes its appearance, namely its colour if wetted. As can be observed in daily life, most trees’ bark becomes dark when wetted. This change of the albedo is due to total internal reflection within a liquid film on rough surfaces as well as on a change of the reflection coefficient due to different refractive indices of water and the wetted material. As can be seen in Figure 1D, a dry Dysodius specimen would be easily detectable on wet bark. As wetted bark is quite common in tropical rain forests, the animal has to cope with that. Most terrestrial insects possess a highly hydrophobic cuticula (Holdgate, 1955), mainly due to waxes rendering these animals not wettable. For most insects, this water repellent behaviour is advantageous. This does not apply for the Dysodius species investigated here. Unlike most other insects, these flatbugs are easily wettable and change colour when wetted. This allows the animals to adapt to the moisture induced colour change of the bark accordingly. This was first shown by Silberglied and Aiello (1980) and can be seen in Figure 1C to F. In the present study we tried to analyse this unusual wetting behaviour of insects. We characterised the epicuticular waxes chemically as well as morphologically and describe specialised wax glands of these animals. We describe the wetting behaviour theoretically and finally we succeed in reproducing the wetting behaviour of the insect’s cuticle by using specially structured polymer foils. B io lo gy O pe n • A dv an ce a rt ic le by guest on November 2, 2017 http://bio.biologists.org/ Downloaded from Materials and Methods Video-analysis Videos for the investigation of liquid spread on the native surface of Dysodius were obtained with a Keyence highspeed video microscope (Model VW-9000; Keyence, Japan) with an attached 50x magnification lens (VHZ00R; Keyence, Japan). Videos were recorded at a frame rate of 125 fps. Videos for the investigation of liquid spread on laser-generated structures on PET substrates were obtained with a digital camera Nikon D5300 (Nikon Corp., Japan) with an attached 60 mm macro lens ( AF-S Micro Nikkor 1:2,8, Nikon Corp., Japan). Scanning and transmission electron microscopy (SEM & TEM) SEM images were obtained with a Philips SEM 525 (Philips, Germany) at 15 kV. Air-dried bug samples were additionally dried on silica gel for 24 hours before sputter coating with gold using a Polaron E5200 auto-coating device (former Polaron Unlimited, Great Britain) at 1 kV for a duration of 180 s. For TEM analysis, pieces of dried specimen were first fixated for 12 hours at 4°C in 2.5% glutaraldehyde solution (SERVA, Germany), made from 0.2 M cacodylate buffer, followed by washing two times 15 minutes in pure cacodylate buffer. For further fixation and contrast enhancement, this was followed by 1 h incubation in 1% osmiumtetroxide (Fluka, USA) solution, again in cacodylate buffer, at 4°C. Two times washing for 15 minutes in pure cacodylate followed by two times 15 minutes in distilled water. Afterwards samples were dried in an ascending alcohol series (30%-70%). Negative staining was then applied by incubating in 1% uranylacetate (Serva, Germany, in 70% ethanol) for 2 hours, before washing twice in 70% ethanol and finishing the drying row up to 100% ethanol. Afterwards ethanol was replaced by 2 times 30 minutes in pure propyleneoxide (Serva, Germany) and followed by overnight incubation in a mix of two parts propylene oxide and one part epon (Epon quick mix, Serva, Germany). The samples were finally embedded in pure epon and polymerized for 72 hours at 70°. Ultra-thin slices for TEM were cut with an ultra-microtome (OM 43, C.Reichert, Austria) and diamond knives. After-staining of the slices was achieved by 0.2% lead citrate (Fluka, USA) and 0.2% uranyl acetate. The former incubated for 7 minutes, the latter for 20, followed again by washing of the samples after each step. For image acquisition, a Zeiss EM 10 (Zeiss, Germany) was used at 60 kV. Reflectance Measurements One dried specimen of D. magnus was mounted in an integrating sphere (15 cm diameter, Gamma Scientific, USA), normal to the incident laser beam. A supercontinuum source, i.e. a white light laser (NKT SuperK, NKT Photonics, Denmark) equipped with a Varia filter box allowed color tuning. The diffuse reflection of the bug is recorded via a custom photodiode facing the backside of the sample B io lo gy O pe n • A dv an ce a rt ic le by guest on November 2, 2017 http://bio.biologists.org/ Downloaded from mount. The laser power was kept well below 1 mW to avoid optical and thermal damage. An overview of the setup used is given in the supplementary material (Fig. 4S) Chemical analysis of cuticle waxes Part I: GC-MS (Gas chromatography with mass spectrometry) We used chloroform (Serva, Germany) to dissolve the topmost layers of cuticle waxes on one specimen of Dysodius lunatus as well as D. magnus, where we expected to find surface active components that interact with water. TEM images in the supplement material show that this attempt was successful (Fig. S3 A and B). The samples’ abdomen were carefully rinsed three to four times with 5 ml of chloroform, in order to dissolve surface components. Samples were then dried in a heating module (Reacti-Therm Heating Modul, PIERCE, USA), down to a volume of 50 μl, before derivatization with 10 μl pyridine (Sigma-Aldrich, Germany) and 10 μl N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA, Sigma-Aldrich, Germany) was done. Samples then were again dried at 70° for 30 minutes before being re-dissolved in 50 μl chloroform and conveyed into auto-sampling veils. For D. magnus (first tested samples), this was the sole procedure. For gas chromatography, an Agilent 7890 A GC system (Agilent Technologies, USA) with mass spectrometer detector was used. A sample volume of 1 μl was injected via cool-on-column method into a DB1 column with a length of 30 m, 320 mm diameter and a stationary phase thickness of 100 nm. Carrier gas was helium, which was introduced for the first 5 minutes at 50 kP pressure before ascending to a final pressure of 150 kP in 3 kP/min steps. 150 kP were then held for 90 minutes. The heating protocol was 50 °C at injection for two minutes, raised by 40 °C min till 110 °C, held for two minutes at 110 °C, raised by 3° min till 320 °C and then held for 50 min at this final temperature. Chemical analysis of cuticle waxes Part II: HPLC-MS (High-performance liquid chromatography with mass spectrometry) For HPLC-MS analysis the chloroform solution used to dissolve the topmost layers of cuticle waxes of two dried specimen of Dysodius magnus was brought to dryness with a gentle stream of nitrogen in a sampling vial and 0.5 mL acetonitrile was added to dissolve low molecular compounds. The resulting solution was used for injection. An Agilent 1100 series HPLC system (Waldbronn, Germany) equipped with a degasser, a quaternary pump and an autosampler was applied. Analytes were separated on a Kinetex C18 column (50 mm x 4.6 mm, particle size 2.6 μm; Phenomenex, Aschaffenburg, Germany) using a water/acetonitrile gradient. Starting conditions were set to 75% solvent A (water with 0.1% formic acid) and 25% solvent B (acetonitrile with 0.1% formic acid). Mobile phase B was linearly increased to 90% within 19 minutes, which was then kept constant from minute 19 to 23. The gradient was changed back to starting conditions for re-equilibration; these conditions were held for 5 minutes. Flow rate was set to 1.0 mL min-1, temperature of the column heater was fixed at 30 °C B io lo gy O pe n • A dv an ce a rt ic le by guest on November 2, 2017 http://bio.biologists.org/ Downloaded from and an injection volume of 20 μL was selected. MS detection was carried out using an Agilent 6520 QTOF equipped with electrospray ionization (ESI) and operated in the positive mode. MS parameters were as follows: drying gas flow 10.5 L min-1, drying gas temperature 350 °C, nebulizer pressure 50 psi, capillary voltage 3750 V, fragmentor voltage 175 V. The HPLC eluent flow was split (0.33 mL min1) before entering the ESI source. Laser-generated structures on PET substrates The experiments were performed on 50 μm thick flat, biaxially stretched Polyethylene terephthalate (PET) foils (Dupont, Mylar ). They were used without any pre-treatment. Nap structures (small, pillar like microstructures, which cover a surface) on PET foils were produced using the KrF excimer laser LPX 300 (Lambda Physik, Germany), operated at a wavelength of l = 248 nm and a pulse length of about 20 ns. The repetition rate was set at 10 Hz. The naps were produced with typically 20 pulses using the full beam profile of the laser of a few cm at a distance of 25 cm from the laser output. The resulting laser fluence is 150 mJ/cm with some deviations due to intensity variations over the beam profile. Test fluid for the PET foils was distilled water, dyed with 0.5% (w/w) Ponceau S (Carl-Roth GmbH + Co. Kg., Germany). Contact angles were measured with a droplet volume of 5 μl, using a custom made contact angle measurement setup. Results Wetting of flatbugs As shown in Figure 1, Dysodius magnus changes its colour when wetted to adapt to the colour change of bark dependent on its humidity. About 5 to 10 μl of water are sufficient to completely wet the dorsal surface of a bug. For Dysodius lunatus an identical behaviour could be observed (not shown). A detailed view on the time course of wetting is depicted in Figure 2 and is shown in the supplementary video (Vid.S1). Within 10 seconds, a drop of 5μl water spreads virtually all over the dorsal abdomen of the animal. The local velocities of the liquid front show a very highly variability. One can measure from 0.1 up to about 2 mm s dependent on the exact location where measured. As evident from Figure 2 and from the corresponding video, the water spreads onto the surface of the connexival plates and the glabrous areas with rather low local velocities the water enters capillary channels in between the connexival and glabrous areas (indicated by arrows in Figure 2). In the open channels, the transport is obviously much faster than the spreading on the surface. However, although faster transport can be achieved in the channel, it appears for the water to be energetically favourable to spread on the surface. As can be seen, the water occasionally leaves the channels and spreads B io lo gy O pe n • A dv an ce a rt ic le by guest on November 2, 2017 http://bio.biologists.org/ Downloaded from (slowly) on the surface at the corresponding positions resulting in a thin water film covering the whole dorsal surface of the animal. This goes hand in hand with a change of the colour. All these observations were identically made with Dysodius lunatus. Both animals show the same behaviour and the same morphology and chemistry (see below). Thus for shortness and simplicity only Dysodius magnus will be treated in detail in the following sections. As already described, a wetting of the surface be it bug or bark will result in the change of optical properties (as shown in Fig 1 C-F). Measurements of the diffuse reflection of both surfaces verify this observation. Figure 3A shows that, especially for wavelengths between 500 and 725 nm, a change in reflectance can be observed. From Figure 3B one can see that the relative diffuse reflection reduction (= scattering reduction, when comparing dry = bright, versus wet = darker) of bark and bug is also following a very similar trend throughout the investigated spectra. Surface morphology To understand how the wetting can be achieved, the surface morphology of the flat bugs was investigated by scanning electron microscopy (SEM). A typical scan of the dorsal surface of Dysodius lunatus is shown as overview in Figure 4A and in detail in Figure 4B. Clearly, a micro structure covers the connexival plates as well as the glabrous areas. Furthermore, the sutures separating segmental borders are visible. These are here functioning like capillary channels. At higher magnifications one can see a rough surface, i.e. that the microstructures are small naps. These naps have a diameter of 1.96 ± 0.37 μm and an average seperation of 4.88 ± 1.38 μm (n=100). The height of the naps can be measured to be in the range of 2.5 to 4μm. In between the naps occasionally circular structures can be seen such as the one marked by an arrow in Figure 2B. These structures have typical diameters of 21 ± 3 μm and are spaced by about 90 ± 40 μm (n=100). The naps were assumed to be made of cuticular waxes and it was tempting to speculate that these wax-structures are involved in the observed wetting behaviour, since they represent the surface coming in contact with water (even though waxes usually show a hydrophobic behaviour, an explanation for this was found later and is given in detail in the following chapters of this work). In order to test this, two bugs, which initially exhibited clear wetting (spreading of the water), were treated with hot 10% KOH-solution for 60 s, to remove wax by means of hydrolysis leaving only the underlying chitin surface intact. After this treatment, the bugs were not wettable anymore but became water repellent (showing contact angles >90°). To give an estimation of the contact angle situation found on these bugs, refer to supplemental Figure S4. The surface morphology of these de-waxed animals is shown in Figure 4C and D. Clearly the naps are removed. The circular structures, which are under native conditions covered with wax-naps at high B io lo gy O pe n • A dv an ce a rt ic le by guest on November 2, 2017 http://bio.biologists.org/ Downloaded from density (Figure 3B, indicated by the arrow) are again visible (indicated in Figure 4C by arrows) but the wax is removed. In the higher magnification (Figure 4D) clearly the naps are gone. It seems that the orifice of a secretion channel can be seen which exhibits a sinus-shape. We assume at this stage, that these circular structures with the sinus-shaped openings in the centre are the orifices of wax-glands, which produce the nap-shaped cuticular waxes. Taken together we see that by KOH-treatment the waxy naps can be removed which concomitantly results in loss of the wetting capability of the cuticle. Thus, the waxy naps are vital for the wetting. Wax analysis In order to be able to get a first idea of the physical and chemical properties of the cuticular wax, especially the outer layers which directly come into contact with water, the wax had to be removed gently from the cuticle without dissolving deeper layers or substances from within the bug’s body. This was done by washing the animals with chloroform. Transmission electron micrographs of so treated animals show that in fact only the outer layers of the cuticular wax was removed but basal wax was still present on the bug’s body (shown in supplement, Fig. S3 A and B). Notable at this point is that the removal of just only these outermost waxes was enough to render the bugs hydrophobic. The chloroform containing the dissolved waxes was dripped onto glass cover slips and was then allowed to evaporate, resulting in a thin white film sustaining on the glass. SEM revealed that the surface of this wax-film was virtually flat (not shown). When applying water onto the wax film, a surprisingly low contact angle of 43.9° ± 4,7° (n=15) was measured. A typical example of such a contact angle measurement is shown in Figure 5A. To determine what molecules are present in the bugs` waxes, they were dissolved and subjected to gas chromatography with subsequent mass spectrometry (GC-MS). This first analysis yielded some main substance classes that could be identified. Besides monodiand triglycerides with chain-length ranged from C 12 to C 29, many free fatty acids (monoand di-carboxylic acids) were detectable. We could not determine whether these are really fatty acids or soaps, i.e. salts of the fatty acid. An exact quantification of the amounts of the different substances was not possible due to the small amounts of samples available. Furthermore, high-performance liquid chromatography mass spectrometry (HPLC-MS) was performed in order to verify the GC-MS findings. Derivatization was not required in case of HPLC-MS. Detection was carried out in positive mode to be able to detect the glycerides. Glycerides were observed almost exclusively as sodiated [M+Na] pseudomolecular ions. As expected fatty acids were detected as protonated [M+H] as well as (in some cases more abundant) sodiated [M+Na] species. Saturated and unsaturated fatty acids (monoand di-carboxylic acids) were detected. For some compounds, a characteristic water loss due to in-source fragmentation was observed indicating that hydroxylated fatty acids are also present in the sample. Overall, the B io lo gy O pe n • A dv an ce a rt ic le by guest on November 2, 2017 http://bio.biologists.org/ Downloaded from complementary HPLC-MS measurements confirmed the previous GC-MS findings, but in total, a smaller number of compounds was detected. Additionally to the fatty acids and glycerides, significant amounts of erucamide (cis-13-Docosenoamide) were identified in HPLC-MS. This substance is well known to be an often-found contamination in analytical chemistry. We are aware of this and therefore all treatment of the substances was done with glass equipment only avoiding any labplastics. All glassware was initially cleaned with chloroform (analytical grade) in order to avoid contaminations. Still, fatty acids, soaps and the dubious erucamide, had at least one commonality: All these substances have polar and non-polar sides (amphiphilic character) and make therefore good candidates to act as mediators between hydrophobic insect wax on the one side, and polar water molecules on the other. Different researchers already showed that aliphatic molecules can be arranged perpendicularly orientated (with regard to their longitudinal axis) on cuticular waxes of plants (Domínguez, Heredia-Guerrero, & Heredia, 2011; Graça, 2002). This gives a high possibility for amphiphilic substances to do the same on insect cuticle waxes. To test whether the erucamide and/or the free fatty acids or soaps are able to render insects wax hydrophobic, bees’ wax was modified. As can be seen in Figure 5B, bees’ wax is hydrophobic, exhibiting a contact angle of about 108°. If simply melting the bees’ wax and applying soaps, fatty acids or erucamide at small amounts to the liquefied material, the contact angle could only be changed marginally by a few degrees (not shown). Only when adding strong detergents like octoxinol9, could the contact angle be lowered significantly. However, when an initial film of bees’ wax was made and then soaps or erucamide was applied on top of the film as solution in chloroform or alcohol, a significant effect on the contact angle could be observed. Figure 5C shows the effect of erucamid-application on a pre-established wax film. Here saturated erucamid-solution was applied and was allowed to dry on the wax. Then the surface was rinsed with water. Clearly the contact angle drops to very hydrophilic values of about 30°. However, in the SEM even the modified wax appears to be flat and no spontaneous nap-formation could be observed (not shown). An exact quantification of the contents was not possible due to the layered structure resulting from the production of the modified wax layer. Nevertheless, it appears clear that the found substances from GC-MS can reduce the contact angle to values observed on natural flatbug’s wax. Morphology of wax glands To understand if a layered structure of the cuticular waxes can be explained, morphological analyses of the presumable wax glands, i.e. of the circular structures observed in the SEM were performed. Freeze fracture through one of the circular structures, which are presumably wax glands, can be seen in Figure 6 A and B. In between the superficial microstructures (Ms) a rim with a filamentous tissue in the middle can be seen. This becomes rather obvious when looking at higher magnification in Figure B io lo gy O pe n • A dv an ce a rt ic le by guest on November 2, 2017 http://bio.biologists.org/ Downloaded from 6B. Additionally TEM-images of an ultra-thin section through the cuticula in the vicinity of a gland were recorded. One typical image is shown in Figure 6C where the typical layered structure of the cuticle can be seen. In the exocuticula small channels can be seen (arrows). Additional TEM-images of the cuticle are shown in the supplementary material (Fig. S1 and S2). Taken together the freeze fracture and the TEM-images, we propose a morphology of the wax glands as depicted in Figure 6D as will be discussed below. Mimicking the wetting behaviour on polymer foils In order to see if the wetting (water spreading) behaviour which can be seen on native Dysodius lunatus can be fully explained by the findings so far, an attempt was made to mimic the effect. For this purpose, naps on PET-foils were produced by irradiation with nanosecond UV laser pulses. The spontaneous formation of the quasi-periodic naps occurs self-organized due to the release of the biaxial stress-fields and crazing in the irradiated but not ablated surface (Arenholz et al., 1991). Figure 7 shows the surface morphology of PET-foil as seen in the SEM. While the untreated PET is virtually flat, areas which were irradiated with the laser are covered by naps. These naps have a diameter of about 2 μm. The PET-naps are slightly denser (average distance 3.5 μm) than the naps observed on Dysodius but on the other hand they are not that high (about 1μm) resulting in similar surface enlargement when compared to the natural archetype (see discussion section). The native PET-foil is slightly hydrophobic. In order to mimic the wax material with an intrinsic contact angle (contact angle on the flat surface) of about 40°, we treated the foils with a sputter coater. Small amounts of gold were deposited in an argon-plasma until a contact angle of 40° was reached (same coating procedure as stated for SEM: 180 s at 1 kV, resulting in a layer thickness of approx. 20 nm). The wetting behaviour of such a foils is shown in Figure 8. The glossy part (upper, left and right margins) of the foil shown there is not irradiated, corresponding to the flat foil shown in Figure 7A. However, the opaque part in the lower centre of Figure 8 is structured by laser processing as shown in Figure 7B. Initially a drop of dyed water was placed on the unstructured PET. A stable droplet is formed which stays at a contact angle of about 40°. On the contrary, if water is applied onto the laser-structured part, the water immediately begins to spread and to form a film of water. The apparent contact angle is below 10°; thus the surface is superhydrophilic. At the water front, a margin with slightly different colour can be observed. A video of this experiment can be found in the supplementary material (Vid. S2). B io lo gy O pe n • A dv an ce a rt ic le by guest on November 2, 2017 http://bio.biologists.org/ Downloaded from Discussion In the present study we found that the excellent wetting properties of Dysodius lunatus, which are a prerequisite for its self-adaptive camouflage are due to a micro structured wax surface. The waxes are hydrophilic due to the superficial addition of amphiphilic molecules like soaps, fatty acids, monoglycerides or erucamide. It was shown, that superficial addition of these substances to typical insects’ waxes can render these waxes hydrophilic. Unfortunately, due to the very limited access to the native wax, we cannot give a quantitative analysis of the components and we cannot definitely show that some fatty acid anions appear as soaps. As shown in the supplementary material, only the superficial layers of the cuticular wax can be removed by chloroform-treatment while removal of the basal wax layers requires hot KOH. Thus the chemical composition is not uniform. The morphology of the presumable wax glands found at Dysodius lunatus fits to this observation: In the vicinity of the presumable wax glands, small channels through the exocuticle can be found, as they are also described from Lockey (1985) to be the insects main body-wax secretion instrument. These are assumed to support the stable but hydrophobic basal wax. The complexly built wax glands with the sinus-shaped secretion opening on the other hand, appear to supply wax onto this basal wax layers. Bloomquist & Jackson (1979) found that cuticle waxes of insects usually are compounds of longer chain hydrocarbons, which obviously need to be produced somewhere in the insects body. Foldi (1981) showed that insect wax glands exhibit so called support-cell structures often directly attached to channel systems of glands. As shown in the supplementary material, we found such support cells for Dysodius glands too, and cell organelles like endoplasmic reticula, oenocytes and lamellar bodies were observed in high quantities. These are all structures, which usually are directly involved in a cells fat/lipid household, as shown by different authors (Lockey, 1988; Schmitz & Müller, 1991). Additionally somehow these glands manage to form the nap-structure when secreting the wax. This mechanism is not fully explainable so far, but it has been shown that similar complex glands are able to produce highly sophisticated wax structures for different insects, as also shown by Foldi. It is worth noting, that the nap structure is not self-assembled in any of our experiments. The removed and readsorbed native waxes as well as the artificially modified bees’ waxes never spontaneously formed naps similar to those observed at the surface of the flatbugs under investigation. How do the hydrophilic waxes with contact angles of about 40° in combination with the observed naps yield a superhydrophilic surface? Observing the natural archetype or the structured PET-foil, one can see that in fact a drop of water applied to these surfaces sits on a substrate while the liquid appears to penetrate the interconnected grooves in between the naps. Thus the droplet faces a patchwork of solid and liquid. This is exactly the scenario of the so called Cassie-Baxter impregnating B io lo gy O pe n • A dv an ce a rt ic le by guest on November 2, 2017 http://bio.biologists.org/ Downloaded from wetting or also called hemi wicking. This wetting has to be distinguished from the pure Wenzel wetting (Wenzel, 1936). In the latter case, the solid outside of the triple line is dry, whereas in the Cassie-Baxter impregnating situation the grooves, acting as capillaries, are wetted. The apparent contact angle θof the drop sitting atop of the structure results to be cos(θ) = 1 − fs + fs ⋅ cos(θ) (1) where θ is the contact angle of the unstructured solid material and fs is the relative fraction of the solid underneath the droplet. As explained in detail in Bormashenko (2013), Cassie-Baxter impregnating wetting is possible if and only if the contact angle of the unstructured material θ fulfils cos(θ) > 1 − fs r − fs (2) with r denoting the Wenzel-roughness. This is the ratio of the total surface area and the projected area, i.e. the factor of surface enlargement due to the structuring. If a regular array of cylindrical naps of radius R spaced by a distance d occurs, the ratio fs follows to be