Passive water collection with the integument: mechanisms and their biomimetic potential

There are numerous studies investigating the adaptations in nature to limited resources. Some reptiles, amphibians, arthropods, birds and even mammals have been found to survive restrictions on water supply by using their body surface to collect water from various sources (Louw, 1972; Rijke, 1972; Lillywhite and Licht, 1974; Gans et al., 1982; Lillywhite and Stein, 1987; Sherbrooke, 1990; Cardwell, 2006; Tracy et al., 2011). Collecting water in arid environments might appear to be contradictory at first, but nevertheless many such areas are known to provide water sources. For example, dew is regularly found in most deserts early in the morning, originating from significant day–night temperature differences (Louw, 1972). The Namib desert is famous for its fog (Shanyengana et al., 2002); depending on the location, fog events occur 40–200 days per year (Seely, 1979; Shanyengana et al., 2002). Furthermore, there are also infrequent rain falls that must be considered (Comanns et al., 2016a).

Maintaining a water balance in xeric habitats (see Glossary) often requires significant reduction of cutaneous water loss. Reptiles commonly have an almost water-proof skin owing to integumental lipids, amongst other components (Hadley, 1989). In some snakes, for example, the chemical removal of lipids has been shown to increase transepidermal water permeation by a factor of 35–175 (Burken et al., 1985).

Amphibians, by contrast, typically lack a significant resistance to cutaneous water loss (Bentley and Schmidt-Nielsen, 1966; Shoemaker and Nagy, 1977; Toledo and Jared, 1993; Maderson et al., 1998; Lillywhite, 2006). The low resistance to cutaneous water loss can be seen as a cost for transcutaneous water uptake and respiration requiring a moist skin (Chew, 1961). In some arboreal hylid frogs, however, a significant reduction of evaporative water loss is provided by a cutaneous secretion of lipids (Shoemaker et al., 1972; Blaylock et al., 1976; McClanahan et al., 1978; Toledo and Jared, 1993; Amey and Grigg, 1995; Tracy et al., 2011). Similarly, arthropods achieve reduced water loss by protective lipid and wax layers of the cuticle (Beament, 1964; Edney, 1977; Hadley, 1989, 1991).

Despite such reduction of water loss, additional water demand has been found in many species. In desert reptiles, a number of species require additional water collection, although they often rely to a large degree on the water content of their diet to cover their water demand (Bentley and Blumer, 1962; Nagy, 1987; Maderson et al., 1998; Lillywhite, 2006). For some desert lizards, this necessity results from their mainly myrmecophagous diet (see Glossary), which demands special mechanisms to support excretion of high concentrations of electrolytes (Bradshaw and Shoemaker, 1967; Withers and Dickman, 1995). Lately, some studies on diet-based water demand discuss the uptake of free water as a general strategy for a number of desert reptiles, in particular carnivorous reptiles, which often show a more or less distinct need for uptake of free water in order to obtain a net gain of water (Wright et al., 2013; Lillywhite, 2017). In many amphibians, water collection is required to replenish the water loss from mucus secretion, in particular for thermoregulation and counteracting dehydration of the epidermis (see Glossary) at higher temperatures (Lillywhite and Licht, 1975; Lillywhite et al., 1998). Some toads prevent dehydration of the skin directly by collecting water from their environment (Lillywhite and Licht, 1974). In general, the need for water collection is often for rehydration, but it is also needed for water adsorption, which prevents dehydration of the skin (Lillywhite and Licht, 1974). Furthermore, collected water serves the thermoregulation in elephants and wharf roaches (Hoese, 1981; Lillywhite and Stein, 1987), is transported by adult sandgrouse from water sources to hydrate the young (Cade and MacLean, 1967), and yields reduced reflectivity for camouflage in flat bugs (Silberglied and Aiello, 1980; Hischen et al., 2017).

Passive water collection takes place from sources of the animals' environment; for example, from water puddles, infrequent precipitation, moist substrate, air humidity and dew or fog (Table 1). In cases of water collection from infrequent precipitation or fog, a number of species, such as snakes (Louw, 1972; Robinson and Hughes, 1978; Andrade and Abe, 2000; Cardwell, 2006; Repp and Schuett, 2008; Glaudas, 2009), tortoises (Auffenberg, 1963), lizards (Schwenk and Greene, 1987; Sherbrooke, 1990; Veselý and Modrý, 2002) and beetles (Hamilton and Seely, 1976; Nørgaard and Dacke, 2010), expose themselves to the water source by using a stereotypic behaviour (Table 1), while the actual water collection remains passive (Joel et al., 2017).

In this Review, I focus on mechanisms to passively collect water with the integument (see Glossary) of animals. As the collection and handling of water has also been of particular interest regarding their potential for nature-inspired improvement of technical applications, a comprehensive overview of analyzed animal species can catalyze further studies and innovations. Several water-collection mechanisms have been adapted for technical applications using a biomimetic approach, and examples are subsequently discussed.

The ability to passively collect water with the body surface can be found in a broad variety of genera. Mechanisms have evolved to access various sources, such as rain, fog, water vapour in air and moisture from the animal's surrounding substrate. The collection of water is followed by a transcutaneous uptake (amphibians), direct drinking (snakes), spreading (see Glossary) over the body surface (toads, elephants, flat bugs and beetles), transport to other parts of the body surface (lizards, tortoises and wharf roaches) or storage in the plumage (sandgrouse) (Table 1). In the end, water is incorporated in some way in all described cases, but not in flat bugs and elephants.

These collection processes and subsequent handling of collected water involve different, specific chemical or structural adaptations of the body surface. Based on these general considerations, six basic mechanisms have been identified as being involved in passive water collection (Fig. 1): (1) increased surface wettability; (2) increased spreading area; (3) transport of water over relatively large distances; (4) accumulation and storage of collected water; (5) facilitating condensation; and (6) utilization of gravity (see Glossary for the terms described).

In more detail, an increased surface wettability of the integument (1), i.e. smaller contact angles (see Glossary), results from either chemical properties and/or microstructures (Fig. 2), such as different kinds of pillar and hexagonal dimples (Bormashenko, 2010; Comanns et al., 2014). The spreading area (2) can be increased by similar microstructures, which allow subsequent drinking of water (Bico et al., 2002; Chandra and Yang, 2011). Further structures, i.e. grooves or channels, are required to transport water (3) over larger distances or precisely in one direction (Berthier and Silberzan, 2010; Comanns et al., 2015). It is important to note the difference between wetting effects and capillarity. The former is the interaction of liquids with (structured) surfaces, whereas the latter is the force acting on the liquid within channel structures (Berthier and Silberzan, 2010). In the examples given in Table 1, transportation distances are typically in the range of few millimetres to several centimetres. Besides these structure-based mechanisms to collect and/or transport water, hairy surface appendages such as feathers facilitate a mechanism for passive accumulation and storage of water (4) (Joubert and MacLean, 1973). Condensation (5) is facilitated by changing the microhabitat to establish a thermal gradient to ambient conditions, i.e. making the body cooler than the surrounding (Tracy et al., 2011). And finally, there are several instances of gravity (6) being utilized by adopting a particular body posture to channel the water to the mouth for drinking (Auffenberg, 1963). In some cases, combinations of these basic mechanisms are found.

The ability to passively collect water with the integument appears to require hydrophilic surfaces (see Glossary), independent of species (Table 1). Hydrophilicity means an increased wettability of the integument, which results from chemical properties and often exists in combination with certain microstructures (Box 1).

In desert lizards, a number of species within the genera Phrynosoma, Phrynocephalus, Trapelus, Moloch, Pogona, Cordylus and Uromastix are known to passively collect water from their environment (Ditmars, 1933; Pianka and Pianka, 1970; Gans et al., 1982; Fitzgerald, 1983; Schwenk and Greene, 1987; Sherbrooke, 1990, 1993, 2004; Withers, 1993; Peterson, 1998; Veselý and Modrý, 2002; Yenmiş et al., 2015). To rationalize the nomenclature for the different sources of water acquisition, they have been termed ‘moisture-harvesting lizards’ (Comanns et al., 2011). The keratinous skin of these moisture-harvesting lizards is hydrophilic and exhibits hexagonal microstructures on the scale surfaces (Fig. 1A) (Comanns et al., 2011). Typical dimensions of these structures are diameters of 10–30 µm and depths of 1–5 µm (Comanns et al., 2011). Once in contact with tiny amounts of water, the microstructures get filled with water and render the skin surface ‘superhydrophilic’ (Fig. 3) (Comanns et al., 2016a). Such water-filling has been considered as pre-wetting, which is likely a preparation for faster uptake of water from more-efficient sources, such as damp sand (Comanns et al., 2016a).

Chemical modifications of the integument are found in secretions of hylid frogs (Toledo and Jared, 1993; Tracy et al., 2011) or ‘hydrophilizing’ components in waxes of flat bugs (Hischen et al., 2017). Some arboreal hylid frogs, such as Phyllomedusa sauvagii or Litoria caerulea, significantly reduce evaporative water loss by a cutaneous secretion of lipids (Shoemaker et al., 1972; Blaylock et al., 1976; McClanahan et al., 1978; Toledo and Jared, 1993; Amey and Grigg, 1995; Tracy et al., 2011). After secretion, the secreted fluid is spread over the body surface by wiping behaviour to obtain a protective coating (Toledo and Jared, 1993; Barbeau and Lillywhite, 2005). The lipid secretions have been considered as having hygroscopic (see Glossary) properties and resulting in a wettable skin surface (Toledo and Jared, 1993). Some toads modify their skin-wetting properties in a comparable way. For example, in their parotoid glands, Anaxyrus sp. toads produce a venomous secretion. Although this secretion primarily serves as defensive liquid, it contains inter alia glycosaminoglycans, which are hygroscopic substances that again are considered to play a role in water balance (Toledo et al., 1992). In contrast, many sun-basking species, such as the American bullfrog Rana catesbeiana, appear to use mucus secretions for maintaining the water balance of the epidermis rather than for (direct) collection of water (Lillywhite and Licht, 1975).

The South American flat bug species Dysodius lunatus and Dysodius magnus collect water for camouflage, in which they reduce their surface reflectivity, rather than rehydration (Silberglied and Aiello, 1980; Reiswich, 2013; Hischen et al., 2017). Immediate spreading of water droplets is facilitated by chemical and structural properties of the integument. Unlike most other insects, the cuticle of these bugs is covered by a hydrophilic wax layer imparted by the amphiphilic component erucamide (Hischen et al., 2017).

Surface structures do not only affect the surface wetting properties in terms of contact angle, but wetting phenomena in general. Wetting phenomena include, for example, water repellence and self-cleaning (Barthlott and Neinhuis, 1997), pinning (see Glossary) of the advancing water front at surface microstructures (Bico et al., 2002; Lai et al., 2012; Srinivasan et al., 2014) or penetration into the structured surface area (Bico et al., 2002; Chandra and Yang, 2011). Water penetration into surface structures with moderate chemical hydrophilicity can cover greater areas than are possible by spreading on a smooth surface (Quéré, 2008). Such water penetration has been found, for example, in toads (Lillywhite and Licht, 1974), elephants (Lillywhite and Stein, 1987) and flat bugs (Hischen et al., 2017).

Some hylid toads of the genus Anaxyrus (e.g. A. boreas, A. woodhousii, A. punctatus) collect water from moist substrates (McClanahan and Baldwin, 1969; Fair, 1970; Lillywhite and Licht, 1974; Toledo and Jared, 1993). Their granular skin contains numerous grooves in which capillary forces occur that suck water from the substrate. Accumulated water is transported even to the dorsal body parts, most likely to maximize effective wetting of the skin to enlarge the area for water uptake (amphibians typically absorb water through the skin) and to prevent dehydration of the epidermis (Lillywhite and Licht, 1974). Comparable surface structures for water penetration have been described for elephants (Loxodonta africana, Elephas maximus); collecting and spreading of water takes place in numerous small grooves of the granular skin (Lillywhite and Stein, 1987). In the flat bug species D. lunatus and D. magnus, pillar-like surface structures support the hydrophilic wetting properties and spreading of water (Fig. 4A). Spreading on the surface is slower than within intersegmental channels, but energetically favourable (Fig. 1B). Hence, a passive spreading of water over the body surface is enabled (Hischen et al., 2017).

The penetration of liquids into surface microstructures [also known as hemiwicking (Bico et al., 2002; Quéré, 2008)] has been described as an interplay between pinning and capillary action (Blow et al., 2009). Microstructures within capillary channels can increase capillarity by hemiwicking (Fig. 4B) (Bico et al., 2002). This has been found, for example, as protrusions in skin channels of Australian thorny devils (Moloch horridus) (Comanns et al., 2017). In cases in which such microstructures are asymmetric, for example triangular pillars, they can achieve directional penetration of a liquid (Fig. 4C) (Blow et al., 2009). The dynamics of hemiwicking have been described using different approaches, and all approaches contribute to the geometric dimensions of the surface structures one way or another (Bico et al., 2002; Quéré, 2008; Chandra and Yang, 2011).

The spreading of collected water can be interpreted as a straightforward way to distribute water away from a collection site. However, passive transportation over greater distances requires capillary action. Capillary liquid transport can take place in small cavities, such as tubes, ridges or channels, where the capillary forces dominate other major contributing forces such as viscosity, friction or gravitational force (Berthier and Silberzan, 2010). Corresponding surface structures can be found in the granular skin of some toads and elephants (Lillywhite and Licht, 1974; Lillywhite and Stein, 1987), surface channels of moisture-harvesting lizards (Gans et al., 1982; Withers, 1993; Veselý and Modrý, 2002; Sherbrooke et al., 2007; Comanns et al., 2015), flat bugs (Hischen et al., 2017) and wharf roaches (Hoese, 1981; Horiguchi et al., 2007; Ishii et al., 2013), and cavities between feather structures of sandgrouse (Rijke, 1972; Joubert and MacLean, 1973; Rijke and Jesser, 2011). In the case of moisture-harvesting lizards, transportation in channels avoids wetting much of the body surface and hence losing volume by evaporation from a larger area (Comanns et al., 2011; Yenmiş et al., 2015).

Moisture-harvesting lizards possess a skin channel network between the scales that extends over the entire body surface (Fig. 1C,D). It allows collection and transport of collected water by capillarity. The channels have a width of 100–300 µm in the basal part (i.e. scale hinges), and narrower openings towards the surface, typically <50 µm (Fig. 5A,B) (Withers, 1993; Sherbrooke et al., 2007). In vertical orientation, these dimensions reflect the measured transport distance of 9.9 cm where capillary forces become equal to gravity in capillaries of about 220 µm in width (Withers, 1993).

Theoretically, the channels must be filled for drinking to occur. However, two structural modifications have been found that most likely enable the lizards to drink even smaller amounts of water than are sufficient for complete filling of the channels (Comanns et al., 2015). First, in the pronounced hierarchical channel structure that has been described particularly for the Australian thorny devil, large cavities can quickly absorb water, whereas sub-capillary structures yield an extension of the transport distance by about 39% (Fig. 5C) (Comanns et al., 2017). Water transportation in these skin channels has been modelled using an adapted dynamics function that closely reflects the channel morphology, hence the combination of water penetration and capillary transport (Chandra and Yang, 2011; Comanns et al., 2017). Second, a directional water transport towards the mouth has been found in two desert lizard species, Phrynosoma cornutum (Comanns et al., 2015) and Phrynocephalus horvathi (Yenmiş et al., 2015), and results from a combination of asymmetric channel geometry and specific network structure: a narrowing of single channels between two neighboured scales yields a local directional water flow, whereas specific interconnections preclude the inhibition of water transport towards the mouth, but not in the rearward direction (Fig. 5D) (Comanns et al., 2015).

A number of crustacean wharf roaches, such as Ligia exotica and Ligia oceanica, passively collect water from wet surfaces of their coastal habitat (Hoese, 1981; Horiguchi et al., 2007). Water is then transported in open structures of the cuticle of the legs, which act as capillaries. Further examination has revealed more detail: hair- and paddle-like microstructures on two neighboured legs (i.e. pereiopods VI and VII) collect and transport the adhered water; the water is then transported further along the swimming limbs (pleopods) and to the hindgut, near the anus, for uptake by absorption (Horiguchi et al., 2007; Ishii et al., 2013). Collected water also establishes a water film on the integument and evaporation is regularly used for thermoregulation (Hoese, 1981).

Various kinds of surface structures can form cavities, in which collected water can accumulate. Such accumulation can be as simple as entrapping rain between the body coils of some snakes from the genera Crotalus, Bothrops and Bitis (Louw, 1972; Robinson and Hughes, 1978; Andrade and Abe, 2000; Cardwell, 2006; Repp and Schuett, 2008; Glaudas, 2009) (Table 1). The collected water is then licked off the body surface. Accumulation of water can also take place in cavities of integumental surface structures of other animals, which has been determined as integumental water holding capacity (Table 2). The highest values have been found for the lizard M. horridus (9.19 mg cm⁻²). Interestingly, the water holding capacity of elephants is much lower (L. africana: 1.27 mg cm⁻²; E. maximus: 0.81 mg cm⁻²), although the granular skin of elephants has a much coarser structure, with depths up to 5 mm (Lillywhite and Stein, 1987) (Table 2). The difference potentially results from stronger hierarchical skin structures or a higher density of skin channels in M. horridus (Comanns et al., 2017). Furthermore, it is worth noting that the Texas horned lizard P. cornutum (5.9 mg cm⁻²) has a similar integumental water holding capacity as the aquatic file snake Acrochordus granulatus (5.39 mg cm⁻²; Table 2).

Many sandgrouse species (e.g. Pterocles alchata, Pterocles bicinctus, Pterocles namaqua) have wettable breast feathers to accumulate and store water (Fig. 1E) (Meade-Waldo, 1896; Cade and MacLean, 1967; Joubert and MacLean, 1973). This is in contrast to other birds, where quasi-hierarchical structures of feathers enable water repellence and protection against water penetration (Rijke and Jesser, 2011; Srinivasan et al., 2014). Sandgrouse feed on dry seeds and inhabit mostly arid regions, which demands daily water uptake, hence drinking. Water is carried in the ventral plumage from ponds up to 80 km away in volumes up to 30 ml per flight, i.e. 5–15% body mass, to water the young (Maclean, 1968; de Juana, 1997). In fibrous structures, such as feathers, water is not held by a single (structured) surface. Future studies may therefore determine the values in relation to the volume of the feathers. The volume of sandgrouse breast feathers is created by a different structure compared with that of other pennaceous feathers (see Glossary): the barbules (radii) are twisted into each other without hooks. In wet conditions, the barbules unfold independently, enclosing the accumulated water volume (Maclean, 1968; Rijke, 1972). Besides suggested reversible physicochemical changes in the feather keratin (Rijke, 1972), the reason for that structural change remains unclear and forms part of current studies (Heiko Schmied, personal communication).

Condensation of water vapour into liquid droplets on the animal has been observed for some hylid frogs, such as L. caerulea and P. sauvagii (Fig. 1F) (Toledo and Jared, 1993; Tracy et al., 2011). The required thermal gradient is achieved by the ectothermic properties and temporal changing of the microhabitat. These frogs need not only to minimize water loss, but also to take up water. Some species of the above-mentioned genera manage this by passively collecting water present as air humidity (Shoemaker et al., 1972; Toledo and Jared, 1993; Tracy et al., 2011). As ectotherms, the frogs cool down in the open but, when they enter warm and humid tree holes, the temperature difference leads to condensation on the colder body surfaces of the frog. Determined condensation rates have been up to 3 mg cm⁻² body surface at a temperature differential of 15°C and duration of 20 min; such a rate is higher than water loss by evaporation (Tracy et al., 2011). In combination with the lipid secretions, it is not surprising that speculations have been made regarding the potential role of these secretions for modifying hygroscopic properties and increasing the capability for condensing water (Toledo and Jared, 1993).

Lasiewski and Bartholomew (1969) have surmised that condensation underlies the water ecology of the desert-dwelling spadefoot toads Scaphiopus hammondii (synonym: Spea hammondii) and Scaphiopus couchii. Based on their considerations, it has been speculated that condensation also serves as a potential water source for some moisture-harvesting lizards (Gans et al., 1982; Schwenk and Greene, 1987; Comanns et al., 2011, 2016a). The spikes of the skin, it is thought, act as condensation foci, in particular for dew, which is regularly found in deserts (Gans et al., 1982; Beysens, 1995). Although laboratory conditions facilitate greater condensation than natural habitats, it has been shown that condensation on lizard skin is very unlikely to provide water in sufficient quantities for drinking (Comanns et al., 2011, 2016a). Thorny devils (M. horridus) can collect about 0.2% body mass by condensation (Withers, 1993; Comanns et al., 2016a), whereas 0.75% body mass gain by condensation has been reported for the western banded gecko Coleonyx variegatus (Lasiewski and Bartholomew, 1969). By comparison, thorny devils have been observed drinking when their skin channels are filled, which relates to a water collection ratio of about 3.2% body mass (Comanns et al., 2016a). However, condensation has been found to be sufficient to pre-wet the skin of the lizards and enable faster collection of water from more-efficient sources (Fig. 3) (Comanns et al., 2016a).

Despite general linguistic usage, fog cannot be ‘condensed’. This is because it comprises numerous fine droplets of already condensed water, light enough to remain suspended in the air (Gultepe et al., 2007). Therefore, fog can be collected without condensation, meaning without a sufficient temperature gradient. Several species of tenebrionid beetles as well as the sand-diving lizard Aporosaura anchietae and the viper Bitis peringueyi collect fog by exposing their body to fog-transporting wind in the Namib desert (Louw, 1972; Hamilton and Seely, 1976; Parker and Lawrence, 2001; Nørgaard and Dacke, 2010). Tenebrionid beetles of the genera Stenocara, Physosterna and Onymacris exhibit a mosaic of hydrophilic and hydrophobic structures (see Glossary) on the elytra that facilitate the collection of the fine mist from air; the hydrophilic islands act as condensation foci, whereas hydrophobic areas allow accumulated droplets to roll off easily (Hamilton and Seely, 1976; Parker and Lawrence, 2001; Nørgaard and Dacke, 2010). Besides physicochemical properties of the body surface, water collection can be assisted by gravity, i.e. a stereotypic body inclination. However, this behaviour is known only in two species of Onymacris (Hamilton et al., 2003).

Some land tortoise species (e.g. Psammobates tentorius trimeni, Kinixys homeana, Homopus areolatus) have been found to benefit from gravity in a quite remarkable way (Table 1). By lifting their hind limbs, water is channelled between the large ridges on the carapace of the tortoise to the mouth for drinking (Fig. 1G) (Auffenberg, 1963). In effect, the animals expose themselves to an external force, which transports collected water towards the mouth. The stereotypic body posture of most moisture-harvesting lizard species depicted in Table 1 includes similar stretching of the hind limbs and lowering of the head (Schwenk and Greene, 1987; Sherbrooke, 1990; Veselý and Modrý, 2002). Here, gravity appears to be utilized to support the functionality of skin microstructures and channels. Tenebrionid beetles also use their body inclination to let droplets of collected fog roll off their elytra to the mouth for drinking (Hamilton and Seely, 1976; Nørgaard and Dacke, 2010). For Onymacris unguicularis the elevation angle has been measured as 23 deg (Nørgaard and Dacke, 2010).

Passive water collection has inspired a lot of studies, and many of the functional structures that are presented above have been considered for adaptation to technical applications. Particular progress has been made for materials for fog collection, surface structures for directional transport of liquids for various material–liquid combinations, and textiles for separation of oil–water emulsions.

The collection of fog (i.e. condensed water suspended in air) has been considered as a means of water supply in regions satisfying certain criteria. Inspiration for development of ‘fog collectors’ (Shanyengana et al., 2002; Dorrer and Rühe, 2008a; Ahmad et al., 2010) has not only been taken from the mosaic pattern of the cuticle of tenebrionid beetles in the Namib desert, but also several plants (Andrews et al., 2011; Azad et al., 2015a,b). Examples of devices include structured material surfaces (Fig. 6A–D) (Dorrer and Rühe, 2008a; White et al., 2013) or fibres for textile applications (Fig. 6E–K) (Sarsour et al., 2010; Dong et al., 2012; Cao et al., 2015; Azad et al., 2017). Biomimetic approaches are less often found for thermally facilitated condensation. However, higher rates of water condensation have been found for uniformly hydrophilic surfaces (as for tree frogs) than for mosaic patterns of hydrophilic/hydrophobic surfaces (as for tenebrionid beetles) (Lee et al., 2012). The maximum condensation rate of 25 mg cm⁻² h⁻¹ that has been measured for artificial hydrophilic surfaces at a temperature differential of 15°C and 92.5% relative humidity is more than twice the rate that has been found for tree frogs under comparable conditions (Tracy et al., 2011; Lee et al., 2012).

Surface channels can transport water by capillarity. The channels on the legs of wharf roaches have been considered, for example, when designing spacecraft water-management systems (Thomas et al., 2010; Ishii et al., 2013). For two lizard species, water transport in skin channels is directional towards the mouth (Comanns et al., 2015; Yenmiş et al., 2015). In a technical context, directional fluid transport is often referred to as ‘liquid diodes’. Technologies for such microfluidic diodes include movable parts such as flaps (Adams et al., 2005) or cylindrical discs (Sochol et al., 2013). However, inspired by the Texas horned lizard, surface structures have been fabricated for passive, directional transport (Comanns et al., 2015). Channel asymmetry in the longitudinal direction as well as specific interconnections have been abstracted and transferred to materials such as polymethyl methacrylate (PMMA) and steel, maintaining directional transport of water and lubricants, even against gravity for a few centimetres (Comanns et al., 2015, 2016b). Lately, other surface structures for passive, directional fluid transport have been derived from transport of the oily defensive liquid by the flat bug D. lunatus (Plamadeala et al., 2017) and from the morphology of the spermatheca of fleas (Buchberger et al., 2018). All such surface structures for passive, directional liquid transport have been considered to be of particular interest for similar applications, such as in fields of lubrication (Comanns et al., 2016b). Here, the aspect of abrasion plays an important role (Uddin and Liu, 2016). Further considerations have been made regarding heat exchangers, microfluidics, distilleries, e-ink displays and hygiene products (Buchberger et al., 2015; Comanns et al., 2015, 2016b; Buchberger et al., 2018). In general, directional wetting phenomena or transportation have been found to result from directional surface structures or their orientation (Hancock et al., 2012; Xia et al., 2012; Chen et al., 2016).

The utilisation of gravity for water transport can be further abstracted to external forces in general that are within reach and utilizable for fluid transport. For example, the movement of machine components can be used to passively transport lubricants into bearings (Sathyan et al., 2011).

As described above, sandgrouse can store water enclosed within their breast feathers (Rijke, 1972; Joubert and MacLean, 1973; Rijke and Jesser, 2011). Another, rather different, mechanism of liquid storage in fibrous systems has been described for oil bees (Buchmann, 1987; Schäffler and Dötterl, 2011); here, branched, hair-like protrusions of the cuticle with oleophilic (see Glossary) properties collect and store the oil from flowers (Rüttgers et al., 2015). Investigations of this mechanism have led, for example, to the development of textiles for oil collection from sea water (Yao et al., 2011). The separation of emulsions is also possible with coated textiles, such as a Janus membrane and others (Feng et al., 2004; Tian et al., 2014; Brown and Bhushan, 2015).

In general, investigations of biological role models provide knowledge about the specific functionality reflecting adaptations to the conditions of the environment of the animals. Adaptations involve a compromise between parameters of various influences and requirements on the animal body surface. Abstraction and transfer to artificial surface structures reduces such compromises, but limitations occur in the challenge of other restrictions; for example, relating to specific conditions of the technical environment.

Transferring abstracted functional principles of wetting phenomena, water collection or water transport to artificial materials commonly requires fabrication of rather filigree structures in the micrometre range. This is non-trivial, but several technologies have been used to fabricate with sufficient accuracy; for example, direct laser structuring (see Glossary) (Klocke, 2007; Poprawe, 2011; Chen et al., 2013; Comanns et al., 2016b). For somewhat larger structures, micromilling (see Glossary) might be considered (König and Klocke, 2002; Biermann and Krebs, 2011), whereas large areas could require methods such as replication, embossing, injection moulding and additive manufacturing (Abbott and Gaskell, 2007).

The ability to collect water using the integument serves as a major adaptation of several animal species to their dry habitat to aid coping with conditions of limited water supply. The complex nature of surface structures itself suggests an important adaptational role of passive water collection performed by the integument. The morphology and function of integumental structures also show some convergence in this regard for the species considered above.

Passive water collection involves hydrophilic surfaces, at least to some extent, and one could even argue that wettable surfaces (contact angles <90 deg) might be required for such ability. Surface structures are similarly involved in water collection, and two conclusions can be made. First, all structures appear to significantly increase the body surface area responsible for collecting water from the animal's environment. This is less pronounced in the case of fog-basking beetles where surface chemistry plays a more important role. Second, the structures appear to offer specific functionality: different kinds of micro-pillars or hexagonal micro-ornamentation can alter the wetting properties of the integument in order to subsequently obtain some degree of spreading, grooves or channels can collect and transport water, and hair-like structures are found in the cases of accumulation and storage of water.

Although many species exhibit an accompanying behaviour – that is, active body movements – the actual process of water collection remains passive. The behaviour can instead be regarded as positioning the body surface towards the source from which water is obtained or to assist gravity-mediated water collection.

If we investigate biological role models, we often find specific functionality reflecting adaptations to the conditions of the animal's environment. Such specific functionalities, especially those resulting from surface structures, can be used for biomimetic approaches. Many of the described structures have been considered for technical applications, in particular mosaic patterns to collect fog for water supply, and surface channels to improve lubrication, micro-fluidics or hygiene products.

I thank my colleagues and friends Anna-Christin Joel, Benjamin Ott and Heiko Schmied for discussions and their feedback.