Joachim Mogdans, Institute of Zoology, University of Bonn, Meckenheimer Allee 169, Poppelsdorfer Schloß, 53115 Bonn, Germany.

You are watching: What is the function of the lateral line

Corresponding Author

Institute of Zoology, University of Bonn, Bonn, Germany

Correspondence

Joachim Mogdans, Institute of Zoology, University of Bonn, Meckenheimer Allee 169, Poppelsdorfer Schloß, 53115 Bonn, Germany.

*
PDF
Please review our Terms and Conditions of Use and check box below to share full-text version of article.
Shareable Link

Use the link below to share a full-text version of this article with your friends and colleagues. Learn more.


Fishes are able to detect and perceive the hydrodynamic and physical environment they inhabit and process this sensory information to guide the resultant behaviour through their mechanosensory lateral-line system. This sensory system consists of up to several thousand neuromasts distributed across the entire body of the animal. Using the lateral-line system, fishes perceive water movements of both biotic and abiotic origin. The anatomy of the lateral-line system varies greatly between and within species. It is still a matter of debate as to how different lateral-line anatomies reflect adaptations to the hydrodynamic conditions to which fishes are exposed. While there are many accounts of lateral-line system adaptations for the detection of hydrodynamic signals in distinct behavioural contexts and environments for specific fish species, there is only limited knowledge on how the environment influences intra and interspecific variations in lateral-line morphology. Fishes live in a wide range of habitats with highly diverse hydrodynamic conditions, from pools and lakes and slowly moving deep-sea currents to turbulent and fast running rivers and rough coastal surf regions. Perhaps surprisingly, detailed characterisations of the hydrodynamic properties of natural water bodies are rare. In particular, little is known about the spatio-temporal patterns of the small-scale water motions that are most relevant for many fish behaviours, making it difficult to relate environmental stimuli to sensory system morphology and function. Humans use bodies of water extensively for recreational, industrial and domestic purposes and in doing so often alter the aquatic environment, such as through the release of toxicants, the blocking of rivers by dams and acoustic noise emerging from boats and construction sites. Although the effects of anthropogenic interferences are often not well understood or quantified, it seems obvious that they change not only water quality and appearance but also, they alter hydrodynamic conditions and thus the types of hydrodynamic stimuli acting on fishes. To date, little is known about how anthropogenic influences on the aquatic environment affect the morphology and function of sensory systems in general and the lateral-line system in particular. This review starts out by briefly describing naturally occurring hydrodynamic stimuli and the morphology and neurobiology of the fish lateral-line system. In the main part, adaptations of the fish lateral-line system for the detection and analysis of water movements during various behaviours are presented. Finally, anthropogenic influences on the aquatic environment and potential effects on the fish lateral-line system are discussed.


1 INTRODUCTION

Sensory ecology is a discipline that focuses on the study of animal sensory systems in order to understand how environmental information is perceived, how this information is processed and how this affects interactions between the animal and its environment (Dangles et al., 2009). Animals live in distinct habitats that are governed by certain physical relationships and this provides constraints for the development of physically based sensory systems. The understanding of the interplay between physical principles and sensory system morphology and function is key to the question whether particular features of a sensory system are of adaptive value to the individual.

The lateral-line system is a sensory system found in fishes and aquatic amphibians. With the lateral-line system, fishes measure the relative movements between their body and the surrounding water at each of up to several thousand sensory organs, the neuromasts (Dijkgraaf, 1952, 1963). To understand the functional significance and any potential adaptations of the lateral-line system to the sensory environment, it is essential to know the physical properties of biologically relevant and irrelevant stimuli, the anatomical organisation of the lateral-line system in different fishes, the neurophysiological basis of principles of operation and the behavioural context in which the lateral-line system is used.

2 NATURAL HYDRODYNAMIC STIMULI

Our knowledge of naturally occurring biologically relevant or irrelevant lateral-line stimuli is still very limited. Measuring natural stimuli in field studies is not an easy task. Pressure waves can be recorded with hydrophones, which are useful for gross measurements in both the laboratory and field environments, but generally they are too big to measure the small-scale pressure changes that are relevant for the lateral-line system (Mogdans & Bleckmann, 1998). Hot-wire anemometers or laser-Doppler anemometers are much better suited to measure these small-scale water motions (Coombs et al., 1989a; Blickhan et al., 1992); however, they measure flow velocity only at a single point in space, i.e., they cannot provide spatial information. In addition, they are very fragile and expensive and thus not well suited for field work. Visualisation of spatio-temporal patterns of water flow in two or even three dimensions can be achieved with particle image velocimetry (Adrian, 2005; Adrian & Westerweel, 2011), which reveals information about flow direction, velocity and vorticity (Hanke et al., 2000). This, however, requires the seeding of the water with large amounts of small, neutrally buoyant glass or polyamide particles that are difficult, if not impossible, to remove once dispensed in the natural environment.

Hydrodynamic stimuli that can be detected by the lateral-line system can occur at the water surface or in midwater (Figure 1). Surface waves of biotic origin are for example caused by terrestrial insects falling into the water or by aquatic animals contacting the water–air interface in order to breathe or feed (Bleckmann, 1988). Subsurface water disturbances may be caused by swimming or opercular (respiratory) movements of fishes or other aquatic animals. Such stimuli can be used in many ways. Flow fields produced by fish during swimming can be used to obtain information about the environment (von Campenhausen et al., 1981; Hassan, 1985, 1986) and oscillatory stimuli generated by body vibrations may provide important communication signals during social behaviour (Satou et al., 1991, 1994). Then again, self-generated stimuli can be disadvantageous for a fish since they allow for detection by predators and may also interfere with the detection of potentially relevant novel stimuli. Strategies to avoid the generation of self-generated water movements have been observed in certain fish behaviours. For instance, black carp Mylopharyngodon piceus (Richardson 1846) (Xenocyprodidae), spend substantially less time moving and exhibit an overall shorter total distance of movement in the presence of predatory snakehead Channa micropeltes (Cuvier 1831) (Channidae; Tang et al. 2017).

*

Examples of biotic water movements: water surface waves generated, from top to bottom, by (a) wind, (b) the clawed frog Xenopus laevis, (c) Carassius auratus and (d) the fly Calliphora vicina. Water movements were recorded with a laser-Doppler anemometer (from Bleckmann et al., 1989); subsurface water movements generated by, (e) the ostracod Tetrdeium crassum, (f) the amphipod Paradexamine houtete (from Montgomery, 1989), (g) male and (h) female spawning Oncorhynchus nerka (from Satou et al., 1991). Water movements from ostracods and amphipods were recorded with an optoelectric transducer and those from salmon with a piezoelectric acceleration transducer
The various hydrodynamic stimuli generated by abiotic sources are generally regarded as unwanted background noise. Usually, noise is defined as unwanted sound (described in terms of sound pressure) that is judged to be unpleasant, loud or disruptive to hearing. For the lateral-line system, noise can be defined as any kind of water movement (described either in terms of particle motion or pressure gradient) that interferes with and even impairs the detection of biologically more relevant water movements. For example, wind or leaves falling onto the water produce surface waves of abiotic origin that may impede the detection of surface waves generated by animals. Below the water surface, currents, tides, changes in temperature, salinity gradients and gravity are abiotic sources of water movements (Wetzel, 1983). Fishes that live in ponds, lakes, or the deep ocean tend to be confronted with less such hydrodynamic noise compared with fishes that live in a fast flowing river or along the ocean shoreline. In these habitats, highly turbulent water would clearly interfere with the detection of other, biologically more relevant signals like those generated by prey, predators or conspecifics. Nonetheless, water currents may still provide important sensory information that may be used by fishes, such as for orientation, station holding and the reduction of energetic costs (Montgomery et al., 1997; Liao et al., 2003; Liao, 2007; Przybilla et al., 2010).

3 ANATOMY OF THE LATERAL-LINE SYSTEM

Neuromast sensory organs of the lateral-line system can be distributed across almost the entire fish body (Figure 2). They consist of a macula comprising sensory hair cells, supporting cells and mantle cells (Münz, 1979). The hair cells are similar in function and morphology to those in the auditory and vestibular system of vertebrates (Roberts et al., 1988). The ciliary bundles of the hair cells are embedded in a gelatinous dome-like structure, the cupula (Figure 2). Water movements cause deflections of the cupula resulting in the shearing of the ciliary bundles (van Netten & Kroese, 1987, 1989; McHenry et al., 2008; van Netten & McHenry 2006), which leads to a change in the hair cells’ membrane potential (Görner, 1963; Harris et al., 1970; Sand et al., 1975).

*

(a) Distribution of neuromasts in a teleost, Carassius auratus:
*
, superficial neuromasts;
*
, canal pores. Typically, a canal neuromast is located between two adjacent canal pores. (b) Schematic drawings of a superficial neuromast and (c) a canal neuromast. While superficial neuromasts are stimulated directly by water flow across the fish surface, canal neuromasts are responsive to water flow inside the canal which results from pressure differences between canal pores
The most salient feature of the peripheral lateral-line system is the division into a population of superficial neuromasts and a population of canal neuromasts (Figure 2). Superficial neuromasts (SN) occur directly on the surface of the skin, where they are arranged in lines or clusters on the head, trunk and tail fin. Functionally, SNs are velocity detectors; i.e., their neuronal responses are proportional to the velocity of the water flowing around the cupula. In contrast, canal neuromasts (CN) occur in canals on fishes’ heads and trunk. The fluid inside the canals contacts the water surrounding the fish through a series of canal pores. In bony fishes, especially teleosts, one CN is typically found between two adjacent canal pores (Webb & Northcutt, 1997). Consequently, CNs function as pressure gradient detectors, i.e., they respond to pressure differences between neighbouring canal pores (Coombs & Montgomery, 1999). Outside the canal, the pressure gradient is proportional to the acceleration of the water. Thus, CNs may also be regarded as acceleration detectors of water motions outside the canal (Kalmijn, 1989a).

The cephalic lateral-line canal system of bony fishes comprises the supra and infraorbital, the otic and postotic and the mandibular and preopercular canals. The supraorbital and infraorbital canals meet behind the eye where they continue as the otic canal. The mandibular canal merges with the preopercular canal and the latter meets the otic canal just rostral to the operculum from where they continue as the postotic canal. The postotic canal meets the trunk canal, which extends along the side of the fish. Finally, the supratemporal commissure connects the lateral-line canals of the two body sides by crossing the top of the head (Webb, 1989a,b, 2014a,b).

The anatomy of the peripheral lateral-line system varies greatly across species (Figure 3; Coombs et al., 1988; Webb, 2014a,b). For instance, SNs can be located on the skin, recessed in pits, or elevated on papillae (Dijkgraaf, 1952, 1963). In addition, SN number and size vary greatly among species (Beckmann et al., 2010; Schmitz et al., 2014; Watanabe et al., 2010). The number and structure of lateral-line canals is also highly variable. The ways in which canals differ in the number of branchings, diameter, or number and size of canal pores have been described by Webb (2014a,b). Canals can be reduced in length or modified in position. For example, they can be arched, disjunct, incomplete or multiplied. For detailed reviews of the phylogenetic distribution and morphological variation of the peripheral lateral-line system see Coombs et al. (1988), Northcutt (1989), Webb (1989a,b) and Webb (2014a,b).

*

Distribution of superficial neuromasts (
*
) and canal pores (
*
) in (a) Rhodeus sericeus, (b) Oncorhynchus mykiss and (c) Ancistrus sp

4 FILTER PROPERTIES OF THE LATERAL-LINE SYSTEM

Different peripheral morphologies of a sensory system provide different filter properties. In other words, morphology determines the range of stimuli to which a sensory system is most sensitive. A classic example of the lateral-line system is how canals function as high pass filters for hydrodynamic stimuli, with narrow canals exhibiting high and widened canals exhibiting low cut-off frequencies (Denton & Gray, 1988, 1989; Bleckmann & Münz, 1990). The filter properties of the lateral-line system not only depend on canal morphology, but also on radius and length of the cupula, on cupula sliding stiffness, on the stiffness of the ciliary bundles of the hair cells and thus also on the number of hair cells within a neuromast. Additionally, they are affected by the density and viscosity of the fluid surrounding the cupula; i.e., water in the case of SNs and canal fluid in the case of CNs (Denton & Gray, 1989; van Netten, 1991, 2006; Coombs & van Netten, 2006). These variables strongly determine how information from the water surrounding the cupula is transferred to the lateral-line system. Finally, number and placement of SNs as well as number and placement of canal pores may affect the nature of hydrodynamic information that is received by the lateral-line system (Klein et al., 2013).

Without any doubt, the interspecific variation in lateral-line system anatomy is to some extent subjected to developmental and morphological constraints (Webb, 1989a,b). However, the many examples of convergent evolution of peripheral lateral-line morphology (Coombs et al., 1988; Webb, 2014a,b) and the obvious relationship between peripheral morphology and filter properties both suggest that the various morphological patterns of this sensory system represent, at least partially, adaptations to prevailing hydrodynamic conditions that are encountered in the habitats of different species. This hypothesis was supported by investigations on the morphology of the lateral-line system in the Pacific staghorn sculpin Leptocottus armatus (Girard 1854) (Cottidae), the tidepool sculpin Oligocottus maculosus (Girard 1856) (Cottidae) and the tadpole sculpin Psychrolutes paradoxus (Günther 1861) (Psychrolutidae) (Vischer, 1990). These species live in distinctly different habitats where they are exposed to different hydrodynamic stimuli ranging from slow to highly turbulent water flows. At the same time, they exhibit discrete differences in their lateral-line systems in canal configuration (in particular on the head) and in the number and placement of superficial neuromasts. This suggests that the lateral-line systems are morphologically adapted to the different hydrodynamic environments in which these fishes live. In other studies, the neuronal responses to hydrodynamic stimuli of lateral-line neurons in goldfish Carassius auratus (L. 1758) (Cyprinidae) and trout Oncorhynchus mykiss (Walbaum 1792) (Salmonidae) were compared (Engelmann et al., 2002, 2003). Carassius auratus is a slow-moving still-water fish with an abundance of superficial neuromasts distributed across the head, trunk and tail fin (Puzdrowski, 1989; Schmitz et al., 2008). In contrast, O. mykiss live in fast flowing rivers, possess only a few superficial neuromasts and have lateral-line canals which are narrower than those in C. auratus (Engelmann et al., 2002). In neurophysiological experiments, water flow affected the responses of C. auratus lateral-line neurons more strongly than the responses of O. mykiss neurons. In addition, C. auratus possess more neurons sensitive to water flow than O. mykiss. While running water masked neuronal responses to local vibratory stimuli generated by a mechanical dipole source in both species, responses were affected more strongly in C. auratus (Engelmann et al., 2002, 2003). These physiological differences indicate that the lateral-line systems of C. auratus and O. mykiss are adapted to different hydrodynamic conditions. In contrast, no apparent differences were found in the frequency response functions of anterior lateral-line nerve fibres in six species of Antarctic fishes of the suborder Notothenioidei, despite these fishes exhibiting distinct differences in the dimensions of cranial lateral-line canals (Montgomery et al., 1994). Finally, in a study assessing the abundance and spatial distribution of superficial neuromasts in twelve common European cypriniforms (Beckmann et al., 2010) no differences were found between rheophilic and limnophilic species. These data argue against correlations between lateral-line system morphology and habitat preference.

5 INTRASPECIFIC VARIATION IN LATERAL-LINE SYSTEM ANATOMY

Intraspecific variations in lateral-line anatomy and their origins are not well studied. Differences can be attributed to epigenetic effects or to phenotypic plasticity. While the former entails changes that affect gene activity and expression without altering the DNA sequence (Dupont et al., 2009), the latter refers to the ability of a given genotype to produce more than one phenotype (Price et al., 2003). In many cases, epigenetic effects play a role in phenotypic plasticity.

Intraspecific differences in lateral-line morphology were found between wild-caught and hatchery-reared migratory O. mykiss juveniles. Wild animals had significantly more SNs than hatchery-reared juveniles, although the number of hair cells within individual neuromasts was not significantly different between groups (Brown et al., 2013). In addition, wild and hatchery-raised migratory O. mykiss had different otolith composition and brain mass, which may have other behavioural consequences. In the wild, salmon Oncorhynchus spp. grow up in turbulent rivers and streams containing pools, riffles and cascades, whereas hatchery Oncorhynchus spp. are raised in raceways that are barren, uniform-depth tanks that are flushed by rather low-velocity systems (Kihslinger & Nevitt 2006; Kishlinger et al., 2006). This supports the notion that different hydrodynamic conditions during development can result in differences in the anatomy of a sensory system. In the case of migratory O. mykiss, the reported differences predict a reduced sensitivity to biologically important biotic and abiotic hydrodynamic signals and consequently a reduced survival fitness after release (Brown et al., 2013).

In farm-reared gilthead sea bream Sparus aurata (L. 1758) (Sparidae), distinct lateral-line system deformations were found. The fish exhibited zigzag and wavy trunk lateral-line canals with parts of the canal even missing, compared with the otherwise rather straight and continuous trunk canals found in wild S. aurata (Carillo et al., 2001). Farmed sea bass Dicentrarchus labrax (L. 1758) (Moronidae) and S. aurata) exhibited a so-called scale-pocket deformity in which the lateral-line scales were missing while the underlying canal was still present, whereas in the somatic-scale deformity the lateral-line canal was missing but covered with normal somatic scales (Sfakianakis et al., 2013). Morphological abnormalities of these types are not necessarily a consequence of different hydrodynamic conditions experienced during rearing but could also be caused by the high density of animals in hatcheries, which results in more interactions with a concomitant higher rate of deformations and ablations (Brown et al., 2013).

Intraspecific differences in lateral-line system morphology were also reported for the three-spine stickleback Gasterosteus aculeatus L. 1758 (Gasterosteidae), a species that occupies a wide range of aquatic habitats (Wark & Peichel, 2010). While the arrangement of SN lines on the G. aculeatus body is largely the same in different populations, the number of neuromasts within these lines varies across individuals and populations occupying different habitats. For example, stream G. aculeatus have more neuromasts than G. aculeatus living downstream in the same catchment. Wark and Peichel (2010) also found that G. aculeatus from two different lakes had more trunk neuromasts than sympatric limnetic G. aculeatus, providing evidence for parallel evolution of the lateral-line system. These data indicate that the lateral-line system in a given species may experience different selection pressures in alternative natural habitats and may therefore develop differently under different hydrodynamic conditions.

Consistent with this idea are data collected from guppies Poecilia reticulata (Peters 1859) (Poeciliidae) suggesting that risk of predation is a selective pressure affecting lateral-line system phenotype. Fischer et al. (2013) compared the lateral-line systems of wild-caught Trinidadian P. reticulata (Poeciliidae) from high and low-predation populations in two different river drainages and found that fish in high-predation populations had overall more neuromasts than fish from low-predation populations. Interestingly, laboratory-reared fish from a low-predation population of a third river drainage had more neuromasts than laboratory-reared fish from a high-predation population of a fourth drainage. However, within both populations, fish exposed to chemical cues from a pike cichlid Crenicichla sp. predator had more neuromasts than fish housed in tanks containing only natural water. These data show that in P. reticulata the distribution of neuromasts varies between populations and is influenced by both genetic and environmental factors with exposure to an ecologically relevant stimulus.

6 NEUROBIOLOGY OF THE LATERAL-LINE SYSTEM

Afferent nerve fibres are contacting the hair cells of neuromasts and connect them to the central nervous system (CNS). The fibres course in at least three distinct lateral-line nerves (depending on neuromast location) and terminate predominantly in the brainstem. From there, secondary ascending fibres reach distinct regions in the midbrain and forebrain, indicating that lateral-line information is processed at all levels of the CNS (Striedter, 1991). A detailed account of the organisation of the central nervous system with reference to the lateral-line system is given by Wullimann and Grothe (2014).

Numerous neurophysiological studies have described the representation of lateral-line information by primary afferent nerve fibres as well as brainstem and midbrain neurons (Chagnaud & Coombs 2014; Mogdans & Bleckmann, 2012; Bleckmann & Mogdans, 2014). Afferent nerve fibres are highly sensitive to local water motions like those generated by a sinusoidally vibrating sphere (Coombs et al., 1996), to complex water motions generated for example by a moving object (Mogdans & Bleckmann, 1998), to toroid vortices (Chagnaud et al., 2006) and to bulk water flow (Engelmann et al., 2000, 2002). In addition, the discharges of many afferent fibres represent the shedding frequency of vortices created by obstacles in the flow (Chagnaud et al., 2007a). Based on their responses, central neurons appear to be more selective than primary afferents. For instance, many brainstem and midbrain neurons are not very sensitive to sinusoidal water movements but respond readily to a moving source (Mogdans & Goenechea, 2000; Engelmann et al., 2003; Plachta et al., 2003). Similarly, some brainstem neurons are unresponsive to bulk water flow whereas others are flow sensitive (Mogdans & Kröther, 2001). Central lateral-line neurons may even encode the frequency of the vortices that are shed by a cylinder in the flow (Klein et al., 2015; Winkelnkemper et al., 2018). While it is known that ascending lateral-line information reaches the forebrain of fishes (Striedter, 1991), there are hardly any data on forebrain responses to hydrodynamic stimuli. There is also little knowledge on the function of the efferent connections within the central lateral-line system (Flock & Russell, 1976; Roberts & Russell, 1972; Weeg et al., 2005).

See more: Bill Gates 11 Rules Of Life Pdf, Bill Gates 11 Rules

7 ADAPTATIONS OF THE LATERAL-LINE SYSTEM FOR THE DETECTION OF WATER SURFACE WAVES

The lateral-line systems in different fish species are adapted to specific hydrodynamic signals provided in distinct behavioural contexts or environments (Coombs & Montgomery, 2014; Webb et al., 2008). The most clear-cut example of sensory adaptation to a specific type of hydrodynamic stimulus is the peripheral lateral-line system of surface-feeding fishes. Species such as the topminnow Aplocheilus lineatus (Valenciennes 1846) (Aplocheilidae) or the African butterflyfish Pantodon buchholzi (Peters 1876) (Pantodontidae) have flattened heads bearing a specialised cephalic lateral-line system consisting of six rows each containing acceleration-sensitive neuromasts (Figure 4). As such, the cephalic lateral-line system in these species is particularly well suited for the detection of water-surface waves (Bleckmann et al., 1989; Montgomery et al., 2014).