The Visual System of Seahorses and Pipefish: a Study of Visual Pigments and Other Characteristics
The visual system of seahorses and pipefish: A study of visual pigments and other characteristics.
Virginia Jan Mosk
Bachelor of Science, University of Melbourne Post Graduate Diploma of Science, The University of Western Australia
This thesis is presented for the degree of Master of Science
THE UNIVERSITY OF WESTERN AUSTRALIA School of Animal Biology Zoology
December 2004 Abstract
Syngnathidae (seahorse, pipefish, pipehorses & seadragons) are highly visual feeders with different species feeding on specific types of prey, a behaviour that has been related to snout length. Worldwide, many species have become threatened by habitat destruction, collection for the aquarium trade and exploitation for traditional medicine, as well as recreational and commercial bycatch. Attempts to establish aquaculture programs have been of limited success. Little is known about their visual capabilities in detail.
The visual systems of fishes are known to have evolved specific adaptations that can be related to the colour of water in which they live and specific visual tasks such as predator detection and acquisition of food. This study examined the ocular and retinal morphology, photoreceptor structure and spectral sensitivity of adult individuals of a local pipefish (S. argus), local seahorse (Hippocampus subelongatus) which both inhabit green water seagrass beds, and a tropical species of seahorse (Hippocampus barbouri) from blue water coral reefs. Some juveniles were also investigated. Accordingly, we developed an understanding of the features that are common to all syngnathids and those that have evolved for specific environments.
Cryosections of the eyes were taken to determine morphological distinctions of this group. Lens characteristics measured using a spectrophotometer determined 50% cut-off wavelengths below 408nm for all 3 species, hence no transmission of UV light to the retina. Histological examination determined a cone dominated fovea in the ventro-temporal retina and very large rods concentrated in the peripheral retina and adjacent to the optic nerve. Microspectrophotometry measured the absorption characteristics of the visual pigments within the photoreceptors showing the presence and maximum sensitivity (λmax) of rods, SWS single cones, and a broad, complex array of LWS double/twin cones. The results are discussed in relation to the light environment inhabited by each species and their feeding requirements. The implications for the design of suitable light environments for aquarium and aquaculture programs for the Syngnathidae are also discussed. Rearing success of this family of fish, for both the aquarium trade and re-stocking programs, would be advised to take lighting regimes and specifics of the animals’ vision into account.
i Table of contents
Abstract i Acknowledgements vi List of Abbreviations vii
Chapter 1: INTRODUCTION 1 The underwater light environment 1 Visual system structural organisation in fish 4 Eye and optics 4 Optics: Cornea and Lens 4 Retinal and photoreceptor structure 6 Colour sensitivity of photoreceptors 8 Development of the fish retina 11 The Syngnathidae 12 Vision in syngnathids 15 Rationale and broad aims of this study 17
Chapter 2: MATERIALS AND METHODS 19 Animals - Stigmatopora argus 19 Hippocampus subelongatus 19 Hippocampus barbouri 20 Pugnaso curtirostris 21 Seahorse and pipefish collection, holding and experiments 22 Morphology 24 Pupil response 24 Measurements of ocular dimensions from cryosections 25 Lens transmission spectra 26 Microspectrophotometry (MSP) 27 Tissue preparation 27 Measurement of visual pigment absorbance spectra 28 Analysis of the visual pigment absorbance spectra 28
Chapter 3: RESULTS 30 External morphology and orientations of the body 30 Adults - Stigmatopora argus 30 Hippocampus subelongatus 31 Hippocampus barbouri 32
ii
Snout length as a proportion of head length, and eye diameter as a proportion of snout length 33 Ocular Structure and Dimensions 35 Lens colour and shape 36 Lens transmission spectra 37 Pupil response 38 Retinal structure 38 Histology 40 The photoreceptors and their visual pigments 45 Adults 45 Rods 46 Cones 50 Pairings 60 Juveniles 62
Chapter 4: DISCUSSION 65 Morphology and eye structure 65 Snout length to head length, and eye diameter to snout length 65 Irideal flaps 66 Pupil Dilation 67 Ocular Dimensions and function 68 Lens properties 69 Shape 69 Filtering Effects 69 Photoreceptor structure and photoreceptors 71 Rods 72 Cones 72 Colour sensitivity of the photoreceptors 73 Rods 73 Cones 74 Mechanisms of visual pigment absorbance of double cones 77 Juveniles 77 Future directions 79
REFERENCES 81
iii Acknowledgements
This study and thesis only eventuated with the combined assistance of many individuals whom I have the greatest pleasure in thanking:
Special thanks to my supervisors Dr Julia Shand and Professor Lyn Beazley. Lyn, for her positive support, enthusiasm and assistance in realising this unusual dream, attending the Seadragon Festival and trips to AQWA. Julia, for providing the quiet, consistent assistance, and warmth and support that I needed throughout the project and her understanding about issues involved in juggling single-parenting of a young child, and other difficulties that life throws at you. We both leant much and I will always have deep respect and admiration. Both wonderful women gave me the space and freedom to peak and trough as I needed.
Dr Nicole Thomas and Dr Nathan Hart for their immense assistance during many hours in the dark room with the MSP and analysing the visual pigment data. Warm thanks to Michael Archer for his brilliant expertise in everything computer related, also teaching me and assisting with the histological slides. Michael endured much frustration during my steep learning curve. The lovely Brenda Loney, and the rest of the Neuroscience Lab. for their support, encouragement and kindness. I had many breaks during this study and their faces were always welcoming and positive when I returned.
Dr Alan Kendrick for his kindness and generosity in sharing whatever information he had on these wonderful animals’, and for helping me through the seasickness. Dr Glenn Hyndes for his talks on seagrasses and positive interest. Andy Grice and Dr Mic Payne, Dr Craig Lawerence, Glenn Moore and Paul Groves for their passion for seahorses, sharing of information and generosity in providing animals. Also, Project Seahorse, Dragon Search, Victoria & NSW Museums for sharing information. Rudie Kuiter for his informal chats and emails, his zeal and individuality in the study of these beautiful animals. Professor Dale Roberts for critical and very helpful comments on the thesis. Dr Julian Partridge for his optimistic and constructive help with the project generally, assistance with lens transmission data, fishing and newspaper interviews.
Dr Ian Williams (and wife Judy) for their kind, gentle, warm and encouraging support throughout the project, guidelines for writing the thesis, completion of the Rottnest swim together, pep talks, training of wilful dogs and breakfasts at ‘Beaches’ and ‘Beverley’. Dr Jasper Trendall for his interest in the project, ‘Leafies’, and valuable comments on a very rough draft of the thesis. Also, the hospitality shown by Jasper and his beautiful lady, Lindy Leggett, on trips to Albany.
The inner circle…. Cate Kandiah, Claire D’Angelo, Jackie Hooper, David Worthy, Trish Grindrod, Marion Raad, Lorna Stewart, Karen Dewar, Tim Carew-Reid for their friendship, support and conviction that I really am quite mad…….and Aaron Zee for his sacrifices and patience in allowing me to complete another higher degree whilst mothering him, and zipping him around to his activities. This study spanned from years 5 to 9 of his young life, and the wisdom, interest and support he has shown has always been way beyond his years.
iv List of Abbreviations
λ ………………………………………………………………………...wavelength
λmax ………….…………….wavelength of maximum absorption (lambda max) INL……………………………………………………..….…….inner nuclear layer LWS…………………………………………………….long wavelength-sensitive MSP ……………………………………………………..microspectrophotometry MWS……………………………………………….medium wavelength-sensitive ONL……………………………………………………………..outer nuclear layer OS……………………………………………………………………outer segment PBS ……………………………………………………phosphate-buffered saline RH1…………………………………………………………………...rod rhodopsin RH2………………………………………………………………rod rhodopsin-like SWS……………………………………………………short wavelength-sensitive TCM …………….……….…………………………Traditional Chinese Medicine UV …………….…………………………………………………………..ultraviolet
v Chapter I - Introduction
Chapter 1. INTRODUCTION
The underwater light environment
The ocean’s epipelagic zone (from the surface to around 200 meters depth) is considered the photic zone where sunlight penetrates sufficiently to support primary production by photosynthesis and supports the food web (Castro & Huber, 1997). The sun’s electromagnetic radiation is visible to humans when it is of a wavelength in the range 400 to 700 nanometers (nm) (Figure 1). Light of the wavelength of approximately 400nm appears violet to humans. Ultraviolet wavelengths are those shorter than 400nm. Light approaching 700nm appears red. Wavelengths beyond 700nm are termed infrared.
Figure 1. The electromagnetic spectrum. (Goldstein, 1996)
At the water’s surface, the radiation from the sun extends in a spectral band ranging from approximately 300nm in the ultraviolet to around 1100 nm in the infrared. Light is rapidly attenuated through the depth of the water column by the absorption of both short and long wavelengths. The rate of absorption in different regions of the spectrum depends on the amounts of plankton (chlorophyll) and dissolved organic matter (Gelbstoffe or yellow substances) present. In nutrient-poor, clear oceanic water, away from land, sunlight penetrates deeper into the ocean. Here, very short wavelengths (ultraviolet and violet) and the longer wavelengths of light (such as red and infrared) are filtered out leaving predominantly a monochromatic blue wavelength between 470 and
1 Chapter I - Introduction
480 nm penetrating to the mesopelagic zone (beneath the photic zone) (Fig. 2A). Many marine organisms inhabiting extreme ocean depths utilise light in the form of bioluminescence (Douglas et al. 1998a). Ultraviolet (UV) photons exist at depth however, and in clear water they may be available for vision to several hundred metres (Frank & Widdler 1996).
Figure 2. Transmission of light by water is dependent on the colour or wavelength of the light. In clear oceans and lakes (a) light becomes increasingly monochromatic and blue as its path length increases. In fresh water that carries organic matter (b) light at all wavelengths is absorbed more quickly than it is in clear water, but the light becomes greener with path length. In rivers, swamps and marches that carry large amounts of plant and animal decay products (c), absorption is rapid and the spectral distribution of the light shifts to the red. Such waters are called infrared-transmitting water (black) because the human eye is relatively insensitive to light at long wavelengths. The depths given for the maximum penetration of light are typical, but they vary widely (After Levine & MacNichol, 1979). 2 Chapter I - Introduction
Coral reefs are present in relatively shallow, nutrient-poor water in a bright light environment where down-welling sunlight contains a broad spectrum of light including relatively large amounts of UV (less than 390nm) and the far-red (600- 700nm) wavelengths (Fig. 3A). Coastal waters, where phytoplankton and land run-off increase levels of chlorophyll and particulate matter, transmit a narrower spectrum with a maximum wavelength in the green around 510-550nm (Partridge 1990) (Fig 3B & 4B). In contrast many freshwater and estuarine waters containing products of vegetable decay and tannins (such as the Swan River estuary in West Australia) transmit longer (red) wavelengths but inhibit the transmission of blue light (Fig. 2C). These environments provide an array of light regimes in which the resident animals have responded with a variety of eye structure and functional adaptations.
Figure 3. Colour of water along a horizontal line of sight in: (A) a tropical reef coral environment compared with: (B) a temperate, green seagrass environment.
Visual system structural organisation in fish
Animals have evolved adaptations in their visual system to subserve two main visual functions: to collect light and to analyse the focused image (Ferald 1990). Animals use light to perceive the objects around them, but may also interact using visual signals. Animal signalling extends the world of the individual animal to encompass a community, allowing its participants to make decisions
3 Chapter I - Introduction based not only on their own state, but also upon the physiology, morphology and behaviour of others (Endler 1993). Signals based on colour require a certain threshold of ambient illumination in order to be visible, and as a consequence their use tends to be restricted to diurnal animals. The eyes of vertebrates show many adaptations that are related to the intensity and spectral composition of light in which the species are active and the specific visual tasks they need to perform (Walls, 1942; Lythgoe, 1979). Predation by most diurnal fishes is primarily visual and determined by photic conditions (Spotte, 1992). Fish have been extensively studied and provide examples of specialisations in optics, retinal structure and physiology, due to the diversity of light environments and habitats in which they are found and the tasks which they must perform (Douglas & Djamgoz 1990). Additionally, many species undergo developmental changes in retinal structure that can be related to changes in lifestyle, such as the changes in spectral sensitivity to the changing light environment (Shand, 1997; Shand et al. 2002).
Eye Structure Optics: Cornea and Lens
A B
Figure 4. (A) Schematic representation of the fish eye and retina and: (B) the photoreceptor-types found in fish retinae. (Modified from Lythgoe, 1979).
Light enters the eye through the cornea, is focused by the lens and passes through the cavity occupied by the vitreous humour onto the retina (Fig. 4A). Light must then pass through the neural retina which is comprised of the ganglion cell layer, inner nuclear layer, which contains horizontal cells, amacrine cells and bipolar cells, then finally to the photoreceptor layer. The cornea is a relatively thin, avascular, transparent structure consisting of two curved and parallel
4 Chapter I - Introduction surfaces. Its refractive function is dependent on the existence of media of unequal refractive index in front and in back (Sivak 1980). The aquatic habitat of ancestral vertebrates limited the original function of the cornea to that of a transparent window (Ferald 1990).
In the aquatic habitat, the lens is the only refractive element and lens optical quality represents that of the whole eye. Fish possess lenses with high dioptric (refractive) power (Ferald 1990). The spherical shape of the lens is considered to be the most appropriate for aquatic vertebrates (Walls 1942; Duke- Elder 1958; Pumphrey 1961) including teleosts (Matthiessen 1882), although exceptions have been found (Sivak 1990). Yellow lenses and corneas function as UV filters, and are particularly common in coral reef fishes. The transmittance of the ocular media of 211 coral reef fish species (Siebeck & Marshall 2001) established that the ocular media of 50.2% of species strongly absorbed light of wavelengths below 400nm, and in 80% of these species, the lens was the limiting filter for the whole eye. UV vision was very unlikely in these species. Those animals which do not block UV from reaching the retina may still have a UV absorbing filter (macular pigment) in the retina (Nussbaum et al. 1981). As fish move into different environments with age, blocking compounds may accumulate protecting the retina from UV at the expense of sensitivity to UV light (Losey et al. 2000). Coloured filters in the light path could also enhance the detection of contrasting colours and images. They are believed to enhance image quality by reducing the effect of optical imperfections of the ocular media that cause chromatic aberration and scatter (Kennedy & Milkman 1956; Denton 1956; Motais 1957; Moreland & Lythgoe 1968; Muntz 1976; Siebeck & Marshall 2001; Nelson et al. 2003). Both chromatic aberration and scatter are strongest for short wavelengths (Lythgoe, 1979). Paradoxically, lenticular filters have also been found in deep sea fishes (Douglas & Thorpe, 1992), but may play a role in the detection of bioluminescence against down-welling light (Douglas, Marshall & Partridge 1998a).
In fish, the focal length (distance from the centre of the lens to the retina) divided by the lens radius is termed the Matthiessen’s Ratio (Matthiessen 1882) (Fig. 5). The value is thought to be constant within any given species and has
5 Chapter I - Introduction been found to vary between species with a range of 2.2 – 2.8 and an average of 2.5 (Sivak 1990; Collin et al. 1998; Shand et al., 1999; Collin & Collin 1993). As the fish grows, the eye grows in a scaled fashion, allowing increased light to enter the larger eye focusing on a more distant retina, so the light per unit area remains constant. A higher Matthiessen’s Ratio generally provides an image that is projected to the retina and sampled by higher numbers of photoreceptors giving better resolution of that image. A larger Matthiessen’s Ratio also results in less light reaching the retina (Land 1981), conversely a lower Matthiessen’s Ratio allows more to the retina due to the larger aperture in relation to the focal length analogous to a low F number of a camera (focal length/aperture diameter).
radius lens
focal length
retina
Figure 5. Cryosection, (horizontal plane) of the eye of a typical teleost, the black bream Acanthopagrus butcheri (Shand et al. 1999). In most teleosts the Matthiessen’s Ratio (MR = focal length divided by lens radius) is 2.5.
Retinal and photoreceptor structure
The visual signal begins with the photoreceptors, differentiated nerve cells in the outer retina that are specialised for the absorption of photons and the conversion of the light energy into electrical activity. All vertebrate photoreceptors are constructed upon the same basic plan (Fig. 4B), consisting of an outer segment (containing the light sensitive visual pigment), an inner
6 Chapter I - Introduction segment, a nucleus, and a synaptic terminal (Rodieck, 1973). As the fish pupil is usually immobile, the control of incoming light reaching photoreceptors takes place by means of retinomotor mechanisms involving cellular changes at the level of the retinal pigment epithelium and/or the photoreceptors themselves (Walls 1942; Ali & Wagner 1975).
On the basis of morphological, biochemical and functional criteria, photoreceptors are classified into two categories, the cones, sub-serving colour vision at high light levels and the rods, mediating vision at low light levels (Walls 1942; Cohen, 1972). Most diurnal fish retinae are duplex, having both cone and rod photoreceptors. Fish possess several sub-types of cones, classified by their size, association with neighbours (double cones), and the region of the spectrum to which their visual pigments are maximally sensitive. The function of multiple classes of cones is to provide greater hue discrimination in colour vision. The types, densities and sizes of fish photoreceptors can be related to the habitat and/or period of activity of the species. For example, nocturnal species or those inhabiting the dark world of the deep-sea have a rod-dominated retina. Species that are primarily active around dawn and dusk have large cones that increase the probability of photon capture at low levels of light (Lythgoe, 1979). Many coral reef species possess high densities of small cones for increasing spatial resolution, allowing them to feed on plankton (planktivorous) during the day (Shand, 1997).
Two members of a double cone are fused along their inner segment and may or may not possess identical visual pigments in their outer segments. The function of double cones is not understood but they are thought to be involved in contrast and movement detection, or to increase sensitivity over a broader spectral range (Shand et al. 2001; Cummings & Partridge 2001). The arrangement of double and single cones within the retina can also be highly variable ranging from random placement to precisely organised rows or square mosaics (Lyall 1957; Engstrom 1963). Many teleost retinae mosaics consist of four double cones forming a square around one single cone, with an additional four single cones occupying the corners of the double cones.
7 Chapter I - Introduction
In addition, regional specialisations in photoreceptor and ganglion cell density are correlated with the main visual axis (direction of view) and can be related to feeding behaviour (Collin, 1997; Shand et al., 2000). For example, some planktivorous species, including syngnathids, require high acuity and possess a fovea. A fovea is a specialised region in which the ganglion cell and inner nuclear layers may be laterally displaced to allow light direct access to cone photoreceptors exposed within a depression in the retina (Fig. 6); an equivalent structure is present in the human retina but is most pronounced in birds of prey where acuity is paramount. Vision in the Syngnathidae has been explored only to a limited degree with some studies on the structure of the fovea. A fovea has been found in the genera Hippocampus (Carriere 1885; Krause 1886; Slonaker 1897; Verrier 1928; Kahmann 1934; 1936) and a deep pit fovea in the pipefish, Corythoichthyes paxtoni (Collin & Collin 1999).
Figure 6. Schematic representation of a fovea, a specialised area of the retina. If found in the ventro-temporal retina, it provides the highest acuity in frontal regions of the visual field. (Modified from Kandel, Schwartz & Jessell , 2000).
Colour sensitivity of photoreceptors
Visual pigment absorption characteristics are governed in two main ways: by the amino acid sequence of the transmembrane protein (the opsin), and/or whether the chromophore that binds the opsin is retinal (vitamin A1-based) or 3,4- didehydroretinal (vitamin A2-based), and changes in either moiety affect visual pigment spectral sensitivity (for a review, see Loew 1995). The visual pigment molecule selectively absorbs light and therefore appears coloured. The majority
8 Chapter I - Introduction of diurnal marine fish possess a single, short wavelength (“blue”) sensitive cone and double cones that are sensitive to medium (“green”) or long (“red”) wavelengths. Shallow-water species may contain a second smaller type of single cone sensitive to ultra-violet wavelengths. It is the combination of specific visual pigment types that forms the basis for the colour vision capacities of fish, as for all vertebrates.
The colour or depth of water inhabited by a species has consequences for the functioning of the visual system. Visual ecologists have investigated the variation in spectral sensitivities in aquatic organisms inhabiting different optical waters (Wald et al. 1955; Denton & Warren 1956; Lythgoe 1972, 1979; Bowmaker et al. 1994; Lythgoe et al. 1994; McDonald & Hawryshyn 1995; Cummings & Partridge 2001; Parry & Bowmaker 2000; Shand et al. 2001; Shand et al. 2002). Generally, a pattern of spectral sensitivities correlating with the photic environment has emerged (Lythgoe 1966, 1968; McFarland & Munz 1975; Loew & Dartnall 1976; Loew & Lythgoe 1978; Munz & McFarland 1973, 1977; Bowmaker 1995). Species-specific differences have been reported in habitat usage at microhabitat scales accounting for differences in photoreceptor peak absorbance among closely related sympatric species (Cummings & Partridge 2001). Visual communication may coevolve with spectral sensitivities. Marshall et al. (2003a) found that yellow and blue colours on their own are strikingly well matched to backgrounds on the reef such as coral and bodies of horizontally viewed water. These colours were viewed as important in camouflage as well as conspicuousness. The spectral characteristics of fish colours were found to be correlated to the known spectral sensitivities in reef fish single cones and were tuned for maximum signal reliability when viewed against known backgrounds (Marshall et al. 2003b).
It is now recognized that the colours absorbed by visual pigments in fish double cones are sensitive to the colour of the surrounding water (Bowmaker et al. 1994; Lythgoe et al. 1994). For example, freshwater and estuarine species, such as the black bream Acanthopagrus butcheri, live their larval stage in shallow water and possess short wavelength-absorbing cones which may aid the detection of plankton. When they mature to their adult stage, they move to
9 Chapter I - Introduction deeper, tannin-stained water with lower light intensity. This tannin-stained water transmits only longer wavelengths of light (red) whilst filtering out shorter wavelengths. In these adults maximum sensitivity (λmax) of their double cones shifts towards longer (red) wavelengths as the fish migrates from surface to deeper tannin-stained waters (Shand et al., 2001). The shift is accompanied by a change in feeding strategy from planktivory to feeding from the substrate (Shand et al. 2002). The shift is underpinned by a change in opsin expression (the amino acid sequence of the trans-membrane protein of visual pigments, Shand et al. 2002). In contrast, oceanic species inhabiting water that predominantly transmits blue wavelengths of light possess short wavelength-shifted visual pigments in the double cones (Lythgoe and Partridge, 1989; Lythgoe et al., 1994). Deep-sea oceanic species that lack cones possess rod visual pigments absorbing at shorter wavelengths, increasing sensitivity to the predominantly short wavelengths of dim down-welling light at depth (Douglas et al., 1998b).
Investigations of fish with cones have found that in each species studied the single cones have shorter wavelength (blue or ultraviolet) absorbing visual pigments than the double cones and fulfil the task of contrast detection. In addition to the sensitivity of cones to the surrounding water, prey detection against the optical background has been found to correlate with the λmax value for fish cones. Their location in the retina and the λmax of each member of the double cone in relation to each other has also been found to be important for predators in the detection of prey and its ability to contrast it against the background (Collin et al. 2003; Marshall et al. 2003b; Shand et al. 2002). Combinations of visual pigments have been found that cause an object to appear bright or dark against the background light. However, the precise placement of the visual pigments
λmax within the spectrum is determined by the water type in which the fishes live and the visual tasks of the species. Several authors have discussed the advantages of ‘offset’ visual pigments in contrast enhancement (Lythgoe 1972; Partridge 1990; Marshall et al. 2003b). Cone types that have visual pigments offset from surrounding colour of the light are implicated in the detection of contrasts. In general, ‘offset’ visual pigments increase contrast for grey targets where a target is seen against a darker background, i.e. when the predator is looking downwards to its prey (Partridge 1990). Conversely, where a dark target
10 Chapter I - Introduction is seen against a lighter background (i.e. when the predator is looking upwards at its prey against the down-welling sunlight), a visual pigment ‘matching’ the ambient spectrum will maximize contrast.
The potential for UV vision in the form of UV-sensitive cones was first discovered in fish in 1983 (Avery et al. 1983; Harosi & Hashimoto 1983). The perception of UV patterns has been reported in many surface-living larval species (Bowmaker 1990; Losey et al 1999; Loew et al. 2002). In some species, such as the yearling trout which feed on plankton near the water’s surface (Douglas et al. 1989), UV-sensitive cones are a major component of the receptor mosaic suggesting good acuity at the UV and short wavelengths. The fish lose UV- sensitive cones with age as they move down in the water column to feed on small invertebrates (Bowmaker & Kunz 1987). The transmission spectra of 2 zooplankters commonly eaten by larval trout, perch and sunfish, both absorb substantially in the near-UV and would be expected to appear darker against their UV and/or blue coloured background. The potential for ‘colour’ differences can occur as planktonic prey scatter light and can appear darker or lighter against their background depending on illumination direction, shape and refractive index differences (Muntz 1990).
Development of the fish retina
During early larval stages, the retinae of many marine fishes contain cones but no rods, and in a number of species, these are single cones. Double cones are formed at later stages of larval life and rods are first differentiated around the time of metamorphosis from larval to juvenile stages, when fin rays form and body pigmentation occurs (Blaxter & Jones, 1967; Blaxter & Staines, 1970; Ahlbert 1973; Evans & Fernald, 1993; Pankhurst & Eagar, 1996; Shand et al., 1999). The formation of double cones and rods, together with an increase in eye size, improves visual sensitivity and allows exploitation of deeper waters (Shand et al., 1999). However, not all species follow a similar sequence of photoreceptor differentiation and inter-specific differences can be related to the ecology and behaviour of the early life history stages (Shand, 1997; Shand et al., 1999). The
11 Chapter I - Introduction visual pigments can also change during development in line with changing habitat and feeding behaviour (Shand et al., 1988; Shand, 1994; Shand et al., 2001).
The Syngnathidae
The Syngnathidae (Fig. 7), commonly referred to as seahorses, pipefish, pipehorses and seadragons, are a large family of marine teleost fish that are distributed world-wide in shallow-water marine and estuarine habitats, although there are also both freshwater and pelagic species (Pollard, 1984a; Dawson, 1985; Gordina et al. 1991). Half of the world’s species of Syngnathidae are distributed around Australia’s temperate and tropical shores comprising at least 14 species of seahorse, 90 species of pipefish and two endemic seadragons. Members of all four groups are important components of temperate sea grass beds; in addition seahorses and pipefish are represented in tropical coral reef and mangrove habitats. A high proportion of species that occur in temperate Australian waters are endemic (Dawson 1985; Lourie et al., 1999a; Kuiter 2000). The 2 species of seadragons are endemic to temperate Australian waters, the weedy (Phyllopteryx taeniolatus) and the leafy (Phycodurus eques), and have taken camouflage to extremes with leafy appendages mimicking seagrass and kelp in which they live.
The seadragons and some species of seahorse and pipefish are listed on the 2000 IUCN (World Conservation Union) Red List of Threatened Species, including 2 of the species studied in this thesis, namely Hippocampus barbouri and H. subelongatus. All seahorses have recently been listed on Appendix II of the Convention in International Trade in Endangered Species of Wild Fauna and Flora (CITES), which was implemented on the 15th May 2004 (Sarah Foster pers. comm.). All members of the syngnathid family are under threat from exploitation for traditional Chinese medicine (TCM), the aquarium trade and curiosities. In addition, these fishes are frequently taken in bycatch of recreational and commercial trawl fishing, and are vulnerable to destruction and degradation of their coastal habitats (seagrass beds, mangroves, corals and estuaries).
12 Chapter I - Introduction
The family Syngnathidae are generally small, crypto-benthic, highly specialised species that feed diurnally, and rely on their vision to locate and capture prey (Kuiter 2000). The skeleton of all syngnathids is modified to form many semi-rigid dermal plates that encase the body and limit flexibility, while the jaws are fused to a tubular snout (Branch 1966). The fish possess small, undulating fins (Consi et al., 2001) and are weaker swimmers than most teleosts. Particular fins may be absent and those species with a reduced or absent caudal fin may possess a prehensile tail. They generally conform to one of four body forms that are commonly referred to as pipefish, seadragons, seahorses and pipehorses, differing primarily by the angle of the head in relation to the body (Dawson, 1985; Lourie et al., 1999a; Kuiter 2000). The most divergent forms are the pipefish, which has a head that is in line with the horizontally or vertically orientated body, and the seahorse, whose head is angled ventrally at ca 90o to the vertically orientated body (Fig. 7). They may possess extra skin filaments to imitate kelp or seaweed, such as in the weedy and leafy seadragons, or imitate fronds of seagrass, such as the vertical resting position adopted by pipefish. Some species also become encrusted with bryozoans and algae. Slow, non- abrupt movements also enhance their ability to remain unseen.
Figure 7. A. Leafy seadragon (Phycodurus eques) length to 350mm (photo by N. Conlon); B. Weedy seadragon (Phyllopteryx taeniolatus) length to 450mm; C. Brushtail pipefish (Leptoichthys fistularius), length to 650mm; D. Zebra-snouted seahorse (Hippocampus barbouri), length to 150mm (photographs by Dennis Sarson). Note the differences in relative snout lengths, and position and structure of eyes.
13 Chapter I - Introduction
Syngnathids have unique reproductive characteristics. There is evidence that seahorses are site faithful (establishing small home ranges), and have elaborate courtship rituals (Vincent & Sadler 1995; Moore 1999). Fecundity is low for a fish with as few as 100-300 eggs being laid at one time, compared with millions in fish such as cod. The clutches of eggs produced by female syngnathids are brooded by the males, either by attachment directly to their ventral surface, tail or by enclosure within a pouch, the complexity of which varies between species (Herald 1959). Males then contribute to the nourishment of the brood via a ‘pseudo-placental’ vascularisation of the brood area (Berglund et al. 1986; Carcupino et al. 1997). When released from the pouch, the young are then entirely independent and hatch with well-developed vision, displaying immediate feeding capabilities and several specific, intricate behaviours in predation and predator avoidance. The high degree of male parental care means that the potential reproductive rate of females may exceed that of males and thus syngnathid species may be sex-role reversed, such that females possess secondary sexual characters and actively compete for mates (Clutton-Brock & Vincent 1991; Vincent et al. 1992; Pagel 2003). Syngnathid species vary in the degree of sexual dimorphism, an aspect which is likely to be related to the intensity of sexual selection, and monogamous, polygynandrous and polyandrous mating systems have been identified (Kvarnemo et al. 1999; Jones & Avise 2001). Some monogamous species have protracted and complex social rituals associated with male recognition and pair-bond affirmation (Gronell 1984; Vincent & Sadler 1995; Masonjones & Lewis 1996; Moore 1999).
Feeding is restricted to prey of specific sizes in different species due to the jaw structure being highly specialised, with jaws fused into a conical snout, adaptations in snout length and a very limited gape (Fig. 7) (Paulus, 1994a; Teixeira & Musick, 1995). There is partitioning of space and food resources within their ecosystem as a consequence of their uniquely adapted mouth parts along with the specificity of their prey (Paulus 1994b). Previous research has also shown that maximum gut fullness of pipefish species occurred when the sun was directly overhead, indicating that foraging success was best under high light and constrained by low light. However, some species have the ability to capture prey under low light conditions (James & Heck 1994) and there is at least one
14 Chapter I - Introduction nocturnal species, Hippocampus comes (Lourie et al. 1999a; Perante et al. 2002) indicating that crepuscular and nocturnal foraging may be more common in the Syngnathidae than currently recognised.
Syngnathids are known to use highly mobile, independent eye movements to locate, track and capture individual fast moving prey (Steffe et al, 1989, Campolmi et al, 1995; Collin & Collin 1999). The fish use independent eye movements and utilise a stationary or saltatory (stop and start) ambush foraging strategy and swallow their prey of small crustaceans, juvenile fish, copepods and mysids with a very rapid inhalant action that is completed within 5-7 ms of the onset of the strike, making these species among the fastest feeding fishes reported to date (Bergert & Wainright, 1997, Flynn & Ritz, 1999). Adult seahorses and pipefish can swallow thousands of small, whole prey in one feeding session with several sessions occurring per day, e.g. Stigmatopora argus (spotted pipefish) was found to have up to 1,200 prey items in their gut (Kendrick 2002). Furthermore, there are changes in prey choice during growth (Diaz-Ruiz et al., 2000; Kendrick 2002). At hatching and with development, species occupy differing habitats, some moving to surface waters whilst others take refuge in algal cover (Kuiter, 1988). Syngnathids are also masters of camouflage with some species having the ability to change skin colour to blend with their environment.
Vision in syngnathids
The important role that vision plays in syngnathid behaviour patterns, including tracking prey, feeding with great accuracy, changing skin colour for camouflage, and also locating mates, indicate a highly specialised visual system at all stages of development (Ryer & Boehlert 1983; Howard & Koehn 1985; Tipton 1987; Bennet & Branch 1990; James & Heck 1994; Bergert & Wainright 1997; Collin & Collin 1997; Flynn & Ritz 1999). Although syngnathids exhibit precise, swift and abundant foraging behaviour which must rely on acute vision, surprisingly little is known of the visual system of these fishes and how this aspect relates to feeding and no data at all are available in relation to colour vision.
15 Chapter I - Introduction
As previously noted, members of the syngnathid family are predominantly active during the day and rely primarily on vision for the capture of prey (Paulus, 1994a; Kendrick, 2002). The young of most species hatch with relatively well developed visual systems, some feeding immediately thereafter, others relying on a yolk sac for nutrition for the first few days (Collin & Collin 1999; Kuiter 2000; Woods 2000a). For the first time a live birth of H. subelongatus was photographed in the wild during a night dive in the Swan River estuary (Fig. 8 - 17 Mar 2004). The young were seen to move towards the surface (possibly following moonlight), where they fill their swim bladders (to aid buoyancy control and move up and down in the water column). Although there are some studies on the structure of the fovea in selected syngnathids (see Collin & Collin, 1999), no investigations have been carried out to examine the visual pigments within the photoreceptors in either adults or juveniles.
Fig. 8. Live birth of H. subelongatus in the Swan River, Perth, 17 Mar 2004. Note: Juveniles immediately moving toward the surface. (Photo by Chris Cunnold).
16 Chapter I - Introduction
Rationale and broad aims of this study
In recent years a limited number of species of seahorse have been reared and are undergoing re-stocking trials in the Philippines and other Asian countries. Feeding juveniles has always been problematic both due to adequate sourcing of nutritious prey and more importantly prey detection of the juvenile and its survival in an artificial environment. For the majority of seahorses, seadragons, pipefish and pipehorses, however, breeding and rearing has yet to become successful on a small or large scale. Special permits have been issued to selected breeders to allow collection of egg-carrying male leafy and weedy seadragons from the wild. Nevertheless, aquariums worldwide, including Aquarium WA in Perth, have encountered problems maintaining the adults and rearing newly-hatched juveniles. Breeding and rearing have been inhibited by the inability to identify suitable lighting regimes for feeding, particularly for juveniles, and difficulties with establishing appropriate lighting systems for breeding, to enable the animals to experience their diurnal cycles and achieve successful egg transfer from female to male parent.
For successful rearing of any species it is imperative to provide a light environment suitable for an animal’s visual system to develop optimally and enable prey capture at different stages of maturation. It is also highly important for culturing prey, syngnathid mate selection and conditioning adults for breeding. However, to date there has been no investigation of the colour sensitivity of the visual system in syngnathids. This project is the first study to explore this, and the influence of the colour of ambient water and prey detection on the spectral position of the visual pigments will be investigated and discussed. Artificial light may impact upon diurnal cycles and hence these animals’ ability to achieve breeding condition, as well as prey detection, consequently influencing survival of syngnathid hatchlings in an artificial light environment.
This study:
• Examined 3 species of Syngnathidae in detail and one to a limited degree, investigating adaptations to their feeding requirements compared with ambient water colour of their natural habitat. Two local, temperate species were used, the spotted pipefish Stigmatopora argus and the West
17 Chapter I - Introduction
Australian seahorse Hippocampus subelongatus, both of which inhabit shallow coastal habitats with similar light environments (green water, seagrass beds) but feed on different prey items. One tropical species is used, the zebra-snouted seahorse Hippocampus barbouri, which inhabits the blue water, tropical coral reef environment.
• Investigated the structure of the visual system in mature adults of these species, with the aim of understanding the visual specialisations common to the group and specific to individual species. Eye dimensions and retinal structure are investigated to establish whether any specialisations for improving visual acuity such as a fovea, are present.
• Investigated spectral absorbance characteristics of the lenses and measured retinal visual pigments with a microspectrophotometer (MSP) to examine the potential for colour vision. The results are interpreted in the context of how the influences of habitat water colour and/or feeding regimes may influence the absorption charactersistics of the visual pigments. The influence of the colour of the water versus prey detection on the spectral position of the visual pigments, and possible changes concomitant with growth and variations in environment and diet will be discussed. It is hoped that this knowledge will aid the design of the light environment for aquaculture programs and the introduction of appropriate feeding regimes, and enhance the rearing success of the threatened species for both the aquarium trade and re-stocking programs.
18 Chapter 2 – Materials and Methods
Chapter 2. MATERIALS AND METHODS
Animals Three species of Syngnathidae were investigated in detail:
Stigmatopora argus (spotted pipefish) occur extensively along Australia’s south coast from central New South Wales to Shark Bay, Western Australia. They inhabit sheltered, green-water seagrass beds in bays and estuaries, depth range 1-8m. They are very fast horizontal swimmers for short bursts of around 20m, after which they seek camouflage in a vertical resting position in seagrass beds (Kendrick 2002). Their long slender tails wrap around seagrass for stability. These pipefish possess very long snouts compared with mouth gape and feed predominantly on mobile calanoid and cyclopoid copepods. Both calanoid and cyclopoid copepods are usually translucent (Jones & Morgan, 1994) and lengths range from 0.4–1.4mm and 0.4-0.7mm respectively (Kendrick 2002).
20mm
Fig. 9. A, Spotted pipefish (Stigmatopora argus) (Photo by Dennis Sarson) – Note very long snout length and position of eye. Body length to 280mm. B, Characteristically dense Posidonia sinuosa seagrass bed at Owen Anchorage, Cockburn Sound (depth ca 8 m).
Hippocampus subelongatus (West Australian seahorse) occurs in the temperate waters of south-western Australia. In early summer they migrate to the lower reaches of the tannin-stained Swan River in Western Australia where
19 Chapter 2 – Materials and Methods they release offspring to feed on crustacean prey (Kuiter 2000). The species has also been found to inhabit sheltered, green-water seagrass beds in coastal bays, depth range 1-25m (Lourie et al. 1999a) and may move to deeper water during non-breeding periods. They are slow swimmers covering an area of around 1m2 only per swim (Kendrick 2002). In the wild they feed on a characteristic mixture of crawling gammarid and caprellid amphipods and free- swimming smaller carid shrimp and mysids (Kendrick, 2002). Size of prey can be up to 5mm, and colour can be variable ranging from white to pale or brightly coloured.
20mm
Fig. 10. A, West Australian seahorse (Hippocampus subelongatus) (Photo by Dennis Sarson) – Note medium snout length and more dorsal position of the eye. Body length to 200mm. B, Distribution of S. argus (green) and H. subelongatus (yellow) in the temperate o cean of southern Australia.
Hippocampus barbouri (Zebra-snout seahorse) occurs in the Philippines and northern Indonesia and Malaysia (Lourie et al 1999). Population declines are evident based on substantial exploitation, decreased availability commercially, and habitat degeneration. H. barbouri is sold particularly in the aquarium trade, but also as a curiosity, and/or Traditional Chinese Medicine (Pajaro, unpubl. data). It is also a by-catch in other fishing gears used in its shallow habitat (Pajaro, unpubl. data). In addition to the substantial commercial demand for this species, its coral reef habitats are also threatened. The species possesses a spiny coronet and numerous thick
20 Chapter 2 – Materials and Methods spines including one in the front/nasal side of the eyes. Unlike most seahorses, it is often found clinging to hard Acropora-like corals at a shallow depth range of 6-10m (Kuiter, 2000). Information about the diets of H. barbouri in the wild is lacking although they have been successfully reared to juvenile age on calanoid copepods, and then to adult age on mysid shrimp (Mic Payne, pers comm.)
20mm
Fig. 11. A, Zebra-snout seahorse (Hippocampus barbouri) (Photo by Dennis Sarson) – Note similarly medium snout length and more dorsal position of eye, to Western Australian seahorse. Body length to 150mm. B, Distribution of H. barbouri (yellow) around islands of Indonesia, Malaysia and Philippines.
A fourth species was investigated to a limited degree due to reduced availability:
Pugnaso curtirostris (Pugnose pipefish) occurs along Australia’s southern coast from south Western Australia to the Bass Strait region. It occurs on algal reefs, often over large rubble sand, and on broad-leaved seagrass such as Posidonia and Amphibolis spp growing amongst low reef. Juveniles are often amongst decaying leaf litter (Kuiter 2000). They hunt small crustaceans on the substrate and mysids are also an important part of their diet (Kuiter 2000). The species are found in sheltered estuaries and bays to about 10m depth.
Fig 12. Pugnose 10mm pipefish (Pugnaso curtirostris) (Photo by R. Kuiter) – Note relatively wider and shorter snout. Body length to 170mm.
21 Chapter 2 – Materials and Methods
Seahorse and pipefish collection, holding and experiments
All experiments were conducted at The University of Western Australia. Experimental procedures were approved by the University of Western Australia Ethics Committee, approval number RA/3/100/122. and followed the guidelines of the National Health and Medical Research Council of Australia.
Pipefish, S. argus, were collected by trawling over meadows of seagrass Posidonia sinuosa (characterised by blade-like leaves clustered densely at the surface of the substrata – Fig. 9B) and Posidonia coriacea (characterised by narrower leaves and patchy coverage surrounded by unvegetated sand) (Kendrick 2002) on Success Bank, Cockburn Sound, Western Australia in February 2001, under licence from Fisheries WA (July 2000). They were also sampled by purse seine and then maintained in captivity under similar light intensities to their natural habitat. All S. argus were a typical brown spotted colour and were maintained with 95 lux illumination. A pregnant male collected, released its young in captivity one week after collection from the wild. Juveniles were separated from the adult male and were kept alive for 1-2 days.
Temperate seahorses, H. subelongatus, were sourced from a local breeder who had originally captured the adults from the wild. They varied in colour including grey, brown and white striped, yellow, orange and purple. They were maintained with 95 lux illumination.
Tropical seahorses, H. barbouri, were obtained from local aquarist and scientist, Dr Mic Payne, and maintained at a water temperature of 26oC under fluorescent lighting (Philips Coolwhite, 36 W) similar to natural habitat (light meter reading was 550-650 lux). In the wild this species can be white, yellow, brown and red-brown; the animals studied were cream/white. A pregnant male from the local aquarist, released its young in captivity and the young were donated to this study. Juveniles were maintained for varying periods from 1 to 30 days.
22 Chapter 2 – Materials and Methods
Pugnose pipefish, P. curtirostris, were collected from the wild by trawling over meadows of seagrasses, P. sinuosa and P. coriacea on Success Bank, Cockburn Sound, Western Australia in February 2001, under licence from Fisheries WA (July 2000). They were also sampled by purse seine and then maintained in captivity under similar light regimes (at 95 lux) to their natural environment. All animals were a dark brown colour. A pregnant male collected, gave birth in captivity one week after collection from the wild. Juveniles were separated from father and maintained for 43 days.
In an attempt to ensure necessary nutrients were supplied, fish were fed a mixed diet of just hatched Artemia nauplii, live adult Artemia enriched on microalgae Nannochloropsis oculata, Isochrysis galbana and Chaetoceros gracili, and frozen Artemia and mysid shrimp (See Kendrick 2002 for dietary compositions of wild animals).
Fish were maintained in the University of Western Australia’s recirculating marine aquarium system. Each species was held in separate tanks receiving gentle aeration and were subject to a 12L:12D cycle. Seawater was filtered to 5μm, of 30g/l salinity, ambient temperature was around 21.5oC – 23 ± 0.5oC and a pH was 8.22 at all times unless otherwise noted. Daily water exchange in each tank occurred constantly. Substrate for seahorses was provided in each tank in the form of polypropylene rope weighted down with bleached coral. Detritus was siphoned from the bottom of each tank daily. Water quality in the system was measured twice per week. Dissolved oxygen, pH, ammonia and nitrate measured 95-99%, 8.1-8.3, <0.1mg/l and <0.1mg/l, respectively, throughout the experiments. A UV water sterilisation unit is fitted to the recirculation system.
Five individuals of each of S. argus and H. subelongatus, and 11 of H. barbouri were used for MSP within 3 weeks of procurement (lenses from these animals were removed and retained for lens transmission measurements). Two individuals each of H. subelongatus and H. barbouri were used to prepare cryosections sections and one individual of H.
23 Chapter 2 – Materials and Methods subelongatus was used for semi-thin sections. Two juveniles of S. argus, H. barbouri and one of P. curtirostris, were used for MSP. See table 1A.
Table 1A: Numbers of animals of each species and life stage used for each procedure.
Number of animals used Retinal Lens Cryosections morpholoy transmissions for ocular from & MSP scans Species name - life stage dimensions histology (total 1248) S. argus - adults 0 0 5 S. argus - juveniles 0 0 2 H. subelongatus - adults 2 1 5 H. subelongatus - juveniles 0 0 0 H. barbouri - adults 2 0 11 H. barbouri - juveniles 0 0 2 P. curtirostris - juveniles 0 0 1
Morphology
Fish were terminally anaesthetised by immersion in a lethal concentration of methanesulfonate (MS222, Sigma-Aldrich, Pty, St. Louis, MO; 1:2000 in seawater). Fish length (snout to base length)(BL), head length (HL), snout length (SL) and eye diameter (ED) were measured with callipers prior to eye removal. Lenses were removed and placed in phosphate-buffered saline (PBS) solution, (pH 7.4) for determination of colouration. Retinae were fixed in Karnovsky’s fixative (2.5% glutaraldehyde, 2% paraformaldehyde, 2% sucrose in 0.1M cacodylate buffer (pH 7.2)), and post-fixed in 1-2% osmium tetroxide. Retinae were embedded in epoxy resin following the procedures of Shand et al., (1999). Serial semi-sections were cut from nasal to temporal regions and stained with Toluidine blue and viewed in a light microscope, and photographed. Retinae from hatchlings was not prepared for morphology, however, the presence of photoreceptors was established from MSP (page 27).
Pupil Response
To determine whether a pupil light response is present in any of our study species, pupillary constriction in response to an increase in ambient illumination was tested. Fish were dark-adapted in a light-proof box for 1hr,
24 Chapter 2 – Materials and Methods before being transferred to the transparent container. They were filmed using the 'nightshot' facility of a SONY DCR-PC100E digital handycam. The inbuilt light source was not used as it is too bright, and 'red', and would allow the fish to detect it. Instead an infrared (IR) light source from the darkroom was used. To approach the eye, a +4 filter was placed in front of the camera lens. All filming was conducted by looking down the viewfinder, rather than using the camera's LCD screen. Fish were initially filmed in the dark and then exposed to a 'white' tungsten light source (headtorch), which was shone onto the animal. All filming was done over a 10 minute period using the Nightshoot facility.
Measurement of ocular dimen sions from cryosections
Whole eyes of each species were snap-frozen by immersion in a mixture of dry ice and ethanol. The technique reduces the different effects of fixation and embedding artefacts o n the tissues of the eye (Sivak, 1978; Shand et al. 1999). Rupturing of the eye cup due to increased intraocular pressure during freezing was not observed. The frozen eye was then mounted on a sledge microtome, fitted with a freezing stage (Peltier heat-exchange device), using 30% sucrose in phosphate buffer for cryoprotection. The eyes were mounted and oriented to facilitate sectioning in either the vertical or horizontal plane and viewed with an Olympus binocular microscope mounted above the microtome stage. The whole eye was frozen and cryosections were cut from nasal to temporal, or dorsal to ventral regions to locate the presence and position of the fovea (Shand et al., 1999). Following the removal of each section (10 µm in thickness), photographs of the remaining block face were taken using an Olympus camera mounted on the microscope. Photographic enlargements of the region of the eye with the greatest lens diameter (the geometric or optical axis) were used for measurements of eye dimensions. The distance from the centre of the lens to the centre of the retina was measured along the horizontal axis: photographs of the eye cut in the horizontal plane providing nasal, temporal and optical axes (example, Fig. 13). Matthiessen's Ratio was calculated by dividing the focal length by the lens radius (Fig. 13).
25 Chapter 2 – Materials and Methods
radius lens focal length Fig. 13. Cryosection, (horizontal focal length plane) of the eye of a typical teleost, the black bream Acanthopagrus butcheri. In most teleosts the Matthiessen’s Ratio (MR = focal length divided by lens radius) is 2.5. retina Scale bar = 200 µm.
Lens transmission spectra
Ocular lenses were removed from the eye, placed into PBS, pH 7.4 and retained for transmission measurements. For measurement, lenses were placed in a cuvette (Eppendorf UVette) in PBS. The lens was orientated in the v-shaped bottom of the cuvette, adjacent to the wall of the cuvette furthest from the light source and such that light was transmitted along the natural optic axis of the lens. Light from a Xenon flash light (Ocean Optics PX-2) was transmitted to the cuvette via a 600 micron diameter fibre-optic (Ocean Optics P600-025-SR), a variable attenuator (Ocean Optics FVA-UV), a 400 micron fibre-optic (Ocean Optics P400-025-AL-SR) and fused silica collimating lens (Ocean Optics 74-UV). In order to restrict the diameter of the collimated beam to less than that of the ocular lens, the light was directed to the ocular lens via a small aperture made in aluminium foil held between the silica lens and the cuvette. To reduce the effect on spectral measurements of scattering and focusing by the ocular lens, light transmitted by it was collected with a cosine corrected fibre optic irradiance probe (Ocean Optics CC-3) mated to a 600 micron diameter fibre-optic (Ocean Optic P600-025-SR) which was terminated in the entrance port of a miniature spectrometer (Ocean Optics USB2000; 100micron entrance slit). Measurements were made with Ocean Optics OOIIBase32 software and transmission spectra recorded over the range 300-
26 Chapter 2 – Materials and Methods
800nm using the empty cuvette to obtain a reference (tissue free) spectrum. Data were averaged for ca. 30 repeated measurements of each lens, were binned into 1nm bins, smoothed with a 21-nm running average, and normalized for display.
Microspectrophometry (MSP)
Tissue Preparation
Fish were dark-adapted for at least 2 hours before being euthanased by immersion in methanesulfonate (MS-222; Sigma-Aldrich Pty, St. Lois, MO; 1:2000 in seawater), and entire MSP preparation and scanning procedures were performed in darkness. Retinae were dissected under infrared illumination with the aid of an image-converter (FJW Industries USA) to remove and hemisect the eyes. Lenses were retained for transmission measurements and placed into PBS solution, pH 7.4. Sections of unfixed retinal tissue were teased apart to detach individual photoreceptors from other retinal tissue in teleost Ringer solution. Excess Ringer solution was removed by applying a piece of filter paper to the edge of the droplet. Dextran (Sigma 250k RMM) diluted to 10% in teleost Ringer solution was added on a rectangular 50 x 22 mm No.1 coverslip (to increase the viscosity of the solution and reduce cellular movement). The preparation was covered with a smaller (19mm2) No. 1 coverslip and sealed with nail varnish to prevent dehydration. Several preparations were made at a time, and those not for immediate use were stored in a lightproof aluminium container in a refrigerator at approximately 4oC.
The absorbance characteristics of individual rod and cone photoreceptors, including both partners of double cones, were measured using MSP. Hatchlings of egg-bearing male H. barbouri, S. argus and P. curtirostris were used to establish the existence and range of retinal photoreceptor cells to compare with adults. MSP investigations were performed discovering the presence of rods, single cones and double cones in adults, and rods and double cones in the juveniles.
27 Chapter 2 – Materials and Methods
Measurement of visual pigment absorbance spectra
A single-beam wavelength-scanning microspectrophotometer was used to measure the absorption characteristics of the photoreceptor outer segments. The MSP procedure for use and analysis has been described previously by Shand et al. (2002). Briefly, spectral absorbance measurements were made by placing the outer segment of individual photoreceptors in the path of the measuring beam and scanning over the wavelength range 350 to 750 nm. Data were recorded at each odd wavelength on the ‘downward’ long- wavelength to short-wavelength spectral pass and at each interleaved even wavelength on the ‘upward’ short-wavelength to long-wavelength spectral pass. One sample scan was made of each outer segment combined with two separate baseline scans from an area adjacent to the outer segment being scanned. The two absorbance spectra obtained were averaged to improve the signal-to-noise ratio of the absorbance spectra used to determine the λmax values. Following these ‘pre-bleach’ scans the outer segment was then bleached with white light from the monochromator for 2-4 mins and an identical number of sample and baseline scans were subsequently made. The post-bleach average scan was subsequently obtained and deducted from the pre-bleach average to produce a difference spectrum for each outer segment and confirm the presence of photolabile visual pigment. Photoreceptor dimensions were measured for each cell scanned.
Analysis of the visual pigment absorbance spectra
Baseline and sample data were converted to absorbance values at 1nm intervals and the upward and downward scans were averaged together by fitting a weighted three-point running average to the absorbance data (Hart et al., 2000). Absorbance spectra were normalised to the peak and long- wavelength offset absorbances, obtained by fitting a variable-point unweighted running average. Following the method of MacNichol (1986), a regression line was fitted to the normalized absorbance data between 30% and 70% of the normalized maximum absorbance at wavelengths longer than that of the absorbance peak. The regression equation was then used to find the
28 Chapter 2 – Materials and Methods wavelength of 0.5% absorbance, which was used to predict the wavelength of maximum sensitivity (λmax) and fit the visual pigment template following the methods of Govardovskii et al. (2000). Acceptable pre-bleach spectra (Levine and Mac Nichol, 1985; see Partridge et al., 1992) had a characteristic ‘bell- shaped’ curve with a clear alpha peak, low noise and flat long wavelength tail above the wavelength at which the absorbance had fallen to less than 0.5% normalised maximum absorbance and those of similar λmax were averaged and reanalysed to create mean absorbance spectra for each category of visual pigment. For display, averaged spectra were overlain with an A1 visual pigment template of the same λmax, generated using the equations of Govardovskii et al. (2000).
29 Chapter 3 - Results
Chapter 3. RESULTS
External morphology, behaviour and orientations of the body
Adults
Stigmatopora argus appeared to mimic seagrass when resting by wrapping its tail around and floating with both its body and head in line in a vertical orientation. As indicated in Figs. 14A & B, other photographs captured on film and personal observations of behaviour in the tank, there were many orientations of the head as the fish moved around the tank. When pursuing its prey, it was observed to be an energetic and accurate feeder, moving into several orientations quickly to pursue prey and optimise feeding (Fig. 14A & B). Its eyes moved independently and switched quickly from dorsal to ventral binocular vision, with the animal having obvious capabilities to see prey or predators approaching from above or below.
20mm
20mm
Figs. 14 A & B. Stigmatopora argus with head held in typical positions - animal wrapped tail around seaweed which it mimicked. Note: Independent eye movements as it tracks prey. Note also the possibility of binocular vision in ventral field of view.
S. argus lacked irideal flaps (Fig. 15) and did not exhibit a pupil response after maintenance in the dark and then exposure to bright light. Ocular dimensions from frozen sections were not obtained for S. argus.
30 Chapter 3 - Results
Dorsal
Fig. 15. The eye of pipefish, S. argus. Note: Lacked irideal flaps but pigment is located in the dorsal and ventral positions of the iris.
Ventral
Hippocampus subelongatus had independent eye movements and binocular vision in frontal and ventral areas (Fig. 16 & 17). Its head was angled at ca 90o to the vertically orientated body whilst attached to seagrass. H. subelongatus possessed irideal flaps which may direct the light to the ventro- temporal position of the retina (Fig. 19). No pupil response was noted when the animal was transferred from darkness to bright light.
Fig. 16 . Hippocampus subelongatus Fig. 17. H. subelongatus using showing binocular vision in the frontal binocular vision to track swimming visual field as it looks at the camera. prey, Artemia. 31 Chapter 3 - Results
Fig. 18. Typical position of head Fig. 19. Close up of H. as this seahorse wraps its tail subelongatus eye. Arrow shows around seaweed. main visual axis with light from above shaded but frontal visual field is unobscured. Note: irideal flaps in dorsal and ventral position.
H. barbouri also displayed binocular vision using both eyes to focus on its prey. Its head was angled ventrally at ca 90o to the vertically orientated body when swimming and feeding (Fig. 20 & 21). It possessed irideal flaps which may direct the light to the ventro-temporal position of the retina (Fig. 22b). No pupil response was noted when the animal was transferred from darkness and to bright light.
Fig. 20. H. barbouri wraps tail Fig. 21. H. barbouri - Indicating around Acropora-like corals typical position in which head is (although this is artificial held. seaweed in an aquarium). This photograph indicates head held in typical position.
32 Chapter 3 - Results
Fig. 22a & b. Close up of eye. Arrow shows clearly how the light from above is shaded by the irideal flaps from the frontal field and directed to the ventro- temporal region of the retina.
Snout Length as a proportion of head length, and eye diameter as a proportion of snout length
Morphological dimensions for adult individuals of each species are represented in Table 1 and Figs. 23 & 24. S. argus body length was markedly longer compared to head length than the 2 seahorse species, but they also possessed a much longer snout length relative to head length at 0.78 (Table 1 & Fig. 23). Although H. barbouri’s head length was a smaller average size (17.56mm) compared with H. subelongatus (29.75mm), the ratio of mean snout length as a proportion of head length is very similar with H. subelongatus at 0.51, and H. barbouri at 0.50. H. subelongatus had a much larger eye diameter in contrast with the other 2 species (Table 1 and Fig. 24). Comparing mean eye diameter to mean snout length in each species, S. argus exhibited a much smaller ratio of 0.14, compared with the 2 seahorse species which were similar, H. subelongatus at 0.29, and H. barbouri at 0.33.
Table 1: Mean morphological dimensions of 3 main species as a comparison. Average Body length (BL), head length (HL), snout length (SL) and eye diameter (ED) (mm) ± SD (standard deviation). Snout length as a proportion of head length, and eye diameter as a proportion of snout length.
Species Ave BL Ave HL Ave SL Ave ED SL/HL ED/SL 168 27.29 21.14 2.89 S. argus 0.78 0.14 (n=7) ±13.02 ±1.15 ±0.88 ±0.27 126 29.75 15.08 4.42 H. subelongatus 0.51 0.29 (n=12) ±4.33 ±0.76 ±0.45 ±0.17 73 17.56 8.82 2.88 H. barbouri 0.50 0.33 (n=9) ±3.15 ±0.55 ±0.29 ±0.21
33 Chapter 3 - Results
Snout length vs head length
30
25
20 S. argus 15 H. subelongatus H. barbouri 10 Snout lengthSnout (mm)
5
0 0 10203040 Head length (mm)
Fig. 23. Morphological dimensions of 3 target species comparing mean snout length to mean head length. Data points are from individual fish.
Eye diameter vs Snout length
6
5
4 S. argus 3 H. subelongatus H. barbouri 2 Eye diameter (mm) diameter Eye 1
0 0102030 Snout length (mm)
Fig. 24. Morphological dimensions of 3 target species comparing eye diameter to snout length. Data points are from individual fish.
34 Chapter 3 - Results
Ocular Structure and Dimensions
Ocular dimensions were measured from cryosections cut in the vertical and horizontal planes for adults of H. subelongatus and H. barbouri. Following the techniques outlined in the methods no rupturing or distortion of the eye cup was observed. Photographs were taken at the region of the eye with the greatest lens diameter and used for ocular dimension measurements (Fig. 25 & 26). The retina was relatively thick with an obviously large vitreal gap between the lens and retina. Matthiessen's Ratio (calculated from the horizontal axis by dividing the focal length by the lens radius) was consistent for 4 individual eyes of each seahorse species (total of 8). For both seahorse species, H. subelongatus and H. barbouri, the value was found to be 3.7.
Dorsal
Vitreal gap
Lens 1.3mm
Retina Lens radius
Ventral
Fig. 25. Cryosection of Hippocampus subelongatus adult eye. Photograph is a sagittal cross-section taken at the region of the eye with the greatest lens diameter. Note: long focal length. Scale bar: 1mm.
35 Chapter 3 - Results
Dorsal
Lens 1cm
Optic nerve
Retina – Note ventral thickening compared with dorsal area.
Ventral
Fig. 26. Cryosection of Hippocampus barbouri adult eye. This micrograph is taken at the level of the optic nerve exiting the retina. At the region of the eye with the greatest lens diameter, cryosections of both seahorses’ eyes were cut in the vertical plane. Scale bar: 1mm.
Lens colour and size
The lenses of H. subelongatus and S. argus appeared a pale yellow colour while H. barbouri appeared almost colourless. Adult S. argus and H. barbouri possessed extremely small lenses of ca 1mm, and H. subelongatus a larger lens of 1.3mm (Figs. 27 & 28a, b).
Scale bar: 1mm Scale bar: 20mm
Fig. 27. Yellow, spherical lens of S. argus (1mm diameter).
36 Chapter 3 - Results
Fig. 28a & b. Note: Yellow, spherical lens of H. subelongatus (1.3 mm diameter), and almost colourless, spherical lens of H. barbouri – (1mm diameter)
Lens transmission spectra
Mean transmission spectra for ocular lenses of two individuals each of three species of syngnathids are shown in Fig. 29. Results of one individual lens per species are represented in the graph. The other lens of the same species yielded identical results. The extremely small lenses (ca. 1 mm diameter) of H. barbouri and S. argus introduced minor artefacts into the transmission spectra particularly at the extremes of the wavelength range that were not evident in measurements of the larger lens (1.3 mm diameter) of H. subelongatus. All lenses showed high transmissions at long wavelengths with a shortwave cut-off, these having wavelengths of 50% transmission at 428 nm for S. argus, 423 nm for H. subelongatus, and 408 nm for H. barbouri. All lenses displayed Class I ocular media transmission spectra (Siebeck & Marshall 2001), with H. barbouri possessing the most steep slope (<30nm between 0 and 100% transmission) resulting in a sharp cut-off. S. argus and H. subelongatus also displayed steep slopes from the 100% transmission point and tapered off to 0% from the 20-30% transmission point.
37 Chapter 3 - Results
Syngnathid lenses and mean transmissions
140
120
100
80
60 H.barbouri 40 H.subelongatus S.argus 20
norm a lise d tra nsm0 ission 350 400 450 500 550 600 650 700 -20
-40
-60 wavelength (nm)
Fig. 29. Calculated transmission spectra of the lenses of individuals of each 3 species. H. subelongatus and S. argus appeared a yellow colour. H. barbouri lens appeared almost colourless. All lenses showed high transmissions at long wavelengths with a shortwave cut-off, these having wavelengths of 50% transmission at 408 nm for H. barbouri, 423 nm for H. subelongatus and 428 nm for S. argus.
Pupil Response
When euthenasing the seahorse H. subelongatus, pupil dilation was thought to be detected when the animal died. However, testing pupillary constriction in response to increased ambient illumination, after the fish were dark adapted for 1 hour and filmed after exposure to bright light for 10 minutes, yielded no conclusive results in any individuals of the 3 species (one per species).
Retinal Structure
The retina of H. subelongatus and H. barbouri (2 individuals of each species) both possess a fovea located in a ventro-temporal position subtending the frontal region of the visual field in front of the elongated snout and mouth. The retina immediately surrounding the foveal pit was thickened (Figs. 30 & 31). As the eyes of these seahorses are highly mobile and can
38 Chapter 3 - Results move independently, they are able to scan the upper and lower regions of their frontal visual field, before fixating on and sucking in their moving prey.
Dorsal
Fig 30. H. subelongatus. Eye sectioned from nasal to temporal. Arrow indicates the position of the fovea located in the ventro-temporal region. Note localised thickening of surrounding retina. Scale bar: 0.5mm
Ventral
Fig. 31. H. barbouri - Eye sectioned from nasal to temporal. Arrow indicates the position of the fovea located in the ventro-temporal region. Note localised thickening of surrounding retina. Scale bar: 0.5mm
Dorsal
Ventral
39 Chapter 3 - Results
Histology
Examination of the H. subelongatus retina using light microscopy (Figs. 32-34) reveals a fovea densely packed with cones and devoid of rods in the ventro-temporal retina, providing highest acuity in frontal regions of the visual field. As we had already detected a fovea when examining the ocular structure via the frozen cryosection preparations, these areas of high cone domination (both single and double) were not unexpected and underlay the foveal invagination. The inner retinal layers were not dispersed laterally as is classically described for some vertebrate foveae, with all retinal classes of neurons lining the inner aspect of the foveal curvature. There was further detail of the photoreceptor differentiation into the ganglion cell layer (GCL), inner nuclear layer (INL) and photoreceptor nuclei were evident in the outer nuclear layer. The boundaries of all neuronal layers within the retina, including the inner (ILM) and outer limiting membrane closely followed the contours of the foveal invagination.
500µm
ON
GCL
INL 200µm
Fig. 32. Light micrographs of the seahorse, H. subelongatus, fovea. It was found in ventro-temporal retina, providing highest acuity in frontal regions of the visual field. There was a lack of rods beneath the retinal pit. Note the other retinal neurons, the inner nuclear layer (INL) and ganglion cell layer (GCL) are not displaced in this region but are greatly reduced in density. Optic nerve (ON). 40 Chapter 3 - Results
COS ROS
Dark stained rod nuclei
200μm
Fig. 33. Light micrograph of the seahorse, H. subelongatus, foveal and perifoveal retina. The fovea was densely packed with cones and lacked rods beneath the retinal pit (rod nuclei within the perifoveal region increase in number to the right of the photo and are depicted by the arrows). Note the marked difference in thickness of the non- foveal retina and high density of rods. ROS: rod outer segment; COS: cone outer segment.
TS
LS
50μm
Fig. 34. Light micrograph of H. subelongatus retina, area moving from the obliquely displaced, tangentially-sectioned (TS) cones in the fovea, to a perifoveal area where cones were longitudinally sectioned (LS) - direction of main arrow.
41 Chapter 3 - Results
Only cones are found under the foveal invagination (Fig. 33). Dark stained rod nuclei begin to appear at around 500μm from the base of the foveal pit. There was a marked decrease in all the neural layers in the foveal area. In the perifoveal region, each of the retinal layers increased to around double its peripheral thickness. The retina is 250μm in thickness within the foveal pit, increasing to around 500μm in the perifoveal region before falling again to approximately 210μm in the periphery (Figs. 32-33).
SC
DC
10μm
Fig. 35. Light micrographs of H. subelongatus perifoveal retina under higher power. Areas of retina with a high density of small single (SC) cones and longer double cone (DC) photoreceptors. Note absence of large rod nuclei.
There were areas of mixed single and double cones in the perifoveal retina (Fig. 35), as well as oblique displacement of cones in the region of the fovea, resulting in tangential sections of cones appearing in Figs. 33 & 34.
42 Chapter 3 - Results
Photoreceptors immediately beneath the foveal pit appeared to form retinal mosaics (RM) with each component of a double cone adopting an optimal hexagonal packing (Fig. 36).
RM 10μm
Fig. 36. Light micrograph of H. subelongatus foveal retina under higher magnification, showing prevalence of retinal mosaics (RM). Note that at this level, the square mosaic does not appear to be present with each component of a double cone adopting an optimal hexagonal packing.
Areas of rod domination occurred in the peripheral retina as well as adjacent to the optic nerve (Fig. 37 & 38). Very large, chunky outer segments and large rod nuclei were found.
ROSROS ROS
100µm 10μm Rod nuclei
Fig. 37. Light micrograph of H. subelongatus peripheral retina is dominated by large rod outer segments (ROS) (up to 33 µm in length).
43 Chapter 3 - Results
ROS
Rod nuclei ON
10μm
Fig. 38. Light micrograph of H. subelongatus retina adjacent to the optic nerve (ON). Note difference in thickness of the retinal layers and proliferation of large rods. Optic nerve (ON), Rod outer segments (ROS).
44 Chapter 3 - Results
The photoreceptors and their visual pigments
Adults
Visual pigment absorbance spectra meeting selection criteria (see Materials and Methods, which included the processing and vetting of around 1248 scans) were obtained giving rise to an average of at least 40 individual cone and 13 rod photoreceptors from each species. On the basis of goodness-of-fit of the mean absorbance spectra to mathematical visual pigment templates (Gardovskii et al. 2000), each of the pigments found in all species was a rhodopsin (vitamin A1-based visual pigments). All adults of the study species were found to possess rods, single short-wavelength sensitive (SWS) and long-wavelength sensitive (LWS) double cones (Figs. 41-46c). The dimensions of the outer segments of rods and cones for adults of each species are shown in Table 2 & 3. Additionally, MSP absorbance measurements of just-hatched S. argus, H. barbouri and Pugnaso curtirostris indicated the presence of rods, single SWS and double LWS cones (Figs. 48 & 49).
20 µm
Double cone
Rod
Fig. 39. Microspectrophotometry (MSP) preparation of photoreceptors of H. subelongatus, showing rod and double cone outer and inner segments.
45 Chapter 3 - Results
30 µm
Rod
Figs. 40a & b. Microspectrophotometry (MSP) preparation of photoreceptors of H. barbouri, showing the presence of very large rods and double cones.
8 µm Double cone
Rods
Rod outer segments were characterised in these 3 species by their very large size, being up to 26.67μm in length in S. argus and H. subelongatus, and up to 33.33μm in length in H. barbouri (Tables 2 & 3, & Figs. 39 & 40a) compared with other teleost species, such as the surfperch of up to 7μm (Cummings & Partridge 2001), and the black bream up to 8μm (Shand et al. 2002). Rod lengths varied greatly for S. argus and H. subelongatus and most extremely between 5 – 33.33µm. Regional differences were noted in the distribution of rod photoreceptors throughout the retina. The most pronounced difference in average measurements between the species is the average length of the rods of H. subelongatus being 20.48µm (Table 2), showing that the area sampled was dominated by long, chunky rods, e.g. near the optic nerve and the peripheral region (Figs. 37-38). Other areas throughout the retina were dominated by cones, e.g. fovea and perifovea (Fig.34-36).
46 Chapter 3 - Results
A single class of rod contained a medium wavelength-sensitive visual pigment with a mean wavelength of maximum sensitivity (λmax) at 502nm for S. argus (SD ±3.47nm, N=10), 500nm for H. subelongatus (SD ±2.40nm, N=8) and 499nm for H. barbouri (SD ±3.92nm, N=21). Frequency histograms of MSP measurements of rod outer segments and average absorption curves for the 3 species are represented in Figs 41 & 42.
Frequency histograms of rod λmax distribution show absorbances close to 500nm for all records.
Table 2. Mean dimensions of outer segments (OS) of 3 photoreceptor cell classes scanned for 3 species ± SD (standard deviation) (µm). Single short-wavelength sensitive (SWS) and medium/long-wavelength sensitive (M/LWS) double cones. Single Double SWS cones M/LWS cones Rods Length Width Length Width Length Width 4.08 1.33 5.37 1.33 12.13 2.08 S. argus ±1.10 ±0.0 ±1.03 ±0.32 ±7.73 ±0.83 (n=3) (n=22) (n=10) 4.67 1.67 6.49 1.49 20.48 3.10 H. subelongatus ±0.47 ±0.0 ±1.85 ±0.21 ±5.33 ±0.16 (n=7) (n=31) (n=8) 5.81 1.60 5.78 1.31 12.15 2.02 H. barbouri ±2.31 ±0.59 ±1.72 ±0.27 ±7.24 ±0.81 (n=16) (n=44) (n=21)
Table3. Range of cell dimension of 3 photoreceptor classes scanned for 3 species (µm). Sample sizes are as for Table 2. Single Double SWS cones LWS cones Rods Length Width Length Width Length Width S. argus 3.3 – 5.6 1.33 3.33 – 6.67 1.3 – 2.3 4 – 26.67 1.67 – 3.33 H. subelongatus 4.3 - 5 1.67 4 – 8.33 1.33 – 1.67 11.67 – 26.67 3 - 3.3 H. barbouri 4.3 - 5 1.67 4 – 8.33 1.33 – 1.67 5 – 33.33 1.6 – 3.33
47 Chapter 3 - Results
Stigmatopora argus
3
2
1 No. cells scanned n=5 0 450 475 500 525 550
Hippocampus subelongatus
3
2
1
No. cells scanned n=5 0 450 475 500 525 550
Hippocampus barbouri
9
8
7
6
5
4
3
2
No. scanned cells n=11 1
0 450 475 500 525 550 Wavelength m ax. sens itivity
Fig. 41. Frequency histograms of all MSP Rod samples of the peak absorbance wavelength λmax estimates included in the average absorbance files for each of the three species (1nm bins). Note the tightness of Rod distribution around 500nm.
48 Chapter 3 - Results
1.2
1 0.8
0.6 0.4 S. argus n=10 0.2 0 Normalised absorbance 350 450 550 650 750 -0.2
1.2
1
0.8
0.6
0.4 H. subelongatus n=8 0.2
0 Normalised absorbance 3 50 4 50 550 6 50 750 -0.2
1.2
1
0.8
0.6
0.4 H. barbouri 0.2 n=21
Normalised Absorbance 0 350 400 450 500 550 600 650 700 750 -0.2 Wavelength m ax sensitivity (nm )
Fig. 42. Average spectral absorbance curves (dots) for the Rod photoreceptors of adult animals of each of the three species. Best fitting template for a rhodopsin A1 based visual pigment (line) was used throughout.
49 Chapter 3 - Results
Cones
Both LWS double and SWS single cones were present in all adults studied. Mean dimensions of photoreceptor cell classes scanned from MSP preparations, for 3 species are presented in Table 2. All species had typically small SWS and slightly longer LWS cones. The range of cell diameters (Table 3) indicates a regional variation in the photoreceptors. Single SWS cones maintained a constant width for each species and a small amount of variation in length, whilst double LWS cone lengths had greater variation. Generally, thinner cells were found in the fovea and thicker cells in the periphery, concurring with results found in other species.
Mean λmax values and standard deviation values are presented in Table 4. The wavelength of maximum sensitivity of SWS single cones, MWS, M/LWS and LWS double cones (ordered into 20nm bins) are presented in Table 4. Frequency histograms of both single and double cone records for the 3 species are represented in Fig 43. Cone average absorption curves for S. argus are shown in Figs. 44a,b, for H. subelongatus in Figs. 45a,b, and for H. barbouri in Figs. 46a,b,c.
A class of SWS single cones in the blue close to 460nm occurred in all 3 species (Figs. 43, 44a, 45a, 46a). Temperate seagrass species S. argus and H. subelongatus and tropical seahorse, H. barbouri were found to have this SWS single cone at 457nm, 466nm and 461nm, respectively. The tropical seahorse, H. barbouri possessed an additional SWS single cone in the violet at 430nm (Table 4 and Figs. 43 & 46a). As seen in Fig. 43, SWS single cone scans discretely fell around 430nm and 460nm wavelengths.
The spread of LWS double cones was unusually broad and complex and did not fall into discreet classes. For this reason the scans were divided into 20nm bins and the average spectral absorbance curves fitted to the A1 template. The intermediate LWS cones which were slightly broader than the
A1 template, were overlain with the A2 visual pigment template to test for a better fit. However, the A1 template best fit all rod and cone data.
50 Chapter 3 - Results
Table 4. Mean λmax ± SD (standard deviation) for each species for each photoreceptor class. SWC=short wavelength sensitive single cone. LWS=long wavelength sensitive double cone (ordered into 20nm bins). Average optical density (OD) is given in brackets. Note: Low SD for all records and high Optical Density of Rod records.
Lambda max wavelength ± SD MWS double M/LWS double LWS double LWS double VS single SWS single 506-525 526-545 546-565 566-585 Rod Colour of Yellow- wavelength Violet Blue Green Green-Yellow Orange Orange-Red Species 457±1.3856 519±3.8604 532±6.1209 553±4.3133 577±6.4239 502±3.4703 (OD=0.0108) (OD=0.0172) (OD=0.0162) (OD=0.0185) (OD=0.0153) (OD=0.0290) S. argus n=3 n=8 n=7 n=2 n=5 n=10 466±3.8166 519±3.5668 534±2.9841 559±5.9469 582±0.0 500±2.3979 (OD=0.0193) (OD=0.0184) (OD=0.0164) (OD=0.0138) (OD=0.0269) (OD=0.0437) H. subelongatus n=7 n=11 n=13 n=6 n=1 n=8 430±4.8273 461±5.7620 519±4.4698 535±6.0703 556±5.5119 573±4.4548 499±3.9216 (OD=0.0153) (OD=0.0138) (OD=0.0136) (OD=0.0119) (OD=0.0104) (OD=0.0074) (OD=0.0414) H. barbouri n=3 n=13 n=11 n=24 n=6 n=3 n=21
51 Chapter 3 - Results
Stigmatopora argus
4
3 n=5 2
1 0 No. cells scanned 400 425 450 475 500 525 550 575 600
Hippocampus subelongatus
4
3
2 n=5
1 No. cells scanned cells No.
0 400 425 450 475 500 525 550 575 600
Hippocampus barbouri
4
3
2 n=11
1 No. cells scanned cells No. 0 400 425 450 475 500 525 550 575 600 Wavelength of max. sensitivity
Fig. 43. Frequency histograms of SWS Single and LWS Double/Twin Cone samples λmax values included in the average absorbance files for each of the three species. Histograms include measurements of cone outer segments only (1nm bins). The short-wavelength (below 500nm) records are obtained from single cones, while the longer wavelength records from the double cones.
52 Chapter 3 - Results
S. argus SWS 457nm
n=3 1.2
1
0.8
0.6
0.4
0.2
0
Normalised absorbance 350 400 450 500 550 600 650 700 750 -0.2
S. argus LWS 519nm
n=8 1.2
1
0.8
0.6
0.4
0.2
0 Normalised absorbance Normalised 350 400 450 500 550 600 650 700 750 -0.2
S. argus M/LWS 532nm
n=7 1.2
1
0.8
0.6
0.4
0.2
0 Normalised absorbance Normalised -0.2350 400 450 500 550 600 650 700 750 Wavelength max sensitivity (nm)
Fig. 44a. Stigmatopora argus - Average normalised spectral absorbance curves (dots) representing 1 SWS and 2 LWS Cones. Best fitting template for a rhodopsin A1 based visual pigment (line) was used throughout.
53 Chapter 3 - Results
S. argus LWS 553nm
n=2 1.2 1 0.8 0.6 0.4 0.2 0
Normalised absorbance 350 400 450 500 550 600 650 700 750 -0.2
S. argus LWS 577nm n=5 1.2
1
0.8
0.6
0.4
0.2
0
Normalised absorbance 350 400 450 500 550 600 650 700 750 -0.2 Wavelength max sensitivity (nm)
Fig. 44b. Stigmatopora argus - Average normalised spectral absorbance curves (dots) representing 2 LWS Cones. Best fitting template for a rhodopsin A1 based visual pigment (line) was used throughout.
54 Chapter 3 - Results
H. subelongatus SWS 466nm
n=7 1.2
1
0.8
0.6
0.4
0.2
0
Normalised absorbance 350 400 450 500 550 600 650 700 750 -0.2
H. subelongatus MWS 519nm
n=11 1.2
1
0.8
0.6
0.4
0.2
0 Normalised absorbance Normalised -0.2350 400 450 500 550 600 650 700 750
H. subelongatus M/LWS 534nm n=13 1.2
1
0.8
0.6
0.4
0.2
0 Normailised absorbance Normailised -0.2350 400 450 500 550 600 650 700 750 Wavelength max sensitivity (nm)
Fig. 45a. Hippocampus subelongatus - Average normalised spectral absorbance curves (dots) representing 1 SWS and 2 LWS Cones. Best fitting template for a rhodopsin A1 based visual pigment (line) was used throughout.
55 Chapter 3 - Results
H. subelongatus LWS 559nm
n=6 1.2
1
0.8
0.6
0.4
0.2
0
Normalised absorbance 350 400 450 500 550 600 650 700 750 -0.2
H. subelongatus LWS 582nm
n=1 1.2
1
0.8
0.6
0.4
0.2
0
Normalised absorbance 350 400 450 500 550 600 650 700 750 -0.2 Wavelength max sensitivity (nm)
Fig. 45b. Hippocampus subelongatus - Average normalised spectral absorbance curves (dots) representing 2 LWS Cones. Best fitting template for a rhodopsin A1 based visual pigment (line) was used throughout.
56 Chapter 3 - Results
H. barbouri SWS 430nm
1.2 n=3
1
0.8
0.6
0.4
0.2
Normalised absorbance 0 350 400 450 500 550 600 650 700 750 -0.2
H. barbouri SWS 461nm
1.2 n=13 1
0.8
0.6
0.4
0.2
0
Normalised absorbance 350 400 450 500 550 600 650 700 750 -0.2 Wavelength max sensitivity (nm)
Fig. 46a. Hippocampus barbouri - Average normalised spectral absorbance curves (dots) representing 2 SWS Cones. Best fitting template for a rhodopsin A1 based visual pigment (line) was used throughout.
57 Chapter 3 - Results
H. barbouri MWS 519nm
n=11 1.2
1 0.8
0.6
0.4
0.2
ormalised absorbance 0 N 350 400 450 500 550 600 650 700 750 -0.2
H. barbouri M/ LWS 535nm
1.2 n=24
1
0.8
0.6
0.4
0.2
0 350 400 450 500 550 600 650 700 750 -0.2
Wavelength max sensitivity (nm)
Fig. 46b. Hippocampus barbouri - Average normalised spectral absorbance curves (dots) representing 2 LWS Cones. Best fitting template for a rhodopsin A1 based visual pigment (line) was used throughout.
58 Chapter 3 - Results
H. barbouri LWS 556nm
n=6 1.2
1
0.8
0.6
0.4
0.2
0 Normalised absorbance 350 400 450 500 550 600 650 700 750 -0.2
H. barbouri LWS 573nm n=3 1.2
1
0.8
0.6
0.4
0.2
0 Normalised absorbance 350 400 450 500 550 600 650 700 750 -0.2 Wavelength max sensitivity (nm)
Fig. 46c. Hippocampus barbouri - Average normalised spectral absorbance curves (dots) representing 2 LWS Cones. Best fitting template for a rhodopsin A1 based visual pigment (line) was used throughout.
59 Chapter 3 - Results
Pairings
In all 3 species M/LWS double and twin cones were found between 515 and 580nm with up to 4 classes within this range (Table 4), with outer segments joined together. Double/twin cones that were obviously paired together in MSP preparations revealed identical and non-identical pairings occurring in at least 4 combinations in each species. These have been plotted with the mean λmax of the second member against that of the first member (Fig. 47) reflecting diverse combinations of visual pigments in all 3 species. A line of best fit was plotted for S. argus and H. barbouri where there seemed to be potential patterns forming. It revealed that for S. argus, cones with shorter LWS pigments paired with cones with longer LWS pigments. For H. barbouri, cones with shorter paired with shorter and longer LWS pigments paired with longer. There were fewer records for H. subelongatus and these revealed no particular trend. These patterns were very vague due to the small number of records and may produce clearer trends with further investigation.
Values of LWS double cone pairings were:
S. argus - 515/515nm (green/green), 520/570nm (green/yellow- orange/red), 520/580nm (green/red), 530/580nm (green- yellow/red).
H. subelongatus – 515/520nm (green/green), 520/560nm (green/yellow- orange), 530/530nm (green-yellow/green-yellow), 530/540nm (green-yellow/yellow).
H. barbouri – 520/530nm (green/green-yellow), 530/530nm (green- yellow/green-yellow), 530/540nm (green-yellow/yellow), 540/555nm (yellow/yellow-orange), 555/570nm (yellow- orange/orange-red).
60 Chapter 3 - Results
S. argus double cone pairs
600
580 R2 = 0.503951 560
540
520 wavelength (nm) pair 2nd Fig. 47. Graphical 500 representation of pairing 500 520 540 560 580 600 of LWS double/twin cone samples with λmax values of first member plotted H. subelongatus double cone pairs against λmax values of the 2nd member for each of the three species. 600
580
560 R2 = 0.337359
540
w avelength 2nd pair 520
500 500 520 540 560 580 600
H. barbouri double cone pairs
600
580
R2 = 0.82723 560
540
520 w avelength 2nd pair
500 500 520 540 560 580 600 w avelength 1st pair
61 Chapter 4 – Discussion
Chapter 4. DISCUSSION
The aim of this thesis was to investigate the structural and visual specialisations of the eyes and the photoreceptors of 3 species of Syngnathidae, using a range of methods. A fourth species was investigated to a limited degree. These enquiries included external morphological features, measurements of ocular dimensions, lens transmissions, and the absorptive properties of the photoreceptors. The following discussion will address the results and how they relate to natural behaviours and habitat of these fishes. Mechanisms whereby different visual pigments are sensitive to different light wavelengths will also be discussed. A final section will propose future work arising from the current study.
The characteristics of the eyes of S. argus, H. subelongatus and H. barbouri suggest they have evolved specialised visual systems. Although these fishes are relatively weak swimmers and possess fused snouts limiting their gape and hence prey size, they are able to locate, track and capture fast moving prey. The specialisation of the visual system of these 3 species is crucial in facilitating such feeding behaviour in aquatic environments with different light intensities, transmission properties and turbidities, and results presented here reflect this.
Morphology and eye structure
Snout length to head length, and eye diameter to snout length Differences between the pipefish and seahorses in relative snout/head length have been observed by Kendrick (2002), with those over 0.53 being able to successfully pursue and consume >70% mobile prey. Those within the ratio range of 0.34 to 0.53 consumed ≤20% relatively mobile prey and the remainder of the diet being comprised of harpacticoid copepods, gammarid and crapellid amphipods and carid shrimp, all of which crawl on the sediment or phytal surfaces. These observations indicated different feeding strategies for creating or increasing feeding opportunities by bringing prey items into the strike zone, with a longer snout being more efficient in feeding on mobile prey.
65 Chapter 4 – Discussion
Morphological and behavioural observations established in this study concurred with the features that Kendrick reported, with S. argus being an extremely fast and energetic predator, pursuing small, fast moving calanoid and cyclopoid copepods. The 2 seahorse species maintained a more saltatory approach, detecting their prey from a distance and moving carefully towards the target trying not to create disturbance. They then consumed prey with a rapid vertical ‘head flick’.
Although S. argus pursued the fastest and most energetic prey, it interestingly had the smallest ratio of eye diameter to snout length (0.14) indicating a more difficult visual challenge than the seahorses. The seahorse ratios were much larger and similar, with H. subelongatus and H. barbouri at 0.29 and 0.33, respectively (Table 1 & Fig. 24), indicating similar and lesser visual challenges compared to the pipefish, however, they were the ones that pursued the slower and more sedentary prey. In addition, the smaller eye (and lens) sizes of S. argus and H. barbouri, may imply reduced acuity (Land, 1981). As S. argus and H. barbouri live in shallower water than H. subelongatus, there may be more light availability and a smaller lens which provides less acuity may not be an impediment. As discussed below (page 69), both species also possess double cones and rods of lesser dimensions than H. subelongatus (Table 3), thereby increasing acuity by this mechanism instead.
Irideal flaps The presence of an irideal flap on the dorsal part of the eye of adult lamprey, G. australis, was found to represent an adaptation for life in surface waters (Collin et al. 1999). A comparable role in increasing the resolution of images was proposed for the irideal flap found in the eyes of certain teleosts that live in brightly-lit environments (Nicol, 1989). The structure also reduces the amount of extraneous light falling on the ventral part of the eye and thus also the degree of intraocular flare. As the lamprey pupil is considered to be able to undergo only slight movement (Kleerekoper, 1972; Dickson and Graves, 1981), the possession of an irideal flap would be of particular value to the adult during the time it spends in well-lit surface waters. The absence of
66 Chapter 4 – Discussion
irideal flaps in pipefish, S. argus, in the current study implies that the direction of light to a specific area of the retina may not be as important as in the 2 species of seahorse that possess these flaps. As indicated in Figs. 14 & 15 and also personal observations of behaviour in the aquarium, there were many orientations of the head of this pipefish as it moved around the tank. Swimming freely and swiftly towards its prey, the animal can see prey or predators approaching from above or below whilst using independent eye movements. Due to its variable orientation and its smaller eye diameter, the need for protection from excess light and intraocular flare by an irideal flap may be reduced. The pigment on the dorsal and ventral sides of the iris may camouflage the fish.
Conversely, the two seahorse species, H. subelongatus and H. barbouri, possess one dorsal and one ventral irideal flap which direct the light to the most important area of the retina for image resolution, the fovea. As previously noted, irideal flaps assist in reducing intraocular flare during life in surface waters in other species. In this way extraneous rays of light are removed and the remaining is better directed to the fovea (found in the ventro- temporal position of the retina in both seahorse species) therefore increasing acuity. Being highly mobile, their eyes subtend both monocular and binocular regions of the frontal visual field in front of the mouth, which was consistent with the behaviour of these syngnathids.
Pupil Dilation In regulating the amount of light reaching the photoreceptors, the pupil enables control of sensitivity (Muntz, 1999). As in many teleosts, the lack of pupillary constriction in response to an increase in ambient illumination in any of these syngnathid species indicates that immediate light regulation is not a primary function of the pupil, and that their lifestyle does not call for an instantaneous adaptation to varying light levels. Douglas et al. (2002) found that in the suckermouth armoured catfish, Liposarcus pardalis, pupil dilation occurred over a period of 60 minutes after the animals had been dark-adapted over night. Our animals were dark-adapted for only 1 hour, having been taken from a brightly lit aquarium environment and were filmed for only 10 mins,
67 Chapter 4 – Discussion
therefore sufficient time to adjust to both light intensities may have been lacking in our experiment precluding a more convincing response. Many species of fish possess complex mechanisms for retinal adaptation, such as retinomotor movements of the rods, cones and retinal pigment epithelium (Martin, 1999), providing optical isolation of the photoreceptors and also enhancement of visual acuity (Tansley, 1965). Immediate light regulation may instead be controlled by these mechanisms in syngnathids.
Ocular Dimensions and function Vertebrates possess a number of mechanisms to increase photon capture (Land 1981; Warrant 1999 for review) including increasing photoreceptor size (van der Meer 1994; Shand 1997; Pankhurst & Hilder 1998); changes to relative eye dimensions resulting in a reduction of Matthiessen’s Ratio (McFarland 1991a; Shand et al. 1999b; Job & Bellwood 2000); and also increasing convergence ratios of photoreceptors to higher order neurones (van der Meer et al. 1995; Fuiman & Delbos 1998). We found an increase in rod size in this study but a higher Matthiessen’s Ratio that reduces light reaching the retina.
Prior to this study the value of a fish’s Matthiessen’s Ratio (MR) was thought to be constant in adults of any given species and found to vary between species with a range of 2.2 – 2.8 (Sivak 1990; Collin et al. 1998; Shand et al., 1999; Collin & Collin 1993). A Matthiessen's Ratio of 2.55 (Walls, 1942; Pumphrey, 1961) is commonly used to calculate focal length in the estimation of visual acuity in fish (see Wanzenböck, Zaunreiter, Wahl, & Noakes, 1996 for review). MR values of our 2 seahorse species differ markedly from this constant. Ocular dimensions revealed that both the temperate and tropical seahorses possess very high MR values of 3.7. The result implies that the focal length compared to lens radius in these species is extremely large and is the first time such high values have been found in an adult fish. An increased MR assists in projecting the image onto more photoreceptors for a given visual angle than in an animal with a smaller MR, and hence better resolution (Fig. 50). The creation of a larger image and improved image resolution would help to locate small prey at a distance and
68 Chapter 4 – Discussion
enable more accurate strikes. This is compatible with their highly visual feeding habit and may be a useful capitalisation of visual acuity at the target range of the long, fused snout and the suction area beyond that. The benefit of large eyes may be at a cost of decreased sensitivity to low light.
* typical teleost * seahorse
Fig. 50. Diagram represents increased image resolution of * the image of the prey * onto the retina.
Lens properties
Shape Spherical lenses found in each of the three study species are consistent with most other teleosts and considered to be the most appropriate design for the lens of an aquatic vertebrate (Sivak, 1990). As in most teleosts, these lenses are the only refractive element in the eye and account for the total power of the optical system. Lenses in the seahorses were found to protrude through the pupil and lie in close proximity to the cornea (Figs. 26 & 27). Typically, a spherical lens compensates for the loss of refractive power of the cornea underwater and occurs in most shallow-water species (Collin 1997) as in our study species.
Filtering Effects Spectrophotometric measurements of the yellow lenses of various animals, including some fish, have long been available (Kennedy & Milkman 1956). The lenses act as cut-off filters, passing long wavelengths and
69 Chapter 4 – Discussion
absorbing short wavelengths, and the spectral position at which they absorb varies between species (Muntz 1976; Collin 1997; Siebeck & Marshall 2001).
Reduction of chromatic aberration and contrast through scattering are 2 aspects of vision improved by yellow filters. The contrast between an object and its background is less underwater than in air. This reduction in contrast occurs because the image forming light from the object is attenuated in its passage through the water by being both absorbed and scattered out of the light beam, while at the same time the water between the object and the observer scatters diffuse veiling light into the eye (Duntley 1962). Under these circumstances yellow filters should increase contrast of objects against their backgrounds, since the scattering is occurring predominantly at shorter wavelengths. Muntz (1976) found a correlation in the presence of yellow filters in fish species with conditions of high light intensity, when the loss of sensitivity may be relatively unimportant. He concluded also that when water is turbid and particles are large, the scattering of particles becomes independent of wavelength consequently the advantage of possessing yellow filters is reduced.
It was unexpected, therefore, to find yellow lenses in the two green- water seagrass species, spotted pipefish S. argus and the Western Australian seahorse H. subelongatus. These two species possessed yellow filters cutting out wavelengths of light below 425nm (Fig. 29). One would assume that a green-water environment acts as a filter to the shorter wavelengths of light, thus precluding the need for a yellow filter for protection from UV. However, the lenses may aid in reduction of intraocular flare from bright down-welling light. The occurrence of the yellow filter in these 2 species may provide contrast differentiation of prey to its background improving its visual resolution, even though it may also be associated with a loss of sensitivity.
The shallow dwelling coral reef seahorse, H. barbouri, on the other hand, possessed a lens with a 50% cut-off wavelength of 408nm (Fig. 29) thus allowing the transmission of light to an additional SWS cone with a λmax at
70 Chapter 4 – Discussion
430nm (violet). The additional SWS pigment may be advantageous for feeding on plankton in the clear coral reef waters with broad spectrum light.
Retinal structure and photoreceptors
The finding that all 3 species possessed a retina of rod, single and double cone photoreceptors is consistent with the species being diurnal and/or crepuscular feeders. This study confirms that H. subelongatus and H. barbouri possess a cone-dominated fovea in the ventro-temporal retina within a retinal thickening, consistent with previous syngnathid studies (see Collin & Collin 1999). The steep-sided retinal pit provides image magnification (Walls 1942; Snyder & Miller 1978), detection and maintenance of accurate fixation (Pumphrey 1948), monocularly-mediated directional focus (Harkness & Bennett-Clark 1978) and perception of depth (Munk 1975; Locket 1992), which will assist these animals possessing independent eye mobility to increase sensitivity to small angular movements. This is similar for birds of prey and anolid lizards that possess deep and convexiclivate fovea for locating small, moving prey (Fite & Lister 1981; Barbour et al. 2002).
We do not know if S. argus has a fovea but a steep-sided (convexiclivate) fovea was found in the pipefish, Corythoichthyes paxtoni, which was also characterised by the exclusion of rods, with increased density of photoreceptors and a regular square mosaic of four double cones surrounding a single cone (Collin & Collin 1999). The authors concluded that diurnal vision in the frontal region of the visual field was optimised and the square mosaic was used primarily for the perception of movement of small approaching prey, in conjunction with high visual acuity provided by an increased photoreceptor density. Areas of S. argus retina processed for MSP displayed areas devoid of rod and dominated by cone photoreceptors, and other areas of high rod domination, indicating it is likely that S. argus also possesses a fovea although the precise location is unknown.
71 Chapter 4 – Discussion
Rods H. subelongatus displayed consistently long rod outer segments (OS) compared to S. argus and H. barbouri, which both had a broader range from short through to long OS. S. argus and H. barbouri also displayed narrower rods than H. subelongatus. This may be due to an adaptive mechanism by H. subelongatus to deal with lower light levels at greater depths (down to 25m).
The presence of extremely long rod OS (up to 33.33µm) in each of the 3 species was striking and very unusual (Table 2) being up to 4-5 times the length of rods of other shallow-living teleost species, e.g. the surfperch of up to 7μm (Cummings & Partridge 2001), and the black bream up to 8μm (Shand et al. 2002). Areas of the retina dominated by these large, chunky rods would provide a greater probability of photon capture increasing sensitivity in low light levels. This development may help to off-set the lack of light reaching the retina a result of a high Matthiessen’s Ratio (Land1981). The diameter of the OS is the important component of the optical sensitivity of the eye since light reaches the receptor perpendicular to the cross-section of the OS (Land 1981; Warrant 2004), hence, the larger the OS, the more photons sampled. Long photoreceptors are also a feature of nocturnal and deep-sea teleosts, increasing sensitivity in low light levels (Wagner 1990; Warrant 2004, for reviews).
Cones All 3 species had typically small single and double cones with a small amount of variation in width and length. The thinner cone cells generally occurred in the tightly packed fovea where visual acuity is increased with high photoreceptor density. A retinal mosaic was suggested in the histological sections of H. subelongatus (Fig. 36) but not clearly established. Many species of fish possessing a regular square mosaic, strike moving prey with precision in a three-dimensional environment (Collin 1989;; Fineran & Nichol 1974; Kunz et al. 1985; Braekevelt 1985; Collin & Collin 1999; Shand et al. 1999). Further histological examination with an electron microscope would be necessary to determine whether a retinal mosaic is truly present. Identical twin cones found are likely to be involved in both brightness and colour
72 Chapter 4 – Discussion
discrimination (Burkhardt et al. 1980). Furthermore, cones are implicated in spatial resolution (Zaureiter et al. 1980) and motion perception (Gegenfurter et al. 1999). Double/twin cones could provide optimal function in the interface between bright and dim light vision, being of prime importance to these fish that must have functional vision throughout a range of light levels.
Colour sensitivity of the photoreceptors
The variety of visual pigments and spectral sensitivities that have been observed in this study could be correlated with many factors such as phylogeny, feeding strategy, behaviour and prevailing light conditions in the microhabitat (Lythgoe, 1972; Levine & MacNichol, 1979; Bowmaker et al. 1994). This study included a tropical clear-water species, and 2 temperate green-water species, all were diurnal and relatively shallow living, preying on a range of mobile and crawling crustaceans of varied colours. Unfortunately no detailed spectral data of the microhabitats inhabited by these species are available. However, the tropical marine shallow-water habitat of H. barbouri has a broad spectrum available, and the shallow green-water seagrass beds of S. argus and H. subelongatus would also have a broad spectrum at the surface, attenuating to a greener transmission with depth, which appear to correlate with the visual pigments found.
Rods High optical density of rod scans for all species indicates a high absorbance of light energy for those cells. The λmax wavelength of rods of all species showed little variation, falling close to 500nm which is consistent with other shallow-water marine species (Lythgoe et al. 1994). Visual pigments of shallow water coral reef inhabiting fish maximally absorb at shorter wavelengths, compared to those from shallow estuarine and coastal water where chlorophyll and Gelbstoffe shift the λmax to longer wavelengths of yellow-green and yellow light (Partridge 1990). Despite this previous finding the current study found λmax of the rods in the 2 seagrass species S. argus (502nm) and H. subelongatus (500nm) as well as the coral reef seahorse, H. barbouri (499nm) were all surprisingly similar (Table 4 and Fig. 41 & 42).
73 Chapter 4 – Discussion
Cones At the beginning of the study there was a tentative expectation that a UV pigment may be found in H. barbouri due to the broader spectrum of light in shallow-water coral reefs, as well as their lenses initially appearing clear. Investigation of lens transmissions displayed the 50% cut-off filter at 408nm wavelength, and further MSP exploration found no UV visual pigment in H. barbouri. As the lens scans showed 50% cut-off wavelengths of 423nm in H. subelongatus, and 428nm in S. argus, it was also unlikely a UV visual pigment would be found in these other 2 species. Lack of UV visual pigment implies that in these species, detection of plankton by increasing the contrast of short wavelength-absorbing or reflecting zooplankton, is not a requirement for successful feeding on zooplankton. Nevertheless, the presence of SWS single cones absorbing in the blue at around 460nm in all 3 species was similar to other planktivorous teleosts (Bowmaker & Kunz, 1987; Browman et al. 1994; McFarland & Leow, 1994; Britt et al., 2001). H. barbouri possessed an additional SWS single cone in the purple at 430nm. Violet perception may be important in detecting opaque plankton against a bright background and via reflectance from prey (Losey et al. 1999; Siebeck & Marshall 1999). The SWS pigment in H. barbouri extends the visual range of this species in clear coral reef waters where more shortwave light is available. During development coastal species, such as the black bream or Pollack, lose violet- sensitive pigments as they move to deeper water (Shand et al. 1988; Shand et al. 2002)
Despite the environment of H. subelongatus and S. argus containing more chlorophyll and organic matter being subject to tannin staining and land run-off from the river, the visual pigment in the double cones are all very similar and did not show a close parallel with habitat. Many coastal fish have been found to be dichromatic, with a short wavelength-sensitive cone and identical visual pigment in both portions of the long wavelength-sensitive cones (Lythgoe and Partridge 1989, Lythgoe & Partridge 1991 for review). Other coastal species from intermediate depths (10-50m), including the West
Australian dhufish (Shand et al. 2001), possess cone pigments with λmax of the single of 460 and 522nm in the double. These values contrast with those of
74 Chapter 4 – Discussion
trichromatic fish inhabiting clear coral reefs with pigment combinations of single, 424 and double 494/518 nm pairs for blue water snapper, Lutjanus bohar (Lythgoe et al. 1994). These values also differ from those of estuarine fish such as the black bream, Acanthopgrus butcheri, with a single at 480nm and longer double pairing at 545/560nm (Shand et al. 2001). The maximum number of cone classes is 4 found in coral reef dwelling damselfish
(Hawryshyn et al. 2003) including one located in the UV. Each λmax value of cones in previously studied species has been unambiguous and discrete at a particular point. In contrast, our study species possessed a range of λmax values covering the spectrum from 515-580nm. These did not fall into discrete classes (histograms in Fig. 43) but appeared to be a mixture of λmax values of all of the above-mentioned fish, reflecting very complex combinations in the double cones with the potential for complex colour and hue discrimination not previously seen in any other fish.
In many teleosts, spectral sensitivity of the double cones has been found to match the maximum transmission of the water (Lythgoe 1979; 1984; Reckel et al. 2002), which concurred to a certain extent in the species in the current study. Adults of all 3 species exhibited LWS double and twin cones with λmax between 515 and 580nm, exhibiting at least 4 classes within this range. They showed different spectral sensitivities in the 2 outer segments of the double cones, hence trichromacy, tetrachromacy and higher are possible with this species. Fritsches et al. (2000) deduced that in the Billfish (Xiphiidae), 2 cone types have evolved in the ventral retina (which looks up into the bright light) to enhance contrast of dark objects against space light. And in the dorsal retina (which points downwards into the darker water below the animal) twin cones predominated at the wavelength of the ambient light and therefore appeared to be specialised for sensitivity rather than hue discrimination (Fritsches et al. 2003a; 2003b). It was not possible to investigate the regional differences in the retinae of the study species.
Topographical variations in visual pigments may be expected due to the different orientation of the heads of the 3 species. S. argus position themselves in seagrass with their long snouts orientated vertically feeding
75 Chapter 4 – Discussion
predominantly on mobile calanoid and cyclopoid copepods (small, between 0.7-1.4mm and colourless). These copepods must be in constant motion to prevent sinking (Ritz 2000) and are chased vigorously by the pipefish. Therefore, a general upward visual orientation (viewing darker targets against the bright down-welling sunlight) occurs with this species. By contrast, the deeper dwelling H. subelongatus has a strongly prehensile tail (Hale 1996) and swimming capabilities more suited to manoeuvrability than speed (Consi et al 2001). Their body form is better adapted to inhabiting complex habitats and rarely venture out into the open when feeding (Flynn & Ritz 1999), holding their heads generally with the snout facing 90o to the vertical. There was a prevalence of crawling gammarid and crapellid amphipods, and carid shrimp in its diet (up to 5mm, pale or brightly coloured) reflecting it feeding on more sedentary prey in the seagrass beds (Kendrick 2002). Paucity of feeding studies on H. barbouri inhibits our ability to draw conclusions regarding the relationship between ambient light, prey and spectral sensitivities. However, it appeared to have an increased array of double cones paired with offset pigments compared to the other 2 species, to perhaps fulfil the need for this species to contrast brightly reflecting targets against a darker background. Coral reef dwellers are the most colourful and varied assemblages of fish due to territorial marking, sexual display, camouflage through object or background matching, or disruption, aposomatism and temperature control (See Marshall et al. 2003).
Colour vision results from a comparison of two different types of cone cells with different pigments. In all of our species these LWS double cones occurred in a surprisingly larger range and unusual spread of identical and non-identical pairing combinations than ever before reported in a fish. If used for colour vision, multiple combinations may allow the animal the capacity to discriminate wavelengths and distinguish hues, suggesting that double cones are used to match their visual pigments to the ambient light for optimal visual sensitivity. It was difficult to determine a pattern in this regard due to the small number of pairings found, more data would hopefully reveal clearer trends.
76 Chapter 4 – Discussion
There is evidence that double cones are involved in movement detection in both birds (Campenhausen & Kirschfeld 1998) and fish (Levine & McNichol 1982), especially when the cells are arranged in a square mosaic (Collin and Collin 1999). For that reason, the propensity for syngnathids to feed on highly mobile prey in clear and turbid water may account for unusual array of double cone pairings.
Mechanisms of visual pigment absorbance of double cones
As all our average spectral absorbance curves appear to best-fit the A1 template, it can be deduced that our species are similar to the majority of other marine vertebrates in the use of retinal to form ‘rhodoposins’ in their photoreceptors (Partridge & Cummings, 1999). Given the fact that neither the rod nor cone data, are better-fitted by the A2 template, it is likely that there is little or no A2 chromophore in the retina of these species. Considering the binned and averaged data, there was some indication that the intermediate
LWS cones had bandwidths bigger than an A1, i.e. slightly broader than the template (46a,b, 47b, 48b). It may be concluded that 2 or possibly 3 opsins could be co-expressed in the same cell, as suggested in the black bream (Shand et al. 2002), and that the intermediate cells must have mixtures of these. This conclusion is necessarily tentative and the exact mechanism underlying the λmax shift must remain a matter of conjecture until molecular biological studies are completed.
Juveniles
Ontogenetic development of double cones has often been associated with a change in lifestyle from brightly lit surface waters to inhabiting dimmer, deeper waters (Boehlert 1978: Shand et al. 1997). Photoreceptor development in marine teleosts varies between species. In most species studied so far, the fish begins life with a single cone retina then double cone formation occurs and finally the onset of rod differentiation corresponding with exogenous feeding a few days after hatching (Lara 2001; see Shand et al., 1999 for review; Shand et al. 2001a; 2001b; Pankhurst & Eager 1996;
77 Chapter 4 – Discussion
Pankhurst et al. 1993). Double cones are generally formed at later stages of larval life and rods are first differentiated around the time of metamorphosis, when body pigmentation occurs (Shand et al. 1999, for review). Therefore, most marine fish at larval stages possess only one class of visual pigment, and once double cones and rods are formed, there can be ontogentic shifts in spectral sensitivity of cone photoreceptors in juvenile teleosts, associated with changes in habitat and or/or diet (Shand et al. 1988; Shand 1993). Marked changes in light environment and feeding behaviour are likely to occur during settlement in reef fishes as the juveniles take up their reef-associated mode of life (Job & Shand 2001). The upwelling spectral distribution, in particular, varies between pelagic larval and benthic juvenile habitats, becoming more prevalent in long wavelength light (McFarland & Munz 1975; McFarland 1991).
Just-hatched S. argus, H. barbouri and P. curtirostris fed instantaneously upon hatching in our aquariums. MSP performed on these animals at 1-2 days old (H. barbouri and S. argus), and 43 days old (P. curtirostris) revealed the presence of rods in all three species which fit tightly around 500nm (Fig. 48), identical to the adults. Due to limited MSP data, the presence of SWS single cones was not established at this time. LWS double cones were found in all species, and even though only limited MSP scans were performed (Fig. 49), these results contrast with the early larval stages of many marine teleosts. Double cone λmax values, although incomplete, are what we would expect with H. barbouri inhabiting bright coral waters, S. argus floating above and within seagrasses, and P. curtirostris showed longer λmax wavelengths which may be expected given their foraging in seagrass and decaying leaf litter habitats, respectively.
The presence of double cones and rods at hatching indicates that the Syngnathidae have a very specialised visual system and are able to feed instantaneously at lower light levels compared with species in which these events occur later during development. It also indicates a highly developed ability to discriminate colour and contrast at hatching. The sensitivities of their visual pigments occurred at progressively longer wavelengths with the reduction of and the different spectral properties of light.
78 Chapter 4 – Discussion
Future directions
Investigations of the Syngnathidae vision and retina have opened a number of avenues for future study. Further histological examination is required to characterise the nature of the retina mosaic of just-hatched, juvenile and adults.
Photoreceptors transform light information to neural impulses through a complicated network of intermediate cells, and the signal is then received by ganglion cells which transmit it to the visual centres of the brain. Although we have found specialised areas of densely packed cones in different orientations, areas of extremely large, thin or chunky rods, and other areas of mixed rods and cones, which will all influence the perception of an image, it is the density of ganglion cells per unit area of retina that provides the ‘bottleneck’ for the fineness of an image reaching the brain (Pettigrew et al. 1988; Collin 1989). A future investigation of the neurones in the ganglion cell layer will allow an estimate of the highest spatial resolution of these species.
Non-subjective colour measurements by way of spectroradiometer readings to quantify luminance and spectral reflectance determining contrast between prey, habitat (e.g. coral reef or seagrass), conspecifics and the water would provide valuable detail. These reflectances may then match certain sensitivity maxima of the study species. This provides focus for future studies.
Ritz et al. (1997) found no obvious response by mysid swarms to approaching seahorses, suggesting that seahorse ambush techniques have evolved to minimize warning of the predator's approach. Also, when attacking, seahorses can suppress release of kairomones (a natural hormone emitted by seahorse) in order to remain chemically inconspicuous to their prey (mysid shrimp) when they came into visual contact (Cohen & Ritz 2003). This discovery is further evidence that vision plays an extremely important part in, not only detection of prey, but also sets off other biochemical mechanisms to conceal the fact that the predator is near, which can further be explored.
79 Chapter 4 – Discussion
Carotenoid pigmentation of the cornea and lens determines the transmission property of the whole eye (Siebeck & Marshall 2001). Some artificial fish feeds contain carotenoid pigments for basic animal nutrition, but also to increase the skin colour of animals to appeal to aquarists. The level of carotenoids transferred to the eye and its subsequent affect on vision, should be taken into account when considering light and feeding regimes in aquaculture. This would be an important consideration in its ability to contrast prey against its background, and therefore its feeding success and food conversion ratio (FCR - the amount of feed offered to animals which is then converted into increased body mass, i.e. the lower FCR, the less wastage of costly feed and less fouling of water).
Previous studies have noted the importance of light intensity for prey search, capture and duration of feeding in fish (Batty 1987) and specific light regimes for breeding seahorses (Lawrence 1998; Vincent 1995). Behavioural feeding trials of each species experimenting with varying light regimes including different intensities and colours will be the next step in developing an applied approach to this research to facilitate the most favourable feeding regimes (particularly for the juveniles). Trials with different coloured tanks and artificially coloured prey have been explored to a certain extent with the big bellied seahorse H. abdominalis (Woods 2000b).
Aquariums worldwide, including AQWA in Perth, have encountered problems maintaining the adults and rearing newly-hatched juveniles, including reports of adults startling when lights are extinguished often leading to sickness and death of the individuals. Large amounts of energy are expended by females in egg production only to have the eggs fall to the bottom of the tank when mating and egg transfer (from female to male parent) are unsuccessful. Ascertaining the reasons why this happens and supplying suitable light environments in terms of colour and intensities will also allow both juvenile and adult animals to experience their diurnal cycles improving breeding and rearing success of these and other rare species.
80 References
REFERENCES
Ahlbert IB. 1973. Ontogeny of double cones in the retina of perch fry (Perca fluviatus, Teleostei). Acta Zool. (Stockh.) 54, 241-254.
Ali MA & Wagner HJ.1975. Distribution and development of retinomotor responses, in Vision in Fishes (ed. MA. Ali) Plenum Press, New York, pp. 369-96.
Avery JA., Bowmaker JK., Djamgoz MB., Downing JE. 1983. Ultraviolet sensitive receptors in freshwater fish. Journal of Physiology, London 334, 23-24.
Azzarello MY. 1991. Some questions concerning the Syngnathidae brood pouch. Bulletin of Marine Science 49 (3), 741-747.
Batty RS. 1987. Effect of light intensity on activity and food-search of larval herring, Clupea harengus: a laboratory study. Marine Biology. 94, 323-327.
Bennet BA & Branch GM.1990. Relationships between production and consumption of prey species by resident fish in the Bot, a cool temperate South African estuary. Estuarine, Coastal and Shelf Science 31, 139-155.
Bennett ATD, Cuthill IC, Partridge JC & Lunau K. 1997. Ultaviolet plumage colours predict mate preferences in starlings. Proceedings of the National Academy of Sciences, USA 94, 8618-8621.
Bergert BA & Wainright PC. 1997. Morphology and kinetics of prey capture in the syngnathid fishes Hippocampus erectus and Syngnathus floridae. Marine Biology 127, 563-570.
Berglund A., Rosenqvist G. & Svensson I.1986. Mate choice fecundity and sexual dimorphism in two pipefish species Syngnathidae. Behavioural Ecology & Sociobiology 19 (4), 301- 307.
Blaxter JH & Jones M. 1967. The development of the retina and retinomotor responses in the herring. Journal of Marine Biology 47, 677-697.
Blaxter, JH & Staines M. 1970. Pure-cone retinae and retinomotor responses in larval teleosts. Journal of the Marine Biological Association of the UK 50,449-460.
Bowmaker JK. 1990. Visual pigments of fishes. In:. The Visual System of Fish. (eds. Douglas R & Djamgoz M) Chapman & Hall publishers. University Press Cambridge, pp. 81.
Bowmaker JK. 1995. The Visual Pigments of Fish. Progress in Retinal and Eye Research 15, 1-31.
Bowmaker JK, Govardovskii VI, Zueva LV, Hunt DM and Sideleva V. 1994. Visual pigments and the photic environment: the Cottoid fish of Lake Baikal. Vision Research 34, 591- 605.
81 References
Bowmaker JK & Kuntz YW. 1987. Ultraviolet receptors, tetrachromatic colour vision and retinal mosaics in the brown trout (Salmo trutta): age dependent change. Vision Research 27, 2101-2108.
Branch GM. 1966. Contributions to the functional morphology of fishes Part III. The feeding mechanism of Syngnathus acus Linnaeus. Zoologica Africana 2 (1), 69-89.
Britt LL, Leow ER. & McFarland WN. 2001. Visual pigments in the early life history stages of Pacific northwest marine fishes. Journal of Experimental Biology. 204, 2581-2587.
Burkhardt 1980. DA, Hassin G, Lenive JS & MacNichol EF. 1980. Electrical responses and photopigments of twin cones in the retina of the walleye. Journal of Physiology 309, 215- 28.
Campenhausen MV & Kirschfeld K. 1998. Spectral sensitivity of the accessory optic system of the pigeon. Journal of Comparative Physiology A 183, 1-6.
Campolmi M, Franzoi P & Mazzola A. 1996. Observations on pipefish (Syngnathidae) biology in the Stagnone lagoon (west Sicily). Publicaciones Especiales Instituto Español de Oceanografía 21, 205-209.
Carcupino M, Baldacci A, Mazzini M & Franzoi P. 1997. Morphological organisation of the male brood pouch epithelium of Syngnathus abaster Risso (Teleostea, Syngnathidae) before, during, and after egg incubation. Tissue & Cell 29(1), 21-30.
Carriere J. 1885. Die Sehorgane der Thiere, vergleichend-anatomisch dargstellt. Munich R. Othenburg.
Castro P. & Huber ME. 1997. Marine Biology. Wm. C. Brown Publishers, Dubuque, IA, USA.
Chung STL & Pease PL. 1999. Effect of yellow lens filters on pupil size. Optometry & Vision Science 76 (1), 59-62.
Clutton-Brock TH. & Vincent ACJ.1991. Sexual selection and the potential reproductive rates of males and females. Nature 351, 58-60.
Cohen AI. 1972. In: Fourtes MGF (ed): Handbook of Sensory Physiology, Vol. VII/2. pp 63- 110. Springer-Verlag, New York.
Cohen PJ & Ritz DA. 2003. Role of kairomones in feeding interactions between seahorses and mysids. Journal of the Marine Biological Association of the United Kingdom. 83(3), 633-638.
Collin SP. 1989. Topographic organization of the ganglion cell layer and intraocular vascularization in the retinae of two reef teleosts. Vision Research 29, 765-75.
Collin SP. 1997. Specialisations of the teleost visual system: adaptive diversity from shallow water to deep-sea. Acta Physiology. Scand. 161, S638, 5-24.
82 References
Collin SP. 1999. Behavioural ecology and retinal cell topography. In: Archer SN, Djamgoz MBA, Loew ER, Partridge JC, Vallerga S (eds) Adaptive mechanisms in the ecology of vision. Kluwer, Dordrecht, pp 509–535.
Collin SP & Collin HB. 1993. The visual system of the Florida garfish, Lepisosteus platyrhincus (Ginglymodi): II. Cornea and lens. Brain, Behaviour & Evolution 42, (2) 98-115.
Collin SP & Collin, HB. 1999. The foveal photoreceptor mosaic in the pipefish, Corythoichthyes paxtoni (Syngnathidae, Teleostei). Histology & Histopathology. 14, 369- 382.
Collin SP, Hart, NS, Shand, J & Potter, IC, 2003. Morphology and spectral absorption characteristics of retinal photoreceptors in the southern hemisphere lamprey (Geotria australis). Visual Neuroscience 20, 119-130.
Collin SP, Hoskins, RV, Partridge, JC. 1998. Seven retinal specializations in the tubular eye of the deep-sea pearleye, Scopelarchus michaelsarsi: A case study in visual optimization. Brain, Behaviour & Evolution 51 (6), 291-314.
Collin SP, & Pettigrew JD. 1989. Quantitative comparison of the limits on visual spatial resolution set by the ganglion cell layer in twelve species of reef teleosts. Brain, Behaviour & Evolution 34 (3), 184-192.
Colson DJ, Patek SN, Brainerd EL & Lewis SM. 1998. Sound production during feeding in Hippocampus seahorses (Syngnathidae). Environmental Biology of Fishes 51(2), 221- 229.
Consi TR, Seifert PA, Triantafyyllou MS & Edelman, ER. 2001. The dorsal fin engine of the seahorse (Hippocampus sp.). Journal of Morphology 248, 80-97.
Cummings, ME & Partridge, JC. 2001. Visual pigments and optical habitats of surfperch (Embiotocidae) in the Californian kelp forest. Journal of Comparative Physiology A 187, 875-889.
Dawson CE. 1984. A new pipehorse Syngnathidae from Western Australia with remarks on the subgenera of Acentronura. Japanese Journal of Ichthyology 31 (2), 156-160.
Dawson CE. 1985. Indo-Pacific pipefishes. Gulf Coast Research Laboratory, Ocean Springs, Mississippi, USA.
Denton EJ. 1956. Rescherches sur l’absorption de la lumière par le cristallin des poisons. Bull. Inst. Oceanogr. Monaco 1071, 1-10.
Denton EJ. 1990. Light and vision at depths greater than 200 metres. In: Herring PJ, Campbell AK, Whitfield M, Maddock L (eds) Light and life in the sea. Cambridge University Press, Cambridge, pp 127–148.
Denton EJ & Warren FJ. 1956. Visual Pigments of deep-sea fish. Nature (Lond) 178, 1059.
83 References
Diaz-Ruiz, S, Aguirre-Leon, A, Perez-Solis, O. 2000. Distribution and abundance of Syngnathus louisiane and S. scovelli in Tamiahua Lagoon, Gulf of Mexico. Ciencias Marinas 26, 125-143.
Dickson DH & Graves DA. 1981. The ultrastructure and development of the eye. In: The biology of lampreys, Vol. 3 (ed. by M.W. Hardisty and I. C. Potter), Academic Press. London. pp. 43-94.
Douglas, RH, Collin, SP & Corrigan J. 2002. The eyes of suckermouth armoured catfish (Loricariidae, subfamily Hypostomus): pupil response, lenitcular longitudinal spherical aberration and retinal topography. The Journal of Experimental Biology 205, 3425-3433.
Douglas R & Djamgoz M. 1990. The Visual System of Fish. Chapman & Hall publishers. University Press Cambridge.
Douglas, RH, Harper, RD & Case, JF. 1998a. The pupil response of a teleost fish, Porichthys notatus: description and comparison to other species. Vision Research 38, 2697-2710.
Douglas, RH & McGuigan, CM. 1989. The spectral transmission of freshwater teleost ocular media: An interspecific comparison and a guide to potential ultraviolet sensitivity. Vision Research 29, 871-879.
Douglas, RH, Partridge, JC & Marshall, NJ. 1998b. The eyes of deep-sea fish. I. Lens pigmentation, tapeta and visual pigments. Progress Retinal & Eye Research 17, 597- 636.
Douglas, RH & Thorpe A. 1992. Short-wave absorbing pigments in the ocular lenses of deep- sea teleosts. Journal of the Marine Biological Association of the UK 72, 93-112.
Duke-Elder S. 1958. The Eye in Evolution. Vol. I, System of Ophthalmology, Henry Kimpton, London, pp. 1-843.
Duntley, SQ. 1962. Underwater visibility. In: The Sea, Vol I, ed. M.N. Hill, Wiley & Sons, Inc.
Endler, JA. 1993. Some general comments on the evolution and design of animal communication systems. Philosophical Transactions of the Royal Society of London, Series B: Biological Sciences 340, 215-225.
Engstrom A. 1963. Cone types and cone arrangements in teleost retinae. Acta. Zool. Stockholm. 42, 179-243.
Evans BI. & Fernald, RD. 1993. Retinal transformation at metamorphosis in the winter flounder (Pseudopleuronectes americanus). Visual Neuroscience 10, 1055-1064.
Ferald RD. 1990. The optical system of fishes. In:. The Visual System of Fish. (eds. Douglas R & Djamgoz M) Chapman & Hall publishers. University Press Cambridge, pp. 45.
Fineran BA & Nichol JAC. 1974. Studies on the eyes of New Zealand parrot-fishes (Labridae). Proceedings of the Royal Society of London. B. 186, 217-247.
84 References
Fite KV & Lister BC. 1981. Bifoveal vision in anolis lizards. Brain, Behaviour & Evolution 19 (3- 4), 144-54.
Flynn AJ & Ritz DA. 1999. Effect of habitat complexity and predatory style on the Capture success of fish feeding on aggregated prey. Journal of the Marine Biological Association of the United Kingdom 79, 487-494.
Frank, TM & Widder, EA. 1996. UV light in the deep-sea: in situ measurements of downwelling irradiance in relation to the visual threshold sensitivities of UV-sensitive crustaceans. Marine & Freshwater Behavioural Physiology 27, 189-197.
Franzoi P, Maccagnani R, Rossi R & Ceccherelli VU. 1993. Life cycles and feeding habits of Syngnathus taenionotus and S. abaster (Pisces, Syngnathidae) in a brackish bay of the Po River Delta (Adriatic Sea). Marine Ecology Progress Series 97(1), 71-81.
Fritsches KA, Partridge JC, Pettigrew JD, Marshall, NJ. 2000. Colour vision in billfish. Philosophical Transactions of the Royal Society of London B Biological Sciences. 355(1401), 1253-1256.
Fritsches KA, Marshall, NJ, Warrant, EJ. 2003a. Retinal specializations in the blue marlin: eyes designed for sensitivity to low light levels. Marine and Freshwater Research 54,1-9.
Fritsches, KA, Litherland L, Thomas N & Shand J. 2003b. Cone visual pigments and retinal mosaics in the striped marlin. Journal of Fish Biology 63 (5), 1347.
Fuiman LA & Delbos BC. 1998. Developmental changes in visual sensitivity of red drum, Sciaenops ocellatus. Copeia p. 936-943.
Garcia, CM & de Perera TB. 2002. Ultraviolet-based female preferences in a viviparous fish. Behavioural Ecology &. Sociobiology 52: 1-6.
Gasparini JL & Teixeira RL.1999. Reproductive aspects of the gulp pipefish Syngnathus scovelli (Teleostei: Syngnathidae), from south-eastern Brazil. Revista Brasileira de Biologia 59(1), 87-90.
Gegenfurter KR, Mayser H, & Sharpe LT. 1999. Seeing movement in the dark. Nature 398, 475-6.
Gordina AD., Oven LS., Tkach AV & Gilmova TN. 1991. Distribution, reproduction and diet of the pelagic pipefish, syngathus schmidti, in the Black Sea. Journal of Ichthyology 31 (2), 110-119.
Govardovskii V I, Fyhrquist, N. Reuter, T. Kuzman, DG. & Donner, K. 2000. In search of the visual pigment template. Visual Neuroscience 17, 509-528.
Gronell AM. 1984. Courtship, spawning and social organisation of the pipefish, Corythoichthys intestinalis (Pisces: Synganthidae) with notes on two congeneric species. Zeitschrift für Tierpsychologie 65, 1-24.
85 References
Hale ME. 1996. Functional morphology of ventral tail bending and prehensile abilities of the seahorse, Hippocampus kuda. Journal of Morphology 227.
Hannan JC & Williams RJ. 1998. Recruitment of juvenile marine fishes to seagrass habitat in a temperate Australian estuary. Estuaries 21(1), 29-51.
Harkness L & Bennett-Clark HC. 1978. The deep fovea as a focus indicator. Nature London. 272, 814-816.
Harosi FI & Hashimoto Y. 1983. Ultraviolet visual pigment in a vertebrate: a Tetrachromatic cone system in the dace. Science 222, 1021-1023.
Hart NS, Partridge JC, Bennett ATD, Cuthill IC. 2000. Visual pigments, cone oil droplets and ocular media in four species of estrildid finch. Journal of Comparative Physiology A 186, 681-694.
Hawryshyn CW. 2000. Ultraviolet polarization vision in fishes: possible mechanisms for coding e-vector. Philosophical Transactions of the Royal Society of London B Biological Sciences 355, 1187-1190.
Hawryshyn CW, Moyer HD, Allison WT, Haimberger TJ & McFarland WN. 2003. Multidimensional polarization sensitivity in damselfishes. Journal of Comparative Physiol A-Neuroethology Sensory Neural and Behavioural Physiology 189 (3), 213-220.
Herald ES. 1959. From pipefish to seahorse – a study of phylogenetic relationships. Proceedings of the California Academy of Sciences 29 (13), 465-473.
Howard RK & Koehn JD. 1985. Population dynamics and feeding ecology of pipefish (Syngnathidae) associated with eelgrass beds of Western Port, Victoria, Australia. Australian Journal of Marine & Freshwater Research 36(3), 361-370.
Hunt DM, Kanwaljit SD, Partridge JC, Cottrill P & Bowmaker JK. 2001 The molecular basis for spectral tuning of rod visual pigments in deep-sea fish. The Journal of Experimental Biology 204, 3333-3344.
James PL & Heck KL. 1994. The effects of habitat complexity and light intensity on ambush predation within a simulated seagrass habitat. Journal of Experimental Marine Biology & Ecology 176, 187-200.
Job SD & Bellwood DR. 2000. Light sensitivity in larval fishes; implications for vertical zonation in the pelagic zone. Limnol. Oceanogr. 45, 362-371.
Job SD & Shand J. 2001. Spectral sensitivity of larval and juvenile coral reef fishes: implications for feeding in a variable light environment. Marine Ecology Progress Series 214, 267-277.
Jones AG & Avise JC. 1997. Microsatellite analysis of maternity and the mating system in the Gulf pipefish Syngnathus scovelli, a species with male pregnancy and sex-role reversal. Molecular Ecology 6(3), 203-213.
86 References
Jones AG & Avise JC. 2001. Mating systems and sexual selection in male-pregnant pipefishes and seahorses: insights from microsatelitte-based studies of maternity. The Journal of Heredity 92 (2), 150-158.
Jones AG, Rosenqvist G, Berglund A & Avise JC. 1999. The genetic mating system of a sex- role-reversed pipefish (Syngnathus typhle): A molecular inquiry. Behavioural Ecology & Sociobiology 46(5), 357-365.
Kahmann H. 1934. Über die Vorkommen einer Fovea centralis im Knochenfischauge. Zool. Anz. 106, 49-55.
Kahmann H. 1936. Über das foveale Sehen der Wirbeltiere. I. Über die Fovea centralis und die Fovea lateralis bei einigen Wirbeltieren. Arch. F. Ophthalmol. (Copenhagen) 135, 265-276.
Kendrick AJ. 2002. Resource utilisation and reproductive biology of syngnathid fishes in a seagrass-dominated marine environment in south-western Australia. PhD Thesis. Murdoch University.
Kennedy D. & Milkman RD. 1956. Selective light absorption by the lenses of lower vertebrates, and its influence on spectral sensitivity. Biological Bulletin 111, 375-86.
Kleerekoper H. 1972. The Sense Organs. In: The Biology of Lampreys, Vol. 2 (ed. By M.W. Hardisty and I.C. Potter). Academic Press. London, pp. 373-404.
Kornienko ES, Drozdov AL. 1999. Gametogenesis of the Far Eastern pipefish, Syngnathus acusimilis. Biologiya Morya (Vladivostok) 25(4), 323-326.
Krausee W. 1886. Die Retina. II. Retina der Fische. Intern. Monatsschr. F. Anat. U. Hist. 3, 8- 73.
Kuiter, R. 1988. Birth of a leafy sea-dragon. Australian Geographic 12, 91-95.
Kuiter, R. 2000. Seahorses, pipefish and their relatives. A comprehensive guide to Syngnathiformes. TMC Publishing, Chorleywood, UK.
Kunz YW, Shuilleabhain MN. & Callaghan E. 1985. The eye of the venomous marine teleost Trachinus vipera with special reference to the structure and ultrastructure of visual cells and pigment epithelium. Experimental Biology 43, 161-178.
Kvarnemo C, Moore GI, Jones AG, Nelson WS & Avise JC 1999. Monogamous pair bonds and divorces in the Western Australian Seahorse Hippocampus subelongatus. Proc. R. Soc. Lond. (Ser. B)
Land MF.1981. Optics and vision in invertebrates, In: Handbook of Sensory Physiology, vol. VII 6B, Autrum, HJ. (ed), Springer-Verlag, Berlin, pp. 471-592.
Land MF & Nilsson DE. 2002. Animal Eyes. Oxford, UK: Oxford University Press.
87 References
Lara MR. 2001. Morphology of the eye and visual acuities in the settlement-intervals of some coral reef fishes (Labridae, Scaridae). Environmental Biology of Fishes 62, 365-378.
Lawrence, C. 1998. Breeding seahorse – facts and fallacies. West. Fish., Autumn, 39-40.
Levine JS & McNichol EFJ. 1985. Microspectrophotometry of primate photoreceptors: art, artefact, and analysis. In: The Visual System (ed. A. Fein and JS Levine), pp. 73-87. New York: Liss.
Locket NA. 1996. The multiple bank rod fovea of Bajacalifornia drakei an alepocephalid deep-sea teleost. Proceedings of the Royal Society of London 224(1234), 7-22.
Loew ER. 1995. Determinants of visual pigment spectral location and photoreceptor cell sensitivity, Ch. 3, pp. 57-77. In: Neurobiology and clinical aspects of the outer retina. M. B. A. Djamgoz, S. N. Archer and S. Vallegra (eds.). Chapman and Hall, London.
Loew ER & Dartnall HJA. 1976. Vitamin A1/A2-based visual pigment mixtures in cones of the rudd. Vision Research 16, 891-896.
Loew ER, Govardovskii VI, Röhlich P, Szél A. 1996. Microspectrophotometric and immunocytochemical identification of ultraviolet photoreceptors in geckos. Visual Neuroscience 13, 247–256
Loew ER & Lythgoe JN. 1978. The ecology of cone pigments in teleost fishes. Vision Research 18, 715-722.
Losey GS, Cronin TW, Goldsmith TH, Hyde D, Marshall NJ & McFarland, WN. 1999. The UV world of fishes: a review. Journal of Fish Biology 54, 921-943.
Losey GS, Nelson PA & Zamzow JP. 2000. Ontogeny of spectral transmission in the eye of the tropical damselfish, Dascyllus albisella (Pomacentridae) and possible effects on UV vision. Environmental Biology of Fishes 59(1), 21-28.
Lourie SA, Pritchard JC, Casey SP, Truong SK, Hall HJ & Vincent ACJ. 1999b. The taxonomy of Vietnam's exploited seahorses (family Syngnathidae). Biological Journal of the Linnean Society 66(2), 231-256.
Lourie SA, Vincent,ACJ, & Hall HJ. 1999a. Seahorses. An identification guide to the world’s species and their conservation. Project Seahorse, London, UK.
Lyall AH. 1957. Cone arrangement in teleost retinae. Q. J. Microsc. Sci. 98, 189-201.
Lythgoe, JN. 1966. Visual pigments and underwater vision. In: Bainbridge R, Evans GC, Rackham O (eds) Light as an ecological factor, pp 375-390, Wiley, New York.
Lythgoe, JN. 1968. Visual pigments and visual range underwater. Vision Research 8, 997- 1011.
88 References
Lythgoe, JN. 1972. The adaptation of visual pigments to their photic environment. In: Dartnall HJA (ed) Handbook of sensory physiology, vol VII/1. Springer, Berlin Heidelberg New York, pp 578-603.
Lythgoe, JN. 1979. The Ecology of Vision. Clarendon Press, Oxford.
Lythgoe JN, Muntz WRA, Partridge JC, Shand J & Williams DM. 1994. The ecology of the visual pigments of snappers (Lutjanidae) on the Great Barrier Reef. Journal of Comparative Physiology A 174, 461-467.
Lythgoe JN & Partridge JC. 1989. Visual pigments and the acquisition of visual information. Journal of Experimental Biology 146, 1-20.
Lythgoe JN & Partridge JC. 1991. The modelling of optimal visual pigments of dichromatic teleosts in green coastal waters. Vision Research 31, 361-371.
Maier EJ. 1994. Ultraviolet vision in a passeriform bird: from receptor spectral sensitivity to overall spectral sensitivity in Leiothrix lutea. Vision Research 34, 1415-1418.
McDonald CG & Hawryshyn CW. 1995. Intraspecific variation of spectral sensitivity in threespine stickleback (Gasterosteus acueatus) from different photic regimes. Journal of Comparative Physiology A 176, 255-260.
McFarland WN. 1991. The visual world of coral reef fishes. In: Sale PF (ed) The ecology of fishes on coral reefs. Academic Press, San Diego, p 16-38.
McFarland WN & Leow ER. 1994. Ultraviolet visual pigments in marine fishes of the family Pomacentridae. Vision Research 34, 1393-1396.
McFarland WN & Munz FW. 1975. Part III: The evolution of phototopic visual pigments in fishes. Vision Research 15, 1071-1080.
MacNichol EF.1986. A unifying presentation of photopigment spectra. Vision Research 26, 1543-1556.
Manooch CS, Mason DL. 1983. Comparative food studies of yellowfin tuna Thunnus- albacares and blackfin tuna Thunnus-atlanticus Pisces Scombridae from the Southeastern and Gulf coasts of the USA. Brimleyana 9, 33-52.
Marshall NJ. 2000. The Visual Ecology of Reef Fish Colours. In: Espmark, Y., Amundsen, T. & Rosenqvist, G. (Eds.) Animal Signals: Signalling and Signal Design in Animal Communication, pp. 83-120.
Marshall NJ, Jennings K, McFarland WN, Loew ER & Losey GS. 2003a. Visual Biology of Hawaiian Coral Fishes. II. Colours of Hawiian Coral Reef Fish. Copeia 3, 455-466.
Marshall NJ, Jennings K, McFarland WN, Loew ER & Losey GS. 2003b. Visual Biology of Hawaiian Coral Fishes. III. Environmental Light and an integrated approach to the ecology of reef fish vision. Copeia 3, 467-480.
89 References
Martin G. 1999. Optical structure and visual fields in birds: their relationship with foraging behaviour and ecology. In: Adaptive mechanisms in the ecology of vision. pp. 485-508. Archer, SN, Djamgoz, MBA, Loew, ER, Partridge, JC & Vallerga, S. (Eds). Dordrecht: Kluwer Academic.
Masonjones HD & Lewis SM. 1996. Courtship behaviour in the dwarf seahorse, Hippocampus zosterae. Copeia 3, 634-640.
Masonjones HD & Lewis SM. 2000. Differences in potential reproductive rates of male and female seahorses related to courtship roles. Animal Behaviour 59(1), 11-20.
Matthiessen 1882. Über die Beziehung, welche zwischen dem Brechungsindex des Kernzentrums der Kristallinse und den Dimensionen des Auges bestehen. Plugers Archiv für die gesamte Physiologie des Menschen und der Tiere. 27, 510-23.
Moore GI. 1999. Reproductive Biology of West Australian Seahorse Hippocampus subelongatus. Master’s Thesis, The University of Western Australia.
Motias R. 1957. Sur l’absorption de la lumière oar le cristallin de quelques poisson de grand profondeur. Bull. Inst. Oceanog. Monaco 1094, 1-4.
Movchan YV. 1988. Fecundity of fish from the family Syngnathidae from the Black Sea and the Sea of Azov USSR. Voprosy Ikhtiologii 28 (4), 626-631.
Munk O. 1975. On the eyes of two foveate notosudid teleosts, Scopelosaurus hoedti and Ahliesaurus berryi. Vidensk. Meddr. Dansk naturh. Foren. 138, 87-125.
Muntz WRA. 1976. The visual consequences of yellow filtering pigments in the eyes of fishes occupying different habitats. In: GC Evans, R Bainbridge & O Rackham ed.) Light as an Ecological Factor II, pp. 272-287. Blackwell Scientific, Oxford.
Muntz WRA. 1976. The visual consequences of yellow filtering pigments in the eyes of fishes occupying different habitats. In: Light as an Ecological Factor II. Eds. S.C. Evans, R. Bainbridge, O. Rackham. Blackwell Scientific, Oxford.
Muntz WRA. 1990. Stimulus, environment and vision in fishes. In:. The Visual System of Fish. (eds. Douglas R & Djamgoz M) Chapman & Hall publishers. University Press Cambridge, pp. 491.
Muntz WRA. 1999. Visual systems, behaviour and environments in cephalopods. In: Archer SN, Djamgoz MBA, Loew ER, Partridge JC & Vallerga S. (Eds) Adaptive mechanisms in the ecology of vision, pp. 467-483. Kluwer Academic, Dordrecht.
Munz FW & McFarland FW. 1973. The significance of spectral position in the rhodopsins of tropical marine fishes. Vision Res. 13, 1829-1874.
Munz FW & McFarland FW. 1977. Evolutionary adaptations of fishes to the photic environment. In: Crescietelli F (ed) Handbook of sensory physiology: the visual system in vertebrates, vol VII/5. Springer, Berlin Heidelberg New York, pp 193-274.
90 References
Nelson PA, Zamzow JP, Erdmann SW & Losey Jr GS. 2003. Ontogenetic changes and environmental effects on ocular transmission in four species of coral reef fishes. Journal of Comparative Physiology A. 189, 391-399.
Nichol JAC. 1989. The Eyes of Fishes. Oxford, Clarendon Press.
Nussbaum JJ, Pruett RC & Delroi FC. 1981. Historic perspectives – Macular yellow pigment – the first 200 years. Retina 1, 296-310.
Pagel, M. 2003. Evolutionary Biology: Polygamy & parenting. Nature 424, 23-24.
Pankhurst, T. 2000. Feeding behaviour of greenback flounder larvae, Rhombosolea tapirina (Gunther) with differing exposure histories to live prey. Aquaculture 183(3-4), 285-297.
Pankhurst T. 2001. Visual morphology and feeding behaviour of the daggertooth. Journal of Fish Biology 58(5), 1427-1437.
Pankhurst PM. Eager R. 1996. Changes in the visual morphology through life history stages of the New Zealand snapper, Pagrus auratus. New Zealand Journal of Marine and Freshwater Research 30(1), 79-90.
Pankhurst PM & Hilder PE. 1998. Effect of light intensity on feeding of striped trumpeter Latris lineata larvae. Marine & Freshwater Research 49, 363-368.
Pankhurst PM, Pankhurst NW & Montgomery JC. 1993. Comparison of behavioural and morphological measures of visual acuity during ontogeny in a teleost fish, Forsterygion- varium, Tripterygiidae (Forster, 1801). Brain, Behaviour & Evolution 42, 178-188.
Parry JWL & Bowmaker JK. 2000. Visual pigment reconstruction in intact golfish retina using synthetic retinaldehyde isomers. Vision Research 40, 2241-2247.
Partridge JC.1990. The colour sensitivity and vision of fishes. In: Life and Light in the Sea. (eds. P J Herring, AK Campbell, M Whitfield and L Maddock) pp. 167-184. (Cambridge University Press: Cambridge.)
Partridge JC & Cummings ME. 1999. Adaptations of visual pigments to the aquatic environment. In: Adaptive mechanisms in the Ecology of Vision (ed. By S.N. Archer, M. B. A. Djamgoz, ER. Loew, JC. Partridge & S. Vallerga), Kluwer, Dordrecht, pp. 251-283.
Partridge JC, Shand J, Archer SN, Lythgoe JN, Groningen-Luyben WAHM van. 1989. Interspecific variation in the visual pigments of deep-sea fishes. Journal of Comparative Physiology A 164, 513–529
Paulus T. 1994a. Morphology of the head skull and the tube snout of the Syngnathidae from the Red Sea (Pieces: Teleostei). Senckenbergiana Maritima 21, 53-62.
Paulus T. 1994b. Calculation of the niche breadth and niche overlapping based on digestive system content analysis of three synoptic Syngnathidae from the Red Sea. Senckenbergiana Maritima 25, 63-74.
91 References
Payne M. & Rippingale R & Longmore, RB. 1998. Growth and survival of juvenile pipefish (Stigmatopora argus) fed live copepods with high and low HUFA content. Aquaculture 167, 237-245.
Perante NC, Pajaro MG, Meeuwig JJ & Vincent AC. 2002. Biology of a seahorse species, Hippocampus comes in the central Philippines. Journal of Fish Biology 60, 821 837.
Pettigrew JD, Dreher CS, Hopkins, MJ, McCall MJ & Brown, M. 1988. Peak density and distribution of ganglion cells in the retinae of microchiropteran bats: Implication for visual acuity. Brain, Behaviour & Evolution 32, 39-56.
Pham Thi Mi ESK,. Drozdov AL. 1998. Embryonic and larval development of the seahorse Hippocampus kuda. Biologiya Morya (Vladivostok) 24(5), 315-318.
Pinto L & Punchihewa NN. 1996. Utilisation of mangroves and seagrasses by fishes in the Negombo Estuary, Sri Lanka. Marine Biology (Berlin) 126(2), 333-345.
Pollard DA. 1984a. A review of ecological studies on seagrass-fish communities, with particular reference to recent studies in Australia. Aquatic Botany 18, 3-42.
Pollard DA. 1994b. A comparison of fish assemblages and fisheries in intermittently open and permanently open coastal lagoons on the south coast of New South Wales, South- eastern Australia. Estuaries 17(3), 631-646.
Prochazka K. 1998. Spatial and trophic partitioning in cryptic fish communities of shallow subtidal reefs in False Bay, South Africa. Environmental Biology of Fishes 51(2), 201- 220.
Pumphrey RJ. 1948. The theory of the fovea. Journal of Experimental Biology 25, 299-312.
Pumphrey RJ. 1961. Concerning vision. In: The Cell and the Organism (eds JA. Ramsay and VB Wigglesworth) Cambridge University Press, Cambridge, pp. 193-208.
Reckel F, Melzer RR, Parry JWL, Bowmaker JK. 2002. The retina of five Atherinomorph teleosts: Photoreceptors, patterns and spectral sensitivities. Brain, Behaviour & Evolution 60 (5), 249.
Reyes-Bustamante H, Ortega-Salas AA.1999. Rearing Hippocampus ingens (Pisces: Syngnathidae) under artificial conditions. Revista de Biologia Tropical 47(4), 1045-1049.
Ritz D.A. 2000. Is social aggregation in aquatic crustaceans a strategy to conserve energy? Canadian Journal of Fisheries & Aquatic Sciences. 57( 3), 59-67.
Ritz D.A, Obsborn J E & Ocken, A E J. 1997. Influence of food and predatory attack on mysid swarm dynamics. Journal of the Marine Biological Association of the United Kingdom. 77(1), 31-42.
Rodieck RW.1973. The Vertebrate Retina. Principles of structure and function. San Francisco. California. WH Freeman & Company.
92 References
Ryer CH & Orth RJ. 1987. Feeding ecology of the Northern pipefish, Syngnathus fuscus, in a seagrass community of the lower Chesapeake Bay. Estuaries 10 (4), 330-336.
Shand J. 1993. Changes in the absorption spectrum of cone visual pigments during settlement of the goatfish Upeneus tragula: loss of red sensitivity as a benthic existence begins. Journal of Comparative Neurology. A, 173, 115-121.
Shand J. 1997. Ontogenetic changes in retinal structure and visual acuity: a comparative study of coral-reef teleosts with differing post-settlement lifestyles. Environ. Biol. Fish. 49, 307-322.
Shand J, Archer MA & Collin SP. 1999. Ontogenetic changes in the retinal photoreceptor mosaic in a fish, the black bream, Acanthopagrus butcheri. Journal Comparative Neurology 412, 201-217.
Shand J, Archer MA, Thomas N, Cleary J. 2001. Retinal development of the West Australian dhufish, Glaucosoma hebraicum. Visual Neuroscience 18, 711-724.
Shand J, Chin SM, Harman AM, Moore S & Collin SP. 2000. Variability in the location of the retinal ganglion cell area centralis is correlated with ontogenetic changes in feeding behaviour in the black bream, Acanthopagrus butcheri. Brain, Behaviour & Evolution 55, 176-190.
Shand J, Doving K & Collin SP. 1999. Optics of the developing fish eye: Comparisons of Matthiessen's ratio and the focal length of the lens in the black bream Acanthopagrus butcheri (Sparidae, Teleostei).Vision Research 39(6), 1071-1078.
Shand, J, Hart, NS, Thomas, N & Partridge, JC. 2002. Developmental changes in the visual pigments of back bream. Acanthopagrus butcheri. The Journal of Experimental Biology 205, 3661-3667.
Shand J, Partridge JC, Archer SN, Potts GW & Lythgoe JN. 1988. Spectral absorbance changes in the violet/blue sensitive cones of the juvenile pollack, Pollachius pollachius. Journal of Comparative Physiology 163, 699-703.
Siebeck UE & Marshall NJ. 2001. Ocular media transmission of coral reef fish – can coral reef fish see ultraviolet light? Vision Research 41, 133-149
Silveira RB. 2000. Osteologic development of Hippocampus reidi Ginsburg (Pisces, Syngnathiformes, Syngnathidae), under laboratory conditions. I. Embryonic phase. Revista Brasileira de Zoologia 17(2), 505-513.
Sivak JG. 1980. Accommodation in vertebrates: a contemporary survey. In: JA Zadunaisky & H. Davson (eds.) Current Topics in Eye Research, Volume 3, pp 281-300. Academic Press, New York.
Sivak JG. 1990. Optical variability of the fish lens. In:. The Visual System of Fish. (eds. Douglas R & Djamgoz M) Chapman & Hall publishers. University Press Cambridge, pp. 63.
93 References
Slonaker JR. 1897. A comparative study of the area of acute vision in vertebrates. Journal of Morphology 13, 445-492.
Snyder AW & Miller WH. 1978. Telephoto lens system of falconiform eyes. Nature, London 275, 127-129.
Spotte S. 1992. Captive Seawater Fishes – Science and Technology. John Wiley Sons, Inc. New York.
Steffe AS, Westoby M & Bell JD. 1989. Habitat selection and diet in two species of pipefish from seagrass: sex differences. Marine Ecology Progress Series 55, 23-30.
Svensson I. 1988. Reproductive costs in two sex-role reversed pipefish species Syngnathidae. Journal of Animal Ecology 57(3), 929-942.
Tansley K. 1965. Vision in vertebrates. London: Chapman & Hall.
Teixeira RL & Musick JA. 1995. Trophic ecology of two congeneric pipefishes (Syngnathidae) of the lower York River, Virginia. Environmental Biology of Fishes 43(3), 295-309.
Teixeira RL, Musick JA. 2001. Reproduction and food habits of the lined seahorse, Hippocampus erectus (Teleostei: Syngnathidae) of Chesapeake Bay, Virginia. Brazilian Journal of Biology 61(1), 79-90.
Tipton K. 1987. The feeding ecology of two species of Syngnathidae in a Tampa Bay, Florida, seagrass bed, with special reference to harpacticoid copepods. Department of Biology, University of South Florida, 1-62.
Tomasini JA. Quignard JP. Capape C & Bouchereau JL. 1991. Reproductive success factors of Syngnathus-abaster Risso 1826 Pisces Teleostei Syngnathidae in a Mediterranean lagoon environment Mauguio Lagoon France. Acta Oecologica 12(3), 331-355.
Upton SJ, Stamper MA, Osborn AL, Mumford SL, Zwick L, Kinsel MJ & Overstreet RM. 2000. A new species of Eimeria (Apicomplexa, Eimeriidae) from the weedy sea dragon Phyllopteryx taeniolatus (Osteichthyes: Syngnathidae). Diseases of Aquatic Organisms 43(1), 55-59.
Van der Meer HJ. 1994. Ontogenetic change of visual thresholds in the cichlid fish Haplochromis sauvagei. Brain, Behaviour & Evolution 44, 115-132.
Van der Meer HJ, Anker Gc & Barel CDN. 1995. Ecomorphology in the cichliud fish Haplochromis sauvagei. Brain, Behaviour & Evolution 44, 4049.
Verrier ML. 1928.Recherches sur les yeux et al vision des poisons. Bull. Biol. Fr. Belg., Supp., 11, 1-222.
Vincent ACJ. 1995. Seahorse keeping. J. Macquaculture 3, 1-13.
Vincent A, Ahnesjö I, Bergland A & Rosenqvist G. 1992. Pipefishes and seahorses: are they sex role reversed? Trends in Ecology and Evolution 7 (7), 237-241.
94 References
Vincent ACJ & Sadler LM. 1995. Faithful pair bonds in wild seahorses: Hippocampus whitei. Animal Behaviour 50, 1557-1569.
Wald G, Brown PK & Smith PH. 1955. Iodopsin. Journal of General Physiology 38, 623-681.
Walls G. 1942. The Vertebrate Eye and its Adaptive Radiation, Cranbrook Institute of Science, Bloomfield Hills, Michigan.
Walls GL & Judd, HD. 1933. The intraocular filters of vertebrates. British Journal of Ophthalmology 17, 641-675.
Wanzenböck J, Zaunreiter M, Wahl C, Noakes DLG. 1996. Comparison of behavioural and morphological measures of visual resolution during ontogeny of roach (Rutilus rutilus) and yellow perch (Perca flavescens). Canadian Journal of Fisheries & Aquatic Sciences 53(7), 1506-1512.
Warrant EJ. 1999. Seeing better at night: lifestyle, eye design and the optimum strategy of spatial and temporal summation. Vision Research 39, 1611-1630.
Warrant EJ. 2000. The eyes of deep-sea fishes and the changing nature of visual scenes with depth. Philosophical Transcripts Royal Society of London B Biol Sci 29, 355(1401), 1155-9.
Warrant EJ. 2004. Vision in the dimmest habitats on Earth. Journal of Comparative Physiology A, 190 (10), 765-789
Warrant EJ, Collin SP, Locket NA. 2003. Eye design and vision in deep-sea fishes. In: Collin SP, Marshall NJ (eds) Sensory processing of the aquatic environment. Springer, Berlin Heidelberg New York, pp 303–322
Watanabe S, Watanabe Y & Okiyama M. 1997. Monogamous mating and conventional sex roles in Hippichthys penicillus (Syngnathidae) under laboratory conditions. Ichthyological Research 44(3), 306-310.
Wetzel J & Wourms JP. 1995. Adaptations for reproduction and development in the skin- brooding ghost pipefishes, Solenostomus. Environmental Biology of Fishes 44(4), 363- 384.
Whitfield AK. 1994. Fish species diversity in Southern African estuarine systems: An evolutionary perspective. Environmental Biology of Fishes 40(1), 37-48.
Whitfield AK & Smith JLB. 1995. Threatened fishes of the world: Syngnathus watermeyeri Smith, 1963 (Syngnathidae). Environmental Biology of Fishes 43(2), 152.
Woods CMC. 2000a. Preliminary observations on breeding and rearing the seahorse Hippocampus abdominalis (Teleostei: Syngnathidae) in captivity. New Zealand Journal of Marine & Freshwater Research 34(3), 475-485.
Woods CMC 2000b. Improving initial survival in cultured seahorses, Hippocampus abdominalis Leeson, 1827 (Teleostei: Syngnathidae). Aquaculture 190 (3-4), 1.
95 References
Wong JM & Benzie, JAH. 2003. The effects of temperature, Artemia enrichment, stocking density and light on the growth of juvenile seahorses, Hippocampus whitei (Bleeker, 1855), from Australia. Aquaculture 228, 107-121.
Zaunreiter M, Junger H & Kotschal K. 1991. Retinal morphology of cyprinid fishes: a quantitative histological study of ontogenetic changes and interspecific variation. Vision Research 31, 383-94.
96