Multisensory Integration in Shark Feeding Behavior Jayne M

University of South Florida Scholar Commons

Graduate Theses and Dissertations Graduate School

January 2012 Multisensory Integration in Shark Feeding Behavior Jayne M. Gardiner University of South Florida, [email protected]

Follow this and additional works at: http://scholarcommons.usf.edu/etd Part of the American Studies Commons, Other Education Commons, Physiology Commons, and the Social and Behavioral Sciences Commons

Scholar Commons Citation Gardiner, Jayne M., "Multisensory Integration in Shark Feeding Behavior" (2012). Graduate Theses and Dissertations. http://scholarcommons.usf.edu/etd/4046

This Dissertation is brought to you for free and open access by the Graduate School at Scholar Commons. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Scholar Commons. For more information, please contact [email protected].

Multisensory Integration in Shark Feeding Behavior

Jayne Michelle Gardiner

A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Integrative Biology College of Arts and Sciences University of South Florida

Co-Major Professor: Philip J. Motta, Ph.D. Co-Major Professor: Robert E. Hueter, Ph.D. Jelle Atema, Ph.D. Stephen M. Deban, Ph.D. Lynn B. Martin, Ph.D.

Date of Approval: December 8, 2011

Keywords: olfaction, lateral line, electroreception, vision

DEDICATION

“It’s a wonderful thing to get an education” – Eugene C. Everett

This work is dedicated to the memory of my grandfathers, Paul F. Gardiner (1920 –

2010), the great outdoorsman, who inspired my love of nature, and Eugene C. Everett

(1928 – 2009), who nurtured my childhood fascination with aquatic animals and whose final words to me gave me the strength and courage to persevere, even when this goal seemed unattainable. I owe this accomplishment to their constant love, support, and encouragement.

ACKNOWLEDGEMENTS

This study represents a massive undertaking and assistance was provided by an

incredible number of people, some of whom will inevitably be forgotten here, but whose

contributions are nevertheless greatly appreciated. I would first like to acknowledge and

thank my advisory team: my co-major advisors, Drs. Phil Motta and Bob Hueter for their

guidance, encouragement, and support, and Dr. Jelle Atema, who opened the door to

graduate work by taking a chance on a Microbiology and Immunology student who

wanted to study fish. Thank you also to my committee members, Drs. Stephen Deban

and Lynn Martin for their guidance, and to Dr. Jason Rohr for advice on the statistical

analyses. I am forever indebted to Jack Morris, who designed, built, and maintained the

flume system, collected and maintained animals, helped troubleshoot equipment and procedures, and somehow tolerated my many mistakes along the way. Logistical assistance with much of the above was also provided by Mercedes Smith. Thank you to

Henry Luciano for troubleshooting the numerous computer and camera issues, even late night and on holidays, and to Karen Taylor for providing computer equipment. My assistants, Mote volunteer Rachel Dryer and USF REU intern Brittany Leigh, contributed greatly to the analysis of all of the video files. Assistance with animal collections and transportation were provided by Rhys Probyn, Ed Enos, Bill Klimm, Nate Goddard,

Bryant Turffs, Pete Hull, Dean Dougherty, Greg Byrd, numerous Mote interns and volunteers. Assistance with animal care, husbandry and animal health issues were provided by Dr. Roxanna Smolowitz, Dr. Andy Stamper, Lynne Bird, and Matt Seguin.

Advice on the design of the project, technical help, and loans of equipment were provided by Dr. Brandon Casper, Dr. David Mann, Dr. Stephen Kajiura, Clint Steffan, and Joe

Gaspard. The use of microscope equipment and technical advice were provided by Dr.

Gary Litman, Dr. Donna Eason, Diana Skapura, Tony Greco, Louis Kerr, and Dr. Jackie

Webb. Years of helpful discussions, brainstorming, advice, logistical help, and comic relief were provided by my USF labmates, Kyle Mara, Samantha Mulvany, Lisa

Whitenack, Laura Habegger, and Janne Pfeiffenberger, and my Mote Center for Shark

Research colleagues, Dr. Carl Luer, Dr. Cathy Walsh, Dr. Nick Whitney, Dr. Jennifer

Yordy, Dr. Jose Castro, Stephanie Leggett, and Jodi Miedema. This project was an interdisciplinary and inter-institutional collaboration, with work conducted at the Marine

Biological Laboratory in Woods Hole, MA and Mote Marine Laboratory in Sarasota, FL.

I would like to thank everyone at both institutions for making me feel welcome, particularly Erin Kinney Dikmanis and Andris Dikmanis, who put a roof over my head on Cape Cod. Finally, thank you to all of my friends and family, particularly my parents,

Michael and Betty Gardiner, for always believing in me and encouraging me when the going got tough, I couldn’t have done it without you. Funding for this work was provided by the National Science Foundation (IOS-083440, IOS-0841502, and IOS-

0841478), a Nelson Behavior Award from the American Elasmobranch Society, a Raney

Fund Award from the American Society of Ichthyologists and Herpetologists, a Lerner-

Gray Grant for Marine Research from the American Museum of Natural History, and the

Porter Family Foundation.

TABLE OF CONTENTS

List of Tables ...... iv

List of Figures ...... v

Abstract ...... viii

Chapter One. The Sensory Biology of Sharks ...... 1 Introduction ...... 1 Chemoreception ...... 2 Olfaction ...... 3 Gustation ...... 8 Solitary Chemosensory Cells ...... 9 The Common Chemical Sense ...... 10 Hearing ...... 10 Mechanoreception ...... 12 Vision ...... 16 Electroreception ...... 22 Multimodal Integration ...... 25 Experimental Animals ...... 29 The Smooth Dogfish, Mustelus canis ...... 29 The Nurse Shark, Ginglymostoma cirratum ...... 29 The Bonnethead, Sphyrna tiburo ...... 30 The Blacktip Shark, Carcharhinus limbatus ...... 31 Literature Cited ...... 32

Chapter Two. The Function of Bilateral Odor Arrival Time Differences in Olfactory Orientation of Sharks ...... 68 Summary ...... 68 Results and Discussion ...... 69 Experimental Procedures ...... 76 Supplemental Information ...... 79 Supplemental Data ...... 79 Supplemental Experimental Procedures ...... 80 Experimental Animals ...... 80 Headstage Apparatus ...... 81 Odor Stimuli...... 82 Literature Cited ...... 83 i

Chapter Three. Multisensory Integration in Food Search Behaviors ...... 89 Abstract ...... 89 Introduction ...... 90 Materials and Methods ...... 97 Experimental Animals ...... 97 Behavioral Procedures ...... 98 Sensory Deprivation...... 99 Video Analysis ...... 102 Statistical Analysis ...... 103 Results ...... 104 Tracking ...... 104 Orientation and Striking ...... 108 Capture ...... 111 Wall Collisions...... 114 Discussion ...... 115 Detection ...... 117 Tracking ...... 119 Orientation and Striking ...... 122 Capture ...... 126 Navigation and Obstacle Avoidance ...... 129 Summary and Conclusions ...... 130 Literature Cited ...... 131

Chapter Four. Sensory Contributions to Prey Capture Kinematics ...... 172 Abstract ...... 172 Introduction ...... 173 Materials and Methods ...... 177 Experimental Animals ...... 177 Behavioral Procedures ...... 178 Sensory Deprivation...... 179 Video Analysis ...... 182 Statistical Analysis ...... 185 Results ...... 186 Approach Kinematics ...... 186 Capture Kinematics ...... 188 Cranial Elevation ...... 189 Gape Cycle ...... 190 Upper Jaw Protrusion and Labial Cartilage Protrusion ...... 191 Prey Capture and the Ram-Suction Index ...... 192 Discussion ...... 195 Approach Kinematics ...... 195 Capture Kinematics ...... 200 Summary and Conclusions ...... 207 Literature Cited ...... 208

Chapter Five. Multimodal Integration and Sensory Plasticity in Shark Feeding Behavior ...... 241 Literature Cited ...... 254

About the Author ...... End Page

iii

LIST OF TABLES

Table 2.1: Summary of the Proportion of Null Responses Versus Turns for the Various Stimulus Patterns Used...... 79

Table 2.2: Analysis of Sample Trial Showing 10 Stimulus Pairs ...... 80

Table 3.1: Data Summary – Blacktip Sharks ...... 146

Table 3.2: Data Summary – Nurse Sharks ...... 149

Table 3.3: Data Summary – Bonnetheads ...... 154

Table 3.4: Strike Velocities for Successful Captures vs. Misses – Blacktip Sharks ...... 156

Table 4.1: Kinematic Data Summary – Blacktip Sharks ...... 218

Table 4.2: Kinematic Data Summary – Nurse Sharks ...... 223

Table 4.3: Kinematic Data Summary – Bonnetheads ...... 227

LIST OF FIGURES

Figure 1.1: The Olfactory Rosette ...... 51

Figure 1.2: Nasal Ventilation – Active Elasmobranch ...... 52

Figure 1.3: Nasal Ventilation – Sedentary Elasmobranch ...... 53

Figure 1.4: Shark Brain ...... 54

Figure 1.5: Taste Buds and Solitary Chemosensory Cells ...... 55

Figure 1.6: The Elasmobranch Inner Ear ...... 56

Figure 1.7: The Sensory Cells of the Elasmobranch Neuromast ...... 57

Figure 1.8: Elasmobranch Neuromasts ...... 58

Figure 1.9: Modified Placoid Scales ...... 59

Figure 1.10: Distribution of the Superficial Neuromasts (Pit Organs) in Sharks ...... 60

Figure 1.11: Distribution of the Lateral Line Canals in Sharks ...... 61

Figure 1.12: The Visual Fields of Sharks ...... 62

Figure 1.13: The Elasmobranch Eye ...... 63

Figure 1.14: Visual Streaks and Areae ...... 64

Figure 1.15: The Ampullary Electroreceptor Organ of Elasmobranchs ...... 65

Figure 1.16: Geomagnetic Orientation in Sharks ...... 66

Figure 1.17: The Electrosensory Arrays in Sharks ...... 67

Figure 2.1: Proportion of Turns in Response to Different Stimulus Patterns ...... 87

Figure 2.2: Differences in Odor Arrival Time ...... 88 v

Figure 3.1: Diagram of the Multi-modal Stimulus Field ...... 157

Figure 3.2: Experimental Flume ...... 158

Figure 3.3: Strike Angle Measurement ...... 159

Figure 3.4: Swim Velocity ...... 160

Figure 3.5: Turn Velocity ...... 161

Figure 3.6: Turn Frequency ...... 162

Figure 3.7: Standardized Tracking Time ...... 163

Figure 3.8: Swim Time vs. Rest Time ...... 164

Figure 3.9: Orientation Distance ...... 165

Figure 3.10: Striking ...... 166

Figure 3.11: Strike Velocity ...... 167

Figure 3.12: Number of Misses ...... 168

Figure 3.13: Successful Strike vs. Miss Velocity ...... 169

Figure 3.14: Capture Success ...... 170

Figure 3.15: Frequency of Wall Collisions ...... 171

Figure 4.1: Experimental Flume ...... 230

Figure 4.2: Approach Velocity ...... 231

Figure 4.3: Predator-Prey Distance at the Start of Capture ...... 232

Figure 4.4: Capture Kinematics ...... 233

Figure 4.5: Gape Cycle ...... 234

Figure 4.6: RSI Values ...... 235

Figure 4.7: Predator and Prey Movements During Capture ...... 236

Figure 4.8: Prey Velocity ...... 237

Figure 4.9: Predator Velocity ...... 238

Figure 4.10: Frequency of Bites...... 239

Figure 4.11: Hypothetical Capture Angles ...... 240

Figure 5.1: The Sensory Hierarchy of the Blacktip Shark ...... 264

Figure 5.2: The Sensory Hierarchy of the Nurse Shark ...... 265

Figure 5.3: The Sensory Hierarchy of the Bonnethead ...... 266

vii

ABSTRACT

Multimodal sensory input directs simple and complex behaviors in animals. Most research to date has been limited to studies of individual senses rather than multiple senses working together, leading to important advances in our comprehension of the sensory systems in isolation, but not their complementary and alternative roles in difficult behavioral tasks, such as feeding. In the marine environment, a prey item might emit an odor, create a hydrodynamic disturbance, such as from gill movements or swimming, be visible to the predator, produce a sound, and/or produce a weak electrical field.

Therefore, the goal of this study was to investigate the integration of olfaction, mechanoreception by the lateral line system, vision, and electroreception in a marine animal. Sharks were chosen as a model organism in which to investigate multisensory integration because of their sensitivity and acuity, the presence of the same suite of sensory modalities in all species, the availability of experimental animals from different species, habitats and ecologies, and the rich literature on sharks’ prey capture behavior.

Two approaches were used: controlled artificial stimuli, delivered to the animals, were used to determine the spatial and concentration characteristics of odor encounters that guide the initial orientation to an odor plume in the far field in a model elasmobranch, the smooth dogfish, Mustelus canis; and sensory deprivation was used to restrict the availability of natural cues emanating from live prey items in order to elucidate the complementary and alternating roles of the senses in detecting, tracking, orienting to, striking at, and ultimately capturing prey. In the latter experiments, three species of viii

sharks from different ecological niches were investigated: benthic, suction-feeding nurse

sharks (Ginglymostoma cirratum) that hunt nocturnally for fish; ram-biting bonnetheads

(Sphyrna tiburo) that scoop crustaceans off the bottom of seagrass beds; and ram-feeding

blacktip sharks (Carcharhinus limbatus) that rapidly chase down midwater teleost prey.

In orienting to odor patches, bilateral time differences between the nares are more

important than concentration differences, such that animals turn toward the side

stimulated first, even with delayed pulses of higher concentration. This response would

steer the shark into each oncoming odor patch, helping the animal maintain contact with

an odor plume. Sensory deprivation experiments revealed similarities and differences

among species in terms of which senses they choose to focus on for particular behaviors,

likely as a result of differences in the environments that they hunt in, type of prey

consumed, and foraging strategies used, as well as anatomical differences in the central nervous system and the sensory organs. In most cases, multiple senses can be used for the same behavioral task. Thus, sharks are capable of successfully capturing prey, even

when the optimal sensory cues are unavailable, by switching to alternative sensory modalities, which indicates that feeding behavior is plastic. Nurse sharks rely primarily on olfaction for detection. Olfaction in combination with vision, the lateral line, or touch is required for tracking. Nurse sharks orient to prey using the lateral line, vision, or electroreception, but will not ingest food if olfaction is blocked. Capture is mediated by the electrosensory system or tactile cues. Bonnetheads normally detect prey using olfaction, rely on olfactory-based tracking until they are close to the prey, then vision to line up a strike, and finally electroreception to time the jaw movements for capture. They can detect, orient, and strike visually in the absence of olfactory cues. Blacktip sharks

ix also detect prey using olfaction or vision. Olfaction is used in combination with vision or the lateral line system for tracking. Long-distance orientation and striking is visually mediated, but strike precision relies on lateral line cues and an increase in misses occurs when this system is blocked. In the absence of vision, short-range orientation and striking can occur using lateral line cues. Capture is mediated by electroreception or tactile cues. Collectively, these results were used to develop species-specific sensory hierarchies for shark feeding behavior in a captive environment, the first such hierarchies to cover a complete behavioral sequence in a vertebrate.

CHAPTER 1: THE SENSORY BIOLOGY OF SHARKS

Introduction

Our understanding of most animal’s sensory systems, including sharks, is largely due to isolated studies of the individual senses rather than multiple senses working together. This has led to important advances in our comprehension of one sensory system or another but not their complementary and alternative roles. Integration of multimodal sensory information in the elasmobranch CNS ultimately leads to a behavioral response at the level of the whole animal. How animals integrate the complex input of environmental information through their various senses to form an adaptive response is among the most interesting questions in sensory biology. Sharks were selected as the model organisms for this study because of their sensory sensitivity and acuity, the presence of the same suite of sensory modalities in all species, the availability of experimental animals from different species, habitats and ecologies, and the rich literature on sharks’ prey capture behavior.

Sharks are a 400 million year old vertebrate lineage. A great deal of the success of these apex predators is owed to the tremendous sensory capabilities for which they have become legendary. Much of this reputation is greatly exaggerated and sharks have become the subject of several pervasive myths, such as possessing the ability to detect a single drop of blood in an Olympic-sized swimming pool. In addition to their renowned sense of smell (reviewed in Parker, 1914, 1922), sharks also possess binocular vision, 1 directional hearing, mechanoreception, and a very sensitive electrosensory system

(reviewed in Hueter et al., 2004).

Chemoreception

Olfaction

Olfaction has been suggested to be an important, if not the primary, sensory modality involved in shark feeding behavior based on behavioral observations (Parker,

1909; Sheldon, 1909, 1911; Parker and Sheldon, 1913; Parker, 1914) and due to the relatively large size of their olfactory structures, compared to those of other vertebrates

(reviewed in Northcutt, 1978). The two elasmobranch olfactory organs are ellipsoid saclike structures found in laterally placed cartilaginous capsules on the ventral aspect of the head, in front of the mouth. They are open to the environment via nostrils (nares), which are typically divided by skin-covered flaps into a more lateral incurrent nostril

(naris) and a more medial excurrent nostril (Tester, 1963a; Theisen et al., 1986b; Zeiske et al., 1986; Zeiske et al., 1987). In most species, the olfactory organs are entirely separate from the mouth, but in a few species, they are in close association with the mouth or even connected to it via a deep groove, called the nasoral groove, which extends posteriorly from the excurrent naris, forming a virtual tube between the naris and the mouth (e.g. Orectolobidae, Heterodontidae, Tester, 1963a; Bell, 1993). The external nasal morphology varies greatly among species, though some broad trends have been found based on lifestyle. Benthic and sedentary species tend to have large nasal openings, while bentho-pelagic and faster-swimming species tend to have smaller, slit- like openings or large nasal flaps (Schluessel et al., 2008). An anterior depression or

2 groove may be present, helping to channel water into the incurrent opening, and the excurrent opening may be associated with a shallow posterior depression (Tester, 1963a;

Zeiske et al., 1986; Zeiske et al., 1987). In hammerhead sharks (Sphyrnidae), these prenarial grooves are particularly well-developed (Gilbert, 1967). In addition to the deep, narrow (prenarial) grooves, which extend along the anterior edge of each side of the head, linking to the incurrent nares (major nasal grooves), a second set of smaller grooves (minor nasal grooves) run parallel and anterior to each incurrent nostril on the dorsal side of the head, further assisting with channeling water into the incurrent naris

(Abel et al., 2010).

The olfactory sac is nearly completely filled by an olfactory rosette consisting of two rows of stacked wing-shaped plates, called lamellae, which originate from a central ridge (raphe) and attach to the wall of the olfactory cavity (Tester, 1963a; Theisen et al.,

1986b; Zeiske et al., 1986; Zeiske et al., 1987; Kajiura et al., 2005; Meredith and

Kajiura, 2010) (Figure 1A,B). The lamellae are largest in the middle, decreasing in size towards both the medial and lateral ends (Theisen et al., 1986b; Theiss et al., 2009;

Meredith and Kajiura, 2010). Each lamella is covered with secondary folds (secondary lamellae), which greatly increase the surface area of the olfactory epithelium. The olfactory epithelium is divided into sensory and nonsensory areas. The nonsensory, squamous epithelium is composed of cells that bear microvilli only, and numerous goblet cells (Theisen et al., 1986b; Zeiske et al., 1986; Zeiske et al., 1987; Schluessel et al., 2008; Theiss et al., 2009). It is generally found on the margins of the lamellae, though in some species, it extends along the ridges of the secondary folds, and in other species, a patchy, irregular distribution of sensory and nonsensory areas is found

(Schluessel et al., 2008; Theiss et al., 2009) (Figure 1C,D). The much larger, centrally located sensory epithelium is composed of pseudostratified, columnar epithelium. It contains receptor cells, supporting cells (which bear numerous cilia), and basal cells, along with occasional goblet cells. It is similar to that found in olfactory systems of most vertebrates, with the major exception that the elasmobranch bipolar receptor cells are not ciliated but rather have a dendritic knob (olfactory knob) from which extends a tuft of microvilli (Reese and Brightman, 1970; Theisen et al., 1986b; Zeiske et al., 1986; Zeiske et al., 1987; Schluessel et al., 2008; Theiss et al., 2009) (Figure 1E). Similar microvillous receptors have been found along with the “typical” ciliated type in certain bony fishes. Studies on the clearnose skate, Raja eglanteria, identify two types of nonciliated olfactory receptor neurons (Takami et al., 1994). Type 1 is typical of those found in the other fishes (as above); the type 2 cell, so far unique to elasmobranchs, is distinguished from the type 1 by its thicker dendritic knob and microvilli that are shorter, thicker, and more regularly arranged. The functional meaning of the morphological differences in receptor types has yet to be determined.

The olfactory morphology of numerous elasmobranch species has been examined. Olfactory rosette size, lamellar number, and sensory surface area vary by species (Kajiura et al., 2005; Schluessel et al., 2008; Theiss et al., 2009; Meredith and

Kajiura, 2010), these differences can be correlated with habitat type, but not phylogeny or prey type (Schluessel et al., 2008). Bentho-pelagic sharks and rays possess higher numbers of lamellae, larger olfactory surface areas, and larger rosettes than benthic species (Schluessel et al., 2008; Meredith and Kajiura, 2010). The ontogeny of the olfactory system has been examined in only a handful of species, but it appears to be

4 well-developed at birth, undergoing only minor changes as the animal grows. The morphology of the nares and olfactory rosettes, and ultrastructure of the epithelium of juveniles closely resembles that of the adults (Schluessel et al., 2010). The olfactory bulbs undergo growth, increasing with body size, though not proportionally (Schluessel et al., 2010), such that the olfactory bulbs represent a larger proportion of the brain volume in adults as compared with juveniles (Lisney et al., 2007). The olfactory rosettes undergo similar growth and while lamellar surface area increases with body size, lamellar number does not, except in the spotted eagle ray, Aetobatus narinari

(Meredith and Kajiura, 2010; Schluessel et al., 2010).

Interspecific differences in olfactory morphological data have often been used to assess olfactory capability, with increased sensitivity inferred from increased size

(Theisen et al., 1986a; Zeiske et al., 1986; Zeiske et al., 1987; Kajiura et al., 2005;

Lisney et al., 2007; Schluessel et al., 2008; Theiss et al., 2009; Schluessel et al., 2010), but electrophysiological data refute this. The underwater electro-olfactogram (EOG) is a tool for recording the extracellular DC field potentials or analog of the summed electrical activity of the olfactory epithelium in response to chemical stimulation (Silver et al. 1976). EOG responses have been studied in eight elasmobranchs: four sharks, two rays, and two skates (Silver et al., 1976; Hodgson and Mathewson, 1978a; Silver, 1979;

Zeiske et al., 1986; Nikonov et al., 1990; Tricas et al., 2009; Meredith and Kajiura,

2010). Several amino acids, known to be effective stimuli for evoking EOGs in bony fishes and behavioral responses in both bony fishes and elasmobranchs, were tested in these species. The thresholds for individual amino acids varied by species, but in general, neutral amino acids are more stimulatory, while valine, proline, and isoleucine

(also neutral, but with branched side-chains or secondary amine groups), are the least stimulatory. These results are similar for elasmobranchs and teleost fishes (Hara, 1994;

Meredith and Kajiura, 2010). The EOG magnitude increased exponentially with the log of the stimulus concentration and calculated thresholds ranged between 10-6 and 10-11 M.

These levels are similar to those reported for bony fishes (teleosts) (Hara, 1994), as well as the levels of free amino acids in seawater (Pocklington, 1971; Kuznetsova et al.,

2004). Despite differences in lamellar number and surface area, olfactory thresholds do not differ significantly among elasmobranch species. Since behavioral evidence is lacking, the functional significance of these interspecific differences in olfactory morphology is unknown.

The dynamics of nasal water circulation (nasal ventilation) have been analyzed in a series of detailed studies on several sharks. Briefly, water enters the incurrent nostril, passes along the incurrent channel, is drawn through the interlamellar channels, moves into peripheral channels on the outer edges of the lamellae, then enters the excurrent channel and passes back out to the environment via the excurrent nostril

(Theisen et al., 1986b; Zeiske et al., 1986; Zeiske et al., 1987; Abel et al., 2010) (Figure

2). In actively swimming elasmobranchs, this water flow is likely generated by differences in pressure between the incurrent and excurrent nostrils, which are primarily caused by the forward motion of the animal (Theisen et al., 1986b; Zeiske et al., 1986;

Zeiske et al., 1987). In benthic and more sedentary species, nasal ventilation may be aided by a buccopharyngeal pump: as water is pumped into the mouth to ventilate the gills, it is also drawn through the “virtual tube” between the olfactory organ and the mouth, which results in the flow of water into the incurrent nostril and through the

6 olfactory rosette. These structures thus act as a functional internal naris (Bell, 1993)

(Figure 3). Whether the multi-ciliated nonsensory cells act to propel water is unknown, but no nasal currents were observed in stationary lemon sharks, Negaprion brevirostris

(Zeiske et al., 1986).

The olfactory receptors send their output via axons, which form a very short olfactory nerve (cranial nerve I) to the adjacent olfactory bulbs (OB), part of the telencephalon. A pair of olfactory tracts, also part of the forebrain, links the OB to the telencephalic hemispheres (Figure 4). Species differences in the size of the OB relative to total brain mass or volume have been calculated in several elasmobranch species and used to suggest differences in ecology, particularly in reliance on smell in a variety of behaviors, especially feeding and/or social behavior (Northcutt, 1978; Demski and

Northcutt, 1996; Lisney and Collin, 2006; Lisney et al., 2007). It is unclear at this time, however, how much of the variation can be attributed to phylogeny as opposed to interspecific differences in behavior and ecology. Without supporting behavioral and ecological evidence, it is impossible to determine how observed differences in the size of any of the sensory structures relate to differences in performance.

There is a large body of research and consequently a wealth of information on the involvement of olfaction in feeding behavior in sharks (Parker, 1909; Sheldon, 1909,

1911; Parker and Sheldon, 1913; Parker, 1914, 1922; Hobson, 1963; Tester, 1963a, b;

Mathewson and Hodgson, 1972; Kleerekoper et al., 1975; Hodgson and Mathewson,

1978b; Johnsen and Teeter, 1985; Gardiner and Atema, 2007). Only a few studies have focused on the olfactory mediation of other behaviors. Olfaction has been suggested to play a role in mating, but the evidence is indirect and based on observations of “close

7 following” behavior in which the male swims directly behind a female with its snout positioned against her vent, which has been observed in several species (Myrberg and

Gruber, 1974; Johnson and Nelson, 1978; Klimley, 1980; Tricas, 1980; Luer and

Gilbert, 1985; Gordon, 1993; Carrier et al., 1994). At this time, however, there is no direct experimental evidence of olfactory-mediated responses to sex pheromones in elasmobranchs. While olfaction likely functions in predator avoidance in elasmobranchs, as it does in bony fishes, only one study to date provides direct experimental evidence of a response to predator odors. Juvenile lemon sharks, N. brevirostris, can be aroused from tonic immobility by the odors of American crocodiles,

Crocodylus acutus, a potential predator (Rasmussen and Schmidt, 1992).

Gustation

Only limited studies have been conducted on the gustatory system of elasmobranchs. Anatomical studies have identified receptors that closely resemble the taste organs of other vertebrates and behavioral observations suggest that gustation is important for the acceptance of food in sharks (see Sheldon, 1909; and review by Tester,

1963a). Taste buds consist of small papillae, covered with epithelium, with central elongate sensory receptor cells with apical microvilli which form pores (Figure 5A).

The receptor cells are supplied with nerve fibers from branches of the facial (VII), glossopharyngeal (IX), and vagus (X) nerves, and possibly free nerve endings (Norris and Hughes, 1920; Herrick, 1924; Daniel, 1928; Aronson, 1963; Whitear and Moate,

1994a, b). Found over the entire oropharyngal cavity, they are most densely distributed on the roof of the mouth in the spiny dogfish, Squalus acanthias (Cook and Neal, 1921).

In general, the gustatory system in elasmobranchs appears similar to that of other vertebrates (reviewed in Northcutt, 2004).

Solitary Chemosensory Cells

Solitary chemosensory cells (SCCs) are found in a number of lower vertebrate taxa. These spindle-shaped, epidermal sensory cells are found protruding between the squamous cells of the superficial layer of the epidermis, with a single apical process which bears one or a few microvilli (fish, amphibians) or many microvilli (oligovillous cells in lampreys), and are innervated by spinal or cranial (VII, facial) nerves (reviewed in Kotrschal, 1995) (Figure 5B). Their structure resembles that of taste buds, suggesting a chemosensory function, verified through electrophysiological experiments on teleost fish (Silver and Finger, 1984; Peters et al., 1987) and lampreys (Baatrup and Doving,

1985), which demonstrated that they are sensitive to skin washes and bile from other fish, but not amino acids. It has been hypothesized that in rocklings, Gaidropsarus vulgaris, SCCs allow for bulk water sampling, mainly for detecting the presence of predators upstream (Kotrschal et al., 1996), while in sea robins, Prionotus sp., they may be used to find food (Silver and Finger, 1984). SCCs have been examined in only a handful of species, thus their biological function remains poorly understood, particularly in elasmobranchs, and to date no term for the sense that they mediate has been developed. In elasmobranchs, SCCs have only been confirmed in one species, the thornback ray, Raja clavata (Whitear and Moate, 1994a) where they are found in the oral cavity. However, it has recently been suggested that they may be present on the dorso-lateral surface of the skin, near the pit organs, in Port Jackson sharks,

Heterodontus portusjacksoni, and whitetip reef sharks, Triaenodon obesus (Peach,

2005). Further work is needed to determine the distribution and function of SCCs in elasmobranchs.

The Common Chemical Sense

The common chemical sense, the ability to detect irritating substances, is considered separate from olfaction and gustation. Free nerve endings, which in teleosts occur in the oral and nasal cavities, as well as all over the skin, serve as receptors

(Tester, 1963a; Whitear, 1971). Studies in other vertebrates indicate that the nerves involved in such reactions are part of the somatosensory system and appear to represent a subset of temperature- and pain-sensitive fibers, including spinal nerves and cranial nerves V (trigeminal), VII (facial), IX (glossopharyngeal), and vagus (X). Sharks respond behaviorally to chemical irritants on the skin and in the nostrils, even after the olfactory tracts have been severed. (Sheldon, 1909). Presumably, the function of this system in elasmobranchs, as in other vertebrates, is protection from damaging chemicals. It has been suggested that sharks’ adverse reactions to natural toxins, such as that produced by the skin of the Moses sole, Pardachirus marmoratus (Clark, 1974), may be mediated by this category of unmyelinated somatosensory ending.

Hearing

Sharks possess only inner ear labyrinths, and they do not have swim bladders

(Figure 6). Three of the sensory maculae (sacculus, lagena, and utriculus) each consist

10 of a patch of sensory hair cells overlain by otoconia, consisting of calcium carbonate, aragonite, or silicon dioxide granules, as well as other minerals, embedded in a mucopolysaccharide matrix, which act as an inertial mass (Tester et al., 1972; Lychakov et al., 2000; Mills et al., 2011). The composition and shape of the otoconia vary by species (Evangelista et al., 2010), which may be a mixture of endogenous and exogenous materials (such as silicon dioxide), taken up as particles, called otarenae, via the endolymphatic ducts, which connect to the inner ear (Lychakov et al., 2000). A fourth sensory macula found in elasmobranchs and amphobians, the macula neglecta, is found in the posterior semicircular canal. It consists of one or two patches of sensory epithelium and lacks otoconia but has a tympanic connection to the parietal fossa in the cranium via the fenestra ovalis, an area of loose connective tissue, which has been suggested to provide direct pathway for sound transmission (Corwin, 1978, 1981, 1989).

The maculae are sensitive to differential acceleration between the body and sensory maculae, as a result of the particle motion component of sound, and the otoconial mass

(Tester et al., 1972; reviewed in Hueter et al., 2004). In the acoustic near field, the particle motion component of sound dominates. Sensitivity to sound in the far field, where the pressure component of sound dominates, requires a pressure-to-displacement transducer, such as the swimbladder in bony fishes (Kalmijn, 1988). Since they lack a swimbladder, sharks should be most sensitive to sounds in the near field. However, in field studies, sharks have been demonstrated to be attracted to sound sources from the far field (Nelson and Gruber, 1963; Richard, 1968; Myrberg et al., 1969; Nelson et al.,

1969; Myrberg et al., 1972). They have also been demonstrated in the laboratory to be sensitive to acoustic pressure, possibly through compression of the fluids inside the

11 labyrinth of the inner ear causing displacements at the fenestra ovalis (Van den Berg and

Schuijf, 1983), however, this claim has been refuted (Kalmijn, 1988). Shark responses to particle accelerations are primarily attributed to the saccular macula and macula neglecta (Corwin, 1981; Casper and Mann, 2007), while the lagenal macula and utricular macula respond to gravitational stimuli due to tilting of the animal’s body

(Lowenstein and Roberts, 1951).

Mechanoreception

The lateral line is a mechanosensory system found in fish and amphibians that functions as a hydrodynamic detector (reviewed in Coombs and Braun, 2003). In bony fishes, the lateral line mediates behaviors such as rheotaxis (orientation to flow) and station holding, prey detection, schooling, predator avoidance, hydrodynamic imaging for object localization and obstacle avoidance, and social communication. Less is known about the function of the lateral line in elasmobranchs. The basic unit of this system is the neuromast which consists of sensory hair cells that have a central kinocilium and different sized bundles of directionally sensitive stereovilli in a staircase arrangement at their apices (Peach and Rouse, 2000), topped by an acellular gelatinous cupula (Tester and Kendall, 1968) (Figure 7, 8B). Water motion causes displacement of the cupula, which exerts a shearing force on the stereocilia, causing them to bend.

Bending of the stereocilia towards the kinocilium causes excitatory depolarization of the hair cell and the release of neurotransmitters at the synapse with afferent nerves, resulting in an action potential. Bending of the stereocilia away from the kinocilium

12 results in inhibitory hyperpolarization of the hair cell. (Denton and Gray, 1988). The receptor cells of the neuromasts in the cephalic region are innervated by afferents and efferents from the anterior lateral line nerve (VIII) while those of the neuromasts found along the body are innervated by the posterior lateral line nerve (Northcutt, 1978, 1989).

Superficial neuromasts are found distributed along the head and body on the surface of the skin in shallow pits. In most elasmobranchs, they sit beneath two modified placoid scales and are called pit organs (Figure 9). Elasmobranch pit organs are on the order of 40-60 μm wide, 80-120 μm long and 70-100 μm tall, from basement membrane to apex (Peach and Rouse, 2000). The number and distribution pattern of the surface neuromasts varies by species, from only a few per side in the horn shark

(Heterodontus spp.) to over 600 per side in the scalloped hammerhead (Sphyrna lewini)

(Tester and Nelson, 1967; Peach and Marshall, 2000; Peach, 2003) (Figure 10). In general, benthic elasmobranchs have fewer surface neuromasts than pelagic elasmobranchs (Peach and Rouse, 2004).

Neuromasts are also found beneath the skin within fluid-filled canals (canal neuromasts, Figure 8A). Supraorbital, infraorbital, hyomandibular, and mandibular canals are located on the head. The posterior lateral line canal extends caudally from the endolymphatic pores on the dorsal side of the head, along the flanks to the tip of the tail

(Tester and Kendall, 1969; Boord and Campbell, 1977; Roberts, 1978; Chu and Wen,

1979b; Maruska, 2001) (Figure 11). In elasmobranchs, the canals are 0.3 x 0.5 mm and the canal neuromasts within are 30-50 μm wide, 150-300 μm long, and 30-50 μm tall, from basement membrane to apex (Peach and Rouse, 2000). The canal neuromasts are situated adjacent to one another and the gap between them is so small that they form a

13 nearly continuous sensory epithelium, with multiple neuromasts between adjacent pores

(Ewart and Mitchell, 1892; Johnson, 1917; Hama and Yamada, 1977). This is in contrast to bony fish that have large gaps, on the order of 2-5 mm between canal neuromasts, and only a single neuromast between each pore (Webb and Northcutt,

1997). There are two classes of canals in elasmobranchs, pored and non-pored. Pored canals are open to the exterior via neuromast-free tubules that extend to the surface of the skin and terminate in pores. Non-pored canals are found on the head of many shark species (Chu and Wen, 1979a; Maruska, 2001). They are isolated from the environment and therefore cannot respond to water motion but have been suggested to function as tactile receptors by responding to the velocity of skin movements that result from contact with objects in the environment such as the substrate, prey, or conspecifics during social interactions (Maruska, 2001; Maruska and Tricas, 2004).

Electrophysiological studies have shown them to be far more sensitive than cutaneous tactile receptors and responsive to low-frequency (≤10Hz) stimuli (Maruska and Tricas,

2004). The number and distribution pattern of pored and non-pored canals varies among species.

The lateral line system is stimulated by differential movement between the body and the surrounding water and is considered a short-range system, functioning over one- two body lengths. Neuromasts are generally sensitive to low frequency stimuli (≤ 200

Hz) and to velocities in the μm/s range (superficial neuromasts) and accelerations in the mm/s2 range (canal neuromasts) (Münz, 1985; Bleckmann et al., 1989; Coombs and

Janssen, 1990; Maruska and Tricas, 2004). In bony fish, the surface neuromasts and canal neuromasts have been suggested to play different behavioral roles. The canal

14 neuromasts have been suggested to act as low-pass filters with respect to fluid acceleration and high-pass filters with respect to velocity, reducing the effects of low frequency motion, such as from ambient water motion (Denton and Gray, 1988, 1989).

Behaviorally, the canal neuromasts have been shown to be involved in tracking smaller- scale (high frequency) turbulence, such as the wake generated by prey (Coombs et al.,

2001). The surface neuromasts, on the other hand, respond to fluid velocity (Chagnaud et al., 2008) and have been suggested to function in orientation to the mean current, controlling rheotaxis behaviors (Montgomery et al., 1997; Baker and Montgomery,

1999; Montgomery et al., 2000; Baker et al., 2002). Superficial neuromasts may also be involved in rheotaxis behaviors in elasmobranchs. Port Jackson sharks, Heterodontus portjacksoni, have a reduced ability to orient upstream when resting after ablation of the surface neuromasts (Peach, 2001). Recent work (Gardiner and Atema, 2007) has demonstrated that the smooth dogfish requires lateral line information to locate odor sources. This species can navigate upstream through an odor field to the general area of a turbulent odor source using either vision or the lateral line, but the lateral line is necessary to precisely locate the source of coincident odor and flow. This suggests that these animals are tracking the fine-scale structure of the plume, i.e. a turbulent wake, flavored with food/prey odor, shed either by a moving prey item in still water or a still piece of food in flowing water (eddy chemotaxis, Atema, 1996). The vortices shed by a swimming prey item, such as a fish, can persist for several minutes after the fish has passed through the environment (Hanke et al., 2000; Hanke and Bleckmann, 2004), leaving a hydrodynamic trail for the predator to follow.

Vision

The eyes of sharks are generally laterally placed on the head, with the more benthic sharks (e.g., orectolobids, squatinids) having more dorsally positioned eyes, which is believed to be an adaptation for a benthic lifestyle. Eye size in elasmobranchs is generally small in relation to body size, but relatively larger in juveniles (Lisney et al.,

2007) and in some notable species, such as the bigeye thresher shark, Alopias superciliosus. Eye size differences correlate with habitat type, activity level, and prey type. Oceanic species have relatively larger eyes than coastal and benthic species and more active swimmers that feed on active, mobile prey have relatively larger eyes than more sluggish species that feed on sedentary prey (Lisney and Collin, 2007). As with the bony fishes (Warrant and Locket, 2004), relative eye size in mesopelagic deep-sea sharks is often large to allow for enhanced light-gathering properties.

The two eyes oppose each other, which can allow for a nearly 360° visual field in at least one plane of vision, however, the dynamic visual field can be extended beyond

360° because of the yaw of the head during swimming (McComb et al., 2009).

Binocular overlap is generally small, except in the hammerhead sharks (Sphyrnidae), whose enhanced frontal vision comes at the expense of larger blind areas (McComb and

Kajiura, 2008; Litherland et al., 2009; McComb et al., 2009) (Figure 12). Blind areas exist directly in front of the snout or behind the head, the sizes of which depend on the configuration of the head and the separation of the eyes, but typically the forward blind area extends less than one body length in front of the rostrum.

The elasmobranch eye (Figure 13) has a thick cartilaginous outer layer, called the sclera. The transparent cornea is virtually optically absent underwater due to its

16 similarity in refractive index to that of seawater (Hueter, 1991), thus the crystalline lens provides the total refractive power of the eye. Elasmobranch lenses are typically large, relatively free of optical aberration, and ellipsoidal in shape, although the spiny dogfish,

Squalus acanthias, and clearnose skate, Raja eglanteria, have nearly spherical lenses

(Sivak, 1978a, 1991). Accommodation is the ability to change the refractive power of the eye to focus on objects at varying distances. Elasmobranchs that accommodate do not vary lens shape as humans do, but instead change the position of the lens by moving it toward the retina (for distant targets) or away from the retina (for near targets). The lens is supported dorsally by a suspensory ligament and ventrally by the pseudocampanule, a papilla with ostensibly contractile function (Sivak and Gilbert,

1976). Evidence of accommodation in elasmobranchs has been inconsistent across species, and many of the species studied have appeared to be hyperopic (far-sighted) in the resting state of the eye (Sivak, 1978b; Hueter, 1980; Hueter and Gruber, 1982;

Spielman and Gruber, 1983), but this may be an artifact of handling stress as free- swimming, lemon sharks, N. brevirostris, are not hyperopic and can accommodate

(Hueter et al., 2001).

At the back of the elasmobranch eye, behind the retina and in front of the sclera, lies the choroid, the only vascularized tissue within the adult elasmobranch eye. The elasmobranch retina itself is not vascularized and typically contains no obvious landmarks other than the optic disk (corresponding to a small blind spot in the visual field), which contains no photoreceptors and marks the exit of retinal ganglion cell fibers via the optic nerve from retina to CNS. The choroid in nearly all elasmobranchs contains a specialized reflective layer known as the tapetum lucidum, which consists of a series

17 of parallel, platelike cells containing guanine crystals (Gilbert, 1963; Denton and Nicol,

1964). This layer functions to reflect back those photons that have passed through the retina but have not been absorbed by the photoreceptor layer, allowing a second chance for detection of photons, thereby boosting sensitivity of the eye in dim light. In many elasmobranchs, the tapetum is occlusible, via the migration of dark pigment granules within the tapetal melanophores under bright light conditions (Nicol, 1964; Heath,

1991).

The shark retina contains both rods, which set the limits of the visual sensitivity of the eye, and cones, which are responsible for color vision and higher visual acuity.

Both groups of photoreceptors contain visual pigments, which absorb photons of light.

The pigments are composed of protein called opsin and a chromophore prosthetic group related to either vitamin A1 (called rhodopsins or chyropsins) or A2 (called porphyropsins) (Cohen, 1991). Rhodopsins are maximally sensitive to blue-green light, chrysopsins to deep-blue light, and porphyropsins to yellow-red light. Most elasmobranchs possess rhodopsin, which provides maximum sensitivity for clearer, shallow ocean waters associated with epipelagic environments (Cohen, 1991).

Chrysopsin has been found in deep-sea squaliform sharks, which inhabit regions where the little available light is deep blue (Denton and Shaw, 1963). Porphyropsin, which is common in freshwater teleosts, is more suited for turbid, yellowish photic conditions.

This pigment is rare in elasmobranchs, even freshwater species, except in lemon sharks,

N. brevirostris, which possess porphyropsin as juveniles but rhodopsin as adults. The visual pigment apparently changes from a porphyropsin adapted for maximum sensitivity in inshore, shallow waters to a rhodopsin better suited for clearer, bluer

18 oceanic waters. This visual adaptation matches a habitat shift from shallow to oceanic waters that occurs between juvenile and adult stages of this shark (Cohen et al., 1990).

A duplex (rod-cone) retina does not necessarily provide for color vision in all cases. Color discrimination normally requires at least two types of cones, each containing different visual pigments with different spectral sensitivities.

Microspectrophotometry has shown that in the three species of rays examined to date, three different cone pigments with different spectral sensitivities are found, which suggests that these animals are capable of color vision (Hart et al., 2004; Theiss et al.,

2007). By contrast, only one cone pigment per species was found in 17 species of sharks examined, which suggests that these animals may have monochromatic vision, similar to some marine mammals (Hart et al., 2011). Possessing only a single cone pigment does not, however, completely eliminate the capacity for color vision. If the rod and cone pigments have different spectral sensitivities and the retina and brain are capable of comparing signals between them, dichromatic color vision is possible. This may be the case in the blacknose shark, Carcharhinus acronotus, and the bonnethead,

Sphyrna tiburo, as electroretinography has revealed two absorbance peaks (blue and green) in their photoreceptors (McComb et al., 2010). It is thus unclear at this time whether or not sharks are capable of color vision as defined on a behavioral basis.

Sharks have retinal areas (areae) of higher cone and/or ganglion cell density, which are regional specializations for greater visual acuity. Retinal whole-mount techniques have been used to map the topographic distributions of retinal cells in 33 elasmobranch species representing 17 families of sharks, skates and rays and one family of chimaera (Peterson and Rowe, 1980; Collin, 1988; Hueter, 1991; Logiudice and

Laird, 1994; Collin, 1999; Bozzano and Collin, 2000; Bozzano, 2004; Theiss et al.,

2007; Lisney and Collin, 2008; Litherland and Collin, 2008; Litherland et al., 2009).

Most species have regions of increased photoreceptor and ganglion cell density, the isodensity contours of which can be arranged horizontally, forming a “visual streak”, or radially, forming an area centralis (Figure 14). The position and extent of the horizontal streak appears to vary with habitat and ecology. Benthic species generally have dorsally located horizontal streaks, providing increased sampling of the ventral visual field

(Peterson and Rowe, 1980; Collin, 1988; Logiudice and Laird, 1994; Collin, 1999;

Bozzano and Collin, 2000; Bozzano, 2004; Theiss et al., 2007; Lisney and Collin, 2008;

Litherland and Collin, 2008). This is thought to reflect the importance of the horizon at the substrate-water interface in animals that feed off the benthos or bury themselves in the sand (Bozzano and Collin, 2000). Centrally located horizontal streaks have been found in benthopelagic species, providing increased sampling of the lateral visual field

(Hueter, 1991; Collin, 1999; Bozzano and Collin, 2000). Ventral horizontal streaks found in tiger sharks, Galeocerdo cuvier (Bozzano and Collin, 2000) and bigeye thresher sharks, A. superciliosus (Lisney and Collin, 2008) provide increased sampling of the dorsal visual field. This might represent an adaptation for detecting prey from below. Tiger sharks prey on birds, sea turtles, and marine mammals, which are commonly found on or near the sea surface (Lowe et al., 1996); common thresher sharks, Alopias vulpinus, a sister species to the bigeye thresher, were recently demonstrated to attack prey from below, using the elongated dorsal lobe of their caudal fin to stun their prey (Aalbers et al., 2010).

In contrast to horizontal streaks, concentric retinal areae centrales (Figure 9) are more applicable for visualizing a limited spot in the visual field or for operating in complex, three-dimensional visual environments, such as reefs. Areae have been found in a number of phylogenetically and ecologically diverse species of sharks, ranging from pelagic, open ocean environments to reef, coastal, and even riverine habitats

(Lisney and Collin, 2008; Litherland and Collin, 2008; Litherland et al., 2009). Cookie- cutter sharks, Isistius brasiliensis, and white sharks, C. carcharias, are both ambush predators in open water, while ornate wobbegongs, Orectolobus ornatus, are benthic ambush predators, and all three have retinal areae not streaks (Bozzano and Collin,

2000; Litherland, 2001; Litherland and Collin, 2008). It appears, therefore, that habitat is not the only factor selecting for the presence or absence of retinal areae in sharks.

Locomotory style could influence the adaptiveness of visual streaks vs. areae, for example, favoring streaks in species that are constantly moving forward (Hueter, 1991).

The possible ecological and behavioral correlates with elasmobranch retinal topography have been discussed by Bozzano and Collin (2000) and Lisney and Collin (2008).

Relatively few studies have examined visually mediated behaviors in sharks.

Lemon sharks, N. brevirostris, bull sharks, Carcharhinus leucas, and nurse sharks,

Ginglymostoma cirratum, have been trained to locate visual targets for food and to discriminate between shapes and patterns (Clark, 1959, 1963; Wright and Jackson, 1964;

Aronson et al., 1967), but the visual parameters of the targets were not quantified.

Lemon sharks can also be trained to discriminate between targets of different brightness, down to a 0.3log difference (Gruber and Cohen, 1978). Smooth dogfish, Mustelus canis, can perform rheotaxis behaviors using vision to navigate to the vicinity of an odor

21 source when the lateral line system has been chemically ablated (Gardiner and Atema,

2007). Vision has been suggested to mediate the final attack on prey, based mostly on anecdotal reports, with the exception of the Pacific angel shark, Squatina californica, which responds to visual stimuli with ambush attacks after the elimination of chemical, mechanical, and electrical cues (Fouts and Nelson, 1999).

Electroreception

The electroreceptive system of elasmobranchs consists of hundreds of long canals (from a few centimeters to up to 20cm in rays) that are up to 1 mm in diameter

(Waltman, 1966; Murray, 1974) and open to the exterior via a pore in the skin. Each canal terminates in a bulb, called an ampulla of Lorenzini. The ampulla consists of subdermal pouch-like alveoli, each containing hundreds of receptor cells, similar to the hair cells of the lateral line system. Each receptor cell is innervated by afferents (but no efferents) from the anterior lateral line nerve (VIII) (Kantner et al., 1962) (Figure 15).

The walls of the canal have a high electrical resistance while the interior is filled with a conductive mucopolysaccharide jelly (Murray and Potts, 1961), with an ionic composition similar to that of seawater, so that a voltage gradient forms along the interior of the canal (Brown et al., 2005). The ampullae are distributed on the head in sharks, on both the dorsal and ventral side (Zakon, 1988). The ampullae are arranged so that the pores have different orientations that are believed to provide the animal with directionality (Kalmijn, 1974), while the ampullae are grouped into bilateral clusters, forming a common internal potential. The receptors thus detect differences between the

22 internal potential and the seawater at the pore by effectively measuring the voltage drop along the length of the canal (Bennett, 1971). Longer canals will therefore have greater sensitivity because they sample across a greater distance and therefore a larger potential difference (Broun et al., 1979; Sisneros and Tricas, 2000).

The ampullae are sensitive to DC and low-frequency AC fields. These fields can be generated by biotic and abiotic sources. Biotic sources include the bioelectric fields surrounding crustaceans (DC potentials of up to 50 μV/cm when intact and up to 1250

μV/cm when damaged) or generated by the respiratory movements of fish (AC potentials of up to 500 μV/cm, Kalmijn, 1972). The threshold for detection of elasmobranchs has been demonstrated to be in the range of 1nV/cm – 0.1μV/cm

(Murray, 1962; Kalmijn, 1982; Johnson et al., 1984; Haine et al., 2001; Kajiura and

Holland, 2002b; Kajiura, 2003). For fields produced by aquatic animals, this translates into a distance of 0.5m or less from the source (Kalmijn, 1972). The electrosensory system has been shown to detect bioelectric fields produced by prey (Kalmijn, 1982;

Tricas, 1982; Blonder and Alevizon, 1988; Kajiura and Holland, 2002b; Kajiura, 2003;

Jordan et al., 2009; Kajiura and Fitzgerald, 2009), predators (Sisneros et al., 1998), and conspecifics during social interactions (Tricas et al., 1995).

Abiotic sources of electric fields in seawater include manmade structures, such as underwater cables and towed arrays, which sharks have been documented to bite

(Marra, 1989), as well as the Earth’s magnetic field. Elasmobranchs are theorized to orient to perform geomagnetic orientation using two modes, passive and active navigation. In passive navigation, the animal measures the voltage gradients produced by the flow of ocean currents through the Earth’s magnetic field. In active navigation,

23 the animal induces a voltage gradient by swimming through the Earth’s magnetic field

(Figure 16) (Kalmijn, 1978, 1981, 1982, 1984) or it acquires directional information from the modulation in induced currents that occurs as it turns its head, much like a compass (Paulin, 1995). There is evidence that elasmobranchs can use geomagnetic information. Round stingrays, Urobatis halleri, can be conditioned to enter a test enclosure from magnetic east, avoiding magnetic west (Kalmijn, 1982) and scalloped hammerheads, S. lewini, have been observed to aggregate near seamounts, following daily routes that correlate with magnetic anomalies in the seafloor (Klimley, 1993).

However, direct evidence of the use of geomagnetic orientation in long-distance navigation and homing, as has been shown in other animals, is lacking for sharks at this time.

Sharks have three main ampullary clusters: buccal, mandibular, and superficial ophthalmic, which is subdivided into anterior and posterior groups. The posterior superficial ophthalmic cluster, which projects primarily into the posterior-lateral quadrant, contains the longest and therefore most sensitive canals in two species of sharks examined, and has been suggested to function in geomagnetic orientation. The anterior superficial ophthalmic cluster, which projects in the anterior and posterior direction, is shorter but contains the greatest number of ampullae, and is thought to be important in the detection of dipole fields in front of the head, such as those from prey, predators or mates. The buccal cluster projects to the lateral edge of the head and has been suggested to function in prey orientation behaviors. The mandibular cluster is located close to the mouth and has been suggested to function to stimulate strikes during

24 feeding and grasping bites by males on females during courtship and copulation (Rivera-

Vicente et al., 2011) (Figure 17).

Multimodal Integration

Early studies (reviewed in Aronson, 1963) concluded there was little multisensory integration in the elasmobranch brain and those conclusions influenced the naming of the brain regions. For example, the tectum of the mesencephalon was called the optic tectum, as it was presumed to be dominated by vision, and the telencephalon was called the olfactory lobes as it was presumed to be dominated by olfactory inputs

(Ariëns Kappers et al., 1936). However, electrophysiology has revealed areas of the telencephalon that show responses to multiple sensory stimuli. The pallium of the telencephalon can be divided into lateral, medial, and dorsal portions. The lateral pallium has been found to be dominated by olfaction (Smeets, 1983; Hoffman and

Northcutt, 2008), while the dorsal (or general) pallium and medial pallium appear to be multisensory. The dorsal pallium is the site of recordings in response to visual (optic nerve) and trigeminal nerve stimuli, which may represent cutaneous mechanoreceptors or electroreceptors, in the nurse shark, G. cirratum (Cohen et al., 1973; Ebbesson, 1980).

The medial (or hippocampal) pallium has been found to respond to visual, electrosensory, and lateral line stimuli in the little skate, Leucoraja erinacea (Bodznick and Northcutt, 1984), visual and cutaneous (somatosensory or electrosensory) stimuli in other batoids (Veselkin and Kovačević, 1973), and visual, olfactory, and electrosensory stimuli in spiny dogfish, S. acanthias (Nikaronov and Lukyanov, 1980; Nikaronov,

1983; Bodznick, 1991). Units responsive to visual, auditory, and electrosensory stimuli have been recorded from the brains of several galeomorphs, possibly from the pars centralis or medial pallium (Bullock and Corwin, 1979). Additionally, retrograde dye labeling in thornback rays, Platyrhinoidis triseriata, has revealed olfactory areas in the dorsomedial pallium (Hoffman and Northcutt, 2008). Interestingly, the medial portion of the telencephalon is larger in batoids and squalomorphs, while the dorsal pallium is better developed in galeomorphs and myliobatoids (Northcutt, 1978).

The tectum of the mesencephalon is heavily visual, though the highest center of visual processing is the telencephalon (see above), and nurse sharks can still perform some visual discrimination tasks after the tectum has been removed (Graeber et al.,

1973). Most of the retinal efferents project to the tectum of the mesencephalon, particularly the superficial tectal laminae, where they form a topographic map (reviewed in Bodznick, 1991; Hueter, 1991). The deeper layers, however, are multimodal. The electrosensory and mechanosensory medullar nuclei project to a nucleus in the roof of the midbrain, called the lateral mesencephalic nucleus (Boord and Northcutt, 1982,

1988). Recordings from the tectum have been made in response to electrosensory, common cutaneous, and auditory stimuli in several species of rays and sharks (Platt et al., 1974) and single multimodal (visual, electrosensory, tactile/lateral line) neurons have been found in the tectum of the little skate, L. erinacea (Bodznick, 1991). Hoffman and

Northcutt’s (2008) retrograde dye labeling study on thornback rays, P. triseriata, suggested that olfactory, electrosensory, and mechanosensory (lateral line) information converge in the lateral mesencephalic nucleus. Though this has yet to be confirmed with

26 electrophysiology, all of these senses are important for locating prey buried in the substrate.

A biological target (prey item, predator, or potential mate) might simultaneously emit several signals: odor; a hydrodynamic disturbance (sound), such as from gill movements or tail beats (reviewed in Bleckmann, 1994); and a weak electrical field

(Kalmijn, 1972). The sequence with which each of the sensory modalities comes into play depends on a multitude of factors, however, and there is likely no single sensory hierarchy that operates under all circumstances for all elasmobranch species. How animals use sensory information depends not only on what sensory stimuli are available, as determined by the animal’s proximity to the prey, the physics of the stimulus fields, strength of the signals, and the thresholds of detection for each species, but also on which stimulus or stimuli the animal chooses to focus upon when information from multiple senses is available simultaneously.

Multisensory integration in the brain can allow an animal to determine the most salient information for a given task, which has been suggested to occur by weighting the various cues based on the amount of information they provide for a given event

(reviewed in Stein and Stanford, 2008), resulting in sensory switching at the interface of different phases of a behavior. For example, sharks that have been tracking odor plumes switch their focus from an olfactory signal to an electrical signal once it is within the range of detection, with a sudden sharp turn towards, and bite on, the source of the electric field (Kalmijn, 1982; Kajiura and Holland, 2002a; Kajiura, 2003; Jordan et al.,

2009; Kajiura and Fitzgerald, 2009). Multimodal integration can also improve the detection of weak signals. For example, the topographic maps of the visual and

27 electrosensory systems overlap in the tectum of the little skate, L. erinacea, therefore electrosensory and visual stimuli emanating from the same point in space converge on the same neurons (Bodznick, 1990). This may aid in localizing buried prey in low ambient light as the convergence of signals results in a summed response the is greater than the additive responses to the two signals alone (reviewed in Stein and Stanford,

2008). Such convergence of signals may be required for a particular task, with the two sensory modalities playing complementary roles for a given behavior. For example, smooth dogfish, M. canis, require simultaneous input from the olfactory system and the lateral line to precisely locate the source of a turbulent odor plume, through a process known as eddy chemotaxis (Gardiner and Atema, 2007). In other situations, one or more senses may provide sufficient information for a task and thus play alternative roles for a given behavior. For example, navigating large scale flow, such as a current, can be accomplished in M. canis by using either cues from the lateral line system

(hydrodynamic flow field) or vision (visual flow field) (Gardiner and Atema, 2007). For a given task, therefore, one sense may be absolutely required, or the senses may have complementary or alternative roles. Few studies, however, have examined more than one or two senses at a time. The goal of this dissertation work is to examine the complementary and alternative roles of olfaction, vision, mechanoreception, and electroreception in a complex behavioral task, feeding. While hearing with the inner ear is likely contributing to prey localization in sharks, the stimulus field in a closed tank environment is very difficult to control due to echoing off the walls, bottom and surface of the tank, and thus will not be specifically examined in this study.

Experimental Animals

The smooth dogfish, Mustelus canis

The smooth dogfish, M. canis, is an abundant, small coastal shark (~ 1m maximum total length), inhabiting western Atlantic waters off the U.S. northeast coast.

This species feeds primarily on or near the bottom on crustaceans such as crabs, lobsters and shrimp, but it also scavenges opportunistically on other prey (Bigelow and

Schroeder, 1953; Casterlin and Reynolds, 1979). It has been described as locating food without visual cues (Sheldon, 1909; Parker, 1922; Bigelow and Schroeder, 1948). With the nares blocked, it shows no interest in prey (Sheldon, 1911); blocking one naris causes turning behavior to the intact side, suggesting orientation based on concentration comparisons between the two nares (i.e. chemotropotaxis, Fraenkel and Gunn, 1940).

Recent experiments suggest it uses olfaction, lateral line and to a small degree vision

(Gardiner and Atema, 2007) and electric sense (Kalmijn, 1971) to locate and strike live food. It uses a combination of ram (overswimming the prey) and suction to engulf fish and benthic invertebrates (Gerry et al., 2007; Wilga et al., 2007). This species was selected as a sensory generalist emphasizing olfaction and ram-suction feeding.

The nurse shark, Ginglymostoma cirratum

The nurse shark, G. cirratum, is an abundant, demersal species found throughout the year in shallow waters of the tropical Western Atlantic. Birth size is about 30 cm and maximum size approaches 3 m. Nurse sharks feed on or near the bottom and in rocky and reef areas on fishes, mollusks and crustaceans (Castro, 2000) and can rest motionless on the bottom for extended periods of time. They are suction feeders (Motta

29 et al., 2002) with a well-developed olfactory apparatus. It was once believed that this species could locate food using olfaction alone, through chemical gradient searching

(Hodgson and Mathewson, 1971; Mathewson and Hodgson, 1972). Kleerekoper et al.

(1975), however, found that nurse sharks need flowing water to provide a directional vector. Capture occurs rapidly, with peak gape occurring in 40-50ms, and from a distance of only a few centimeters (Motta and Wilga, 1999; Motta et al., 2002). Once the capture sequence has been initiated, there is little to no modulation (Motta et al.,

2002; Robinson and Motta, 2002; Matott et al., 2005). This species was selected as a primarily chemosensory hunter and as a representative of suction feeders.

The bonnethead, Sphyrna tiburo

The bonnethead, S. tiburo, is a small, abundant coastal species of hammerhead inhabiting inshore waters of the southeast U.S. coast, including the Gulf of Mexico.

Birth size is approximately 30 cm and they grow to a maximum size of about 1.4 m

(Castro, 1983). Bonnetheads specialize on crustaceans such as small spiny lobster in the

Florida Keys and portunid crabs along the southwest Florida coast (Cortés et al., 1996).

They feed by depressing the mandible and scooping up prey off the bottom as they swim over it but also will take food in the water column, using ram-biting almost exclusively with the capture cycle occurring in approximately 200 ms (Wilga and Motta, 2000).

They were previously believed to crush hard prey, such as blue crabs, Calinectes sapidus, with their molariform-like rear teeth. Recent bite force evidence, however, indicates that they may not be capable of crushing the entire size range of crabs in their diet but they may seize larger crabs in their jaws and use headshaking behaviors to

30 remove the appendages prior to swallowing the carapace whole (Cortés et al., 1996;

Mara et al., 2009). Their olfactory systems have been well studied (Kajiura et al., 2005;

Meredith and Kajiura, 2010) and the widely spaced electrosensory ampullae on the enlarged cephalofoil are involved in prey detection and localization (Kajiura and

Holland, 2002b; Kajiura, 2003). This species was selected for its electrosensory capabilities due to its laterally expanded head and as a representative of epibenthic ram feeders.

The blacktip shark, Carcharhinus limbatus

The blacktip shark, C. limbatus, is one of the more common large coastal species inhabiting the Gulf of Mexico coast. Juveniles and adults are abundant along the Florida

Gulf coast in spring and summer. The young are born at about 55-60 cm and maximum size is about 2 m (Castro, 1983). Blacktip sharks are fast-swimming predators, feeding primarily on small bony fishes and other elasmobranchs at all levels of the water column and capturing prey using ram-feeding almost exclusively (Frazzetta and Prange, 1987;

Castro, 1996; Heupel and Hueter, 2002). The prey capture kinematics of this species have not previously been accurately measured, but are expected to be similar to those of other species, with capture times on the order of 100-400 ms (Motta et al., 2002; Motta,

2004). This species appears to modulate the kinematics of capture with jaw protrusion occurring during some, but not all, capture events (Frazzetta and Prange, 1987). This species was selected as a primarily visual hunter and as a more pelagic ram-biting feeder on elusive prey.

Literature Cited

Aalbers, S. A., Bernal, D. and Sepulveda, C. A. (2010). The functional role of the caudal fin in the feeding ecology of the common thresher shark Alopias vulpinus. Journal of Fish Biology 76, 1863-1868.

Abel, R. L., Maclaine, J. S., Cotton, R., Xuan, V. B., Nickels, T. B., Clark, T. H., Wang, Z. and Cox, J. P. L. (2010). Functional morphology of the nasal region of a hammerhead shark. Comparative Biochemistry and Physiology, A: Comparative Physiology 155, 464-475.

Ariëns Kappers, C. U., Huber, G. C. and Crosby, E. C. (1936). The Comparative Anatomy of the Nervous System of Vertebrates, Including Man. New York: Hafner Publishing Co.

Aronson, L. R. (1963). The central nervous system of sharks and bony fishes with special reference to sensory and integrative mechanisms. In Sharks and Survival, (ed. P. W. Gilbert), pp. 165-241. Boston: D.C. Heath.

Aronson, L. R., Aronson, F. R. and Clark, E. (1967). Instrumental conditioning and light-dark discrimination in young nurse sharks. Bulletin of Marine Science 17, 249-256.

Atema, J. (1996). Eddy chemotaxis and odor landscapes: exploration of nature with animal sensors. Biological Bulletin 191, 129-138.

Baatrup, E. and Doving, K. B. (1985). Physiological studies on solitary receptors of the oral disk papillae in adult brook lamprey, Lampetra planeri (Bloch). Chemical Senses 10, 559-566.

Baker, C. F. and Montgomery, J. C. (1999). The sensory basis of rheotaxis in the blind Mexican cave fish, Astyanax fasciatus. Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 184, 519-527.

Baker, C. F., Montgomery, J. C. and Dennis, T. E. (2002). The sensory basis of olfactory search behavior in banded kokopu (Galaxias fasciatus). Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 188, 553-560.

Bell, M. A. (1993). Convergent evolution of nasal structure in sedentary elasmobranchs. Copeia 1993, 144-158.

Bennett, M. V. L. (1971). Electroreception. In Fish Physiology, vol. 5 eds. W. S. Hoar and D. J. Randall), pp. 493-574. New York: Academic Press.

Bigelow, H. B. and Schroeder, W. C. (1948). Fishes of the Western North Atlantic: Lancelets, cyclostomes, sharks. New Haven, Connecticut. 32

Bigelow, H. B. and Schroeder, W. C. (1953). Fishes of the Gulf of Maine. Fishery Bulletin of the Fish and Wildlife Service 74, 561.

Bleckmann, H. (1994). Reception of hydrodynamic stimuli in aquatic and semi-aquatic environments. New York: Gustav Fischer Verlag.

Bleckmann, H., Weiss, O. and Bullock, T. H. (1989). Physiology of lateral line mechanoreceptive regions in the elasmobranch brain. Journal of Comparative Physiology, A 164, 459-474.

Blonder, B. I. and Alevizon, W. S. (1988). Prey discrimination and electroreception in the stingray Dasyatis sabina. Copeia 1988, 33-36.

Bodznick, D. (1990). Elasmobranch vision: multimodal integration in the brain. Journal of Experimental Zoology Suppl. 5.

Bodznick, D. (1991). Elasmobranch vision: multimodal integration in the brain. Journal of Experimental Zoology Suppl. 5, 108-116.

Bodznick, D. and Northcutt, R. G. (1984). An electrosensory area in the telencephalon of the little skate, Raja erinacea. Brain Research 298, 117-124.

Boord, R. L. and Campbell, C. B. G. (1977). Structural and functional organization of the lateral line system of sharks. American Zoologist 17, 431-441.

Boord, R. L. and Northcutt, R. G. (1982). Ascending lateral line pathways to the midbrain of the clearnose skate, Raja eglanteria. Journal of Comparative Neurology 207, 274-282.

Boord, R. L. and Northcutt, R. G. (1988). Medullary and mesencephalic pathways and connections of lateral line neurons in the spiny dogfish Squalus acanthias. Brain, Behavior and Evolution 32, 76-88.

Bozzano, A. (2004). Retinal specialisations in the dogfish Centroscymnus coelolepis from the Mediterranean deep-sea. Scientia Marina (Barcelona) 68 (Suppl. 3), 185-195.

Bozzano, A. and Collin, S. P. (2000). Retinal ganglion cell topography in elasmobranchs. Brain, Behavior and Evolution 55, 191-208.

Broun, G. R., Il'inskii, O. B. and Krylov, B. V. (1979). Responses of the ampullae of Lorenzini in a uniform electric field. Neurophysiology 11, 118-124.

Brown, B. R., Hughes, M. E. and Russo, C. (2005). Infrastructure in the electric sense: admittance data from shark hydrogels. Journal of Comparative Physiology A 191, 115-123.

Budker, P. (1958). Les organes sensoriels cutanés des sélaciens. In Traité de Zoologie. Anatomie, Systémique, Biologie, vol. Tome XIII. Agnathes et Poissons (ed. P. P. Grassé), pp. 1033-1062. Paris: Masson et Cie.

Bullock, T. H. and Corwin, J. T. (1979). Acoustic evoked activity in the brain in sharks. Journal of Comparative Physiology 129, 223-234.

Carrier, J. C., Pratt, H. L. and Martin, L. K. (1994). Group reproductive behaviors in free living nurse sharks, Ginlymostoma cirratum. Copeia 1984, 646-656.

Casper, B. M. and Mann, D. A. (2007). Dipole hearing measurements in elasmobranch fishes. Journal of Experimental Biology 210, 75-81.

Casterlin, M. E. and Reynolds, W. W. (1979). Diel activity patterns of the smooth dogfish shark, Mustelus canis. . Bulletin of Marine Science 29, 440-442.

Castro, J. I. (1983). The Sharks of North American Waters. College Station, TX: Texam A&M University Press.

Castro, J. I. (1996). The biology of the blacktip shark, Carcharhinus limbatus, off the southeastern United States. Bulletin of Marine Science 59, 508-522.

Castro, J. I. (2000). The biology of the nurse shark, Ginglymostoma cirratum, off the Florida east coast and the Bahama Islands. Environmental Biology of Fishes 58, 1-22.

Chagnaud, B. P., Brücker, C., Hofmann, M. H. and Bleckmann, H. (2008). Measuring flow velocity and flow direction by spatial and temporal analysis of flow fluctuations. The Journal of Neuroscience 28, 4479-4487.

Chu, Y. T. and Wen, M. C. (1979a). Monograph of fishes of China (No. 2): a study of the lateral-line canals system and that of Lorenzini ampullae and tubules of elasmobranchiate fishes of China. Shanghai: Science and Technology Press.

Chu, Y. T. and Wen, W. C. (1979b). A study of the lateral-line canal system and that of Lorenzini ampullae and tubules of elasmobranchiate fishes of China. In Monograph of Fishes of China, vol. No. 2, pp. 117-126. Shanghai: Science and Technology Press.

Clark, E. (1959). Instrumental conditioning of sharks. Science 130, 217-218.

Clark, E. (1963). Maintenance of sharks in captivity with a report on their instrumental conditioning. In Sharks and Survival, (ed. P. W. Gilbert), pp. 115-149. Boston: D.C. Heath.

Clark, E. (1974). The red sea’s sharkproof fish. National Geographic 146, 719-727.

Cohen, D. H., Duff, T. A. and Ebbesson, S. O. E. (1973). Electrophysiological identification of a visual area in the shark telencephalon. Science 182, 492-494.

Cohen, J. L. (1991). Adaptations for scotopic vision in the lemon shark (Negaprion brevirostris). Journal of Experimental Zoology Suppl. 5, 76-84.

Cohen, J. L., Hueter, R. E. and Organisciak, D. T. (1990). The presence of a porphyropsin-based visual pigment in the juvenile lemon shark (Negaprion brevirostris). Vision Research 30, 1949-1953.

Collin, S. P. (1988). The retina of the shovel-nosed ray, Rhinobatos batillum (Rhinobatidae): morphology and quantitative analysis of the ganglion, amacrine and bipolar cell populations. Experimental Biology 47, 195-207.

Collin, S. P. (1999). Behavioural ecology and retinal cell topography. In Adaptive Mechanisms in the Ecology of Vision, eds. S. N. Archer M. B. A. Djamgoz E. R. Loew J. C. Partridge and S. Vallerga), pp. 509-535. Dordrecht: Kluwer Academic.

Cook, M. H. and Neal, H. V. (1921). Are taste buds of elasmobranchs endodermal in origin? Journal of Comparative Neurology 33, 45-63.

Coombs, S. and Braun, C. B. (2003). Information processing by the lateral line In Sensory Processing of the Aquatic Environment, eds. S. P. Collin and N. J. Marshall), pp. 122-138. New York, N.Y.: Springer-Verlag.

Coombs, S., Braun, C. B. and Donovan, B. (2001). The orienting response of Lake Michigan mottled sculpin is mediated by canal neuromasts. J. Exp. Biol. 204, 337-348.

Coombs, S. and Janssen, J. (1990). Behavioral and neurophysiological assessment of lateral line sensitivity in the mottled sculpin, Cottus bairdi. Journal of Comparative Physiology, A 167, 557-567.

Cortés, E., Manire, C. A. and Hueter, R. E. (1996). Diet, feeding habits, and diel feeding chronology of the bonnethead shark, Sphyrna tiburo, in southwest Florida. Bulletin of Marine Science 58, 353-367.

Corwin, J. T. (1978). The relation of inner ear structure to the feeding behavior in sharks and rays. In Scanning Electron Microscopy, vol. 2 (ed. O. M. Johari), pp. 1105- 1112. Chicago: SEM.

Corwin, J. T. (1981). Peripheral auditory physiology in the lemon shark: evidence of parallel otolithic and non-otolithic sound detection. Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 142, 379-390.

Corwin, J. T. (1989). Functional anatomy of the auditory system in sharks and rays. Journal of Experimental Zoology Supplement 2, 62-74.

Daniel, J. F. (1928). The Elasmobranch Fishes. Berkeley: University of California Press.

Demski, L. S. and Northcutt, R. G. (1996). The brain and cranial nerves of the white shark: an evolutionary perspective. In Great White Sharks: The Biology of Carcharodon carcharias, eds. A. P. Klimley and D. G. Ainley), pp. 121-130. San Diego: Academic Press.

Denton, E. J. and Gray, J. A. B. (1988). Mechanical factors in the excitation of the lateral lines of fishes. In Sensory Biology of Aquatic Animals, eds. J. Atema R. R. Fay A. N. Popper and W. N. Tavolga), pp. 595-617. New York, N.Y.: Springer-Verlag.

Denton, E. J. and Gray, J. A. B. (1989). Some observations on the forces acting on neuromasts in fish lateral line canals. In The Mechanosensory Lateral Line, eds. S. Coombs P. Gorner and H. Munz), pp. 229-246. New York, N.Y.: Springer- Verlag.

Denton, E. J. and Nicol, J. A. C. (1964). The chorioidal tapeta of some cartilaginous fishes (Chondrichthyes). Journal of the Marine Biological Association of the United Kingdom 44, 219-258.

Denton, E. J. and Shaw, T. I. (1963). The visual pigments of some deep-sea elasmobranchs. Journal of the Marine Biological Association of the United Kingdom 43, 65-70.

Ebbesson, S. O. E. (1980). On the organization of the telencephalon in elasmobranchs. In Comparative Neurology of the Telencephalon, (ed. S. O. E. Ebbesson), pp. 1- 16. New York: Plenum Press.

Evangelista, C., Mills, M., Siebeck, U. E. and Collin, S. P. (2010). A comparison of the external morphology of the membranous inner ear in elasmobranchs. Journal of Morphology 271, 483-495.

Ewart, J. C. and Mitchell, H. C. (1892). On the lateral sense organs of elasmobranchs. II. The sensory canals of the common skate (Raja batis). Transactions of the Royal Society of Edinburgh 37, 87-105.

Fouts, W. R. and Nelson, D. R. (1999). Prey capture by the Pacific angle shark, Squatina californica: visually mediated strikes and ambush-site characteristics. Copeia 1999, 304-312.

Fraenkel, G. S. and Gunn, D. L. (1940). The Orientation of Animals: Kineses, Taxes, and Compass Reactions. Oxford: Clarendon Press.

Frazzetta, T. H. and Prange, C. D. (1987). Movements of cephalic components during feeding in some requiem sharks (Carcharhiniformes: Carcharhinidae). Copeia 4, 979-993.

Gardiner, J. M. and Atema, J. (2007). Sharks need the lateral line to locate odor sources: rheotaxis and eddy chemotaxis. Journal of Experimental Biology 210, 1925- 1934.

Gerry, S. P., Dean, M. N., Summers, A. P. and Wilga, C. A. D. (2007). Asynchronous bilateral jaw muscle activity in two species of elasmobranchs In Society for Integrative and Comparative Biology. Phoenix, AZ.

Gilbert, C. R. (1967). A revision of the hammerhead sharks (Family Sphyrnidae). Proceedings of the United States National Museum 119, 1-88.

Gilbert, P. W. (1963). The visual apparatus of sharks. In Sharks and Survival, (ed. P. W. Gilbert), pp. 283-326. Boston, MA: DC Heath & Co.

Gordon, I. (1993). Pre-copulatory behaviour of captive sandtiger sharks, Carcharias taurus. Environmental Biology of Fishes 38, 159-164.

Graeber, R. C., Ebbesson, S. O. E. and Jane, J. A. (1973). Visual discrimination in sharks without optic tectum. Science 180, 413-415.

Gruber, S. H. and Cohen, J. L. (1978). Visual system of the elasmobranchs: State of the art 1960-1975. In Sensory Biology of Sharks, Skates, and Rays, eds. E. S. Hodgson and R. F. Mathewson), pp. 11-105. Arlington, V.A.: U.S. Office of Naval Research.

Haine, O. S., Ridd, P. V. and Rowe, R. J. (2001). Range of electrosensory detection of prey by Carcharhinus melanopterus and Himantura granulata. Marine and Freshwater Research 52, 291-296.

Hama, K. and Yamada, Y. (1977). Fine structure of the ordinary lateral line organ. 2. The lateral line canal organ of spotted shark, Mustelus manazo. Cell. Tiss. Res 176, 23-36.

Hanke, W. and Bleckmann, H. (2004). The hydrodynamic trails of Lepomis gibbosus (Centrarchidae), Colomesus psittacus (Tetraodontidae) and Thysochromis ansorgii (Cichlidae) investigated with scanning particle image velocimetry. Journal of Experimental Biology 207, 1585-1596.

Hanke, W., Brücker, C. and Bleckmann, H. (2000). The ageing of the low frequency water disturbances caused by swimming goldfish and its possible relevance to prey detection. Journal of Experimental Biology 203, 1193-2000.

Hara, T. J. (1994). The diversity of chemical stimulation in fish olfaction and gustation. Reviews in Fish Biology and Fisheries 4, 1-35.

Hart, N. S., Lisney, T. J., Marshall, N. J. and Collin, S. P. (2004). Multiple cone visual pigments and the potential for trichromatic colour vision in two species of elasmobranch. Journal of Experimental Biology 207, 4587-4594.

Hart, N. S., Theiss, S. M., Harahush, B. K. and Collin, S. P. (2011). Microspectrophotometric evidence for cone monochromacy in sharks. Naturwissenschaften 98, 193-201.

Heath, A. R. (1991). The ocular tapetum lucidum: a model system for interdisciplinary studies in elasmobranch biology. Journal of Experimental Zoology Suppl. 5, 41- 45.

Herrick, C. J. (1924). Neurological Foundations of Animal Behavior. New York: Henry Holt and Company; reprint edition 1965 by Hafner.

Heupel, M. R. and Hueter, R. E. (2002). Importance of prey density in relation to the movement patterns of juvenile blacktip sharks (Carcharhinus limbatus) within a coastal nursery area. Marine and Freshwater Research 53, 543-550.

Hobson, E. S. (1963). Feeding behavior in three species of sharks. Pacific Science 17, 171-194.

Hodgson, E. S. and Mathewson, R. F. (1971). Chemosensory orientation in sharks. Annals of the New York Academy of Sciences 188, 175-182.

Hodgson, E. S. and Mathewson, R. F. (1978a). Electrophysiological studies of chemoreception in elasmobranchs. In Sensory Biology of Sharks, Skates, and Rays, eds. E. S. Hodgson and R. F. Mathewson), pp. 227-267. Arlington: Office of Naval Research.

Hodgson, E. S. and Mathewson, R. F. (1978b). Sensory Biology of Sharks, Skates and Rays. Arlington: U.S. Office of Naval Research.

Hoffman, M. H. and Northcutt, R. G. (2008). Organization of major telencephalic pathways in an elasmobranch, the thornback ray Platyrhinoidis triseriata. Brain, Behavior and Evolution 72, 307-325.

Hueter, R. E. (1980) Physiological Optics of the Eye of the Juvenile Lemon Shark (Negaprion brevirostris). University of Miami, Coral Gables, FL, pp.

Hueter, R. E. (1991). Adaptations for spatial vision in sharks. Journal of Experimental Zoology Suppl. 5, 130-141.

Hueter, R. E. and Gilbert, P. W. (1991). The sensory world of sharks. In Discovering Sharks, (ed. S. H. Gruber), pp. 48-55. Highlands: American Littoral Society.

Hueter, R. E. and Gruber, S. H. (1982). Recent advances in studies of the visual system of the juvenile lemon shark (Negaprion brevirostris). Florida Scientist 45, 11-25.

Hueter, R. E., Mann, D. A., Maruska, K. P., Sisneros, J. A. and Demski, L. S. (2004). Sensory biology of elasmobranchs. In Biology of Sharks and Their Relatives, eds. J. C. Carrier J. A. Musick and M. R. Heithaus), pp. 325-368. Boca Raton: CRC Press.

Hueter, R. E., Murphy, C. J., Howland, M., Sivak, J. G., Paul-Murphy, J. R. and Howland, H. C. (2001). Refractive state and accommodation in the eyes of free- swimming versus restrained juvenile lemon sharks (Negaprion brevirostris). Vision Research 41, 1885-1889.

Johnsen, P. B. and Teeter, J. H. (1985). Behavioral responses of bonnethead sharks (Sphyrna tiburo) to controlled olfactory stimulation. Marine Behaviour and Physiology 11, 283-291.

Johnson, C. S., Scronce, B. L. and McManus, M. W. (1984). Detection of DC electric dipoles in background fields by the nurse shark. Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 155, 681-687.

Johnson, R. H. and Nelson, D. R. (1978). Copulation and possible olfaction-mediated pair formation in two species of carcharhinid sharks. Copeia 1978, 539-542.

Johnson, S. E. (1917). Structure and development of the sense organs of the lateral canal system of selachians (Mustelus canis and Squalus acanthias). Journal of Comparative Neurobiology 28, 1-74.

Jordan, L. K., Kajiura, S. M. and Gordon, M. S. (2009). Functional consequences of structural differences in stingray sensory systems. Part II: electrosensory system. Journal of Experimental Biology 212, 3044-3050.

Kajiura, S. M. (2003). Electroreception in neonatal bonnethead sharks, Sphyrna tiburo. Marine Biology 143, 603-611.

Kajiura, S. M. and Fitzgerald, T. P. (2009). Response of juvenile scalloped hammerhead sharks to electric stimuli. Zoology 112, 241-250.

Kajiura, S. M., Forni, J. B. and Summers, A. P. (2005). Olfactory morphology of carcharhinid and sphyrnid sharks: does the cephalofoil confer a sensory advantage? Journal of Morphology 264, 253-263.

Kajiura, S. M. and Holland, K. N. (2002a). Electroreception in juvenile scalloped hammerhead and sandbar sharks. Journal of Experimental Biology 205, 3609- 3621.

Kajiura, S. M. and Holland, K. N. (2002b). Electroreception in juvenile scalloped hammerhead and sandbar sharks. Journal of Experimental Biology 205, 3609- 3621.

Kalmijn, A. J. (1971). The electric sense of sharks and rays. Journal of Experimental Biology 55, 371-383.

Kalmijn, A. J. (1972). Bioelectric fields in seawater and the function of the ampullae of Lorenzini in elasmobranch fishes. Scripps Institute of Oceanography Reference Series 72-83, 1-21.

Kalmijn, A. J. (1974). The detection of electric fields from inanimate and animate sources other than electric organs. In Handbook of Sensory Physiology, vol. III (ed. A. Fessard), pp. 147-200. New York, N.Y.: Springer-Verlag.

Kalmijn, A. J. (1978). Electric and magnetic sensory world of sharks, skates, and rays. In Sensory Biology of Sharks, Skates, and Rays, eds. E. S. Hodgson and R. F. Mathewson), pp. 507-528. Arlington, Virginia: Office of Naval Research.

Kalmijn, A. J. (1981). Biophysics of geomagnetic field detection. IEEE Transactions on Magnetics 17, 1113-1124.

Kalmijn, A. J. (1982). Electric and magnetic field detection in elasmobranch fishes. Science 218, 916-918.

Kalmijn, A. J. (1984). Theory of electromagnetic orientation: a further analysis. In Comparative Physiology of Sensory Systems, eds. L. Bolis R. D. Keynes and S. H. P. Madrell). Cambridge, UK: Cambridge University Press. 40

Kalmijn, A. J. (1988). Hydrodynamic and acoustic field detection. In Sensory Biology of Aquatic Animals, eds. J. Atema R. R. Fay A. N. Popper and W. N. Tavolga), pp. 83-130. New York, NY: Springer-Verlag.

Kantner, M., Konig, W. F. and Reinbach, W. (1962). Bau und Innervation der Lorenzinischen Ampullen und deren Bedeutung als niederes Sinnesorgan. Z. Zellforsch 57, 124-135.

Kleerekoper, H., Gruber, D. and Matis, J. (1975). Accuracy of localization of a chemical stimulus in flowing and stagnant water by the nurse shark Ginglymostoma cirratum. Journal of Comparative Physiology 42, 79-84.

Klimley, A. P. (1980). Observations of courtship and copulation in the nurse shark, Ginglymostoma cirratum. Copeia 1980, 878-882.

Klimley, A. P. (1993). Highly directional swimming by scalloped hammerhead sharks, Sphyrna lewini, and subsurface irradiance, temperature, bathymetry, and geomagnetic field. Marine Biology 117, 1-22.

Kotrschal, K. (1995). Ecomorphology of solitary chemosensory cell systems in fish: a review. Environmental Biology of Fishes 44, 143-155.

Kotrschal, K., Peters, R. C. and Doving, K. B. (1996). Chemosensory and tactile nerve responses from the anterior dorsal fin of a rockling, Gaidropsarus vulgaris (Gadidae, Teleostei). Primary Sensory Neuron 1, 297-308.

Kuznetsova, M., Lee, C., Aller, J. and Frew, N. (2004). Enrichment of amino acids in the sea surface microlayer at coastal and open ocean sites in the North Atlantic Ocean. Limnology and Oceanography 49, 1605-1619.

Lisney, T. J., Bennett, M. B. and Collin, S. P. (2007). Volumetric analysis of sensory brain areas indicates ontogenetic shifts in the relative importance of sensory systems in elasmobranchs. Raffles Bulletin of Zoology Suppl. 14, 7-15.

Lisney, T. J. and Collin, S. P. (2006). Brain morphology in large pelagic fishes: a comparison between sharks and teleosts. Journal of Fish Biology 68, 532-554.

Lisney, T. J. and Collin, S. P. (2007). Relative eye size in elasmobranchs. Brain, Behavior and Evolution 69, 266-279.

Lisney, T. J. and Collin, S. P. (2008). Retinal ganglion cell distribution and resolving power in elasmobranchs. Brain, Behavior and Evolution 72, 59-77.

Litherland, L. (2001) Retinal topography in elasmobranchs: interspecific and ontogenetic variations. University of Queensland, Brisbane, Australia, pp. 41

Litherland, L. and Collin, S. P. (2008). Comparative visual function in elasmobranchs: spatial arrangement and ecological correlates of photoreceptor and ganglion cell distributions. Visual Neuroscience 25, 549-561.

Litherland, L., Collin, S. P. and Fritsches, K. A. (2009). Eye growth in sharks: ecological implications for changes in retinal topography and visual resolution. Visual Neuroscience 26, 397-409.

Logiudice, F. T. and Laird, R. J. (1994). Morphology and density distribution of cone photoreceptor in the retina of the Atlantic stingray, Dasyatis sabina. Journal of Morphology 221, 277-289.

Lowe, C. G., Wetherbee, B. M., Crow, G. L. and Tester, A. L. (1996). Ontogenetic dietary shifts and feeding behavior of the tiger shark, Galeocerdo cuvier, in Hawaiian waters. Environmental Biology of Fishes 47, 203-211.

Lowenstein, O. and Roberts, T. D. M. (1951). The localization and analysis of the responses to vibration from the isolated elasmobranch labyrinth. A contribution to the problem of the evolution of hearing in vertebrates. Journal of Physiology 114, 471-489.

Luer, C. A. and Gilbert, P. W. (1985). Mating behavior, egg deposition, incubation period, and hatching in the clearnose skate, Raja eglanteria. Environmental Biology of Fishes 13, 161-171.

Lychakov, D. V., Boyadzhieva-Mikhailova, A., Christov, I. and Evdokimov, I. I. (2000). Otolithic apparatus in Black Sea elasmobranchs. Fisheries Research (Amsterdam) 46, 27-38.

Mara, K., Motta, P. J. and Huber, D. R. (2009). Bite force and performance in the durophagous bonnethead shark, Sphyrna tiburo. Journal of Experimental Zoology 311A, 95-105.

Marra, L. J. (1989). Sharkbite on the SL Submarine Lightwave Cable System: History, causes, and resolution. IEEE Journal of Ocean Engineering 14, 230-237.

Maruska, K. P. (2001). Morphology of the mechanosensory lateral line system in elasmobranch fishes: ecological and behavioral considerations. Environmental Biology of Fishes 60, 47-75.

Maruska, K. P. and Tricas, T. C. (2004). Test of the mechanotactile hypothesis: neuromast morphology and response dynamics of mechanosensory lateral line primary afferents in the stingray. Journal of Experimental Biology 207, 3463- 3476.

Mathewson, R. F. and Hodgson, E. S. (1972). Klinotaxis and rheotaxis in orientation of sharks toward chemical stimuli. Comparative Biochemistry and Physiology, A: Comparative Physiology 42, 79-84.

Matott, M. P., Motta, P. J. and Hueter, R. E. (2005). Modulation in feeding kinematics and motor pattern of the nurse shark Ginglymostoma cirratum. Environmental Biology of Fishes 74, 163-174.

McComb, D. M., Frank, T. M., Hueter, R. E. and Kajiura, S. M. (2010). Temporal resolution and spectral sensitivity of the visual system of three coastal shark species from different light environments. Physiological and Biochemical Zoology 83, 299-307.

McComb, D. M. and Kajiura, S. M. (2008). Visual fields of four batoid species: a comparative study. Journal of Experimental Biology 211, 482-490.

McComb, D. M., Tricas, T. C. and Kajiura, S. M. (2009). Enhanced visual fields in hammerhead sharks. Journal of Experimental Biology 212, 4010-4018.

Meredith, T. M. and Kajiura, S. M. (2010). Olfactory morphology and physiology of elasmobranchs. Journal of Experimental Biology 213, 3449-3456.

Mills, M., Rasch, R., Siebeck, U. E. and Collin, S. P. (2011). Exogenous material in the inner ear of the adult Port Jackson shark, Heterodontus portusjacksoni (Elasmbranchii). The Anatomical Record 294, 373-378.

Montgomery, J. C., Baker, C. F. and Carton, A. G. (1997). The lateral line can mediate rheotaxis in fish. Nature 389, 960-963.

Montgomery, J. C., Carton, G., Voigt, R., Baker, C. F. and Diebel, C. (2000). Sensory processing of water currents by fishes. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 355, 1325-1327.

Motta, P. J. (2004). Prey capture behavior and feeding mechanics of elasmobranchs. In Biology of Sharks and Their Relatives, eds. J. C. Carrier J. A. Musick and M. R. Heithaus), pp. 596. Boca Raton, FL: CRC Press.

Motta, P. J., Hueter, R. E., Tricas, T. C. and Summers, A. P. (2002). Kinematic analysis of suction feeding in the nurse shark Ginglymostoma cirratum (Orectolobiformes, Ginglymostomidae). Copeia 2002, 24-38.

Motta, P. J. and Wilga, C. D. (1999). Anatomy of the feeding apparatus of the nurse shark, Ginglymostoma cirratum. Journal of Morphology 241, 1-29.

Münz, H. (1985). unit activity in the peripheral lateral line system of the cichlid fish Sarotherodon niloticus L. Journal of Comparative Physiology, A 157, 555-568. 43

Murray, R. W. (1962). Response of ampullae of Lorenzini of elasmobranchs to electrical stimulation. Journal of Experimental Biology 39, 119-128.

Murray, R. W. (1974). The ampullae of Lorenzini. In Handbook of Sensory Physiology, vol. III (ed. A. Fessard). New York, N.Y.: Springer-Verlag.

Murray, R. W. and Potts, T. W. (1961). The composition of the endolymph, and other body fluids of elasmobranchs. Comparative Biochemistry and Physiology 2, 65- 75.

Myrberg, A. A., Jr. (2001). The acoustical biology of elasmobranchs. Environmental Biology of Fishes 60, 31-45.

Myrberg, J., A.A. and Gruber, S. G. (1974). The behavior of the bonnethead shark, Sphyrna tiburo. Copeia 1974, 358-374.

Myrberg, J., A.A., Banner, A. and Richard, J. D. (1969). Shark attraction using a video- acoustic system Marine Biology 2, 264-276.

Myrberg, J., A.A., Ha, S. J., Walewski, S. and Banbury, J. C. (1972). Effectiveness of acoustic signals in attracting epipelagic sharks to an underwater sound source. Bulletin of Marine Science 22, 926-949.

Nelson, D. R. and Gruber, S. H. (1963). Sharks: attraction by low-frequency sounds. Science 142, 975-977.

Nelson, D. R., Johnson, R. H. and Waldrop, L. G. (1969). Responses of Bahamanian sharks and groupers to low-frequency, pulsed sounds. Bulletin of the Southern California Academy of Sciences 68, 131-137.

Nicol, J. A. C. (1964). Reflectivity of the chorioidal tapeta of selachians. Journal of the Fisheries Research Board of Canada 21, 1089-1100.

Nikaronov, S. I. (1983). Electrophysiological analysis of convergent relations between the olfactory and visual analyzers in the forebrain of Squalus acanthias. Journal of Evolutionary Biochemistry and Physiology 18, 268-273.

Nikaronov, S. I. and Lukyanov, A. S. (1980). Electrophysiological investigations of visual afferentation pathways in the forebrain of Squalus acanthias Journal of Evolutionary Biochemistry and Physiology 16, 132-139.

Nikonov, A. A., Illyin, Y. N., Zherelove, O. M. and Fesenko, E. E. (1990). Odour thresholds of the black sea skate (Raja clavata). Electrophysiological study. Comparative Biochemistry and Physiology Part A: Physiology 95, 325-238.

Norris, H. W. and Hughes, S. P. (1920). The cranial, occipital, and anterior spinal nerves of the dogfish, Squalus acanthias. Journal of Comparative Neurology 21, 293- 402.

Northcutt, R. G. (1978). Brain organization in the cartilaginous fishes. In Sensory Biology of Sharks, Skates, and Rays, eds. E. S. Hodgson and R. F. Mathewson), pp. 117-193. Arlington, VA: Office of Naval Research Department of the Navy.

Northcutt, R. G. (1989). The phylogenetic distribution and innervation of craniate mechanoreceptive lateral lines. In The Mechanosensory Lateral Line: Neurobiology and Evolution, eds. S. Coombs P. Görner and H. Münz), pp. 17- 78. New York: Springer-Verlag.

Northcutt, R. G. (2004). Taste buds: development and evolution. Brain, Behavior and Evolution 64, 198-206.

Parker, G. H. (1909). The influence of eyes and ears and other allied sense organs on the movement of Mustelus canis. Bulletin of the United States Bureau of Fisheries 29, 43-58.

Parker, G. H. (1914). The directive influence of the sense of smell in the dogfish. Bulletin of the United States Bureau of Fisheries 33, 61-68.

Parker, G. H. (1922). Smell, Taste, and Allied Senses in Vertebrates. Philadelphia, P.A.: J.P. Lippincott Co.

Parker, G. H. and Sheldon, R. E. (1913). The sense of smell in fishes. Bulletin of the United States Bureau of Fisheries 32, 33-46.

Paulin, M. G. (1995). Electroreception and the compass sense of sharks. Journal of Theoretical Biology 174, 325-339.

Peach, M. B. (2001). The dorso-lateral pit organs of the Port Jackson shark contribute sensory information for rheotaxis. Journal of Fish Biology 59, 696-704.

Peach, M. B. (2003). Inter- and intraspecific variation in the distribution and number of pit organs (free neuromasts) of sharks and rays. Journal of Morphology 256, 89- 102.

Peach, M. B. (2005). New microvillous cells with possible sensory function on the skin of sharks. Marine and Freshwater Behaviour and Physiology 38, 275-279.

Peach, M. B. and Marshall, N. J. (2000). The pit organs of elasmobranchs: a review. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 355, 1131-1134. 45

Peach, M. B. and Rouse, G. W. (2000). The morphology of the pit organs and lateral line canal neuromasts of Mustelus antarcticus (Chondrichthyes: Triakidae). Journal of the Marine Biological Association of the United Kingdom 80, 155- 162.

Peach, M. B. and Rouse, G. W. (2004). Phylogenetic trends in the abundance and distribution of pit organs of elasmobranchs. Acta Zoologica 85, 233-244.

Peters, R. C., van Steenderen, G. W. and Kotrschal, K. (1987). A chemoreceptive function for the anterior dorsal fin in rocklings (Gaidropsarus and Ciliata: Teleostei: Gadidae): electrophysiological evidence. Journal of the Marine Biological Association of the United Kingdom 67, 819-823.

Peterson, E. H. and Rowe, M. H. (1980). Different regional specializations of neurons in the ganglion cell layer and inner plexiform layer of the California horned shark, Heterodontus francisci. Brain Research 201, 195-201.

Platt, C. J., Bullock, T. H., Czéh, G., Kovačević, N., Konjević, D. J. and Gojković, M. (1974). Comparison of electroreceptor, mechanoreceptor, and optic evoked potentials in the brain of some rays and sharks. Journal of Comparative Physiology 95, 323-355.

Pocklington, R. (1971). Physical sciences: free amino-acids dissolved in North Atlantic Ocean waters. Nature 230, 374-375.

Rasmussen, L. E. L. and Schmidt, M. J. (1992). Are sharks chemically aware of crocodiles? In Chemical Signals in Vertebrates, vol. VI eds. R. L. Doty and D. Müller-Schwarze), pp. 335-342. New York: Plenum Press.

Reese, T. S. and Brightman, W. M. (1970). Olfactory surface and central olfactory connections in some vertebrates. In Taste and Smell in Vertebrates, eds. G. E. W. Wolstenhome and J. Knight), pp. 115-149. London: Ciba Foundation Symposium/J. and A. Churchill.

Richard, J. D. (1968). Fish attracted with low-frequency pulsed sound. Journal of the Fisheries Research Board of Canada 25, 1441-1452.

Rivera-Vicente, A. C., Sewell, J. and Tricas, T. C. (2011). Electrosensitive spatial vectors in elasmobranch fishes: implications for source localization. PLoS ONE 6, e6008.

Roberts, B. L. (1978). Mechanoreceptors and the behavior of elasmobranch fishes with special reference to the acoustico-lateralis system. In Sensory Biology of Sharks, Skates, and Rays, eds. E. S. Hodgson and R. F. Mathewson), pp. 331-390. Arlington, VA: U.S. Office of Naval Research. 46

Roberts, B. L. and Ryan, K. P. (1971). The fine structure of the lateral-line sense organs of dogfish. Proceedings of the Royal Society of London. Series B: Biological Sciences 179, 157-169.

Robinson, M. P. and Motta, P. J. (2002). Patterns of growth and the effects of scale on the feeding kinematics of the nurse shark (Ginglymostoma cirratum). Journal of Zoology 256, 449-462.

Schluessel, V., Bennett, M. B., Bleckmann, H., Blomberg, S. and Collin, S. P. (2008). Morphometric and ultrastructural comparison of the olfactory system of elasmobranchs: the significance of structure-function relationships based on phylogeny and ecology. Journal of Morphology 269, 1365-1386.

Schluessel, V., Bennett, M. B., Bleckmann, H. and Collin, S. P. (2010). The role of olfaction throughout juvenile development: functional adaptations in elasmobranchs. Journal of Morphology 271, 451-461.

Sheldon, R. E. (1909). The reactions of the dogfish to chemical stimuli. Journal of Comparative Neurology and Psychology 19, 273-311.

Sheldon, R. E. (1911). The sense of smell in selachians. Journal of Experimental Zoology 10, 51-62.

Silver, W. and Finger, T. E. (1984). Electrophysiological examination of a nonolfactory, non-gustatory chemosense in the sea robin. Journal of Comparative Physiology A 154, 167-174.

Silver, W. L. (1979). Olfactory responses from a marine elasmobranch, the Atlantic stingray, Dasyatis sabina. Marine Behaviour and Physiology 6, 297-305.

Silver, W. L., Caprio, J., Blackwell, J. F. and Tucker, D. (1976). The underwater electro- olfactogram: a tool for the study of the sense of smell of marine fishes. Experientia 32, 1216-1217.

Sisneros, J. A. and Tricas, T. C. (2000). Androgen-induced changes in the response dynamics of ampullary electrosensory primary afferent neurons. Journal of Neuroscience 20, 8586-8595.

Sisneros, J. A., Tricas, T. C. and Luer, C. A. (1998). Response properties and biological function of the skate electrosensory system during ontogeny. Journal of Comparative Physiology A 183, 87-99.

Sivak, J. G. (1978a). Optical characteristics of the eye of the spiny dogfish (Squalus acanthias). Reviews of Canadian Biology 37, 209–217.

Sivak, J. G. (1978b). Refraction and accommodation of the elasmobranch eye. In Sensory Biology of Sharks, Skates, and Rays, eds. E. S. Hodgson and R. F. Mathewson), pp. 107–116. Arlington, VA: U.S. Office of Naval Research.

Sivak, J. G. (1991). Elasmobranch visual optics. Journal of Experimental Zoology Suppl. 5, 13-21.

Sivak, J. G. and Gilbert, P. W. (1976). Refractive and histological study of accommodation in two species of sharks (Ginglymostoma cirratum and Carcharhinus milberti). Canadian Journal of Zoology 54, 1811-1817.

Smeets, W. J. A. J. (1983). The secondary olfactory connections in two chondrichthians, the shark Scyliorhinus canicula and the ray Raja clavata. Journal of Comparative Neurology 218, 334-344.

Spielman, S. L. and Gruber, S. H. (1983). Development of a contact lens for refracting aquatic animals. Ophthalmic and Physiological Optics 3, 255-260.

Stein, B. E. and Stanford, T. R. (2008). Multisensory integration: current issues from the perspective of the single neuron. Nature Reviews Neuroscience 9, 255-266.

Takami, S., Luer, C. A. and Graziadei, P. P. C. (1994). Microscopic structure of the olfactory organ of the clearnose skate, Raja eglanteria. Anatomical Embryology (Berlin) 190, 211-230.

Tester, A. L. (1963a). Olfaction, gustation, and the common chemical sense in sharks. In Sharks and Survival, (ed. P. W. Gilbert), pp. 255-282. Lexington: D.C. Heath and Company.

Tester, A. L. (1963b). The role of olfaction in shark predation. Pacific Science 17, 145- 170.

Tester, A. L. and Kendall, J. I. (1968). Cupulae in shark neuromasts: composition, origin, generation. Science 160, 772-774.

Tester, A. L. and Kendall, J. I. (1969). Morphology of the lateralis canal system in shark genus Carcharhinus. Pacific Science 23, 1-16.

Tester, A. L., Kendall, J. I. and Milisen, W. B. (1972). Morphology of the ear of the shark genus Carcharhinus, with particular reference to the macula neglecta. Pacific Science 26, 264-274.

Tester, A. L. and Nelson, G. J. (1967). Free neuromasts (pit organs) in sharks. In Sharks, Skates, and Rays, eds. P. W. Gilbert R. F. Mathewson and D. P. Rall), pp. 503- 531. Baltimore, M.D.: John Hopkins Press.

Theisen, B., Zeiske, E. and Breucker, H. (1986a). Functional morphology of the olfactory organs in the spiny dogfish (Squalus acanthius L.) and the small- spotted catshark (Scyliorhinus canicula L.). Acta Zoologica (Stockholm) 67, 73- 86.

Theisen, B., Zeiske, E. and Breucker, H. (1986b). Functional morphology of the olfactory organs in the spiny dogfish (Squalus acanthius L.) and the small- spotted catshark (Scyliorhinus canicula L.). Acta Zoologica (Stockholm) 67, 73- 86.

Theiss, S. M., Hart, N. S. and Collin, S. P. (2009). Morphological indicators of olfactory capability in wobbegong sharks (Orectolobidae, Elasmobranchii). Brain, Behavior and Evolution 73, 91-101.

Theiss, S. M., Lisney, T. J., Collin, S. P. and Hart, N. S. (2007). Colour vision and visual ecology of the blue-spotted maskray, Dasyatis kuhlii Müller & Henle, 1814. Journal of Comparative Physiology A 193, 67-79.

Tricas, T. C. (1980). Courtship and mating-related behaviors in myliobatid rays. Copeia 1980, 553-556.

Tricas, T. C. (1982). Bioelectric mediated predation by the swell shark, Cephaloscyllium ventriosum. Copeia 4, 948-952.

Tricas, T. C., Kajiura, S. M. and Summers, A. P. (2009). Response of the hammerhead shark olfactory epithelium to amino acid stimuli. Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 195, 947-954.

Tricas, T. C., Michael, S. W. and Sisneros, J. A. (1995). Electrosensory optimization to conspecific phasic signals for mating. Neuroscience Letters 202, 129-132.

Van den Berg, A. V. and Schuijf, A. (1983). Discrimination of sounds based on the phase difference between particle motion and acoustic pressure in the shark Chiloscyllium griseum. Proceedings of the Royal Society of London. Series B: Biological Sciences 218, 127-134.

Veselkin, N. P. and Kovačević, N. (1973). Nonolfactory afferent projections of the telencephalon of elasmobranchii. Journal of Evolutionary Biochemistry and Physiology 9, 512-518.

Waltman, B. (1966). Electrical properties and fine structure of the ampullary canals of Lorenzini. Acta Physiologica Scandinavica 66 (suppl. 264), 1-60.

Warrant, E. J. and Locket, N. A. (2004). Vision in the deep sea. Biological Reviews of the Cambridge Philosophical Society 79, 671-712.

Webb, J. F. and Northcutt, R. G. (1997). Morphology and distribution of pit organs and canal neuromasts in non-teleost bony fishes. Brain, Behavior and Evolution 50, 139-151.

Whitear, M. (1971). The free nerve endings in fish epidermis. Journal of Zoology 163, 231-236.

Whitear, M. and Moate, R. M. (1994a). Chemosensory cells in the oral epithelium of Raja clavata (Chondrichtyes). Journal of Zoology 232, 295-312.

Whitear, M. and Moate, R. M. (1994b). Microanatomy of the taste buds in the dogfish, Scyliorhinus canicula. Journal of the Submicroscopic Cytology and Pathology 26, 357-367.

Wilga, C. D. and Motta, P. J. (2000). Durophagy in sharks: feeding mechanics of the hammerhead Sphyrna tiburo. Journal of Experimental Biology 203, 2781-2796.

Wilga, C. D., Motta, P. J. and Sanford, C. P. (2007). Evolution and ecology of feeding in elasmobranchs. Integrative and Comparative Biology 47, 55-69.

Wright, T. and Jackson, R. (1964). Instrumental conditioning of young sharks. Copeia 1964, 409-412.

Zakon, H. H. (1988). The electroreceptors: diversity in structure and function. In Sensory Biology of Aquatic Animals, eds. J. Atema R. R. Fay A. N. Popper and W. N. Tavolga), pp. 813-850. New York, N.Y.: Springer-Verlag.

Zeiske, E., Caprio, J. and Gruber, S. H. (1986). Morphological and electrophysiological studies on the olfactory organ of the lemon shark Negaprio brevirostris (Poey). In Indo-Pacific Fish Biology: Proceedings of the Second International Conference on Indo-Pacific Fishes, eds T. Uyeno R. Arai T. Taniuchi and K. Matsuura), pp. 381-391. Tokyo: Ichthyological Society of Japan.

Zeiske, E., Theisen, B. and Gruber, S. H. (1987). Functional morphology of the olfactory organ of two carcharhinid shark species. Canadian Journal of Zoology 65, 2406-2412.

Figures

Figure 1. The Olfactory Rosette. (A) Low power SEM image of stacks of lamellae in the bull shark, Carcharhinus leucas. (B) Whole lamella from the western wobbegong, Orectolobus hutchinisi, scale bar = 1mm. High power SEM showing (C) extension of the non-sensory epithelium along the secondary folds in the spotted eagle ray Aetobatus narinari, scale bar = 100µm; (D) division between the sensory (ciliated region) and non-sensory (non-ciliated region with microvilli) in the blue-spotted maskray Dasyatis kuhlii, scale bar = 100µm (E) an olfactory knob present on the lamellae of O. hutchinsi. SF = secondary folds, NS = non-sensory epithelium, S = sensory epithelium, C = cilia, M = microvilli, OK = olfactory knob. (A, C, and D: Schluessel et al., 2008, B and E: from Dr. Susan Theiss).

Figure 2. Nasal Ventilation – Active Elasmobranch. Schematics of the functional morphology of the nasal region of an active elasmobranch. (A) ventral surface of the head of a lemon shark, Negaprion brevirostris (based on Zeiske et al., 1986). Dotted line: approximate location of the olfactory chamber. (B) Boxed region in (A). Lines labeled “C”, “F”, and “E” indicate positions of the front face of sections in panels (C), (E), and (F), respectively. (C) Sagittal section through the olfactory chamber, towards the medial end of the chamber (based on Zeiske et al, 1986), with secondary lamellae shown on the left lamella only. Scale bar: 5 mm. Inset: outlines of incurrent and excurrent channels created by lamellae and the roof of the olfactory chamber. (D) Transverse section through two lamellae, towards the side wall of the olfactory chamber, showing the convoluted nature of the interlamellar channel. (E) Flow through the olfactory chamber, same view as (C). (F) Cut-away view to one side of the olfactory chamber, showing principle flow (arrowed line) through incurrent and excurrent channels (interlamellar gaps and secondary lamellae omitted for clarity). Gray arrows: incurrent flow; white arrows: excurrent flow; dark arrows: flow in interlamellar channels; EC: excurrent channel; EN: excurrent nostril; iC: interlamellar channel; IC: incurrent channel; IN: incurrent nostril; L: lamella; PC: peripheral channel; R: raphe; SF: secondary fold (Abel et al., 2010). 52

Figure 3. Nasal Ventilation – Sedentary Elasmobranch. Proposed path of water drawn through the olfactory chamber by the buccopharyngeal pump in a sedentary elasmobranch. Ventral view; anterior is up. Water (arrows) enters through the incurrent opening (I), flows through the nasal pouch (olfactory chamber; NP), exits the excurrent opening (E) into the nasoral groove (NG), which is covered by the anteromedial nasal flap (ANF). Water continues through the nasoral groove across the palatoquadrate (PQ) and into the mouth (M). Finally, water exits the pharynx (PH) through the gill slits (GS). The first two complete gills slits and spiracle (S) are shown (Bell, 1993)

Figure 4. Shark Brain. Dorsal view of the brain and olfactory system of the white shark, Carcharodon carcharias. AR, anterior ramus of the octaval nerve; AV, anteroventral lateral-line nerve; BU, buccal ramus of the anterodorsal lateral line nerve; DO, dorsal octavolateralis nucleus; MA, mandibular ramus of the trigeminal nerve; MX, maxillary ramus of the trigeminal nerve; OB, olfactory bulb; OC, occipital nerves; OE, olfactory epithelium (within the olfactory sac); OP, olfactory tract (or olfactory peduncle, cranial nerve I); OT, optic tectum; PL, posterior lateral line nerve; PR, posterior ramus of the octaval nerve; PRO, profundal nerve; SC, superficial ophthalmic ramus of the anterodorsal lateral line nerve; T, telencephalon; 0, terminal nerve; II, optic nerve; III, oculomotor nerve; IV, trochlear nerve; VII, facial nerve; IX, glossopharyngeal nerve; X, vagus nerve. Bar, 3 cm. (Demski and Northcutt, 1996)

Figure 5. Taste buds and Solitary Chemosensory Cells. Line drawings of an elasmobranch taste bud (A) and solitary chemosensory cell (B), after Cook and Neal 1921, Whitear and Moate 1994b (A) and Whitear and Moate 1994a (B).

Figure 6. The Elasmobranch Inner Ear (Myrberg, 2001).

Figure 7. The Sensory Cells of the Elasmobranch Neuromast (Roberts and Ryan, 1971).

150μm

50μm

Figure 8. Elasmobranch Neuromasts. Diagram of the canal lateral line system (A) (Tester and Kendall, 1969) and superficial neuromasts (B) of elasmobranchs (Budker, 1958).

Figure 9. Modified Placoid Scales. Scanning electron micrograph of a pair of modified placoid scales covering a pit organ in the bonnethead, Sphyrna tiburo. Image was taken with a Hitachi S-3500N scanning electron microscope (Hitachi High-Technologies America, Inc., Pleasanton, CA USA).

Figure 10. Distribution of the Superficial Neuromasts (Pit Organs) in Sharks. Each dot represents a single neuromast. (A) Ventral surface of the lemon shark, Negaprion brevirostris. (B) Superficial neuromasts on the nurse shark, Ginglymostoma cirratum. (C) Superficial neuromasts on the bonnethead, Sphyrna tiburo (Tester and Nelson, 1967).

A B

C C

Figure 11. Distribution of the Lateral Line Canals in Sharks. Canals are found on the dorsal (A) and ventral (B) sides of the head and along the flank (C) of the bonnethead, Sphyrna tiburo. HYO = hyomandibular canal, IO = infraorbital canal, MAN = mandibular canal, PLL = posterior lateral line canal, SO = supraorbital canal. (Maruska, 2001).

Figure 12. The Visual Fields of Sharks. Maximum dynamic horizontal visual fields (when the eyes are fully converged and diverged and with maximum lateral head yaw) and static vertical fields of four shark species. Values within the shaded areas represent the monocular fields. Values outside of the shaded areas represent degrees of binocular overlap (anterior and posterior) or blind areas, if in parentheses (McComb et al., 2009).

Figure 13. The Elasmobranch Eye. Cross section through an elasmobranch eye showing ocular and retinal anatomy (modified from Hueter and Gilbert, 1991). Inset: light micrograph of the retina of the giant shovelnose ray, Rhinobatos typus, showing the photoreceptive layer (longer receptors are rods, shorter receptors are cones). gc: ganglion cell layer; h: horizontal cell layer; ipl: inner plexiform layer; p: photoreceptor layer. Scale bar: 100µm. (photomicrograph from Hart et al., 2004).

Figure 14. Visual Streaks and Areae Centrales. Diagrammatic representation of regions of the visual field subserved by regions of higher retinal cell density, depicting: horizontal streaks (lightly shaded bands) with multiple areae (darkly shaded ovals) in (A) the eastern shovelnose ray, Aptychotrema rostrata, and (B) the epaulette shark, Hemiscyllium ocellatum; and concentric retinal areae in (C) the whitetip reef shark, Triaenodon obesus; and (D) the ornate wobbegong, Orectolobus ornatus. N: nasal, T: temporal. (modified from Litherland and Collin, 2008).

Figure 15. The Ampullary Electroreceptor Organ of Elasmobranchs (modified from Waltman, 1966; Hueter et al., 2004).

Figure 16. Geomagnetic Orientation in Sharks. Diagram of the induced electric current in the head and body as the shark swims through the Earth’s magnetic field (Kalmijn, 1988).

Figure 17. The Electrosensory Arrays in Sharks. Horizontal view of the electrosensory arrays in sandbar shark, Carcharhinus plumbeus (A) and the scalloped hammerhead, Sphyrna tiburo (B). Canals from the different ampullary groups are shown in different colors. Buccal = dark blue, mandibular = light blue, anterior superficial opthalmic = green, posterior superficial opthalmic = red (Rivera-Vicente et al., 2011).

CHAPTER 2: THE FUNCTION OF BILATERAL TIMING DIFFERENCES

IN OLFACTORY ORIENTATION OF SHARKS 1

Summary

The direction of an odor signal source can be estimated from bilateral differences

in signal intensity and/or arrival time. The best-known examples of the use of arrival time

differences are in acoustic orientation (Poganiatz et al., 2001). For chemoreception,

animals are believed to orient by comparing bilateral odor concentration differences,

turning toward higher concentrations (Bardach et al., 1967; Atema, 1971; Johnsen and

Teeter, 1985). However, time differences should not be ignored, because odor plumes

show chaotic intermittency, with the concentration variance several orders of magnitude

greater than the concentration mean (e.g., Webster, 2007). We presented a small shark

species, Mustelus canis, with carefully timed and measured odor pulses directly into their

nares. They turned toward the side stimulated first, even with delayed pulses of higher

concentration. This is the first conclusive evidence that under seminatural conditions and

without training, bilateral time differences trump odor concentration differences. This

response would steer the shark into an odor patch each time and thereby enhance its

contact with the plume, i.e., a stream of patches. Animals with more widely spaced nares

would be able to resolve smaller angles of attack at higher swimming speeds, a feature

1 This chapter was previously published as Gardiner, J. M. and Atema, J. (2010). The function of bilateral timing differences in olfactory orientation of sharks. Current Biology 20, 1187-1191. Reproduced with permission. 68

that may have contributed to the evolution of hammerhead sharks. This constitutes a

novel steering algorithm for tracking odor plumes.

Results and Discussion

The notion that animals respond to bilateral odor concentration differences is

based on the fact that odor dilutes and diffuses gradually away from the source and on the

commonly held but erroneous idea that this causes a measurable concentration gradient.

This idea does not take into account the chaotic nature of most odor dispersal processes.

In freely moving fluids, turbulent mixing generates a cascade of ever-smaller eddies, resulting in an odor dispersal field, or plume. An odor plume, often an odorous wake left behind another animal or object, is essentially a stream of mixing eddies (or patches, filaments) with and without odor. These spatial patches appear as temporal pulses to typically fast-adapting olfactory receptor cells, resulting in strong responses to pulse onset and sudden concentration increase, followed by response cessation (Gomez and

Atema, 1996). Odor plumes show chaotic intermittency, with the concentration variance several orders of magnitude greater than the concentration mean (e.g., Webster, 2007).

Therefore, a spatial concentration gradient can be obtained only by averaging. However, it typically requires many minutes for the concentration averaging process to reach a stable mean. For most animals, this is much too slow to be useful for tracking prey or mates (Elkinton and Cardé, 1984; Moore et al., 1991; Webster and Weissburg, 2001;

Webster, 2007). Steering algorithms based on odor concentration differences implemented in an underwater robot were useful only near the source under coherent plume conditions (Grasso et al., 2000).

Sharks are classically believed to respond to differences in odor concentration at the nares (Sheldon, 1911; Parker, 1914; Hasler, 1957; Tester, 1963; Hodgson and

Mathewson, 1971; Mathewson and Hodgson, 1972; Johnsen and Teeter, 1985).

Previously, Johnsen and Teeter (1985) fitted bonnetheads, Sphyrna tiburo, with a stereo headstage to control the delivery of (food) odor stimuli directly in front of the nares

(nostrils). When one naris received a (2x) stronger odor pulse, the animals turned toward the side receiving the stronger stimulus. However, it is likely that concentration and timing differences were confounded. The odors in their experiment were preloaded as a discrete bolus into long tubing, with seawater both ahead of and behind the odor. This caused dilution of the leading and trailing edges of the odor bolus: it reached 50% of the applied concentration 7 s after initiation of odor delivery. Thus, the high concentration side would have reached response threshold before the low concentration side, and the animals received bilateral differences not only in the concentration but also – unintentionally – in the arrival time of detectable levels of odor at each naris.

To evaluate their contributions to steering behavior, both odor arrival time and concentration must be known. This requires accurate control of odor pulse shape, i.e., the concentration-time profile of the stimulus (Moore et al., 1991; Webster and Weissburg,

2001). To accomplish this goal, we fitted a small shark species, Mustelus canis, with a

headstage apparatus designed to separately control the concentration of odor and the

timing of its arrival at the nares via computer-controlled syringe pumps (Figures 2.1A

and 2.1B). All odor pulses were standardized to concentration, volume 0.5 ml, duration

5.22 s, and flow speed 1.5 cm/s. The latter is less than 10% of the estimated natural flow

through the shark nose (see Supplemental Experimental Procedures) that is driven by

70 pressure differences between inflow and outflow nares resulting from forward swimming motion (Theisen et al., 1986; Zeiske et al., 1986; Zeiske et al., 1987; Abel et al., 2010).

Six different patterns of odor pulses were used. Four involved timing differences between the nares, such that one naris received an odor pulse (1) 0.1 s, (2) 0.2 s, (3) 0.5 s, or (4)

1.0 s ahead of the other naris. Two odor stimulus patterns involved concentration differences. A 100-fold dilution of squid odor was delivered to one naris and full-strength squid odor to the other naris either (5) simultaneously to the two nares (0 s delay) or (6) with a 0.5 s delay such that the naris receiving the weak stimulus received it 0.5 s ahead of the naris receiving the full-strength squid odor. To confirm that responses were to the odor component of the pulse, we also tested each animal as above (7) with ambient seawater pulses, delivered with a 0.5 s time delay between the nares.

The seawater control caused a greater proportion of null there was no significant difference between the number of turns toward and away from the water pulses (t test, p

= 0.8); thus, the animals did not respond to seawater pulses. In contrast, each of the odor stimulus patterns caused a greater proportion of turns than null responses (see also Table

2.1); thus, odor and seawater stimuli were significantly different in terms of the proportion of responses observed (repeated measures analysis of variance (RM

ANOVA), p < 0.001, n = 8; Tukey test, p < 0.001). Among the odor stimulus patterns, the proportion of responses observed was not significantly different (Tukey test, p = 1).

These results indicate that the animals did not respond to the tactile sensation of slightly increased flow entering the nares. Of course, one might still argue that the directional response might be caused by the tactile sensation of time-delayed bilateral flow, but only in the presence of nondirectional odor. However, in nature, odor arrival is a far more

71 salient signal than ubiquitous flow variance, which, in our experimental condition, was estimated to be at least an order of magnitude lower than the normal flow through the nose (see Supplemental Experimental Procedures).

For the 0.1 s, 0.2 s, and 0.5 s time delays, the animals turned with a significantly greater frequency toward the side receiving the first stimulus than toward the side receiving the later stimulus. For the 1 s time delay, the animals turned to either side with equal frequency (Figure 2.1C). For the bilateral (0 s delay) concentration differences, the animals again turned toward either side with equal frequency. When the concentration and time differences were combined, the animals once again turned toward the side receiving the first, albeit weaker, stimulus with significantly greater frequency (Figure

2.1D). The results for the 0.1 s, 0.2 s, and 0.5 s delays, and for the concentration difference with 0.5 s delay, were significantly different from those for the concentration difference with 0.5 s delay, were significantly different from those for the 1 s time delay and simultaneous bilateral concentration differences (RM ANOVA, p < 0.001; Tukey test, p < 0.001), but they were not significantly different from one another (Tukey test, p

= 0.98–1).

The results show that odor arrival time differences, and not concentration differences, cause directional turning in the shark M. canis. Even 100x concentration differences were ignored in favor of arrival time differences. Although the concentration of odor is still important in that it must be high enough to trigger a behavioral response, the animals did not use comparisons of odor concentration between the nostrils for orientation as previously hypothesized (Sheldon, 1911; Parker, 1914; Hasler, 1957;

Tester, 1963; Hodgson and Mathewson, 1971; Mathewson and Hodgson, 1972; Johnsen

and Teeter, 1985).

The latency from delivery of the stimulus to the initiation of a turn (1.07 ± 0.1 s

standard error of the mean) did not vary significantly among the different stimulus patterns (RM ANOVA, p = 0.08). Thus, directional turning was limited to a subsecond time window of arrival time differences. This suggests that sharks use a simple brain algorithm that would allow the animal to distinguish between (1) a head-on encounter with an odor patch (no time difference: respond with turning in either direction to obtain new patch encounters with directional information), (2) an oblique encounter (0.1–0.5 s time difference: respond directionally and turn into the odor patch), and (3) an encounter with two separate odor patches on either side, indicating a position likely within the plume (1 s time difference: respond with a turn in either direction to obtain new encounters with directional information). Unilateral stimulation (encounter with odor on one side and a delay of > > 1 s before the next encounter) is likely to indicate a position at the plume edge and results in a turn toward the stimulated side as shown earlier

(Parker, 1914). A related phenomenon is known in flying moths, which, during plume tracking, delay their turning behavior by 300 ms after losing contact with an odor patch.

Apparently this delay provides good probability of hitting a possible next patch, which leads them closer to the plume source (Mafra-Neto and Cardé, 1994; Vickers and Baker,

1994). The overall result is an enhanced ability to stay connected with the chaotic dispersal field of an odor plume, reducing the real danger of losing the plume altogether.

(We note that our smallest temporal resolution of 100 ms internarial delay for M. canis,

73 measured with 20 ms accuracy, was the result of our technical limitation of measurable odor delivery and did not establish the actual lower limit of the animal.)

Differences in bilateral encounter times with a single patch are a function of three factors: the angle of attack (i.e., the angle by which the animal approaches the odor patch boundary (Figure 2.2A), the spacing of the nares, and the swimming speed of the animal with respect to its fluid environment. Whereas the angle of attack is a function of local plume structure and internarial spacing is a species property, swimming speed is regulated by the individual animal. This suggests control over swimming speed during plume tracking, which determines the size scale of patches that can be resolved, because both faster and slower speeds could take them out of the neural detection-time window.

Our sharks track at a typical speed of w1 m/s, which is significantly less than what they are capable of (>3 m/s), despite the potentially competitive nature of food tracking among nearby animals. This speed (10 cm/100 ms) is well suited to resolve odor patches spaced on the order of 10 cm. Similarly, reduced walking speed during plume tracking has been observed in lobsters (Moore et al., 1991). Also, lobsters stop tracking if patch spacing is greater than 10 cm (Kozlowski et al., 2001). It also suggests that in order to obtain the same information, faster swimming animals may have developed greater time resolution and/or more widely spaced nares.

For a given angle of attack, the lag time between nares is a function of internarial distance (Figure 2.2B). Hammerhead sharks (Carcharhiniformes, Sphyrnidae) all possess dorsoventrally compressed and laterally expanded heads, termed cephalofoils. This anatomy increased separation of the nares (Tester, 1963; Gilbert, 1967; Compagno, 1984;

Kajiura et al., 2005) and may confer an olfactory advantage to these animals (Hasler,

1957; Tester, 1963; Nelson, 1969; Johnsen and Teeter, 1985; Kajiura et al., 2005). Their broad nares and long prenarial grooves allow hammerheads to sample a greater area of the medium, increasing the probability of detecting an odor (Tester, 1963; Gilbert, 1967;

Kajiura et al., 2005). However, hammerheads do not possess a greater olfactory epithelial surface area than the typical carcharhinid sharks (Tester, 1963; Kajiura et al., 2005), nor

do they appear to have a greater olfactory sensitivity, as indicated by their thresholds for

detection of single amino acids, which are comparable to other fishes (Hodgson et al.,

1967; Hara, 1992; Meredith and Kajiura, 2009; Tricas et al., 2009). However, as a result of the wider spacing of the nares, hammerheads may be able to perceive a bilateral time difference at a smaller angle or at a greater swimming speed than an animal with a narrow head.

We compared measurements of the head and nares of M. canis to other elasmobranchs, as per Kajiura et al. (2005). Using a swim speed of 1 body length per second and scaling all species to a total length of 1 m reveals that the winghead,

Eusphyra blochii, which has the widest head and the greatest narial separation, would be expected to experience the longest internarial timing delays for a given angle (Figure

2.2C). Assuming that both the concentration detection threshold and the threshold for detection of internarial time differences are the same across all four species, E. blochii

would be capable of orienting to odor patches at a much smaller angle of attack as

compared to the other species, followed by the scalloped hammerhead, Sphyrna lewini,

the sandbar shark, Carcharhinus plumbeus, and the smooth dogfish, Mustelus canis. The

results of our study suggest a new theory for the evolution of the ‘‘hammer’’: enhanced

olfactory tracking capabilities as a result of increased temporal resolution for klinotaxis.

To confirm this new version of the enhanced olfactory hypothesis, we need to examine

the range of internarial timing differences that hammerhead sharks can actually detect.

In conclusion, this study has shown that sharks can use odor information for steering despite the scalar nature of odor. This function may be important and specific: turning into an odor patch. Superficially, steering by a scalar suggests using an intensity gradient, but the physics of odor dispersal make this difficult, and it appears to not be used except in special odor-dispersal circumstances. However, bilateral arrival time differences can give useful directional information at the spatial scale of an odor patch encounter. Because odor plumes can be seen as odor patch dispersal fields, tracking the sequence of patches may be the best way to locate the odor source (e.g., Atema, 1996).

Insects use this approach, but they steer by local wind direction (Elkinton and Cardé,

1984; Mafra-Neto and Cardé, 1994; Vickers and Baker, 1994; Steck et al., 2010); sharks appear to do the same, but they steer by bilateral odor arrival time differences. Again, the physics of dispersal may have enforced the difference. In insects, small antennal separation distance and great flight speed may make time differences more difficult to resolve, whereas visual contact with the ground allows them to use local wind drift for steering, something not possible for many marine animals that cannot see the bottom.

Experimental Procedures

Eight smooth dogfish, Mustelus canis, were captured in the waters surrounding

Woods Hole, MA and transported to the Marine Biological Laboratory (MBL), where

they were maintained according to protocols approved by the Institutional Animal Care

and Use Committees at the MBL (protocol 08-26) and the University of South Florida

(protocol W3211). To control the presentation of odors, we fitted the animals with headstages, attached such that the tubing sat just inside the inflow nares (Figures 2.2A and 2B). Tubing from the left and right sides of the headstage were attached to two syringes, each driven by a programmable syringe pump. These pumps were synchronized via a laptop computer. Squid rinse (Gardiner and Atema, 2007) was used as the odor source in all experiments. Details of this apparatus and procedures are described in the

Supplemental Experimental Procedures.

Experiments were conducted in a flume tank filled to 60 cm and divided into 3 m long 3 2 m wide pens. During trials, seawater flow in the tank was shut off, and experiments were conducted in still water. Each trial began by offering the animal a small piece of squid to confirm its hunger status. If the animal did not consume the squid, testing was delayed for another 24 hr. If the animal consumed the squid, the trial proceeded. In a trial, an animal was presented ten times with one of the seven stimulus patterns (described above), each repetition separated by a 10 s delay between the end of the previous stimulus and the onset of the next. Each trial was simultaneously observed and recorded by overhead video cameras. Each animal was given two trials for each of the six odor stimulus patterns (as well as the seawater control), once with the left naris receiving the first or strongest odor pulse and once with the right naris receiving the first or strongest odor pulse. This was to account for any possible bias on the part of the animal to turn in a particular direction.

Following each trial, the animal was again offered a piece of squid. If it was not consumed, the animal was determined to lack the proper motivation and the prior trial was rejected, resulting in rejection of 3 out of 115 trials. Between trials, the tank was

77 placed on flow-through seawater for 30 min to flush out any lingering odors. Up to four trials per day were conducted in this manner on each animal. Each animal was tested with all six stimulus patterns, as well as the seawater control. The order of the presentation of the stimulus patterns was randomized, and the headstage tubing was flushed with fresh seawater after every trial.

The overhead video was analyzed using MaxTRAQ Standard v.1.93 software

(Innovision Systems). A turn was defined as at least a 30° change in heading from the direction of travel just prior to the delivery of an odor pulse. A null response score was given to any changes in heading of less than 30° or to turns that were initiated more than

6 s after the beginning of the first odor pulse. If the animal was already in a turn when the odor pulses were delivered, the response was removed from the analysis because it could not be determined whether the animal was responding to the pattern of the odor stimulus or simply completing the turn that it had already initiated. The direction of any turn was recorded, as well as the latency, defined as the time from the start of the first odor pulse in each pair to the initiation of a turn. The researcher analyzing the video was provided with the start time for each odor pulse but was otherwise blind to the stimulus pattern being tested. A film record of a sample trial (Movie 2.1) and its analysis (Table 2.2) are available as part of the Supplemental Data.

The data for each stimulus pattern for each animal were pooled, sine square-root transformed, and tested for normality and equality of variance using Kolmogorov-

Smirnoff and Levene Median tests (Sokal and Rohlf, 1995). Within each stimulus pattern used, responses were examined using t tests for dependent samples, with each animal weighted based on the number of data points (turns + null responses). Because the same

animals were tested with each stimulus pattern, data were compared among stimulus

patterns using RMANOVA and Tukey post hoc tests. Analyses were conducted using

Statistica 5.5 (StatSoft) and SigmaStat 3.05 (Systat Software).

Supplemental Information

Supplemental Data

Table 2.1. Summary of the Proportion of Null Responses versus Turns for the Various Stimulus Patterns Used Stimulus Pattern Null Responses Turns n p Value Seawater 0.77 ± 0.04 0.23 ± 0.04 8 0.0008 0.1s timing delay 0.06 ± 0.03 0.94 ± 0.03 8 0.00004 0.2s timing delay 0.06 ± 0.02 0.94 ± 0.02 8 0.00002 0.5s timing delay 0.06 ± 0.02 0.94 ± 0.02 8 0.00002 1.0s timing delay 0.05 ± 0.02 0.95 ± 0.02 8 0.000006 0s delay, concentration difference 0.07 ± 0.03 0.91 ± 0.03 8 0.00001 0.5s delay, concentration difference 0.08 ± 0.03 0.92 ± 0.03 8 0.0001

Data are represented as mean ± SEM. The p values are from t-tests for dependent samples of the null hypothesis that there is no difference between null responses and turns for each of the stimulus patterns. The odor and seawater stimuli were significantly different from one another in terms of the proportion of responses observed (repeated measures ANOVA, p < 0.001, n=8; Tukey post-hoc test, p < 0.001). The odor stimulus patterns were not significantly different from one another in this regard (Tukey post-hoc test, p = 1).

Movie 2.1. Sample Trial. Turning behavior of Mustelus canis in response to odor stimuli received via the headstage apparatus. The animal is receiving a pulse of squid rinse to the left naris 0.1 s ahead of a pulse to the right naris. The L and R icons at the bottom of the screen indicate the timing and duration of the left and right odor pulses, respectively. Video playback has been slowed by 25% to allow the viewer a better opportunity to see the responses (summarized in Table 2.2) http://www.sciencedirect.com/science/MiamiMultiMediaURL/B6VRT-508NK44-1/B6VRT- 508NK44-1- 1/6243/html/S0960982210005919/dafa2977aa8756a4fed88267b1cc5b7d/mmc2.mp4

Table 2.2. Analysis of Sample Trial Showing 10 Stimulus Pairs Stimulus Latency (s) Response Notes 1 (3.7s) 1.9 L 2 (22.8s) 0.9 L 3 (43.2s) 1.6 R 4 (63.7s) 1.7 L 5 (85.4s) 1 L 6 (103.9s) 0.7 L 7 (124.23s) 1.4 L 8 (144.8) NA Animal was in a turn when pulses were delivered 9 (165.1) 1.8 L 10 (185.3) 0.6 L

The responses to the odor stimuli were scored as follows L = change in heading ≥ 30° to the left; R = change in heading ≥ 30° to the right; NR = Null response, change in heading < 30° (none occurred in this trial); NA = Not analyzed. In this movie, all pulse delays are left naris 0.1s ahead of right naris. Note: timestamps (in parentheses) correspond to the time the stimulus was applied and the latency between start of stimulus and start of response seen in Movie 1 The movie was slowed by 25% for clarity.

Supplemental Experimental Procedures

Experimental Animals

Smooth dogfish, Mustelus canis, were captured by otter trawl or rod and reel in the waters surrounding Woods Hole, MA, and transported back to the Marine Biological

Laboratory (MBL) using a flow-through life support system. Animals were housed in 6m long x 1m wide rectangular tanks, filled to a depth of 0.6 m. The tanks were supplied with flow-through seawater at ambient temperature and maintained on a 12 hour light cycle. The animals were maintained on a diet of squid and fish, supplemented with

Mazuri Shark Tabs, and fed to satiation three times per week except during periods of experimentation, during which time they were fed small amounts of squid daily as part of the experimental procedure.

Headstage Apparatus

The headstages were made from thin PVC rods, heated and bent to match the contours of the animal’s head and snout. Stiff polyethylene tubing, ID 1.4mm, OD

1.9mm (Intramedic), was glued to the PVC rods. Connectors, made from 16 gauge needles (BD, Franklin Lakes, NJ) were fitted to the ends of the tubing, allowing for a connection on each side with a 50cm length of flexible PVC tubing, ID 2.4mm, OD

3.2mm. A small styrofoam float was attached to the PVC tubing to prevent the animal from tangling. The entire headstage was fixed to a small piece of fiberglass which was glued to a piece of Velcro.

For experiments, each side of the headstage was connected to a 6m length of flexible PVC tubing, ID 2.4mm, OD 3.2mm, suspended above the experimental tank to prevent animal tangling, connected to a 5mL syringe (BD, Franklin Lakes, NJ) driven by a programmable syringe pump (Braintree Scientific, Braintree, MA). The two syringe pumps were synchronized via a laptop computer, using Win Pump Control software

(Open Cage Software Inc., Huntington, NY). The smallest inter-pump time delay that could be programmed was 100 ms. Algorithms for each stimulus pattern were written in the pump programming language and uploaded to the pumps via the laptop. The syringes and the entire length of tubing, including the headstage, were pre-loaded entirely with the specified odor stimulus prior to each experiment. Prior to use on animals, the entire apparatus was calibrated by filming pulses of seawater containing Rhodamine B dye, using a Photron Fastcam 512 PCI high speed digital video camera (Photron USA Inc.,

San Diego, CA) at 500 frames s-1, a resolution of 2 ms. Dye did not begin to exit the headstage apparatus until the after the syringe pumps were turned on. The delay from the

onset of the pump to the exit of odor from the tips of the tubing on the headstage was

found to be 194 ± 18 ms (SEM) and the left and right sides of the headstage were found

to be synchronized to within 15 ± 7ms (SEM).

To attach and detach the headstage easily we sutured under anesthesia with MS-

222 a small piece of Velcro to the dorsal side of the head using nylon sutures (Look

Surgical Specialties Corporation, Reading, PA). The headstages were attached such that

the polyethylene tubing sat just inside the inflow nares. Prior to testing, animals were allowed at least a week to recover from anesthesia and to acclimate to wearing the headstage.

Odor Stimuli

Squid rinse, made by soaking 100g (wet weight) of freshly thawed Loligo pealei in 1L seawater for one hour (Gardiner and Atema, 2007) was used as the odor stimulus for all experiments. One large stock batch was prepared and aliquots were frozen and stored. For each experiment, an aliquot was thawed and warmed to ambient tank temperature. Odor was delivered to the nares via the headstage in pulses, each of which consisted of 0.5mL of odor (Johnsen and Teeter, 1985), delivered at 5.75mL/min, resulting in a 5.22s pulse duration. From videos of pulses of dye-colored water exiting the headstage (above), we determined that the average velocity of the pulses was 1.5 cm/s.

Using video of the model of Abel et al (Abel et al., 2010) we determined that for an shark swimming at 20 cm/s, the flow rate just inside the incurrent nares was 15 cm/s. M. canis swims at approximately 1 body length per second (average 80 cm/s in this study), but we cannot estimate how much faster the narial flow rate would be at this speed, since this

relationship likely does not scale linearly and the two species are quite different. We

estimate that our odor pulses delivered a flow rate at least an order of magnitude lower

than the natural flow of water through the animals’ nares. One cannot deliver controlled

odor to the animals without some flow component, since the two are inextricably linked.

However, the pulses were designed to be low velocity and thus unlikely to present a

tactile stimulus to the animals.

Literature Cited

Atema, J. (1971). Structures and functions of the sense of taste in the catfish, Ictalurus natalis. Brain Behavior and Evolution 4, 273-294.

Atema, J. (1996). Eddy chemotaxis and odor landscapes: exploration of nature with animal sensors. Biological Bulletin 191, 129-138.

Bardach, J. E., Todd, J. H. and Crickmer, R. (1967). Orientation by taste in fish of the genus Ictalurus. Science 155, 1276-1278.

Compagno, L. J. V. (1984). FAO Species Catalogue. Sharks of the World. An Annotated and Illustrated Catalogue of Shark Species Known to Date. Part 2. Carcharhiniformes. FAO Fisheries Synopsis 125 Vol. 4.

Elkinton, J. S. and Cardé, R. T. (1984). Odor dispersion. In Chemical Ecology of Insects, eds. W. J. Bell and R. T. Cardé), pp. 73-91. London: Chapman and Hall Ltd.

Gardiner, J. M. and Atema, J. (2007). Sharks need the lateral line to locate odor sources: rheotaxis and eddy chemotaxis. Journal of Experimental Biology 210, 1925-1934.

Gardiner, J. M. and Atema, J. (2010). The function of bilateral timing differences in olfactory orientation of sharks. Current Biology 20, 1187-1191.

Gilbert, C. R. (1967). A revision of the hammerhead sharks (Family Sphyrnidae). Proceedings of the United States National Museum 119, 1-88.

Gomez, G. and Atema, J. (1996). Temporal resolution in olfaction: stimulus integration time of lobster chemoreceptor cells. Journal of Experimental Biology 199, 1771- 1779.

Grasso, F. W., Consi, T. R., Mountain, D. C. and Atema, J. (2000). Biomimetic robot lobster performs chemo-orientation in turbulence using a pair of spatially separated sensors: progress and challenges. Robotics and Autonomous Systems 30, 115-131.

Hara, T. J. (1992). Mechanisms of olfaction. In Fish Chemoreception, (ed. T. J. Hara), pp. 150-170. London: Chapman and Hall.

Hasler, A. D. (1957). Olfactory and gustatory senses of fishes. In The Physiology of Fishes, vol. II Behavior (ed. M. E. Brown), pp. 187-209. New York: Academic Press Inc.

Hodgson, E. S. and Mathewson, R. F. (1971). Chemosensory orientation in sharks. Annals of the New York Academy of Sciences 188, 175-182.

Hodgson, E. S., Mathewson, R. F. and Gilbert, P. W. (1967). Electroencephalographic studies of chemoreception in sharks. In Sharks, Skates, and Rays, eds. P. W. Gilbert R. F. Mathewson and D. P. Rall), pp. 491-501. Baltimore: John Hopkins Press.

Johnsen, P. B. and Teeter, J. H. (1985). Behavioral responses of bonnethead sharks (Sphyrna tiburo) to controlled olfactory stimulation. Marine Behaviour and Physiology 11, 283-291.

Kajiura, S. M., Forni, J. B. and Summers, A. P. (2005). Olfactory morphology of carcharhinid and sphyrnid sharks: does the cephalofoil confer a sensory advantage? Journal of Morphology 264, 253-263.

Kozlowski, C., Yopak, K., Voigt, R. and Atema, J. (2001). An initial study on the effects of signal intermittency on the odor plume tracking behavior of the American Lobster, Homarus americanus. Biological Bulletin 201, 274-276.

Mafra-Neto, A. and Cardé, R. T. (1994). Fine-scale structure of pheromone plumes modulates upwind orientation of flying moths. Nature 369, 142-144.

Mathewson, R. F. and Hodgson, E. S. (1972). Klinotaxis and rheotaxis in orientation of sharks toward chemical stimuli. Comparative Biochemistry and Physiology, A: Comparative Physiology 42, 79-84.

Meredith, T. L. and Kajiura, S. M. (2009). Olfactory thresholds of elasmobranchs. Chemical Senses 34, A51-A51.

Moore, P. A., Scholz, N. and Atema, J. (1991). Chemical orientation of lobsters, Homarus americanus, in turbulent odor plumes. Journal of Chemical Ecology 17, 1293-1307.

Nelson, D. R. (1969). The silent savages. Oceans 1, 8 - 22.

Parker, G. H. (1914). The directive influence of the sense of smell in the dogfish. Bulletin of the United States Bureau of Fisheries 33, 61-68.

Poganiatz, I., Nelken, I. and Wagner, H. (2001). Sound-localization experiments with barn owls in virtual space: influence of interaural time difference on head-turning behavior. Journal of the Association for Research in Otolaryngology 2, 1-21.

Sheldon, R. E. (1911). The sense of smell in selachians. Journal of Experimental Zoology 10, 51-62.

Sokal, R. R. and Rohlf, F. J. (1995). Biometry: The Principles and Practices of Statistics in Biological Research. New York: W.H. Freeman and Co.

Steck, K., Knaden, M. and Hansson, B. S. (2010). Do desert ants smell in stereo? Animal Behaviour 79, 939-945.

Tester, A. L. (1963). Olfaction, gustation, and the common chemical sense in sharks. In Sharks and Survival, (ed. P. W. Gilbert), pp. 255-282. Lexington: D.C. Heath and Company.

Theisen, B., Zeiske, E. and Breucker, H. (1986). Functional morphology of the olfactory organs in the spiny dogfish (Squalus acanthius L.) and the small-spotted catshark (Scyliorhinus canicula L.). Acta Zoologica (Stockholm) 67, 73-86.

Vickers, N. J. and Baker, T. C. (1994). Reiterative responses to single strands of odor promote sustained upwind flight and odor source location by moths. Proceedings of the National Academy of Sciences of the United States of America 91, 5756- 5760.

Webster, D. R. (2007). Structure of turbulent chemical plumes. In Trace Chemical Sensing of Explosives, (ed. R. L. Woodfin), pp. 109-129. New York: John Wiley & Sons.

Webster, D. R. and Weissburg, M. J. (2001). Chemosensory guidance cues in a turbulent chemical odor plume. Limnology and Oceanography 46, 1034-1047.

Zeiske, E., Theisen, B. and Gruber, S. H. (1987). Functional morphology of the olfactory organ of two carcharhinid shark species. Canadian Journal of Zoology 65, 2406- 2412.

Figures

Figure 2.1. Proportion of Turns in Response to Different Stimulus Patterns (A) Photograph of a resting smooth dogfish, Mustelus canis, wearing the headstage apparatus. (B) Illustration of the ventral side of a smooth dogfish wearing the headstage apparatus. The tubing orifice is located just inside the entrance of the inflow naris without closing it off and keeps the outflow naris unobstructed for normal function. Close observation of dye pulse delivery showed that the entire pulse is immediately drawn into the inflow naris. (C) Proportion of turns in response to timing differences. The size of the gap between the bars on the y axis is proportional to the difference in timing of arrival of the odor stimulus at the nares. Light bars indicate turns toward the first side stimulated, darker bars indicate turns toward the second side stimulated. Odor concentration is equal in all cases. Data are represented as mean ± standard error of the mean (SEM). (D) Proportion of turns in response to concentration differences. The gap between the bars on the y axis is proportional to the difference in arrival time of the stimulus at the nares. Dark bars indicate stronger odor stimuli (full strength), lighter bars indicate weaker odor concentration (1:100 dilution). Data are represented as mean ± SEM. 87

Figure 2.2. Differences in Odor Arrival Time (A) Differences in odor arrival time based on angle of attack. Illustration of a smooth dogfish, Mustelus canis, swimming into the leading edge of an odor patch oriented at various angles to the head. Assuming a constant swim speed, the time delay of odor arrival at the second naris (arrow) increases as the angle increases. (B) Differences in odor arrival time based on internarial spacing. Comparison of the smooth dogfish, Mustelus canis, with the sandbar shark, Carcharhinus plumbeus, and two species of hammerhead sharks, the scalloped hammerhead, Sphyrna lewini, and the winghead, Eusphyra blochii, swimming into an odor patch at a 45° angle. If the swim speed is constant across all four species, the wider spacing of the nares of the hammerheads results in an increased internarial arrival time difference (arrows). All heads are scaled to animals of 1 m total length. (C) Comparison of internarial arrival time differences among shark species. The expected internarial timing difference (in seconds) resulting from the encounter of the leading edge of an odor patch over a range of angles for four species of sharks: E. blochii (black solid line), S. lewini (gray solid line), C. plumbeus (gray dashed line), and M. canis (gray dotted line). The horizontal dashed lines indicate the largest and smallest timing differences tested in this study, to which M. canis demonstrated a directional response.

CHAPTER 3: MULTISENSORY INTEGRATION IN FOOD SEARCH

BEHAVIORS

Abstract

Our understanding of elasmobranch sensory biology is largely due to studies of individual senses rather than multiple senses working together, leading to important advances in our comprehension of the sensory systems in isolation, but not their

complementary and alternative roles in difficult behavioral tasks, such as feeding. In this

study, three species from different ecological niches were investigated: benthic, suction-

feeding nurse sharks that hunt nocturnally for fish; ram-biting bonnetheads that scoop

crustaceans off the bottom of seagrass beds; and ram-feeding blacktip sharks that rapidly

chase down midwater teleost prey. Animals were deprived of information from the senses

(olfaction, vision, mechanoreception, and electroreception), alone and in combination, to

elucidate their roles in detecting, tracking, orienting to, striking at, and capturing live

prey. Nurse sharks rely primarily on olfaction for detection. Olfaction in combination

with vision, the lateral line, or touch is required for tracking. Nurse sharks orient to prey using the lateral line, vision, or electroreception, but will not ingest food if olfaction is blocked. Capture is mediated by the electrosensory system or tactile cues. Bonnetheads normally detect prey using olfaction, rely on olfactory-based tracking until they are close

to the prey, then vision to line up a strike, and finally electroreception to time the jaw

movements for capture. They can detect, orient, and strike visually in the absence of 89

olfactory cues. Blacktip sharks also detect prey using olfaction or vision. Olfaction is

used in combination with vision or the lateral line system for tracking. Long-distance

orientation and striking is visually mediated, but strike precision relies on lateral line cues

and an increase in misses occurs when this system is blocked. In the absence of vision,

short-range orientation and striking can occur using lateral line cues. Capture is mediated by electroreception or tactile cues. All three species rely on vision or the lateral line for navigation and obstacle avoidance; an increase in collisions with obstacles occurs when both senses are blocked simultaneously.

Introduction

As a fitness-related behavior (reviewed in Arnold, 1983), feeding is arguably one of the most important functions of sensory systems and, therefore, sensory performance is of the utmost importance. Searching for food involves: 1) initially detecting one or more cues that alert the animal to the presence of a food item somewhere in its environment, 2) tracking these cues to the vicinity of the source, 3) orienting to the food item, 4) striking

at the food item, and finally moving the jaws and 5) capturing the food (Curio, 1976).

How animals perform these behaviors depends not only on what sensory stimuli are

available, as determined by the animal’s proximity to the prey, the physics of the stimulus

fields, and the thresholds of detection for each species, but also on which stimulus or

stimuli the animal chooses to focus upon when information from multiple senses is

available simultaneously. As an animal approaches a biological target, the sensory

environment becomes increasing complex and the animal might either integrate new

90 information encountered, if the sensory cues play complementary roles for a particular task, or they may demonstrate sensory switching, changing their focus to the cue that provides the most salient information for a particular behavior. One particular sensory modality may be required for a given task, or two or more cues may provide sufficient information, therefore playing alternative roles for that behavior. Several studies have examined the role of individual senses, or of a few senses for one phase of a behavior, but no study to date has examined multisensory integration across a complete behavioral sequence.

A prey item might emit an odor, create a hydrodynamic disturbance, such as from gill movements or swimming, and/or produce a weak electrical field (Kalmijn, 1972).

The threshold for detection of electrical fields by electrosensitive animals is in the range of 1nV/cm – 0.1μV/cm (Murray, 1962; Kalmijn, 1982; Johnson et al., 1984; Haine et al.,

2001; Kajiura and Holland, 2002b; Kajiura, 2003; Jordan et al., 2009b; Kajiura and

Fitzgerald, 2009). For fields produced by aquatic animals, this translates to a distance of less than a meter from the source (Kalmijn, 1972). The detection limits of the visual system of most aquatic animals are not well known and depend on the amount of available light, the amount of scatter (Duntley, 1963; Mazur and Beauchamp, 2003), and the background contrast in intensity, polarization and pattern of reflected light (Johnsen and Sosik, 2004; Johnsen, 2005), but rarely exceed tens of meters. Sound can be divided into the acoustic near field (primarily particle motion, detected by the otoliths or otoconia as particle acceleration (Kalmijn, 1988; Schellart and Popper, 1992)) and the far field

(primarily pressure, radiated by the swim bladder to the inner ear (Popper and Fay, 1999).

For a dipole sound source, the near field dominates at a distance from the source less than

one sixth of the wavelength (λ/2π); for a sound of 100Hz, a frequency in the hearing

range of many fishes, this translates to a distance of approximately 2.5 m (Kalmijn,

1988). The acoustic regime is, however, frequency-dependent: low frequency sounds

extend over a greater range than high frequency sounds, and have a near/far field

boundary at a greater distance from the source. As they lack a swim bladder,

elasmobranchs should be most sensitive to sounds in the near field. The maximum range

of detection of the lateral line has been shown to be approximately one half to two body

lengths (Denton and Gray, 1983; Denton and Gray, 1988; Coombs, 1999; Braun and

Coombs, 2000; Palmer et al., 2005; Jordan et al., 2009a).

Odor, on the other hand, may be carried a great distance from the source by the mean flow. In flowing water, odors are dispersed by two mechanisms: advection and turbulent mixing (reviewed in Webster, 2007). Advection refers to the transportation of a filament or patch of odor by the mean flow. Turbulent flow generates swirling packets, or eddies, that break up into a series of successively smaller eddies through a process known as the Kolmogorov cascade (reviewed in Weissburg, 2000). The hydrodynamic motion of these eddies can be detected by the lateral line system. Intermolecular viscous

forces dissipate the energy until these eddies reach the smallest size that still contains turbulent energy, known as the Kolmogorov length scale, on the order of millimeters.

Beyond this scale, in the odor far field, only very patchy odor information is available, carried by the mean flow (for a summary, see Figure 3.1). Sharks can orient to these patches, based on differences in the timing of arrival of the odor at the two nares. By turning toward the side receiving the first cue, the animals turn into an oncoming odor patch (Gardiner and Atema, 2010, Chapter 2).

Once an animal has oriented to the patch, it must determine the direction from

which the odor patches are emanating, and then track these patches to their source.

Sharks typically approach their prey from downstream with tight circles and figure-8

patterns (Tester, 1963a, b; Gardiner and Atema, 2007), covering a greater area in the presence of food odors than when these odors are absent (Kleerekoper, 1978, 1982).

Early studies suggested that sharks locate their food by olfaction (Bateson, 1890;

Sheldon, 1909, 1911; Parker and Sheldon, 1913; Parker, 1914, 1922), as animals could locate food sources without visual cues, but animals with their nostrils blocked show no interest in prey (Sheldon, 1911). Blocking one naris causes turning behavior to the intact side, suggesting that these animals use chemotropotaxis, or orientation based on

concentration comparisons between the two nares. Hobson (1963), however, questioned

the involvement of other senses in these odor source localization behaviors, noting that

sharks can readily follow an olfactory trail in running water.

Later experiments suggested two different tracking strategies. When presented

with an attractive odor stimulus, the lemon shark, Negaprion brevirostris, a more

midwater-swimming species, will swim upstream into the strongest current, regardless of

where the odor source was actually located. This behavior continued when the odor was

introduced into an area of sluggish water movement (Hodgson and Mathewson, 1971;

Mathewson and Hodgson, 1972), suggesting that this species’ response to an odor

stimulus is dominated by rheotaxis, or orientation to the mean current, as this would

presumably bring them into proximity of the odor source where then other senses, such as

vision or electroreception, would allow the animal to precisely pinpoint the source. This

93 behavior has been referred to as odor-stimulated rheotaxis (Hodgson and Mathewson,

1971; Mathewson and Hodgson, 1972).

In contrast, the nurse shark, Ginglymostoma cirratum, a benthic species, would begin moving up the odor corridor and was always able to localize the source (Hodgson and Mathewson, 1971; Mathewson and Hodgson, 1972), suggesting that for this species, a chemical stimulus triggered true concentration gradient searching, which was later termed klinotaxis (Fraenkel and Gunn, 1940). A subsequent study (Kleerekoper et al.,

1975) found that nurse sharks were unable to precisely locate an odor source in stagnant water, suggesting that localization is dependent in part on gradient characteristics but that flowing water provides a directional vector which enables the animal to precisely locate the source.

Animals tracking an odor plume were previously thought to perform bilateral comparisons between the two sensors (e.g. nares in fishes or antennules in invertebrates) moving up the concentration gradient to its source, a process termed klinotaxis (Fraenkel and Gunn, 1940). However, within the eddies generated in turbulent flow, turbulent stirring by the random velocity fluctuations results in very steep gradients of concentration that are irregular and unpredictable both in space and in time (Moore and

Atema, 1991). Recent studies employing fluorescent dyes to study the nature of a turbulent chemical plume (Webster et al., 2001; Webster and Weissburg, 2001;

Weissburg et al., 2002; Webster et al., 2003) have visualized the highly dynamic and spatially and temporally intermittent patterns of these plumes. Thus, the time-averaged concentration of these peaks converges too slowly to be of use to an animal tracking a prey item, that is, an animal would have to sample this intermittent concentration field

over a very long period to determine the average concentration. This process would

require several minutes whereas these animals have been demonstrated to be capable of

locating odor sources in tens of seconds (Moore and Atema, 1991; Moore et al., 1991;

Webster and Weissburg, 2001; Weissburg et al., 2002). It has been proposed that

simultaneously tracking the odor and the hydrodynamic wake of eddies could lead an animal to the source, through eddy chemotaxis (Atema, 1996; Gardiner and Atema,

2007). Interneurons in the crayfish brain have been demonstrated to respond to signals

from both the olfactory and mechanosensory systems (Mellon, 2005). These cells

respond to simultaneously presented odor and flow with spike rates that are twice the

summed response to either signal presented alone. Recent work (Gardiner and Atema,

2007) has demonstrated that the smooth dogfish, Mustelus canis, requires lateral line

information to locate odor sources. This species can navigate upstream through an odor

field to the general area of a turbulent odor source using either vision or the lateral line,

but the lateral line is necessary to precisely locate the source of coincident odor and flow.

This suggests that these animals are tracking the fine-scale structure of the plume, i.e., a

turbulent wake, flavored with food/prey odor, shed either by a moving prey item in still

water or a still piece of food in flowing water (eddy chemotaxis, Atema, 1996).

As an animal approaches a biological target, the sensory environment becomes

increasing complex, allowing the animal to use a number of different senses to pinpoint

prey and orient to it in preparation for the final strike and capture. Chemical, hydrodynamic, acoustic, visual, and electrical cues are all available at close range to the prey. Multisensory integration in the brain can allow an animal to determine the most salient information for a given task, which has been suggested to occur by weighting the

95 various cues based on the amount of information they provide for a given event

2009b; Kajiura and Fitzgerald, 2009). Multimodal integration can also improve the detection of weak signals. For example, the topographic maps of the visual and electrosensory systems overlap in the tectum of the little skate, L. erinacea, therefore electrosensory and visual stimuli emanating from the same point in space converge on the same neurons (Bodznick, 1990). This may aid in localizing buried prey in low ambient light as the convergence of signals results in a summed response the is greater than the additive responses to the two signals alone (reviewed in Stein and Stanford, 2008). Such convergence of signals may be required for a particular task, with the two sensory modalities playing complementary roles for a given behavior (e.g., using olfaction and the lateral line for eddy chemotaxis, above). In other situations, one or more senses may provide sufficient information for a task and thus play alternative roles for a given behavior. For example, navigating large scale (mean) flow, such as a current, can be accomplished in M. canis by using either cues from the lateral line system (hydrodynamic flow field) or vision (visual flow field) (Gardiner and Atema, 2007). For a given task, therefore, one sense may be absolutely required, or the senses may have complementary or alternative roles.

The goals of this study were to: 1) examine the integration of information from

the olfactory, mechanoreceptive, visual, and electroreceptive senses at each stage of the

feeding sequence; 2) investigate sensory switching; and to 3) elucidate the

complementary and alternative roles of the senses in each phase of feeding behavior.

Examining three shark species from different habitats, with different feeding strategies

for different prey types, allows for comparisons of the use of various senses in disparate environments.

Materials and Methods

Experimental animals

Eighteen young-of-the-year (YOY) blacktip sharks, Carcharhinus limbatus

(Müller and Henle 1839), 51-65cm total length, were collected from Terra Ceia Bay, FL

using rod and reel and gillnet gear. Ten juvenile nurse sharks, Ginglymostoma cirratum

(Bonnaterre 1788), 67-94cm total length, were collected from the waters near Long Key,

FL using rod and reel gear. Sixteen bonnetheads, Sphyrna tiburo (Linnaeus 1758), 69-

95cm total length, were collected from Terra Ceia Bay, FL, and the waters near Sarasota,

FL using gillnet gear. All animals were transported to Mote Marine Laboratory in

Sarasota, FL in a transport container equipped with a life support system. At the

laboratory, the sharks were held in a 210,000L tank operated on a closed recirculating life

support system with sand filtration and heater/chiller units. The tank was maintained at

spring/summer conditions of daylight (12-14 hours) and temperature (24-26°C). Animals

were fed fish, shrimp, and squid, supplemented with Mazuri Vita-Zu Sharks/Rays

Vitamin Supplement Tablets (PMI Nutrition International, St Louis, MO, USA), to satiation three times per week, except during periods of experimentation, when food was withheld for 48 hours prior to any behavioral trials to ensure that the animals were hungry. Protocols for animal handling and use were approved by the Institutional Animal

Care and Use Committees at the University of South Florida (W3817) and Mote Marine

Laboratory (11-03-RH1).

Behavioral procedures

Experiments were conducted in a near-laminar flow channel (flume) which was constructed within the 210,000 liter oval tank: working area (test arena) was 7.5 m long x

2 m wide, filled to 120 cm depth, with a flow rate of 2.3 cm/s (Figure 3.2). A 4m x 2m holding area at the downstream ends, allowed for an animal containment area behind a mesh gate. As per Gardiner and Atema (2007), for each trial, an individual animal was moved into the flume channel and allowed to acclimate for 30 minutes, then offered a small piece of food to confirm that it was hungry. The animal was then herded into the holding pen. A live prey item from the natural diet of each species (nurse and blacktip sharks: pinfish, Lagodon rhomboides (Bigelow and Schroeder, 1948; Castro, 1996,

2000); bonnethead: pink shrimp, Farfantepenaeus duorarum (Cortés et al., 1996)), was tethered, injured from tether attachment, at the upstream end of the flume using a piece of thin, degradable, cotton thread, in order to restrict it to the area in front of a window in the side of the tank (Figure 3.2). Within each species, the prey items were size-matched to the total length of the predator and prey size was consistent across trials. Prey were suspended midwater (approximately 60cm above the bottom) for the blacktip sharks,

98 approximately 30cm above the bottom for the bonnetheads, and just above the bottom

(approximately 10cm for the nurse sharks. The shark was held in the start box for six minutes to allow a plume of sensory cues emanating from the prey to establish along the length of the flume channel. The shark was then released and a trial proceeded for 10 minutes or until the prey was consumed, during which time the shark’s behavior was simultaneously monitored and filmed from above using a series of three overhead cameras (Sony 1/3 inch CCD). A lateral view of any strikes or bites on the prey was recorded using a fourth camera placed in front of the previously mentioned window in the tank wall (Figure 3.2). The images from these cameras were combined using a multiplexer (Nuvico EV-8250N, Englewood, NJ, USA) and saved digitally via a computer. Animals were examined intact, and following blocks of each of the sensory systems (outlined below), alone and in combination.

Sensory deprivation

The olfactory system was blocked by inserting pieces of cotton soaked in petroleum jelly into the animal’s nares. The hydrophobic nature of the petroleum jelly prevents water, and therefore odor, from reaching the olfactory organs (Atema, 1971;

Basil et al., 2000). To block vision, small pieces of heavy black plastic were glued around the margins of the eyes using cyanoacrylate glue. The sensitivity of the electrosensory system was reduced by painting over the pores of the ampullae of Lorenzini with silicone-rubber paint (bonnetheads; Smooth-On Mold Max Stroke, Smooth-On Inc.,

Easton, PA USA) or with cyanoacrylate glue (blacktip and nurse sharks; The Original

Super Glue, Super Glue Corp., Rancho Cucamonga, CA USA). The location of the pores

99 in all three species has previously been mapped (Kajiura, 2001; Cornett, 2006). Though only the pores were painted over, dripping of the glue or stretching of the silicone-rubber paint may have resulted in occlusion of some of the lateral line canal pores and/or superficial neuromasts on the head. Prior to use on animals, the insulating nature of these two materials was verified by covering one electrode on the prey-simulating electrical stimulus apparatus described in Kajiura and Holland (2002b). The pair of electrodes was then immersed in seawater and a current of up to 200mA was applied; no current was detected at the multimeter, indicating that the paint and glue break the electrical circuit and are therefore insulating. All of these blocks were applied while the animal was under anesthesia with tricaine methanesulfonate (MS-222), with a dose of 100mg/L in buffered seawater for induction and 50mg/L for maintenance. Animals were ventilated using a hose attached to a small recirculating pump while the blocks were applied, then revived using fresh seawater. Animals were allowed to recover in approximately 1000L of seawater in a 244cm diameter round tank for three hours, then moved to the flume channel and allowed to acclimatize for a further 30 minutes as above, prior to a behavioral trial.

The lateral line system was lesioned by holding the animals in a 0.5g/L solution of streptomycin sulfate in seawater for three hours (Montgomery et al, 1997; Faucher et al, 2006; Gardiner and Atema, 2007). An individual animal was held in approximately

1000L of this solution in the 244cm diameter round tank. For combinations of sensory blocks, those requiring anesthesia were first applied, then the animal was moved to the recovery round tank where it was held until it had recovered sufficiently to swim and navigate the tank normally. The streptomycin sulfate was added to the water and the

100 animal was held in this solution for three hours as described above, prior to being moved to the flume channel for behavioral testing. Streptomycin is an ototoxic antibiotic that has been shown to lesion both the surface neuromasts and canal neuromasts in teleosts

(Wersall and Flock, 1964; Kaus, 1987; Montgomery et al., 1997; Faucher et al., 2006a;

Faucher et al., 2006b; Faucher et al., 2010). In amphibians, treatment with this drug results in an increase in spontaneous firing of the afferent nerves, which is linked to direct effects on the membrane of the hair cell, and a large lag phase in the receptor potentials, which may be caused by interference with the motion of the sensory hairs (Kroese and van den Bercken, 1982). It does not affect inner ear function unless applied intraluminally (Matsuura et al., 1971). The duration of the effects of this drug are not completely understood. Since teleosts treated with this drug return to normal behavior in

20–24 hours (Blaxter and Fuiman, 1989), all lateral line blocked trials were completed within six hours of application of the drug. However, since physical damage to the hair cells has been found on scanning electron micrographs of streptomycin-treated neuromasts (Faucher et al., 2006a; Faucher et al., 2006b; Faucher et al., 2010), following lateral line lesion treatments, animals were allowed to recover for a minimum of four weeks prior to any other behavioral testing to allow time for the neuromasts to regenerate

(Coombs et al, 2001; Faucher et al, 2006). While hearing with the inner ear is likely contributing to prey localization in sharks, the stimulus field in a closed tank environment is very difficult to control due to echoing off the walls, bottom and surface of the tank, and thus it was not specifically examined in this study.

101

Video analysis

Videos were digitized using MaxTRAQ Lite+ v.2.2.2.2 software (Innovision

Systems Inc., Columbiaville, MI, USA). As described above, the behavior of these

animals can be divided into five phases: detection, tracking, orientation, striking, and

capture. Detection is defined as the onset time of the first of the behaviors below to be

initiated (i.e., the onset of tracking, or in the absence of tracking, the onset of

orientation/striking). Tracking in other shark species typically begins with a rapid turn

and descent towards the bottom (Tester, 1963b). The animals then approach their prey

from downstream with tight circles and figure-8 patterns (Parker, 1914; Hobson, 1963;

Tester, 1963a, b; Gardiner and Atema, 2007). Orientation, a turn to align the body or head for the strike, is immediately followed by striking. Striking in ram-feeding bony fish is defined as a direct, rapid whole-body acceleration towards the prey, often using an

S-start (Nyberg, 1971; New and Kang, 2000; New, 2002). In ambush predators, such as the Pacific angel shark, Squatina californica, striking begins in a relatively stationary animal with the raising of the head and a short lunge (Fouts and Nelson, 1999). In fishes, capture begins with the onset of jaw depression (Ferry-Graham and Lauder, 2001; Motta and Wilga, 2001; Motta, 2004) and for comparative purposes, in this study was defined as ending when the center of mass of the prey had passed the anterior margin of the

mouth. Thus, in unmanipulated animals, detection is typically indicated by the start of

tracking behavior, which ends with orientation (a turn) and a directed strike, which

culminates in capture. The onset time of each of these behaviors was noted. For the

tracking phase, the following variables were examined: (1) swim velocity, in body

102

lengths/s; (2) turn velocity, in °/s; (3) frequency of turns, in turns/s; and (4) tracking time,

in s, from the start of tracking to the first strike. To account for differences in the

distance at which tracking began, tracking time was standardized by dividing it by

predator-prey distance at the onset of tracking, expressed as a proportion of the total

length of the test arena, i.e., timestd. = time/(distance to prey/test arena length). For the

orientation, striking, and capture phases, (5) orientation distance, predator-prey distance

at orientation, in cm; (6) strike rate, percentage of trials in which strikes occurred; (7)

strike angle, in °, the angle between the midline of the predator and center of mass of the

prey, with respect to the flow (Figure 3.3); (8) strike velocity, in body lengths/s; (9)

number of misses; and (10) capture success rate, percentage of trials resulting in

successful capture, were examined. Over the entire trial, (11) the frequency of wall

collisions was calculated, in collisions/s. For the nurse sharks only, (12) rest time, in s,

the proportion of the trial spent resting on bottom, as well as (13) resting angle, in °, with

respect to the direction of flow, were also measured.

Statistical analysis

Data for each species were regressed against total length using the least squares method to remove the effects of size (Packard and Boardman, 1999) and the standardized residuals were used in all subsequent analyses. All data were tested for normality and equality of variance with Kolmogorov-Smirnov tests and Levene Median tests (Sokal and

103

taken from the wild, individual animals were used in more than one, but not in all,

treatment groups. This design precludes the use of either repeated measures or non-

parametric tests, as there are missing cells for each individual animal. Data for the

different treatments were therefore compared within each species using one-way Analysis

of Variance (ANOVA). When significant differences were found, Tukey post hoc tests were then used to perform pairwise comparisons of the treatments. Data that failed the normality and/or equality of variance tests were analyzed using non-parametric Kruskal-

Wallis one-way ANOVA on ranks and pairwise comparisons were made using Dunn’s

Method when significant differences were found. The Benjamini-Hochberg method was used to control the false discovery rate in multiple statistical tests (Benjamini and

Hochberg, 1995). All analyses were conducted using SigmaPlot 12.0 (Systat Software

Inc., San Jose, CA, USA).

Results

Tracking

In all three species, tracking (rapid, frequent turning) occurred in control, vision

blocked, lateral line blocked, and electrosensory blocked animals, while blocking olfaction, vision + olfaction, or lateral line + vision + olfaction abolished tracking behavior (slow, infrequent turns; more resting in the nurse shark). Blocking the lateral line + vision did not impact tracking behavior in the nurse shark, while in the blacktip shark and the bonnethead it resulted in excitement (fast turns) but an inability to track

(infrequent turns).

104

Blacktip sharks with all of their senses intact typically initiated tracking shortly after exiting the holding pen. During tracking, these animals swam at a mean velocity of

1.19 ± 0.03 body lengths/s (BL/s) and executed turns rapidly (140.2 ± 5.1°/s) and frequently (0.72 ± 0.04 turns/s). Blocking the visual, lateral line, or electrosensory systems individually did not significantly affect the swimming velocity or turn velocity

(Table 3.1, Figures 3.4-3.6), but turns occur significantly more frequently when the visual

(0.88 ± 0.04 turns/s) or electrosensory systems (1.08 ± 0.01 turns/s) were blocked.

Following blocks of the olfactory system, visual + olfactory systems, lateral line + olfactory systems, or lateral line + visual systems, tracking behaviors were absent, as indicated by the lower frequency (0.16 ± 0.03 turns/s, 0.20 ± 0.01 turns/s, 0.23 ± 0.03 turns/s, 0.16 ± 0.01 turns/s, respectively) and slower velocity (54.3 ± 2.3°/s, 52.9 ± 4.3°/s,

47.8 ± 7.2°/s, 48.3 ± 3.4°/s, respectively) of turns (Table 3.1, Figures 3.5, 3.6). When the lateral line and vision were blocked simultaneously, the animals executed rapid turns

(110.8 ± 18.9°/s), not significantly different from the treatments in which the animals displayed tracking behavior, but turned significantly less frequently (0.36 ± 0.03 turn/s), similar to the animals that were not displaying tracking behavior (Table 3.1, Figures 3.4-

3.6). The swimming velocities did not vary among any of the treatments. Among the treatments where tracking behavior occurred, tracking time was compared. The animals with vision blocked took significantly longer, as compared to each of the other treatments

(control, lateral line block, and electrosensory block), to locate the prey. Tracking time did not vary significantly among the other treatments (Table 3.1, Figure 3.7).

Nurse sharks with all of their senses intact initiated tracking shortly after exiting the holding pen. During tracking, they swam at a mean velocity of 0.44 ± 0.04 BL/s and

105 turned rapidly (83.6 ± 4.8°/s) and frequently (0.52 ± 0.03 turn/s). Blocking the visual, electrosensory or lateral line + visual systems did not cause any significant changes in tracking behavior. Following blocks of the olfactory system, visual + olfactory systems, or lateral line + visual + olfactory systems resulted in a lack of tracking behavior, as indicated by the significantly slower (47.8 ± 2.9°/s, 40.0 ± 2.6°/s, 45.2 ± 7.9°/s, 47.5 ±

3.1°/s, respectively) and less frequent (0.11 ± 0.01 turns/s, 0.15 ± 0.02 turns/s, 0.11 ±

0.02 turns/s, 0.17 ± 0.03 turns/s, respectively) turns, which were not significantly different from one another. The swimming velocity was not significantly affected by any of the treatments. (Table 3.2; Figures 3.4-3.6). Blocking the lateral line + olfactory systems resulted in a similar lack of tracking behavior (0.11 ± 0.02 turns/s at

45.2±7.9°/s), but a significantly faster swimming velocity (0.73 ± 0.08 BL/s). Blocking the lateral line system alone also resulted in a significantly faster swimming velocity

(0.76 ± 0.07 BL/s), but turning behavior at an intermediate frequency (0.31 ± 0.04 turns/s) and velocity (58.0 ± 5.6°/s). The frequency and velocity of the turns were significantly lower than that of the control, electrosensory block or visual block treatments, but significantly greater than that of the olfactory, olfactory + vision, lateral line + olfactory, and lateral line + olfactory + vision block treatments. The animals with only the lateral line system blocked took a significantly shorter time to locate the prey

(17.6 ± 4.5s) as compared to the other treatments. Tracking time was longest in the lateral line + vision blocked animals (189.8 ± 79.1s), but this value was only significantly different from the tracking time of the electrosensory blocked animals. The tracking times for control, vision block, and electrosensory block treatments were not significantly different from one another (Table 3.2, Figure 3.7). The proportion of the trial time that

106

the nurse sharks spent resting on the bottom varied among the treatments. Blocking the

olfactory, visual + olfactory, or lateral line + visual + olfactory systems resulted in

significant increases in the proportion of the trial time spent resting on bottom (50.9 ±

12.2%, 75.4 ± 37.2%, 41.1 ± 15.5%, respectively). Little to no resting was observed

among the rest of the treatments, which were not significantly different from one another

(Table 3.2, Figure 3.8).

Bonnetheads also initiated tracking shortly after exiting the holding pen when all

of their senses were intact, swimming at 0.69 ± 0.02 BL/s and executing frequent (0.79 ±

0.01 turn/s), rapid (130.6 ± 1.7°/s) turns (Table 3.3, Figures 3.4-3.6). Blocking the visual or electrosensory systems did not significantly change this behavior. Following blocks of the olfactory or vision + olfactory systems, the animals no longer displayed tracking behavior, as indicated by the significantly slower (47.2 ± 4.9 and 38.7 ± 0.0°/s, respectively) and less frequent (0.13 ± 0.01 and 0.1 ± 0.0 turns/s) turns. Blocking the lateral line resulted in turns that were significantly slower (98.7 ± 3.6°/s), compared to the control and electrosensory blocked animals, but significantly faster compared to the olfactory and vision + olfactory blocked animals, at a frequency that was not significantly different from the control or electrosensory blocked animals (0.69±0.0 turns/s). Blocking the lateral line and vision in combination resulted in a similar decrease in turn velocity

(100.3 ± 13.0°/s, significantly slower compared to the control and electrosensory blocked animals, but significantly faster compared to the olfactory and vision + olfactory blocked animals), which was not significantly different from the lateral line blocked animals, but the frequency of turns was also affected. These lateral line + vision blocked animals displayed an intermediate frequency of turns (0.47±0.07 turns/s), significantly lower than

107

the control or electrosensory blocked animals, but significantly greater than the olfactory

or vision + olfactory blocked animals. They also swam at a velocity that was

significantly slower as compared to the olfactory, lateral line, and electrosensory block

treatments (Table 3.3, Figures 3.5, 3.6). There was no difference in swimming velocity

among the other treatments (Table 3.3, Figure 3.4). Among the treatments where

tracking behavior occurred, the tracking time (from start of tracking to the first strike, standardized for predator-prey distance at the start of tracking) was not significantly different among any of the treatments (Table 3.3, Figure 3.7).

Orientation and Striking

In the blacktip shark, rapid long-distance strikes occurred in 100% of the control, olfaction blocked, lateral line blocked, lateral line + nose blocked, and electrosensory blocked trials. When vision was blocked, 60% of trials resulted in strikes, which were slower and initiated from a closer proximity and from a greater variation in angles. No strikes occurred in the vision + olfaction blocked, lateral line + vision blocked, or lateral line + vision + olfaction blocked animals. In the nurse shark, orientation to the prey and striking occurred at a close proximity to the prey in all control, vision blocked, lateral line blocked, lateral line + vision blocked and electrosensory blocked animals. Orientation to the prey occurred in some trials in the olfaction blocked, vision + olfaction blocked, lateral line + olfaction blocked and lateral line + vision + olfaction blocked animals, but of these, only the lateral line + olfaction blocked trials occasionally resulted in strikes. In the bonnethead, orientation and striking occurred in all of the control, olfaction blocked,

108 lateral line blocked, and electrosensory blocked animals. No orientation or striking occurred in the vision blocked or lateral line + vision blocked animals.

In the control blacktip sharks, tracking was followed in every trial by orientation to the prey, which occurred at a distance (238.9 ± 16.3cm), and was immediately followed by a rapid strike (1.96 ± 0.22 BL/s), driven by a whole-body acceleration, which culminated in ram prey capture (Table 3.1, Figures 3.9-3.11). Blocking the lateral line or the electrosensory system did not significantly affect these behaviors. When the olfactory system was blocked, alone or in combination with the lateral line, tracking was absent (as described above), but when the blacktip sharks’ swimming as they cruised the tank brought them to a closer proximity to the prey, orientation occurred in every trial, followed by striking (Table 3.1, Figure 3.10). These animals oriented at distances from the prey that were not significantly different from the control animals (253.7 ± 2.5cm and

177.5 ± 27.5cm, respectively), but that were significantly different from one another.

Blacktip sharks with the visual system blocked used tracking behavior until they were in a significantly closer proximity to the prey, as compared to all of the other treatments, and only oriented and struck from close range (17.4 ± 2.5cm; Table 3.1, Figure 3.9).

Orientation and striking in the blinded animals occurred in only 60% of the trials, the remainder of the animals continued tracking behaviors until the end of the trial (Table

3.1, Figure 3.10). The strikes that occurred were at a slower velocity as compared to the control animals (Table 3.1, Figure 3.11). Strikes also occurred at a greater angle to the prey in the vision blocked animals as compared to the control, olfactory blocked, lateral line blocked, and lateral line + nose blocked animals (Table 3.1), meaning that the animals were not striking from more or less directly in front of the prey, but rather

109

striking from the side or behind. Strike velocities and strike angles were not significantly different among the other treatments above (Table 3.1, Figure 3.11). No strikes occurred when the olfactory and visual systems were blocked simultaneously, either alone or in combination with the lateral line system, or when the lateral line and visual systems were blocked simultaneously (Table 3.1, Figure 3.10).

Nurse sharks with all of their senses intact displayed tracking behavior until they were in close proximity to the prey (12.7 ± 1.4cm), at which point they oriented to the prey by lifting the head, followed immediately by striking and suction capture in 100% of the trials (Table 3.2, Figures 3.9, 3.10). Similarly, orientation and striking occurred in

100% of the trials with the visual, lateral line, lateral line + visual, and electrosensory systems blocked. Following blocks of the olfactory, visual + olfactory, lateral line + olfactory, or lateral line + visual + olfactory systems, the animals only oriented to the prey in 20.2 ± 10.6%, 8.3 ± 8.3%, 46.7 ± 0.0% and 20.0 ± 12.2% of trials, respectively, significantly less than the other treatments (Table 3.2). Among these latter treatments,

striking occurred only in the lateral line + olfactory blocked animals, and only in 36.7 ±

0.10% of the trials, which was significantly different from all of the other treatments

(Table 3.2, Figure 3.10). No striking occurred in any of the animals in the olfactory, visual + olfactory, or lateral line + visual + olfactory blocked trials. In these treatments groups, when orientation occurred, it was followed by the animals lowering their heads and swimming away. The distance from the prey at which the nurse sharks oriented was not significantly different among any of the treatment groups. Due to the proximity to the prey from which the strikes occurred and the resolution of the video (30 frames s-1), strike velocity was not evaluated in this species.

110

In the control bonnetheads, tracking behavior was followed in every trial by orientation from a distance (73.8 ± 10.7cm), followed immediately by striking and capture. Strike velocity (0.75 ± 0.03BL/s) was similar to swim velocity (0.69 ±

0.02BL/s) in this species. Blocking the lateral line or electrosensory systems did not significantly affect these behaviors (Table 3.3, Figures 3.9-3.11). When the visual system was blocked, the animals were unable to locate the prey to orient and no strikes occurred (Table 3.3, Figure 3.9), instead the animals continued to perform tracking behavior until the end of the trial. When the olfactory system was blocked, tracking was absent (as described above), but the bonnetheads’ swimming as they cruised the tank brought them in proximity to the prey, orientation occurred in every trial, followed by striking (Table 3.3, Figure 3.10). These animals oriented at greater distances from the prey (165.7 ± 31.8cm), as compared to the control animals (Table 3.3, Figure 3.10). In the vision + olfactory and lateral line + vision block treatments, no strikes occurred

(Table 3.3, Figure 3.9). Strike velocity and strike angle did not vary significantly among any of the treatments (Table 3.3, Figure 3.11).

Capture

In the blacktip shark, all control, olfaction blocked, lateral line blocked, lateral line + olfaction blocked, and electrosensory blocked animals successfully captured the prey, but in the lateral line + olfaction and electrosensory blocked animals, these successes often occurred after misses. When the lateral line was blocked (alone or with olfaction), high velocity strikes resulted in a miss while slow velocity strikes resulted in capture. When the electrosensory system was blocked, successful capture occurred if

111

physical contact was made with the prey prior to opening the jaws. In the nurse shark, all

control, vision blocked, lateral line blocked, and lateral line + vision blocked, and

electrosensory blocked animals successfully captured prey; however, similar to the

blacktip shark, misses occurred if the electrosensory blocked animals did not make

physical contact with the prey prior to opening the mouth. In the bonnethead, successful

capture occurred in all control, olfaction blocked, and lateral line blocked animals. No

successful captures occurred in the electrosensory blocked animals, which did not move

their jaws and instead ran into the prey with the mouth closed.

In control and olfactory blocked blacktip sharks, each trial culminated in a single strike, of which 100% were successful. In the remainder of the treatments where strikes occurred (vision, lateral line, electrosensory, and lateral line + olfactory blocks), multiple strikes were observed, and capture only occurred after one or more misses. The number of strikes, and therefore misses, was significantly higher in the electrosensory blocked

(1.17 ± 0.58 misses/trial) and lateral line + olfactory blocked animals (0.58 ± 0.23 strikes/trial). The changes in the number of misses in the lateral line blocked and vision blocked animals were not significant (Table 3.1, Figure 3.12). The strike velocities for successful capture were compared to those for misses using paired t-tests. In the lateral line blocked animals (lateral line alone or lateral line + olfaction blocked), the velocities of these strikes were significantly different, such that high velocity strikes resulted in misses while slow velocity strikes resulted in successful capture. There were no significant differences in strike velocity between successful captures and misses in the electrosensory blocked animals (Table 3.4, Figure 3.13). However in the electrosensory blocked animals, in many of the sequences the blacktip shark made physical contact with

112 the prey prior to moving the jaws. A significant positive correlation was found between a successful strike and the blacktip shark making pre-jaw movement contact with the prey

(Pearson Product Correlation, p = 0.00002). Ultimately, all of the treatments in which strikes occurred culminated in successful prey capture, with the exception of the vision block, in which only 50% of the animals successfully captured the prey (Table 3.1,

Figure 3.14).

In the control and lateral line + vision blocked nurse sharks, every trial culminated in a single strike, 100% of which resulted in a successful capture. In all of the other treatments, a proportion of the strikes resulted in a miss, however, these changes were not significant (Table 3.2, Figure 3.12). In the electrosensory blocked nurse sharks, similar to the blacktip sharks, it was noted that successful strikes were preceded by the nurse shark making physical contact, with either the head or the barbels, with the prey prior to moving the jaws. A significant positive correlation was found between a successful strike and the nurse shark making pre-jaw movement contact with the prey (Pearson

Product Correlation, p = 0.00006). Ultimately, all of the treatments in which strikes occurred eventually resulted in successful prey capture, with the exception of the lateral line + olfactory block, in which only 45.8 ± 4.2% of the strikes that occurred resulted in capture, significantly fewer than in the other treatment groups (Table 3.2, Figure 3.14).

In the control, olfactory blocked, and lateral line blocked bonnetheads, trials typically culminated in a single strike and successful capture, with only a very small number of misses per trial (0.04 ± 0.04, 0.00 ± 0.00, and 0.25 ± 0.14 misses/trial respectively). Ultimately, 100% of these trials eventually resulted in successful capture

(Table 3.3, Figures 3.12, 3.14). These treatments were not significantly different from

113

one another. In the electrosensory blocked animals, a significantly greater number of misses occurred (35.4 ± 11.7 misses/trial) and none of these animals were successful in capturing prey (Table 3.3, Figures 3.12, 3.14). Although the strikes appeared to be well- directed, electrosensory blocked bonnetheads failed to move their jaws in every strike

and thus were unable to capture the prey.

Wall Collisions

In all three species, frequent wall collisions occurred in the lateral line + vision

blocked and lateral line + vision + olfaction blocked animals. In the bonnethead, but not

the blacktip or nurse sharks, an increase in wall collision also occurred when vision was blocked. Wall collisions in all other treatments were very rare. Blacktip sharks with all of their senses intact typically navigated the tank well and very rarely collided with the wall (0.001 ± 0.001 collisons/s). Blocking the olfactory, visual, visual + olfactory, lateral

line, lateral line + olfactory, or electrosensory systems did not result in significant

changes. When the lateral line + visual and lateral line + visual + olfactory systems were

blocked, a significant increase in the frequency of wall collisions was observed (0.018 ±

0.002 and 0.013 ± 0.001 collison/s, respectively). These treatments were not

significantly different from one another (Table 3.1; Figure 3.15). Nurse sharks with all

of their senses intact also very rarely collided with the walls (0.001 ± 0.001 collisions/s).

Blocking the olfactory, visual, visual + olfactory, lateral line, lateral line + olfactory, or

electrosensory systems did not result in any significant changes. Blocking the lateral line

+ visual and lateral line + visual + olfactory systems, however, resulted in a significant increase in the frequency of wall collisions (0.018 ± 0.003 and 0.040 ± 0.006 collisions/s,

114

respectively). These treatments were not significantly different from one another (Table

3.2; Figure 3.15). Bonnetheads with all of their senses also rarely collided with the

walls (0.000 ± 0.00 collisions/s). Blocking the olfactory, lateral line or electrosensory

systems did not result in any significant changes. Blocking the visual, visual + olfactory

or lateral line + visual systems resulted in a significant increase in the frequency of wall

collisions (0.052 ± 0.000, 0.027 ± 0.000, 0.049 ± 0.001 collisions/s, respectively) (Table

3.3, Figure 3.15).

Discussion

Since animals are guided to food by information from the senses, and feeding

performance is strongly correlated with fitness (Arnold, 1983), animals will likely rely on

whichever sense(s) provides them with the best performance for a particular phase of

feeding, in order to maximize the likelihood of success. The senses can therefore play

different roles in different phases of feeding behavior. As an animal approaches a prey

item, the sensory information becomes increasingly rich (Figure 3.1) and an animal might

either integrate the new information encountered in an additive fashion, with the senses

playing complementary roles for a given behavioral task, or they may demonstrate

sensory switching at the interfaces of the different phases of feeding behavior. For

example, muskellunge, Esox masquinongy, detect prey visually but switch to visual and lateral line cues to orient and strike (New et al., 2001). Animals may also face losses of sensory information, for example when cues become unavailable due to natural environmental changes or human disturbances. Vision, for example, can be impacted by

115

the natural variation in light levels from day to night and with cloud cover (McMahon

and Holanov, 1995), while pollution can mask odor cues (Fisher et al., 2006; Tierney et

al., 2010). The ability to perform a behavior using an alternative sensory modality allows

animals to be successful in a variety of environmental conditions. For example, alewife,

Alosa pseudoharangus, can switch senses to use the lateral line, rather than vision, to

strike at and capture prey at night (Janssen et al., 1995). In elasmobranchs, multisensory

integration and sensory switching are likely happening at the level of the central nervous

system. Though the exact location where this occurs is not yet known, areas of multimodal input have been identified in the dorsal pallium (Cohen et al., 1973;

Ebbesson, 1980) and medial pallium (Veselkin and Kovačević, 1973; Nikaronov and

Lukyanov, 1980; Nikaronov, 1983; Bodznick and Northcutt, 1984; Bodznick, 1991) of the telencephalon, as well as the tectum of the mesencephalon (Platt et al., 1974; Boord and Northcutt, 1982, 1988; Bodznick, 1991; Hoffman and Northcutt, 2008). Multimodal areas have also been identified in the optic tecta of other vertebrates, such as salamanders

(Bartels et al., 1990), snakes (Hartline et al., 1978) and owls (Knudsen, 1982), as well as the superior colliculus of mammals (Jassik-Gerschenfeld, 1965; Horn and Hill, 1966).

Most studies on sensory integration have focused on only a few of the senses, e.g., the integration of visual and lateral line information in feeding in fishes (Janssen and

Corcoran, 1993; Janssen et al., 1995; New and Kang, 2000; New et al., 2001; New,

2002), or they have examined the use of all of the senses, but only for one phase of a behavior, e.g., striking in snakes (Haverly and Kardong, 1996; Young and Morain, 2002).

This study is the first to examine multisensory integration across a complete behavioral sequence. Understanding which senses can be used for a particular task, their

116

complementary and alternative roles, as well as their limits, that is, which cues are crucial

for performing a particular behavior, allows for the establishment of a multimodal

sensory hierarchy for feeding behavior. This information is important for understanding

the success of this vertebrate lineage over the past 400 million years and assessing their

potential to survive in a changing environment, as well as for the development algorithms

for steering underwater robotic vehicles (Grasso and Atema, 2002).

Detection

The physics of the dispersal fields of the sensory cues that emanate from a prey

item in water dictate that odor is often the cue available farthest from the prey, as it can

be carried a great distance by the mean flow (Figure 1), and sharks are sensitive to amino acids in the picomolar range (Hodgson and Mathewson, 1978; Zeiske et al., 1986;

Meredith and Kajiura, 2010). Thus, odor is often the first cue detected by a shark that is

hunting for food (Parker, 1909; Sheldon, 1909, 1911; Parker and Sheldon, 1913; Parker,

1914; Gilbert, 1963). The blacktip sharks, bonnetheads, and nurse sharks in these

experiments were all initially alerted to the presence of food in the test arena by the

presence of an attractive odor, similar to experiments on lemon sharks, N. brevirostris

(Gilbert, 1963), which triggered the start of tracking behaviors in all three species. This

scenario corresponds to the natural situation of an animal approaching a food item from

downstream, as is typical for many shark species (Hobson, 1963; Hodgson and

Mathewson, 1971; Mathewson and Hodgson, 1972; Kleerekoper, 1978, 1982; Gardiner

and Atema, 2007). For blacktip sharks and bonnetheads, the odor cues are not absolutely

required. When the olfactory system is blocked, these animals can also detect their prey

117

visually and this visual detection triggers orientation and striking behaviors, similar to the

case in lemon sharks, N. brevirostris (Gilbert, 1963). This scenario could occur in the

natural environment if these animals come across a prey item from upstream, which

could happen by chance as these animals move through their environment, or if they are initially attracted by sound, as has been reported in field studies (Nelson and Gruber,

1963; Myrberg et al., 1969; Myrberg et al., 1972). Similarly, if a prey item is approaching a stationary predator, the prey may only be perceived visually, for example, if the predator is upstream of the prey making the chemical cues unavailable. Strikes are

prompted by visual cues in many lie-and-wait predators, such as the Pacific angel shark,

S. californica, (Fouts and Nelson, 1999), often through the perception of motion, rather than the perception of the presence of the object itself (Honigmann, 1944). Either a visual cue or an olfactory cue is required for the blacktip shark and bonnethead to

identify the prey as food, since animals with both senses blocked cease to feed, similar to

the lemon shark, N. brevirostris (Gilbert, 1963).

For the nurse shark, however, an attractive odor is required for these animals to

feed. Nurse sharks possess eyes with retinal areas of higher cone concentration, a

specialization for greater visual acuity (Hamasaki and Gruber, 1965; Hueter, 1991;

Collin, 1999). They are, therefore, likely seeing the prey, but the visual cues are generally not sufficient to prompt a strike, suggesting that they do not identify their prey visually, though vision is clearly important in this species for other behavioral tasks

(Aronson et al., 1967; Graeber et al., 1973; Graeber et al., 1978). This species has been described as being well-adapted as a stalking or ambush predator, often cornering fish in reef crevices at night (Hodgson and Mathewson, 1971; Mathewson and Hodgson, 1972;

118

Motta et al., 2008). Visual cues may be diminished on a dark night or even unavailable in the case of hidden prey, which may explain why chemical cues are more important than visual cues for feeding in this species. For the blacktip shark and bonnethead, olfactory and visual cues serve alternative roles in prey identification, whereas for the nurse shark, olfactory cues are essential for this task.

Tracking

Tracking odor plumes involves using a combination of olfactory and hydrodynamic cues. This can be accomplished using either the mean flow, through odor- stimulated rheotaxis, which has been described in sharks (Hodgson and Mathewson,

1971; Mathewson and Hodgson, 1972; Gardiner and Atema, 2007), bony fishes (Baker et al., 2002; Carton and Montgomery, 2003), birds (Nevitt et al., 2008), crustaceans

(Zimmer-Faust et al., 1995; Atema, 1996; Keller et al., 2003) and insects (Vickers and

Baker, 1994), or by tracking the fine-scale wake, through eddy chemotaxis, which has been described in sharks (Gardiner and Atema, 2007), bony fishes (Pohlmann et al.,

2001; Montgomery et al., 2002; Pohlmann et al., 2004), insects (Mafra-Neto and Cardé,

1994), and pinnipeds (Dehnhardt et al., 2001; Glaser et al., 2011). In fishes, the

hydrodynamic cues are detected either with the lateral line system (mean flow and fine-

scale turbulence) or vision (mean flow). The lateral line system detects the mean flow

(turbulence of a larger scale than the animal itself), allowing the animals to determine

upstream from downstream and perform rheotaxis behaviors (Montgomery et al., 1997;

Baker and Montgomery, 1999b, a; Montgomery et al., 2000; Peach, 2001; Gardiner and

Atema, 2007; Chagnaud et al., 2008). Rheotaxis can also be visually guided, as the

119

animals perceive and orient to visual features in the environment as they drift by (Lyon,

1904, 1909). Odor-stimulated rheotaxis can thus be either lateral line or visually

mediated (Baker et al., 2002; Gardiner and Atema, 2007). The lateral line alone detects

fine-scale wakes (turbulence smaller than the animal itself), such as the wakes left behind

by swimming prey (Hanke et al., 2000; Hanke and Bleckmann, 2004), so this system is

essential for wake tracking and eddy chemotaxis in fish (Pohlmann et al., 2001;

Pohlmann et al., 2004; Gardiner and Atema, 2007). Wake tracking in animals that lack a

lateral line, such as pinnipeds, is accomplished using mechanosensitive vibrissae

(Dehnhardt et al., 2001; Glaser et al., 2011; Miersch et al., 2011).

All three species examined in this study display similar tracking behavior, which

is characterized by frequent, rapid, turns, similar to other shark species (Parker, 1914;

Tester, 1963b; Gardiner and Atema, 2007). Similar to the smooth dogfish M. canis

(Gardiner and Atema, 2007), in the three species in this study, this behavior is mediated

by olfactory and hydrodynamic cues. In all three species, a lack of olfactory cues results

in slow, unexcited turns, which occur as the animal simply cruises the tank. The frequent

resting behavior of the nurse shark is further evidence of a lack of stimulation, as nurse

sharks typically rest, often on top of one another, when they are not actively hunting.

The typically position themselves with the head oriented against the direction of local

flow, awaiting sensory cues from prey (Gudger, 1921; Hodgson and Mathewson, 1971;

Mathewson and Hodgson, 1972; Castro, 2000). The presence of an appropriate olfactory

cue is therefore required for tracking behavior in all three species.

Vision is not required for tracking behavior, as blocking it does not alter this

behavior in any of these species. Disabling the lateral line system has slightly different

120 effects on each species. In the blacktip shark, disabling the lateral line alone does not alter tracking behavior, which suggests that vision can completely compensate for the loss of the lateral line information, and thus the two systems play alternative roles in this behavior. In the bonnethead, disabling the lateral line alone slows the turns slightly, though they are still faster and therefore more stimulated than when olfactory cues are absent. Turn frequency, however, is still very high and thus the loss of the lateral line only slightly impacts tracking behavior in this species. When both vision and the lateral line system are disabled simultaneously, however, tracking behavior is disrupted in both species. Both species still display excited turns, as indicated by the high velocity and medium velocity turns in the blacktip sharks and bonnetheads, respectively, but they turn less frequently. Without information on the direction of flow, obtained either from the lateral line, or through the visual flow field, these animals cannot navigate from one patch to the next to achieve coherent tracking behavior and cannot find the prey target, similar to the smooth dogfish, M. canis (Gardiner and Atema, 2007).

The nurse shark, on the other hand, changes its behavior in response to the loss of lateral line information. The slow velocity and lower frequency turns, combined with the faster swimming velocity and shorter time taken to reach the prey suggest that these animals are taking a straighter path to the target. Thus, when they are focusing on visual cues to navigate the flow, they are primarily turning upstream and swimming quickly, similar to the odor-stimulated rheotaxis behavior described previously for lemon sharks,

N. brevirostris (Hodgson and Mathewson, 1971; Mathewson and Hodgson, 1972). When vision is blocked in addition to the lateral line, the nurse shark reverts to typical tracking behavior (frequent, rapid turns). Since this species tends to maintain contact with the

121

bottom as it swims, occasionally using its pectoral fins to propel itself (Moss, 1972;

Limbaugh, 1975), it may be using tactile cues to orient to the flow. Sharks possess tactile

receptors, nerve endings from the trigeminal nerve in the skin, with areas of enhanced

sensitivity in the fins (Roberts, 1978). As the animals’ bodies are pushed by the flowing

water, they may be able to feel small bumps and ridges on the bottom of the tank sliding across the skin, and use this information to orient to the direction of flow (Lyon, 1904;

Arnold, 1974; Baker and Montgomery, 1999b). Since the bottom of this test tank is fairly uniform, one would expect this to be a slow process, and indeed, the nurse sharks in this study took significantly longer to reach the prey when tracking without visual or lateral line cues. Therefore, they are taking a more convoluted path to the prey when they rely on tactile cues to track. In contrast, intact nurse sharks turn frequently, but take a more or less direct path upstream (Mathewson and Hodgson, 1972). In the blacktip shark and bonnethead, vision and the lateral line play alternative roles in determining the direction of flow, and two systems play complementary roles with olfaction in tracking behaviors.

In the nurse shark, vision, the lateral line, and possibly tactile information are alternative with one another and complementary with olfaction for this task.

Orientation and Striking

Orientation and striking behaviors differ in the three species examined. Blacktip sharks orient from a distance of a few meters, and execute a direct, rapid strike. Since vision blocked animals can only strike from a distance of a few centimeters and only at a slower velocity, the longer distance orientation and striking are visually mediated.

Furthermore, when visual cues are available, these animals strike from a low angle of

122

attack, that is, the prey is more or less directly in front of the head, within the area of binocular overlap (McComb et al., 2009). Ram feeding involves the predator overtaking

its prey with the mouth open, which inherently requires that the predator pinpoints its

prey from a distance, in order to have sufficient room to accelerate (Gardiner and Motta,

In Press). This is especially true for the blacktip shark, which hunts for elusive teleost

prey (Castro, 1996), which it must rapidly chase down in the water column. The large

accelerations seen in this species may be a result of the rapid and potentially extensive

escape responses of the prey fish. Switching the focus from hydrodynamic + olfactory

cues to visual cues at the start of the orientation phase allows for the long distance strikes in the blacktip shark, since vision allows the animal to precisely localize the prey from a greater distance than the lateral line, which functions over a distance of only 0.4-2 body lengths (Denton and Gray, 1983; Denton and Gray, 1988; Coombs, 1999; Braun and

Coombs, 2000; Palmer et al., 2005; Jordan et al., 2009a) or electroreception, as the electric fields around prey are detectable at a distance of less than half of a meter

(Murray, 1960; Kalmijn, 1972; Kajiura and Holland, 2002b; Kajiura, 2003; Jordan et al.,

2009b; Kajiura and Fitzgerald, 2009).

When vision is blocked, strike angle is much larger, with strikes coming from the side or even behind the prey, similar to bony fish (Conley and Coombs, 1998; New,

2002). In the blinded animals, the lateral line appears to mediate the short-range orientation, as animals with both vision and the lateral line system blocked could not orient or strike, even when they swam in very close proximity to the prey, well within the range of the electrosensory system (Kajiura and Holland, 2002b; Kajiura, 2003), which suggests that electrical cues alone are not sufficient to prompt a strike. Since not all of

123

the blinded animals were successful in orienting to and striking at the prey, vision

provides the best performance for these tasks. Therefore, in the blacktip shark, orientation and striking are primarily visually mediated, as has been described for the lemon shark, N. brevirostris (Gilbert, 1963), but the lateral line can partially compensate, suggesting that it can play a partially alternative role in these tasks.

Orientation and striking appear to be entirely visually mediated in the bonnethead, as the blinded animals were never successful in lining up a strike. The large binocular overlap in this species provides them with enhanced three-dimensional vision as compared to pointed-nosed sharks, such as the blacktip shark (McComb et al., 2009).

When olfactory information is absent and the animals are not tracking, these strikes are initiated from a greater distance. This suggests that bonnetheads have switched their focus from the olfactory and lateral line cues to focus on visual cues. Since the orientation distance was not significantly different between olfactory blocked and lateral line blocked animals, animals using olfaction and vision for tracking may be paying greater attention to the visual cues, and therefore detecting the prey visually earlier and at a greater distance, than those tracking with hydrodynamic cues. At first glance, this may seem counterintuitive, however, bonnetheads typically hunt for swimming crabs and other crustaceans in seagrass beds (Cortés et al., 1996; Bethea et al., 2007), and thus likely cannot detect the prey visually from a distance. Therefore, these animals probably typically rely on odor tracking to bring them into close proximity of the prey. Guppies,

Poecilia reticulate, that have been reared in the dark and are accustomed to finding food using chemical cues show only weak responses to visual cues when foraging in a well-lit environment, which suggests that plasticity in the sensory mediation of behaviors does

124

not occur over the short term, but must be learned (Chapman et al., 2010). Similarly, the

bonnetheads in this study, though capable of visually detecting the prey from a distance,

have learned to focus instead primarily on the chemical cues. This reliance on chemical

cues may have been learned in these individuals from prior experience capturing prey in

the wild, as the bonnetheads in this study were subadult to adult animals, or this may be a

trait that has evolved in this species, which certainly warrants further investigation. In

this species, orientation and striking are completely visually mediated and thus vision is

obligatory for these tasks.

In contrast, for the nurse shark, when vision, the lateral line, or electroreception

are blocked, there are no significant changes in the frequency of orientation and striking,

nor in the distance from the prey at which these behaviors occur. Since suction feeding is only effective over a distance of a few centimeters (Motta et al., 2002; Nauwelaerts et al.,

2006; Lowry and Motta, 2007; Motta et al., 2008; Wilga and Sanford, 2008), suction feeders such as the nurse shark can use any of these senses to line up their short distance strikes. Blocking the olfactory system, alone or in combination with other senses, results in significantly less orientation and striking. However, as discussed above, since these animals display an overall lack of excitation and since orientation in these treatments is rarely followed by striking, it is likely that the animals are simply not identifying the prey item as food, rather than experiencing difficulty in orienting. It would appear that in the nurse shark, vision, the lateral line, and electroreception can play alternative roles in orienting and striking. However, since the finer scale aspects of these behaviors, such as strike velocity, could not be accurately measured for this species, it is possible that they do rely differentially on these sensory systems for these tasks, but those differences

125 cannot be detected by this experimental protocol. Filming these animals with high speed videography may provide sufficient resolution to analyze these behaviors.

Capture

In the blacktip shark, the rapid strikes are typically followed by successful capture; these animals rarely miss. Blocking the lateral line or electrosensory system results in an increase in misses. In the lateral line blocked animals, high velocity strikes tend to result in misses, whereas slower strikes result in successful captures. It would thus appear that the lateral line allows the animals to fine-tune the final moments of the very rapid strikes, however, they can compensate for the lack of lateral line information by slowing down their strikes. The frontal binocular overlap that exists in sharks comes at the cost of a blind area directly in front of the head (McComb et al., 2009). In the two ram feeders (blacktip shark and bonnethead), while the strikes are lined up visually, the final moments just prior to capture are blind, however the lateral line and electrosensory systems function to fill this void. In the blacktip shark, the lateral line functions to fine- tune and provide precision to the final moments of their very rapid strikes. Without lateral line information, they must slow their strikes to successfully capture prey. Since the bonnethead strikes with a slower velocity, the lateral line may not be as critical for strike precision. Slowing the strikes could decrease the reaction distance of the prey

(Viitasalo et al., 1998) by decreasing the bow wave in front of the head, which displaces the prey and alerts it to the approach of the predator, prompting an escape response

(Ferry-Graham et al., 2003). Because suction is effective over only a few centimeters, suction feeding varies over a smaller scale than ram feeding (Ferry-Graham et al., 2001)

126

and as there is little modulation of the strike in the nurse shark (Motta et al., 2002; Matott et al., 2005; Motta et al., 2008), no fine-tuning may be needed. Since they use little to no

forward motion, the prey is primarily alerted to the presence of the predator by the

suction flow (Holzman and Wainwright, 2009), which is often too late for a successful

escape response.

The increase in misses observed in electrosensory blocked blacktip sharks has no relation to strike velocity. Instead, there is a significant positive correlation between the blacktip shark making physical contact with the prey prior to beginning to move the jaws

and successful capture, which suggests that either the electrosensory system or touch can

mediate capture in this species. The lack of both cues generally results in a miss. Prey

contact with the jaws has been shown to elicit biting behavior in the marbled electric ray,

Torpedo marmorata (Belbenoit, 1986). Sharks possess areas of enhanced tactile

sensitivity with higher densities of trigeminal nerve endings, which function as touch

receptors, in the skin around the head and jaws (Roberts, 1978). Stimulation of the teeth

and surrounding tissues has been found to elicit rapid jaw closing in smooth dogfish, M.

canis, when the mouth is open (Roberts and Witovsky, 1975), while in cats, stimulation

of these areas when the mouth is closed results in jaw opening (Sherrington, 1917).

Using tactile information to position the jaws for capture has been previously described in other batoids, which pin the food to the substrate. This is likely mediated by non- pored lateral line canals and possibly the vesicles of Savi (Maruska and Tricas, 1998;

Maruska and Tricas, 2004), which are also found in a number of shark species (Chu and

Wen, 1979) and are far more sensitive to tactile cues than the touch receptors found in the skin (Maruska and Tricas, 2004). Either the non-pored canals, or the touch receptors, or

127 both, could be mediating the initiation of jaw movements upon prey contact observed in blacktip sharks.

Similar to the blacktip shark, there is also a significant positive correlation between a nurse shark making physical contact with the prey prior to beginning to move the jaws and successful capture. This suggests that capture in this species can also be mediated by either the electrosensory system or touch. Since this species is a suction feeder that only initiates capture from a very close (within a few centimeters) proximity to the prey (Motta et al., 2008), it is likely that in many cases, there is contact between the prey and the nurse shark’s snout or barbels. There were only a few strikes in the electrosensory blocked animals in which no contact was made, though they generally resulted in misses, which may explain the lack of significance in the increase in misses in this treatment. In both the blacktip and nurse sharks, repeated strikes eventually resulted in capture in all trials. By contrast, in the bonnethead, despite repeated strikes, no successful captures occurred when the electrosensory system was blocked, even if they made contact with the prey. In every strike, the jaws failed to move. This suggests that in bonnetheads, capture is completely mediated by the electrosensory system and that it is electrical cues that trigger the beginning of jaw movement. Tactile cues do not appear to be sufficient for this species to initiate jaw movements, despite the presence of non-pored canals in the vicinity of the mouth (Maruska, 2001). Because of the laterally expanded cephalofoil, hammerhead sharks have their electroreceptors spread over a greater lateral distance, compared to other shark species, and therefore sample a larger area of the environment (Kajiura, 2001). The enhanced electrosensory hypothesis suggests that this may be one of the factors that drove the evolution of wider heads in the hammerhead

128

sharks (Compagno, 1984) and it may explain the complete reliance of the bonnethead on

electroreception for prey capture.

Navigation and Obstacle Avoidance

In all three species, collisions with the wall are rare in intact animals. In the

bonnethead, there was a significant increase in wall collisions after the visual system was blocked, suggesting that this species may have a greater reliance on vision than the blacktip or nurse sharks. Increases in the frequency of wall collisions were observed in blacktip sharks, nurse sharks, and bonnetheads when the lateral line and vision were blocked simultaneously. This suggests that these animals are using hydrodynamic imaging, mediated by the lateral line, to navigate through their environment and avoid obstacles, when visual information is lacking. A similar behavior has been observed in blind Mexican cave fish, Astyanax fasciatus, which use hydrodynamic imaging to detect walls (Windsor et al., 2010a, b; Windsor et al., 2011). Although walls exist only in the captive environment, sharks must successfully navigate around a number of natural and manmade obstacles in the natural environment. Hydrodynamic imaging would presumably be of particular importance for obstacle avoidance in blacktip sharks and bonnetheads, as both of these species are obligate ram-ventilators (Parsons, 1990;

Carlson et al., 2004) and therefore cannot stop swimming, even at night when visual cues may be diminished.

129

Summary and Conclusions

In summary, the blacktip shark typically initially detects its prey using olfaction, tracks using olfaction in combination with either vision or the lateral line, switches to vision to orient for a long distance strike, then to the lateral line to fine tune the strike, and finally to electroreception or touch for capture. In the absence of an odor, the blacktip shark can detect its prey visually and proceed to orient, strike, and capture as above. In the absence of vision, the lateral line can sometimes mediate close-range orientation and striking. Thus, in the blacktip shark, olfaction and vision play alternative roles in detection; olfaction is complementary with either vision or the lateral line system for tracking, while vision and the lateral line play alternative roles; vision and the lateral line play partially alternative roles in orientation and striking; and electroreception and perhaps touch play alternative roles in capture.

The bonnethead typically initially detects its prey using olfaction, tracks using olfaction in combination with either vision or the lateral line, switches to vision to orient and strike, and finally to electroreception for capture. In the absence of an odor, the bonnethead can detect its prey visually and proceed to orient, strike, and capture as above. In the bonnethead, olfaction and vision play alternative roles in detection; olfaction is complementary with either vision or the lateral line system for tracking, while vision and the lateral line play alternative roles; vision is obligate for orientation and striking; and electroreception is obligate for capture.

The nurse shark detects its prey using olfaction, tracks using olfaction in combination with vision, the lateral line or possibly touch, switches to vision, the lateral line, or electroreception to orient and strike, and finally to electroreception or perhaps

130

touch for capture. In the nurse shark, olfaction is obligate for detection; olfaction is

complementary with either vision, the lateral line system, or touch for tracking, while

vision, the lateral line, and touch play alternative roles; vision, the lateral line, and

electroreception play alternative roles in orientation and striking; and electroreception

and touch play alternative roles in capture.

Though all of these species possess the same suite of sensory modalities, they

differ in terms of which senses they choose to focus on for particular behaviors. These interspecific differences were likely shaped by differences in the ecological niches that these sharks occupy, including the environment that they hunt in and type of prey consumed, as well as the foraging and capture strategies used.

Literature Cited

Arnold, G. P. (1974). Rheotropism in fishes. Biological Review 49, 515-576.

Arnold, S. J. (1983). Morphology, performance and fitness. American Zoologist 23, 347- 361.

Aronson, L. R., Aronson, F. R. and Clark, E. (1967). Instrumental conditioning and light- dark discrimination in young nurse sharks. Bulletin of Marine Science 17, 249- 256.

Atema, J. (1971). Structures and functions of the sense of taste in the catfish, Ictalurus natalis. Brain Behavior and Evolution 4, 273-294.

Atema, J. (1996). Eddy chemotaxis and odor landscapes: exploration of nature with animal sensors. Biological Bulletin 191, 129-138.

Atema, J. (In Press). Aquatic odor dispersal fields: opportunities and limits of detection, communication and navigation. In Chemical Ecology in Aquatic Systems, eds. C. Bronmark and L. A. Hansson). Oxford, U.K.: Oxford University Press.

131

Baker, C. F. and Montgomery, J. C. (1999a). Lateral line mediated rheotaxis in the Antarctic fish Pagothenia borchgrevinki. Polar Biology 21, 305-309.

Baker, C. F. and Montgomery, J. C. (1999b). The sensory basis of rheotaxis in the blind Mexican cave fish, Astyanax fasciatus. Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 184, 519-527.

Bartels, M., Miinz, H. and Claas, B. (1990). Representation of lateral line and electrosensory systems in the midbrain of the axolotl, Ambystoma mexicanum. Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 167, 347-356.

Basil, J. A., Hanlon, R. T., Sheikh, S. I. and Atema, J. (2000). Three-dimensional analysis of odor tracking by Nautilus pompilius. Journal of Experimental Biology 203, 1409-1414.

Bateson, W. (1890). The sense-organs and perceptions of fishes; with some remarks on the supply of bait. Journal of the Marine Biological Association of the United Kingdom N.S. 1, 225-256.

Belbenoit, P. (1986). Fine analysis of predatory and defensive motor events in Torpedo marmorata (Pisces). Journal of Experimental Biology 121, 197-226.

Benjamini, Y. and Hochberg, Y. (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 57, 289-300.

Bethea, D. M., Hale, L., Carlson, J. K., Cortes, E., Manire, C. A. and Gelsleichter, J. (2007). Geographic and ontogenetic variation in the diet and daily ration of the bonnethead shark, Sphyrna tiburo, from the eastern Gulf of Mexico. Marine Biology 152, 1009-1020.

Bigelow, H. B. and Schroeder, W. C. (1948). Fishes of the Western North Atlantic: Lancelets, cyclostomes, sharks. New Haven, Connecticut.

Blaxter, J. H. S. and Fuiman, L. A. (1989). Function of the free neuromasts of marine teleost larvae. In The Mechanosensory Lateral Line: Neurobiology and Evolution, eds. S. Coombs P. Gorner and H. Munz), pp. 481-499. New York, N.Y.: Springer- Verlag.

Bodznick, D. (1990). Elasmobranch vision: multimodal integration in the brain. Journal of Experimental Zoology Suppl. 5. 132

Bodznick, D. (1991). Elasmobranch vision: multimodal integration in the brain. Journal of Experimental Zoology Suppl. 5, 108-116.

Bodznick, D. and Northcutt, R. G. (1984). An electrosensory area in the telencephalon of the little skate, Raja erinacea. Brain Research 298, 117-124.

Boord, R. L. and Northcutt, R. G. (1982). Ascending lateral line pathways to the midbrain of the clearnose skate, Raja eglanteria. Journal of Comparative Neurology 207, 274-282.

Boord, R. L. and Northcutt, R. G. (1988). Medullary and mesencephalic pathways and connections of lateral line neurons in the spiny dogfish Squalus acanthias. Brain, Behavior and Evolution 32, 76-88.

Braun, C. B. and Coombs, S. (2000). The overlapping roles of the inner ear and lateral line: the active space of dipole source detection. Philosophical Transactions of the Royal Society of London Series B 355, 1115-1119.

Carlson, J. K., Goldman, K. J. and Lowe, C. G. (2004). Metabolism, Energetic Demand, and Endothermy. In Biology of Sharks and Their Relatives, eds. J. C. Carrier J. A. Musick and M. R. Heithaus). Boca Raton, FL: CRC Press.

Carton, A. G. and Montgomery, J. C. (2003). Evidence of a rheotactic component in the odour search behaviour of freshwater eels. Journal of Fish Biology 62, 501-516.

Castro, J. I. (1996). The biology of the blacktip shark, Carcharhinus limbatus, off the southeastern United States. Bulletin of Marine Science 59, 508-522.

Castro, J. I. (2000). The biology of the nurse shark, Ginglymostoma cirratum, off the Florida east coast and the Bahama Islands. Environmental Biology of Fishes 58, 1-22.

Chapman, B. B., Morrell, L. J., Tosh, C. R. and Krause, J. (2010). Behavioural consequences of sensory plasticity in guppies. Proceedings of the Royal Society of London. Series B: Biological Sciences 277, 1395-1401.

Chu, Y. T. and Wen, M. C. (1979). Monograph of fishes of China (No. 2): a study of the lateral-line canals system and that of Lorenzini ampullae and tubules of elasmobranchiate fishes of China. Shanghai: Science and Technology Press.

133

Cohen, D. H., Duff, T. A. and Ebbesson, S. O. E. (1973). Electrophysiological identification of a visual area in the shark telencephalon. Science 182, 492-494.

Compagno, L. J. V. (1984). FAO Species Catalogue. Sharks of the World. An Annotated and Illustrated Catalogue of Shark Species Known to Date. Rome: Food and Agriculture Organization of the United Nations.

Conley, R. A. and Coombs, S. (1998). Dipole source localization by mottled sculpin. III. Orientation after site-specific, unilateral denervation of the lateral line system. Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 183, 335-344.

Coombs, S. (1999). Signal detection theory, lateral-line excitation patterns and prey capture behaviour of mottled sculpin. Animal Behaviour 58, 421-430.

Cornett, A. D. (2006) Ecomorphology of Shark Electroreceptors. M.S. Thesis. Florida Atlantic University, Boca Raton, FL, pp. 102.

Curio, E. (1976). The Ethology of Predation. New York: Springer-Verlag.

Dehnhardt, G., Mauck, B., Hanke, W. and Bleckmann, H. (2001). Hydrodynamic trail- following in harbor seals (Phoca vitulina). Science 293, 102-104.

Denton, E. J. and Gray, J. (1983). Mechanical factors in the excitation of clupeid lateral lines. Proceedings of the Royal Society of London. Series B: Biological Sciences 218, 1-26.

Duntley, S. Q. (1963). Light in the sea. Journal of the Optical Society of America 53, 214-233.

Ebbesson, S. O. E. (1980). On the organization of the telencephalon in elasmobranchs. In Comparative Neurology of the Telencephalon, (ed. S. O. E. Ebbesson), pp. 1-16. New York: Plenum Press. 134

Faucher, K., Dutto, G., Covès, D., Aubert, A. and Lagardère, J. P. (2006a). No efficiency of the lateral system on nocturnal feeding in the European sea bass (Dicentrarchus labrax L.). Aquaculture 252, 462-475.

Faucher, K., Fichet, D., Miramand, P. and Lagardère, J. P. (2006b). Impact of acute cadmium exposure on the trunk lateral line neuromasts and consequences on the “C-start” response behaviour of the sea bass (Dicentrarchus labrax L.; Teleostei, Moronidae). Aquatic Toxicology 76, 278-294.

Faucher, K., Parmentier, E., Becco, C., Vandewalle, N. and Vandewalle, P. (2010). Fish lateral system is required for accurate control of shoaling behaviour. Animal Behaviour 79, 679-687.

Ferry-Graham, L. A. and Lauder, G. V. (2001). Aquatic prey capture in ray-finned fishes: a century of progress and new directions. Journal of Morphology 248, 99-119.

Ferry-Graham, L. A., Wainwright, P. C. and Bellwood, D. R. (2001). Prey capture in long-jawed butterflyfishes (Chaetodontidae): the functional basis of novel feeding habits. Journal of Experimental Marine Biology and Ecology 256, 167-184.

Ferry-Graham, L. A., Wainwright, P. C. and Lauder, G. V. (2003). Quantification of flow during suction feeding in bluegill sunfishes. Zoology 106, 159-168.

Fisher, H. S., Wong, B. M. and Rosenthal, G. G. (2006). Alteration of the chemical environment disrupts communication in a freshwater fish. Proceedings of the Royal Society of London, Series B: Biological Sciences 273, 1187-1193.

Fouts, W. R. and Nelson, D. R. (1999). Prey capture by the Pacific angle shark, Squatina californica: visually mediated strikes and ambush-site characteristics. Copeia 1999, 304-312.

Fraenkel, G. S. and Gunn, D. L. (1940). The Orientation of Animals: Kineses, Taxes, and Compass Reactions. Oxford: Clarendon Press.

Gardiner, J. M. and Atema, J. (2007). Sharks need the lateral line to locate odor sources: rheotaxis and eddy chemotaxis. Journal of Experimental Biology 210, 1925-1934.

Gardiner, J. M. and Atema, J. (2010). The function of bilateral timing differences in olfactory orientation of sharks. Current Biology 20, 1187-1191.

Gardiner, J. M. and Motta, P. J. (In Press). Largemouth bass (Micropterus salmoides) switch feeding modalities in response to sensory deprivation. Zoology.

Gilbert, P. W. (1963). The Visual Apparatus of Sharks. In Sharks and Survival, (ed. P. W. Gilbert), pp. 283-326. Boston, MA: DC Heath & Co. 135

Glaser, N., Wieskotten, S., Otter, C., Dehnhardt, G. and Hanke, W. (2011). Hydrodynamic trail following in a California sea lion (Zalophus californianus). Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 197, 141-151.

Graeber, R. C., Ebbesson, S. O. E. and Jane, J. A. (1973). Visual discrimination in sharks without optic tectum. Science 180, 413-415.

Graeber, R. C., Schroeder, D. M., Jane, J. A. and Ebbesson, S. O. E. (1978). Visual discrimination following partial telencephalic ablations in nurse sharks (Ginglymostoma cirratum). Journal of Comparative Neurology 180, 325-344.

Grasso, F. W. and Atema, J. (2002). Integration of flow and chemical sensing for guidance of autonomous marine robots in turbulent flow. Environmental Fluid Mechanics 2, 95-114.

Gudger, E. W. (1921). Notes on the morphology and habits of the nurse shark, Ginglymostoma cirratum. Copeia 1921, 57-59.

Haine, O. S., Ridd, P. V. and Rowe, R. J. (2001). Range of electrosensory detection of prey by Carcharhinus melanopterus and Himantura granulata. Marine and Freshwater Research 52, 291-296.

Hamasaki, D. I. and Gruber, S. H. (1965). The photoreceptors of the nurse shark, Ginglymostoma cirratum and the sting ray, Dasyatis sayi. Bulletin of Marine Science 15, 1051-1059.

Hartline, P. H., Kass, L. and Loop, M. S. (1978). Merging of modalities in the optic tectum: Infrared and visual integration in rattlesnakes. Science 199, 1225-1229.

Haverly, J. E. and Kardong, K. V. (1996). Sensory deprivation effects on the predatory behavior of the rattlesnake, Crotalus viridis oreganus. Copeia 1996, 419-428.

Hobson, E. S. (1963). Feeding behavior in three species of sharks. Pacific Science 17, 171-194.

136

Hodgson, E. S. and Mathewson, R. F. (1971). Chemosensory orientation in sharks. Annals of the New York Academy of Sciences 188, 175-182.

Hodgson, E. S. and Mathewson, R. F. (1978). Electrophysiological studies of chemoreception in elasmobranchs. In Sensory Biology of Sharks, Skates, and Rays, eds. E. S. Hodgson and R. F. Mathewson), pp. 227-267. Arlington: Office of Naval Research.

Hoffman, M. H. and Northcutt, R. G. (2008). Organization of major telencephalic pathways in an elasmobranch, the thornback ray Platyrhinoidis triseriata. Brain, Behavior and Evolution 72, 307-325.

Holzman, R. and Wainwright, P. C. (2009). How to surprise a copepod: strike kinematics reduce hydrodynamic disturbance and increase stealth of suction-feeding fish. Limnology and Oceanography 54, 2201-2212.

Honigmann, H. (1944). The visual perception of movement by toads. Proceedings of the Royal Society of London, Series B: Biological Sciences 132, 291-307.

Horn, G. and Hill, R. M. (1966). Responsiveness to sensory stimulation of units in the superior colliculus and subjacent tectotegmental regions of the rabbit. Experimental Neurology 14, 199-223.

Hueter, R. E. (1991). Adaptations for spatial vision in sharks. Journal of Experimental Zoology Supplement 5, 130-141.

Janssen, J. and Corcoran, J. (1993). Lateral line stimuli can override vision to determine sunfish strike trajectory. Journal of Experimental Biology 176, 299-305.

Janssen, J., Jones, W. R., Whang, A. and Oshel, P. E. (1995). Use of the lateral line in particulate feeding in the dark by juvenile alewife (Alosa pseudoharengus). Can. J. Fish. Aqua. Sci. 52, 358-363.

Jassik-Gerschenfeld, D. (1965). Somesthetic and visual responses of superior colliculus neurones. Nature 208, 898-900.

Johnsen, S. (2005). Visual ecology on the high seas. Marine Ecology Progress Series 287, 281-285.

Johnsen, S. and Sosik, H. M. (2004). Shedding light on light in the sea. Oceanus 43, 24- 28.

137

Jordan, L. K., Kajiura, S. M. and Gordon, M. S. (2009a). Functional consequences of structural differences in stingray sensory systems. Part I: mechanosensory lateral line canals. Journal of Experimental Biology 212, 3037-3043.

Jordan, L. K., Kajiura, S. M. and Gordon, M. S. (2009b). Functional consequences of structural differences in stingray sensory systems. Part II: electrosensory system. Journal of Experimental Biology 212, 3044-3050.

Kajiura, S. M. (2001). Head morphology and electrosensory pore distribution of carcharhinid and sphyrnid sharks. Environmental Biology of Fishes 61, 125-133.

Kajiura, S. M. (2003). Electroreception in neonatal bonnethead sharks, Sphyrna tiburo. Marine Biology 143, 603-611.

Kajiura, S. M. and Fitzgerald, T. P. (2009). Response of juvenile scalloped hammerhead sharks to electric stimuli. Zoology 112, 241-250.

Kajiura, S. M. and Holland, K. N. (2002a). Electroreception in juvenile scalloped hammerhead and sandbar sharks. Journal of Experimental Biology 205, 3609- 3621.

Kajiura, S. M. and Holland, K. N. (2002b). Electroreception in juvenile scalloped hammerhead and sandbar sharks. Journal of Experimental Biology 205, 3609- 3621.

Kalmijn, A. J. (1972). Bioelectric fields in seawater and the function of the ampullae of Lorenzini in elasmobranch fishes. Scripps Institute of Oceanography Reference Series 72-83, 1-21.

Kalmijn, A. J. (1982). Electric and magnetic field detection in elasmobranch fishes. Science 218, 916-918.

Kaus, S. (1987). The effect of aminoglycoside antibiotics on the lateral line organ of Aplocheilus lineatus (Cyprinodontidae). Acta Otolaryngologica 103, 291-298.

Keller, T. A., Powell, I. and Weissburg, M. J. (2003). Role of olfactory appendages in chemically mediated orientation of blue crabs. Marine Ecology Progress Series 261, 217-231.

Kleerekoper, H. (1978). Chemoreception and its interaction with flow and light perception in the locomotion and orientation of some elasmobranchs. In Sensory

138

Biology of Sharks, Skates, and Rays, eds. E. S. Hodgson and R. F. Mathewson). Arlington, VA: U.S. Office of Naval Research.

Kleerekoper, H. (1982). The role of olfaction in the orientation of fishes. In Chemoreception in Fishes: Developments in Aquaculture and Fisheries Science, (ed. T. J. Hara), pp. 201-225. Amsterdam: Elsevier.

Knudsen, E. I. (1982). Auditory and visual maps of space in the optic tectum of the owl. Journal of Neuroscience 2, 1177-1194.

Kroese, A. B. A. and van den Bercken, J. (1982). Effects of ototoxic antibiotics on sensory hair cell functioning. Hearing Research 6, 183-197.

Limbaugh, C. (1975). Field Notes on Sharks. In Sharks and Survival, (ed. P. W. Gilbert), pp. 63-94. Lexington, MA: DC Heath and Company.

Lowry, D. and Motta, P. J. (2007). Ontogeny of feeding and cranial morphology in the whitespotted bambooshark Chiloscyllium plagiosum. Marine Biology 151, 2013- 2023.

Lyon, E. P. (1904). On rheotropism. I. Rheotropism in fishes. American Journal of Physiology 12, 149-161.

Lyon, E. P. (1909). On rheotropism. II. Rheotropism of fish blind in one eye. American Journal of Physiology 24, 244-251.

Mafra-Neto, A. and Cardé, R. T. (1994). Fine-scale structure of pheromone plumes modulates upwind orientation of flying moths. Nature 369, 142-144.

Maruska, K. P. (2001). Morphology of the mechanosensory lateral line system in elasmobranch fishes: ecological and behavioral considerations. Environmental Biology of Fishes 60, 47-75.

Maruska, K. P. and Tricas, T. C. (1998). Morphology of the mechanosensory lateral line system in the Atlantic stingray, Dasyatis sabina: the mechanotactile hypothesis. Journal of Morphology 238, 1-22.

139

Mathewson, R. F. and Hodgson, E. S. (1972). Klinotaxis and rheotaxis in orientation of sharks toward chemical stimuli. Comparative Biochemistry and Physiology, A: Comparative Physiology 42, 79-84.

Matott, M. P., Motta, P. J. and Hueter, R. E. (2005). Modulation in feeding kinematics and motor pattern of the nurse shark Ginglymostoma cirratum. Environmental Biology of Fishes 74, 163-174.

Matsuura, S., Ikeda, K. and Furukawa, T. (1971). Effects of streptomycin, kanamycin, quinine, and other drugs on the microphonic potentials of goldfish sacculus. Japanese Journal of Physiology 21, 579-590.

Mazur, M. M. and Beauchamp, D. A. (2003). A comparison of visual prey detection among species of piscivorous salmonids: effects of light and low turbidities. Environmental Biology of Fishes 67, 397-405.

McComb, D. M., Tricas, T. C. and Kajiura, S. M. (2009). Enhanced visual fields in hammerhead sharks. Journal of Experimental Biology 212, 4010-4018.

McMahon, T. E. and Holanov, S. H. (1995). Foraging success of largemouth bass at different light intensities: implications for time and depth of feeding. Journal of Fish Biology 46, 759-767.

Mellon, D. (2005). Integration of hydrodynamic and odorant inputs by local interneurons of the crayfish deutocerebrum. Journal of Experimental Biology 208, 3711-3720.

Meredith, T. M. and Kajiura, S. M. (2010). Olfactory morphology and physiology of elasmobranchs. Journal of Experimental Biology 213, 3449-3456.

Miersch, L., Hanke, W., Wieskotten, S., Hanke, F. D., Oeffner, J., Leder, A., Brede, M., Witte, M. and Dehnhardt, G. (2011). Flow sensing by pinniped whiskers. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 366, 3077-3084.

Montgomery, J. C., Baker, C. F. and Carton, A. G. (1997). The lateral line can mediate rheotaxis in fish. Nature 389, 960-963.

Montgomery, J. C., Macdonald, F., Baker, C. F. and Carton, A. G. (2002). Hydrodynamic contributions to multimodal guidance of prey capture behavior in fish. Brain, Behavior and Evolution 59, 190-198.

140

Moore, P. A. and Atema, J. (1991). Spatial information in the three-dimensional fine structure of an aquatic odor plume. Biological Bulletin 181, 408-418.

Moore, P. A., Scholz, N. and Atema, J. (1991). Chemical orientation of lobsters, Homarus americanus, in turbulent odor plumes. Journal of Chemical Ecology 17, 1293-1307.

Moss, S. A. (1972). Nurse shark pectoral fins: an unusual use. American Midland Naturalist 88, 496-497.

Motta, P. J., Hueter, R. E., Tricas, T. C., Summers, A. P., Huber, D. R., Lowry, D., Mara, K. R., Matott, M. P., Whitenack, L. B. and Wintzer, A. P. (2008). Functional morphology of the feeding apparatus, feeding constraint and suction performance in the nurse shark Ginglymostoma cirratum. Journal of Morphology 269, 1041- 1055.

Motta, P. J. and Wilga, C. D. (2001). Advances in the study of feeding behaviors, mechanisms, and mechanics of sharks. Environmental Biology of Fishes 60, 131- 156.

Murray, R. W. (1960). Electrical sensitivity of the ampullae of Lorenzini. Nature 187, 957.

Murray, R. W. (1962). Response of ampullae of Lorenzini of elasmobranchs to electrical stimulation. Journal of Experimental Biology 39, 119-128.

Myrberg, J., A.A., Banner, A. and Richard, J. D. (1969). Shark attraction using a video- acoustic system Marine Biology 2, 264-276.

Nauwelaerts, S., Wilga, C., Sanford, C. and Lauder, G. (2006). Hydrodynamics of prey capture in sharks: effects of substrate. Journal of The Royal Society Interface, 1-6.

Nelson, D. R. and Gruber, S. H. (1963). Sharks: attraction by low-frequency sounds. Science 142, 975-977. 141

Nevitt, G. A., Losekoot, M. and Weimerskirch, H. (2008). Evidence for olfactory search in wandering albatross, Diomedea exulans. Proceedings of the National Academy of Sciences 105, 4576-4581.

New, J. G. (2002). Multimodal integration in the feeding behaviors of predatory teleost fishes. Brain, Behavior and Evolution 59, 177-189.

New, J. G., Alborg Fewkes, L. and Khan, A. N. (2001). Strike feeding behavior in the muskellunge, Esox mascuinongy: Contributions of the lateral line and visual sensory systems. Journal of Experimental Biology 204, 1207-1221.

New, J. G. and Kang, P. Y. (2000). Multimodal sensory integration in the strike-feeding behaviour of predatory fishes. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 355, 1321-1324.

Nyberg, D. W. (1971). Prey capture in the largemouth bass. American Midland Naturalist 86, 128-144.

Palmer, L. M., Deffenbaugh, M. and Mensinger, A. F. (2005). Sensitivity of the anterior lateral line to natural stimuli in the oyster toadfish, Opsanus tau (Linnaeus). Journal of Experimental Biology 208, 3441-3450.

Parker, G. H. (1909). The influence of eyes and ears and other allied sense organs on the movement of Mustelus canis. Bulletin of the United States Bureau of Fisheries 29, 43-58.

Parker, G. H. (1914). The directive influence of the sense of smell in the dogfish. Bulletin of the United States Bureau of Fisheries 33, 61-68.

Parker, G. H. (1922). Smell, Taste, and Allied Senses in Vertebrates. Philadelphia, P.A.: J.P. Lippincott Co.

Parker, G. H. and Sheldon, R. E. (1913). The sense of smell in fishes. Bulletin of the United States Bureau of Fisheries 32, 33-46.

Parsons, G. R. (1990). Metabolism and swimming efficiency of the bonnethead shark Sphyrna tiburo. Marine Biology 104, 363-367. 142

Peach, M. B. (2001). The dorso-lateral pit organs of the Port Jackson shark contribute sensory information for rheotaxis. Journal of Fish Biology 59, 696-704.

Pohlmann, K., Atema, J. and Breithaupt, T. (2004). The importance of the lateral line in nocturnal predation of piscivorous catfish. J. Exp. Biol. 207, 2971-2978.

Pohlmann, K., Grasso, F. W. and Breithaupt, T. (2001). Tracking wakes: the nocturnal predatory strategy of piscivorous catfish. Proceedings of the National Academy of Sciences of the United States of America 98, 7371-7374.

Popper, A. N. and Fay, R. R. (1999). The auditory periphery in fishes. In Comparative hearing: fish and amphibians, eds. R. R. Fay and A. N. Popper), pp. 43-100. New York, NY: Springer-Verlag.

Roberts, B. L. (1978). Mechanoreceptors and the behaviour of elasmobranch fishes with special reference to the acoustico-lateralis system. In Sensory Biology of Sharks, Skates, and Rays, eds. E. S. Hodgson and R. F. Mathewson), pp. 331-390. Arlington, Virginia: Office of Naval Research.

Roberts, B. L. and Witovsky, P. (1975). A functional analysis of the mesencephalic nucleus of the fifth nerve in the selachian brain. Proceedings of the Royal Society of London, Series B: Biological Sciences 190, 473-495.

Schellart, N. A. M. and Popper, A. N. (1992). Functional aspects of the evolution of the auditory system of actinopterygian fish. In The evolutionary biology of hearing, eds. D. B. Webster R. R. Fay and A. N. Popper), pp. 295-322. New York, NY: Springer-Verlag.

Sheldon, R. E. (1909). The reactions of the dogfish to chemical stimuli. Journal of Comparative Neurology and Psychology 19, 273-311.

Sheldon, R. E. (1911). The sense of smell in selachians. Journal of Experimental Zoology 10, 51-62.

Sherrington, C. S. (1917). Reflexes elicitable in the cat from pinna, vibrissae, and jaws. Journal of Physiology 51, 404-431.

Sokal, R. R. and Rohlf, F. J. (1995). Biometry: The Principles and Practices of Statistics in Biological Research. New York: W.H. Freeman and Co.

143

Stein, B. E. and Stanford, T. R. (2008). Multisensory integration: current issues from the perspective of the single neuron. Nature Reviews Neuroscience 9, 255-266.

Tester, A. L. (1963a). Olfaction, gustation, and the common chemical sense in sharks. In Sharks and Survival, (ed. P. W. Gilbert), pp. 255-282. Lexington: D.C. Heath and Company.

Tester, A. L. (1963b). The role of olfaction in shark predation. Pacific Science 17, 145- 170.

Tierney, K. B., Baldwin, D. H., Hara, T. J., Ross, P. S., Scholz, N. L. and Kennedy, C. J. (2010). Olfactory toxicity in fishes. Aquatic Toxicology 96, 2-26.

Veselkin, N. P. and Kovačević, N. (1973). Nonolfactory afferent projections of the telencephalon of elasmobranchii. Journal of Evolutionary Biochemistry and Physiology 9, 512-518.

Viitasalo, M., Kiørboe, T., Flinkman, J., Pedersen, L. W. and Visser, A. W. (1998). Predation vulnerability of planktonic copepods:consequences of predator foraging strategies and prey sensory abilities. Marine Ecology Progress Series 175, 129- 142.

Webster, D. R. (2007). Structure of turbulent chemical plumes. In Trace Chemical Sensing of Explosives, (ed. R. L. Woodfin), pp. 109-129. New York: John Wiley & Sons.

Webster, D. R., Rahman, S. and Dasi, L. P. (2003). Laser-induced fluorescence measurements of a turbulent plume. Journal of Engineering Mechanics 129, 1130-1137.

Webster, D. R., Roberts, P. J. W. and Ra'ad, L. (2001). Simultaneous DPTV/PLIF measurements of a turbulent jet. Experiments in Fluids 30, 65-72.

Webster, D. R. and Weissburg, M. J. (2001). Chemosensory guidance cues in a turbulent chemical odor plume. Limnology and Oceanography 46, 1034-1047.

Weissburg, M. J. (2000). The fluid dynamical context of chemosensory behavior. Biological Bulletin 198, 188-202.

Weissburg, M. J., Dusenbery, D. B., Ishida, H., Janata, J., Keller, T., Roberts, P. J. W. and Webster, D. R. (2002). A multidisciplinary study of spatial and temporal 144

scales containing information in turbulent chemical plume tracking. Environmental Fluid Mechanics 2, 65-94.

Wersall, J. and Flock, A. (1964). Suppression and restoration of the microphonic output from the lateral line organ after local application of streptomycin. Life Sciences 3, 1151-1155.

Wilga, C. D. and Sanford, C. P. (2008). Suction generation in white-spotted bamboo sharks Chiloscyllium plagiosum. Journal of Experimental Biology 211, 3128- 3138.

Windsor, S. P., Norris, S. E., Cameron, S. M., Mallinson, G. D. and Montgomery, J. C. (2010a). The flow fields involved in hydrodynamic imaging by blind Mexican cave fish (Astyanax fasciatus). Part I: open water and heading towards a wall. Journal of Experimental Biology 213, 3819-3831.

Windsor, S. P., Norris, S. E., Cameron, S. M., Mallinson, G. D. and Montgomery, J. C. (2010b). The flow fields involved in hydrodynamic imaging by blind Mexican cave fish (Astyanax fasciatus). Part II: gliding parallel to a wall. Journal of Experimental Biology 213.

Windsor, S. P., Paris, J. and De Parera, T. B. (2011). No role for direct touch using the pectoral fins, as an information gathering strategy in a blind fish. Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 197, 321-327.

Young, B. A. and Morain, M. (2002). The use of ground-borne vibrations for prey localization in the saharan sand vipers (Cerastes). Journal of Experimental Biology 205, 661-665.

Zimmer-Faust, R. K., Finelli, C. M., Pentcheff, N. D. and Wethey, D. S. (1995). Odor plumes and animal navigation in turbulent water flow: a field study. Biological Bulletin 188, 111-116.

145

Tables Table 3.1. Data Summary – Blacktip sharks

Tukey Test vs. Nose vs. Eye vs. Nose vs. LLX vs. LLX vs. LLX vs. LLX + vs. Variable Treatment Mean n p value Block Block + Eye + Nose + Eye Nose + Electro- Block Block Block Eye Block sensory Block Control 1.19±0.03 17 Nose Block 1.10±0.04 8 Eye Block 0.99±0.06 12 Swim Nose + Eye Block 0.72±0.03 9 Velocity LLX 1.40±0.13 5 0.384 (BL/s) LLX + Nose Block 1.14±0.05 5 LLX + Eye Block 0.92±0.06 5 LLX + Nose + Eye Block 0.70±0.04 4 Electrosensory Block 1.20±0.26 3 Control 140.2±2.1 17 N.S. <0.001 N.S. <0.001 N.S. <0.001 N.S. Nose Block 54.3±2.3 8 <0.001 N.S. <0.001 N.S. <0.001 N.S. <0.001 Eye Block 163.1±6.9 12 <0.001 N.S. <0.001 N.S. <0.001 N.S. Turn Nose + Eye Block 52.9±4.3 9 <0.001 N.S. <0.001 N.S. <0.001 Velocity LLX 138.8±16.9 5 <0.001* <0.001 N.S. <0.001 N.S. (°/s) LLX + Nose Block 47.8±7.2 5 0.003 N.S. <0.001 LLX + Eye Block 110.8±18.9 5 0.013 N.S. LLX + Nose + Eye Block 48.3±3.4 4 <0.001 Electrosensory Block 163.9±35.1 3 Control 0.72±0.04 17 <0.001 0.013 <0.001 N.S. <0.001 <0.001 <0.001 0.003 Nose Block 0.16±0.03 8 <0.001 N.S. <0.001 N.S. N.S. N.S. <0.001 Eye Block 0.88±0.04 12 <0.001 0.001 <0.001 <0.001 <0.001 N.S. Turn Nose + Eye Block 0.20±0.01 9 <0.001 N.S. N.S. N.S. <0.001 Frequency LLX 0.56±0.08 5 <0.001* <0.001 0.039 <0.001 <0.001 (turns/s) LLX + Nose Block 0.23±0.03 5 N.S. N.S. <0.001 LLX + Eye Block 0.36±0.03 5 N.S. <0.001 LLX + Nose + Eye Block 0.16±0.01 4 <0.001 Electrosensory Block 1.08±0.01 3 Control 35.0±9.5 17 <0.001 N.S. N.S. Nose Block N.A. 8 Eye Block 949.3±236.1 12 <0.001 0.033 9 Tracking Nose + Eye Block N.A. 5 <0.001* Time (s) LLX 13.1±5.4 N.S. LLX + Nose Block N.A. 5 LLX + Eye Block N.A. 5 LLX + Nose + Eye Block N.A. 4 Electrosensory Block 106.4±6.2 3

146

(Table 3.1 Continued)

Tukey Test vs. Nose vs. vs. Nose vs. LLX vs. LLX vs. LLX vs. LLX + vs. Variable Treatment Mean N p value Block Eye + Eye + Nose + Eye Nose + Electro- Block Block Block Block Eye Block sensory Block Control 238.9±16.3 17 N.S. <0.001 N.S. N.S. N.S. Nose Block 253.7±2.5 8 <0.001 N.S. 0.039 N.S. Eye Block 17.4±2.5 12 <0.001 <0.001 0.003 Orientation Nose + Eye Block N.A. 9 Distance LLX 304.8±32.1 5 <0.001* 0.008 N.S. (cm) LLX + Nose Block 177.5±27.5 5 N.S. LLX + Eye Block N.A. 5 LLX + Nose + Eye Block N.A. 4 Electrosensory Block 214.3±16.3 3 Control 100.0±0.0 17 N.S. <0.001 <0.001 N.S. N.S. <0.001 <0.001 N.S. Nose Block 100.0±0.0 8 <0.001 <0.001 N.S. N.S. <0.001 <0.001 N.S. Eye Block 60.4±11.1 12 <0.001 0.001 0.001 <0.001 <0.001 0.013 Nose + Eye Block 0.0±0.0 9 <0.001 <0.001 N.S. <0.001 <0.001 Strike Rate LLX 100.0±0.0 5 <0.001* N.S. <0.001 <0.001 N.S. (%) LLX + Nose Block 100.0±0.0 5 N.S. <0.001 N.S. LLX + Eye Block 0.0±0.0 5 N.S. <0.001 LLX + Nose + Eye Block 0.0±0.0 4 <0.001 Electrosensory Block 100.0±0.0 3 Control 1.95±0.21 17 N.S. 0.027 N.S. N.S. N.S. N.S. Nose Block 1.50±0.04 8 N.S. N.S. N.S. N.S. N.S. Eye Block 1.25±0.31 12 0.027 N.S. N.S. N.S. Strike Nose + Eye Block N.A. 9 Velocity LLX 2.46±0.62 5 0.015* N.S. N.S. N.S. (BL/s) LLX + Nose Block 1.30±0.08 5 N.S. LLX + Eye Block N.A. 5 LLX + Nose + Eye Block N.A. 4 Electrosensory Block 1.51±0.11 3 Control 15.4±2.5 17 N.S. <0.001 N.S. N.S. N.S. Nose Block 17.6±3.9 8 0.002 N.S. N.S. N.S. Eye Block 91.7±15.5 12 <0.001 0.019 N.S. 9 Strike Nose + Eye Block N.A. 5 <0.001* Angle (°) LLX 7.7±11 N.S. N.S. LLX + Nose Block 21.7±5.6 5 N.S. LLX + Eye Block N.A. 5 LLX + Nose + Eye Block N.A. 4 Electrosensory Block 23.5±7.8 3

147

(Table 3.1 Continued)

Tukey Test vs. Nose vs. vs. Nose vs. LLX vs. LLX vs. LLX vs. LLX + vs. Variable Treatment Mean n p value Block Eye + Eye + Nose + Eye Nose + Electro- Block Block Block Block Eye Block sensory Block Control 0.00±0.00 17 N.S. N.S. N.S. <0.05 <0.05 Nose Block 0.00±0.00 8 N.S. N.S. <0.05 <0.05 Eye Block 0.11±0.05 12 N.S. N.S. <0.05 Nose + Eye Block N.A. 9 Misses LLX 0.34±0.19 5 <0.001* N.S. N.S. LLX + Nose Block 0.58±0.23 5 N.S. LLX + Eye Block N.A. 5 LLX + Nose + Eye Block N.A. 4 Electrosensory Block 1.17±0.58 3 Control 100.0±0.0 17 N.S. <0.05 N.S. N.S. N.S. Nose Block 100.0±0.0 8 <0.05 N.S. N.S. N.S. Eye Block 50.0±10.5 12 N.S. N.S. N.S. Capture Nose + Eye Block N.A. 9 Success LLX 100.0±0.0 5 <0.001* N.S. N.S. Rate (%) LLX + Nose Block 100.0±0.0 5 N.S. LLX + Eye Block N.A. 5 LLX + Nose + Eye Block N.A. 4 Electrosensory Block 100.0±0.0 3 Control 0.12±0.09 17 N.S. N.S. N.S. N.S. N.S. <0.001 <0.001 N.S. Nose Block 0.00±0.00 8 N.S. N.S. N.S. N.S. <0.001 <0.001 N.S. Frequency Eye Block 0.11±0.05 12 N.S. N.S. N.S. <0.001 <0.001 N.S. of Wall Nose + Eye Block 0.03±0.02 9 N.S. N.S. <0.001 <0.001 N.S. Collisions LLX 0.00±0.00 5 <0.001* N.S. <0.001 <0.001 N.S. (collisions LLX + Nose Block 0.25±0.25 5 <0.001 <0.001 N.S. /100s) LLX + Eye Block 1.74±0.25 5 0.004 <0.001 LLX + Nose + Eye Block 1.25±0.15 4 <0.001 Electrosensory Block 0.00±0.00 3

Summary of all variables for the blacktip shark, Carcharhinus limbatus with all senses intact, and following blocks of the senses as indicated. LLX: Lateral line system blocked. All means are ±s.e.m. The p values are the results of ANOVA or Kruskal-Wallis tests performed on each variable. Value marked (*) are significant after Benjamini-Hochberg corrections. Tukey Test p values reflect the results of pairwise post-hoc comparisons between treatments. N.A.: Not applicable, parameter was not assessed because behavior did not occur; N.S.: Non-significant at α = 0.05.

148

Table 3.2. Data Summary – Nurse sharks

Tukey Test vs. Nose vs. vs. Nose vs. LLX vs. LLX vs. LLX vs. LLX + vs. Variable Treatment Mean n p value Block Eye + Eye + Nose + Eye Nose + Electro- Block Block Block Block Eye Block sensory Block Control 0.44±0.04 7 N.S. N.S. N.S. 0.004 0.021 N.S. N.S. N.S. Nose Block 0.54±0.06 7 N.S. N.S. N.S. N.S. N.S. N.S. N.S. Eye Block 0.35±0.01 7 N.S. <0.001 <0.001 N.S. N.S. 0.048 Swim Nose + Eye Block 0.41±0.05 5 0.003 0.014 N.S. N.S. N.S. Velocity LLX 0.76±0.07 6 <0.001* N.S. 0.012 N.S. N.S. (BL/s) LLX + Nose Block 0.73±0.08 5 0.048 N.S. N.S. LLX + Eye Block 0.46±0.02 6 N.S. N.S. LLX + Nose + Eye Block 0.53±0.02 5 N.S. Electrosensory Block 0.60±0.02 6 Control 83.6±4.8 7 <0.001 N.S. <0.001 0.042 <0.001 N.S. <0.001 N.S. Nose Block 47.8±2.9 7 <0.001 N.S. N.S. N.S. 0.002 N.S. 0.003 Eye Block 83.8±4.2 7 <0.001 0.039 <0.001 N.S. <0.001 N.S. Turn Nose + Eye Block 40.0±2.6 5 0.026 N.S. <0.001 N.S. <0.001 Velocity LLX 62.1±3.3 6 <0.001* N.S. N.S. N.S. N.S. (°/s) LLX + Nose Block 45.2±7.9 5 0.007 0.005 0.007 LLX + Eye Block 74.8±3.6 6 0.006 N.S. LLX + Nose + Eye Block 47.5±3.1 5 0.009 Electrosensory Block 73.9±5.9 6 Control 0.52±0.03 7 <0.001 N.S. <0.001 0.001 <0.001 N.S. <0.001 N.S. Nose Block 0.11±0.01 7 <0.001 N.S. 0.001 N.S. <0.001 N.S. <0.001 Eye Block 0.49±0.03 7 <0.001 0.005 <0.001 N.S. <0.001 N.S. Turn Nose + Eye Block 0.15±0.02 5 0.025 N.S. <0.001 N.S. <0.001 Frequency LLX 0.31±0.04 6 <0.001* 0.002 N.S. N.S. 0.09 (turns/s) LLX + Nose Block 0.11±0.02 5 <0.001 N.S. <0.001 LLX + Eye Block 0.41±0.04 6 <0.001 N.S. LLX + Nose + Eye Block 0.17±0.03 5 <0.001 Electrosensory Block 0.45±0.07 6 Control 98.4±20.6 7 N.S. <0.001 N.S. N.S. Nose Block N.A. 7 Eye Block 99.4±15.7 7 <0.001 N.S. N.S. Nose + Eye Block N.A. 5 Tracking LLX 17.6±4.5 6 <0.001* <0.001 0.039 Time (s) LLX + Nose Block N.A. 5 LLX + Eye Block 189.8±79.1 6 0.003 LLX + Nose + Eye Block N.A. 5 Electrosensory Block 52.43±15.6 6

149

(Table 3.2 Continued)

Tukey Test vs. Nose vs. vs. Nose vs. LLX vs. LLX vs. LLX vs. LLX + vs. Variable Treatment Mean n p value Block Eye + Eye + Nose + Eye Nose + Electro- Block Block Block Block Eye Block sensory Block Control 0.2±0.2 7 <0.001 N.S. <0.001 N.S. N.S. N.S. 0.024 N.S. Nose Block 50.9±12.2 7 <0.001 N.S. 0.001 N.S. 0.001 N.S. 0.003 Eye Block 0.0±0.0 7 <0.001 N.S. N.S. N.S. 0.022 N.S. Nose + Eye Block 75.4±37.2 5 <0.001 N.S. <0.001 N.S. <0.001 Rest Time LLX 0.0±0.0 6 <0.001* N.S. N.S. 0.030 N.S. (%) LLX + Nose Block 29.5±.12.0 5 N.S. N.S. N.S. LLX + Eye Block 0.0±0.0 6 0.030 N.S. LLX + Nose + Eye Block 41.1±15.5 5 N.S. Electrosensory Block 3.2±3.2 6 Control 145.5±0.0 7 Nose Block 78.3±14.6 7 Eye Block N.A. 7 Nose + Eye Block 53.4±17.2 5 Rest Angle LLX N.A. 6 0.635 (°) LLX + Nose Block 94.9±34.5 5 LLX + Eye Block N.A. 6 LLX + Nose + Eye Block 84.4±9.5 5 Electrosensory Block 167.8±0.0 6 Control 100.0±0.0 7 <0.001 N.S. <0.001 N.S. <0.001 N.S. <0.001 N.S. Nose Block 20.2±10.6 7 <0.001 N.S. <0.001 <0.001 <0.001 N.S. <0.001 Eye Block 100.0±0.0 7 <0.001 N.S. <0.001 N.S. <0.001 N.S. Orientation Nose + Eye Block 8.3±8.3 5 <0.001 N.S. N.S. N.S. <0.001 Rate LLX 100.0±0.0 6 <0.001* <0.001 N.S. <0.001 N.S. (%) LLX + Nose Block 46.7±0.0 5 <0.001 <0.001 <0.001 LLX + Eye Block 100.0±0.0 6 <0.001 N.S. LLX + Nose + Eye Block 20.0±12.2 5 <0.001 Electrosensory Block 100.0±0.0 6 Control 12.7±1.4 7 Nose Block 13.9±1.1 7 Eye Block 15.2±2.2 7 Orientation Nose + Eye Block 17.7±0.0 5 Distance LLX 20.4±3.3 6 0.250 (cm) LLX + Nose Block 11.3±2.2 5 LLX + Eye Block 12.8±1.5 6 LLX + Nose + Eye Block 20.3±0.0 5 Electrosensory Block 28.6±6.5 6

150

(Table 3.2 Continued)

Tukey Test vs. Nose vs. vs. Nose vs. LLX vs. LLX vs. LLX vs. LLX + vs. Variable Treatment Mean n p value Block Eye + Eye + Nose + Eye Nose + Electro- Block Block Block Block Eye Block sensory Block Control 100.0±0.0 7 <0.001 N.S. <0.001 N.S. <0.001 N.S. <0.001 N.S. Nose Block 0.0±0.0 7 <0.001 N.S. <0.001 <0.001 <0.001 N.S. <0.001 Eye Block 100.0±0.0 7 <0.001 N.S. <0.001 N.S. <0.001 N.S. Nose + Eye Block 0.0±0.0 5 <0.001 0.010 <0.001 N.S. <0.001 Strike LLX 100.0±0.0 6 <0.001* <0.001 N.S. <0.001 N.S. Rate (%) LLX + Nose Block 36.7±0.10 5 <0.001 <0.001 <0.001 LLX + Eye Block 100.0±0.0 6 <0.001 N.S. LLX + Nose + Eye Block 0.0±0.0 5 <0.001 Electrosensory Block 100.0±0.0 6 Control 58.4±10.4 7 N.S. N.S. N.S. N.S. N.S. Nose Block N.A. 7 Eye Block 67.4±14.5 7 N.S. N.S. N.S. N.S. Nose + Eye Block N.A. 5 Strike LLX 35.8±5.1 6 0.007* N.S. N.S. N.S. angle (°) LLX + Nose Block 91.9±22.8 5 0.035 N.S. LLX + Eye Block 93.7±6.3 6 0.033 LLX + Nose + Eye Block N.A. 5 Electrosensory Block 38.4±10.5 6 Control 0.00±0.00 7 Nose Block N.A. 7 Eye Block 0.05±0.05 7 Nose + Eye Block N.A. 5 Misses LLX 0.04±0.04 6 0.073 LLX + Nose Block 0.13±0.13 5 LLX + Eye Block 0.00±0.00 6 LLX + Nose + Eye Block N.A. 5 Electrosensory Block 0.47±0.26 6 Control 100.0±0.0 7 N.S. N.S. <0.001 N.S. N.S. Nose Block N.A. 7 Eye Block 100.0±0.0 7 N.S. <0.001 N.S. N.S. Capture Nose + Eye Block N.A. 5 Success LLX 100.0±0.00 6 <0.001* <0.001 N.S. N.S. Rate (%) LLX + Nose Block 45.8±4.17 5 <0.001 LLX + Eye Block 100.0±0.0 6 N.S. LLX + Nose + Eye Block N.A. 5 Electrosensory Block 94.44±5.56 6

151

(Table 3.2 Continued)

Tukey Test vs. Nose vs. vs. Nose vs. LLX vs. LLX vs. vs. LLX + vs. Variable Treatment Mean n p value Block Eye + Eye + Nose LLX + Nose + Electro- Block Block Block Eye Eye Block sensory Block Block Control 0.06±0.05 7 N.S. N.S. N.S. N.S. N.S. 0.007 <0.001 N.S. Nose Block 0.15±0.15 7 N.S. N.S. N.S. N.S. 0.012 <0.001 N.S. Frequency Eye Block 0.53±0.15 7 N.S. N.S. N.S. N.S. <0.001 N.S. of Wall Nose + Eye Block 1.36±0.60 5 N.S. N.S. 0.044 <0.001 N.S. Collisions LLX 0.34±0.34 6 <0.001* N.S. N.S. N.S. N.S. (collisions/ LLX + Nose Block 0.00±0.00 5 0.007 <0.001 N.S. 100s) LLX + Eye Block 1.81±0.28 6 0.001 0.006 LLX + Nose + Eye Block 4.01±0.56 5 <0.001 Electrosensory Block 0.00±0.00 6

Summary of all variables for the nurse shark, Ginglymostoma cirratum with all senses intact, and following blocks of the senses as indicated. LLX: Lateral line system blocked. All means are ±s.e.m. The p values are the results of ANOVA or Kruskal-Wallis tests performed on each variable. Value marked (*) are significant after Benjamini-Hochberg corrections. Tukey Test p values reflect the results of pairwise post-hoc comparisons between treatments. N.A.: Not applicable, parameter was not assessed because behavior did not occur; N.S.: Not significant at α = 0.05.

152

Table 3.3. Data Summary – Bonnetheads

Tukey Test Variable Treatment Mean N p value vs. Nose vs. Eye vs. Nose + vs. vs. LLX + vs. LLX + vs. Electro- Block Block Eye Block LLX Nose Block Eye Block sensory Block Control 0.69±0.02 14 N.S. N.S. N.S. N.S. N.S. N.S. Nose Block 0.79±0.04 5 N.S. N.S. N.S. 0.002 N.S. Eye Block 0.71±0.00 4 N.S. N.S. N.S. N.S. Swim Nose + Eye Block 0.74±0.00 2 N.S. N.S. N.S. Velocity 0.002* LLX 0.74±0.05 4 0.022 N.S. (BL/s) LLX + Nose Block N.R. 3 LLX + Eye Block 0.49±0.03 3 0.007 Electrosensory Block 0.75±0.06 6 Control 137.4±5.7 14 <0.001 N.S. <0.001 0.007 <0.001 N.S. Nose Block 47.2±4.9 5 <0.001 N.S. <0.001 <0.001 <0.001 Eye Block 128.0±0.0 4 <0.001 N.S. N.S. N.S. Turn Nose + Eye Block 38.7±0.0 2 <0.001 <0.001 <0.001 Velocity <0.001* LLX 98.7±3.6 4 N.S. 0.002 (°/s) LLX + Nose Block N.R. 3 LLX + Eye Block 100.3±13.0 3 0.008 Electrosensory Block 148.7±4.6 6 Control 0.81±0.04 14 <0.001 N.S. <0.001 N.S. 0.001 N.S. Nose Block 0.13±0.01 5 <0.001 N.S. <0.001 <0.001 <0.001 Eye Block 0.85±0.00 4 <0.001 N.S. N.S. N.S. Turn Nose + Eye Block 0.10±0.00 2 <0.001 0.003 <0.001 Frequency <0.001* LLX 0.69±0.08 4 N.S. N.S. (turns/s) LLX + Nose Block N.R. 3 LLX + Eye Block 0.47±0.07 3 0.010 Electrosensory Block 0.77±0.03 6 Control 74.3±18.0 14 N.S. N.S. Nose Block N.A. 5 Eye Block N.A. 4 Tracking Nose + Eye Block N.A. 2 0.030 Time (s) LLX 17.74±6.1 4 N.S. LLX + Nose Block N.A. 3 LLX + Eye Block N.A. 3 Electrosensory Block 41.8±23.5 6

153

(Table 3.3 Continued)

Tukey Test Variable Treatment Mean n p value vs. Nose vs. Eye vs. Nose + vs. vs. LLX + vs. LLX + vs. Electro- Block Block Eye Block LLX Nose Block Eye Block sensory Block Control 73.8±10.7 14 <0.001 N.S. N.S. Nose Block 165.7±31.8 5 N.S. N.S. Eye Block N.A. 4 Orientation Nose + Eye Block N.A. 2 N.S. Distance 0.011* LLX 116.6±16.0 4 (cm) LLX + Nose Block N.R. 3 LLX + Eye Block N.A. 3 Electrosensory Block 106.7±22.8 6 Control 100.0±0.0 14 N.S. <0.001 <0.001 N.S. N.S. <0.001 N.S. Nose Block 100.0±0.0 5 <0.001 <0.001 N.S. N.S. <0.001 N.S. Eye Block 0.0±0.0 4 N.S. <0.001 <0.001 N.S. <0.001 Strike Rate Nose + Eye Block 0.0±0.0 2 <0.001 <0.001 <0.001 <0.001 <0.001 (%) LLX 100.0±0.0 4 N.S. <0.001 N.S. LLX + Nose Block 100.0±0.0 3 <0.001 <0.001 LLX + Eye Block 0.0±0.0 3 <0.001 Electrosensory Block 100.0±0.0 6 Control 0.75±0.03 14 Nose Block 0.84±0.04 5 Eye Block N.A. 4 Strike Nose + Eye Block N.A. 2 Velocity 0.034 LLX 0.79±0.05 4 (BL/s) LLX + Nose Block N.R. 3 LLX + Eye Block N.A. 3 Electrosensory Block 0.95±0.08 6 Control 28.1±5.1 14 Nose Block 26.5±8.9 5 Eye Block N.A. 4 Strike Nose + Eye Block N.A. 2 0.940 Angle (°) LLX 30.2±5.5 4 LLX + Nose Block N.R. 3 LLX + Eye Block N.A. 3 Electrosensory Block 25.6±5.6 6

154

(Table 3.3 Continued)

Tukey Test Variable Treatment Mean n p value vs. Nose vs. Eye vs. Nose + vs. vs. LLX + vs. LLX + vs. Electro- Block Block Eye Block LLX Nose Block Eye Block sensory Block Control 0.04±0.04 14 N.S. N.S. N.S. <0.001 Nose Block 0.00±0.00 5 N.S. N.S. <0.001 Eye Block N.A. 4 Nose + Eye Block N.A. 2 Misses <0.001* LLX 0.25±0.14 4 N.S. 0.019 LLX + Nose Block 0.25±0.25 3 0.022 LLX + Eye Block N.A. 3 Electrosensory Block 35.4±11.7 6 Control 100.0±0.0 14 N.S. N.S. N.S. <0.001 Nose Block 100.0±0.0 5 N.S. N.S. <0.001 Eye Block N.A. 4 Capture Nose + Eye Block N.A. 2 Success <0.001* LLX 100.0±0.0 4 N.S. <0.001 Rate (%) LLX + Nose Block 100.0±0.0 3 <0.001 LLX + Eye Block 100.0±0.0 3 Electrosensory Block 0.0±0.0 6 Control 0.00±0.00 14 N.S. <0.001 <0.001 <0.001 <0.001 N.S. Nose Block 0.00±0.00 5 <0.001 <0.001 N.S. <0.001 N.S. Frequency Eye Block 5.17±0.00 4 <0.001 <0.001 N.S. <0.001 of Wall Nose + Eye Block 2.67±0.00 2 <0.001 <0.001 <0.001 Collisions <0.001* LLX 0.05±0.05 4 <0.001 N.S. (collisions/ LLX + Nose Block N.R. 3 100s) LLX + Eye Block 4.89±0.06 3 <0.001 <0.001 Electrosensory Block 0.03±0.03 6

Summary of all variables for the bonnethead, Sphyrna tiburo with all senses intact, and following blocks of the senses as indicated. LLX: Lateral line system blocked. All means are ±s.e.m. The p values are the results of ANOVA or Kruskal-Wallis tests performed on each variable. Value marked (*) are significant after Benjamini-Hochberg corrections. Tukey Test p values reflect the results of pairwise post-hoc comparisons between treatments. N.A.: Not applicable, parameter was not assessed because behavior did not occur; N.R.: Not recorded due to technical difficulties; N.S.: Non-significant at α = 0.05.

155

Table 3.4. Strike Velocities for Successful Captures vs. Misses – Blacktip Sharks

Treatment Outcome Mean Strike Velocity (BL/s) p value Success 1.17 ±0.26 Lateral Line Block 0.031 Miss 3.62 ±1.10 Success 1.31 ± 0.12 Electrosensory block N.S. Miss 1.67 ± 0.39

Comparison of the strike velocities for successful captures vs. misses in the blacktip shark, Carcharhinus limbatus with the lateral line or electrosensory system blocked. Velocities are in body lengths/s (BL/s). All means are ±s.e.m. The p values are the result of t-tests for dependent samples. N.S.: non-significant at α = 0.05.

156

Figures

Figure 3.1. Diagram of the Multi-modal Stimulus Field. Summary of the hypothetical stimulus fields emitted by a biological target (small dark gray circle) in an unbounded, laminar flow environment. In the natural world, any number of environmental, physical, or biological variables could attenuate any of these sensory inputs to the elasmobranch. In very clear, well-lit waters, the visual stimulus could range much farther than depicted and the acoustic regime is frequency dependent, such that low frequency sounds will extend over a greater range, possibly even as far as olfaction, and the near/far field boundary will be found at a greater distance from the source. Green: odor and wake dispersal fields; Blue: acoustic fields; Purple: target visibility range; Red: electric field. (Modified from Atema, In Press)

157

Figure 3.2. Experimental Flume. Diagram is drawn to scale, with a ~1m bonnethead shark in the test arena. The dot represents the location of the prey only and is not representative of size.

158

Figure 3.3 Strike Angle Measurement. A hypothetical path to the prey is shown. The dotted line indicates tracking, while the solid grey line indicates the strike. The strike angle, Ɵ, is measured as the orientation of the animal’s body to the prey, with respect to the direction of flow.

159

Figure 3.4. Swim Velocity. The swimming velocity, during the tracking phase, in three species of sharks, the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo, with all senses intact (control) and following blocks of the sensory systems indicated in the figure legend (LL: lateral line). Velocities are in body lengths/s (BL/s). Data are shown for all treatments. Error bars are ± s.e.m. * denotes treatments that are significantly different from control at α = 0.05; for comparisons among treatments, see Tables 3.1 – 3.3.

160

Figure 3.5. Turn Velocity. The turning velocity, during the tracking phase, in three species of sharks, the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo, with all senses intact (control) and following blocks of the sensory systems indicated in the figure legend (LL: lateral line). Velocities are in degrees/s (°/s). Data are shown for all treatments. Error bars are ± s.e.m. * denotes treatments that are significantly different from control at α = 0.05; for comparisons among treatments, see Tables 3.1 – 3.3.

161

Figure 3.6. Turn Frequency. The frequency of turns (turns/s), during the tracking phase, in three species of sharks, the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo, with all senses intact (control) and following blocks of the sensory systems indicated in the figure legend (LL: lateral line). Velocities are in degrees/s (°/s). Data are shown for all treatments. Error bars are ± s.e.m. * denotes treatments that are significantly different from control at α = 0.05; for comparisons among treatments, see Tables 3.1 – 3.3.

162

Figure 3.7. Standardized Tracking Time. Time spent performing tracking behavior, standardized based on the distance between predator and prey at the onset of tracking, in three species of sharks, the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo. Data are shown for animals with all senses intact (control) and following blocks of the sensory systems, as indicated (LL: lateral line). Treatments in which tracking behavior was absent were omitted from the figure. Error bars are ± s.e.m. * denotes treatments that are significantly different from control at α = 0.05; for comparisons among treatments, see Tables 3.1 – 3.3.

163

Figure 3.8. Rest Time. The proportion of the trial time spent resting in the nurse shark, Ginglymostoma cirratum, in animals with all senses intact (control) and following blocks of the sensory systems as indicated (LL: lateral line). No resting occurred in the blacktip shark, Carcharhinus limbatus, or the bonnethead, Sphyra tiburo, as these species are obligate ram-ventilators. Error bars are ± s.e.m. * denotes treatments that are significantly different from control at α = 0.05; for comparisons among treatments, see Tables 3.1 – 3.3.

164

Figure 3.9. Orientation Distance. The distance between the predator and the prey at orientation, in cm, in three species of sharks, the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo in animals with all senses intact (control) and following blocks of the sensory systems as indicated (LL: lateral line). Treatments in which orientation did not occur have been omitted. Error bars are ± s.e.m. * denotes treatments that are significantly different from control at α = 0.05; for comparisons among treatments, see Tables 3.1 – 3.3.

165

Figure 3.10. Striking. Percentage of trials that result in a strike in three species of sharks, the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo in animals with all senses intact (control) and following blocks of the sensory systems as indicated (LL: lateral line). Error bars are ± s.e.m. * denotes treatments that are significantly different from control at α = 0.05; for comparisons among treatments, see Tables 3.1 – 3.3.

166

Figure 3.11. Strike Velocity. The velocity of the strikes, in body lengths/s, in the blacktip shark, Carcharhinus limbatus, and the bonnethead, Sphyrna tiburo, in animals with all senses intact (control) and following blocks of the sensory systems as indicated (LL: lateral line). Strike velocity in the nurse shark, Ginglymostoma cirratum, was not examined due to the very short distance over which strikes occurred (see Figure 3.9). Treatments in which no strikes occurred were omitted. Error bars are ± s.e.m. * denotes treatments that are significantly different from control at α = 0.05; for comparisons among treatments, see Tables 3.1 – 3.3.

167

Figure 3.12. Number of Misses. The number strikes per trial that did not result in successful capture in three species of sharks, the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo, in animals with all senses intact (control) and following blocks of the sensory systems as indicated (LL: lateral line). Treatments in which no strikes occurred were omitted. The y axis is truncated in order to allow the reader to see the differences between the other treatments. The values of the values that exceed the y-axis, ± SEM, are indicated next to the respective bars in parentheses. * denotes treatments that are significantly different from control at α = 0.05; for comparisons among treatments, see Tables 3.1 – 3.3.

168

Figure 3.13. Successful Strike vs. Miss Velocity. Comparison of the velocity, in body lengths/s (BL/s), of the strikes that resulted in successful capture (light grey) vs. misses (dark grey) for the electrosensory and lateral line blocked treatments in the blacktip shark, Carcharhinus limbatus. Error bars are ± s.e.m. * denotes significant differences between misses and captures α = 0.05.

169

Figure 3.14. Capture Success. The percentage of trials which resulted in successful capture in three species of sharks, the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo. Data are shown for animals with all senses intact (control) and following blocks of the sensory systems as indicated (LL: lateral line). Treatments in which no strikes occurred were omitted. Error bars are ± s.e.m. * denotes treatments that are significantly different from control at α = 0.05; for comparisons among treatments, see Tables 3.1 – 3.3.

170

Figure 3.15. Frequency of Wall Collisions. The frequency of collisions with the walls during each trial, expressed as collision/s, in three species of sharks, the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo. Data are shown all treatments, animals with all senses intact (control) and following blocks of the sensory systems as indicated (LL: lateral line). Error bars are ± s.e.m. The * denotes treatments that are significantly different from control at α = 0.05; for multiple comparisons, see Tables 3.1 – 3.3. * denotes treatments that are significantly different from control at α = 0.05; for comparisons among treatments, see Tables 3.1 – 3.3.

171

CHAPTER 4: SENSORY CONTRIBUTIONS TO PREY CAPTURE

KINEMATICS IN SHARKS

Abstract

Recent work on teleosts suggests that the kinematics of prey capture may be modified in response to changes in pre-strike sensory information. This study is the first to examine these effects in sharks. The feeding modulation hypothesis suggests that ram feeding and biting sharks are capable of a greater degree of modulation than suction feeding sharks, which have highly stereotyped capture events. In this study, three species of sharks that have different feeding modalities were filmed with high-speed videography while feeding on live prey: the ram feeding blacktip shark, the suction feeding nurse shark, and the ram-biting bonnethead. The sharks were examined intact and after deprivation of information from the senses (olfaction, vision, mechanoreception, and electroreception), alone and in combination, to elucidate their contributions to the kinematics of prey capture. In response to sensory deprivation, the blacktip shark demonstrated the greatest amount of modulation, followed by the nurse shark. Little to no modulation was found in the bonnethead. The blacktip shark aligns its strikes from a distance using vision, but fine tunes the strike just prior to capture using the lateral line.

When deprived of olfactory cues, blacktip sharks demonstrate slower jaw opening, suggesting that they are less motivated when odor cues are unavailable. When deprived of vision, both the blacktip shark and nurse shark simultaneously decrease the amount of 172

ram while increasing the amount of suction, however, unlike bony fish, they do not

switch feeding modalities. These results suggest that prey capture is less plastic in

elasmobranchs than in bony fishes, possibly due to anatomical differences, and that while

suction feeders are generally more stereotyped than ram feeders, modulatory ability

varies by species.

Introduction

Feeding behavior culminates with the capture and ingestion of prey. Prey capture

has been rather well-studied in sharks. When a shark is within close range of the prey, it

begins the capture sequence, which is typically very rapid, lasting from about 50-400

milliseconds (reviewed in Motta, 2004). Capture begins when the mouth starts to open

and lasts until the prey is grasped between the teeth or the jaws are closed on the prey

(Motta et al., 2002; Motta, 2004). Sharks capture prey in a variety of ways, such as ram, suction, and biting, with ram feeding being the most common (Motta and Wilga, 2001;

Motta, 2004).

In aquatic animals, the viscosity and density of water dictate that suction is always generated to some degree when an aquatic animal opens its mouth, as the expansion of

the buccal cavity causes the flow of fluid towards the animal, which draws the prey towards its mouth. The amount of suction generated depends on the rate of expansion of the buccal cavity, including the speed of jaw opening and the magnitude of the gape, which result in higher rates of this fluid flow, and the shape of the mouth, with more tubular mouths directing the suction in front of the mouth (Liem, 1993; Carroll et al.,

173

2004; Higham et al., 2006a; Higham et al., 2006b). In sharks, specializations for suction

feeding include rapid buccopharyngeal expansion, small mouths and teeth, hypertrophied

abductor muscles, and large protractible labial cartilages that laterally occlude the mouth

(Motta and Wilga, 2001; Wilga et al., 2007). In pure suction feeding, the predator

remains completely stationary as it rapidly expands the buccal cavity to draw the prey

into its mouth. Most fishes also employ ram, or forward motion of the body or jaws, to some degree (Norton, 1991a; Norton and Brainerd, 1993; Ferry-Graham and Lauder,

2001; Higham et al., 2005). In pure ram feeding, the predator uses a rapid, whole body acceleration to overtake and engulf the completely stationary prey (Liem, 1980a; Norton

and Brainerd, 1993). Ram feeding sharks typically have larger gapes than suction

feeding sharks (Motta and Wilga, 2001; Wilga et al., 2007). These two behaviors are

regarded as a functional continuum and most fishes use a combination of these two

behaviors, falling somewhere in the middle of the spectrum (Norton, 1991a; Norton and

Brainerd, 1993; Ferry-Graham and Lauder, 2001; Higham et al., 2005). In sharks, biting

may accompany ram feeding. In biting, rather than completely engulfing the prey, the

shark will bite into the prey, or bite pieces off of the prey (Motta and Wilga, 2001; Motta,

2004; Wilga et al., 2007).

Prey capture can be divided into three continuous phases. During the expansive

phase, which extends from the start of jaw opening to peak gape, there is cranial

elevation and hyoid and lower jaw depression, as well as branchial expansion. The labial

cartilages at the edges of the mouth extend. During the compressive phase, which

extends from peak gape until the jaws close, the lower jaw elevates and cranial

depression may occur. In most elasmobranchs, protrusion of the upper jaws occurs

174 during the compressive phase. This is in contrast to bony fishes which protrude the jaws during the expansive phase. During the compressive phase, the prey is either engulfed or it is caught between the teeth as the jaws close. The recovery phase extends from jaw closure to the time when all of the cranial elements have returned to their initial (resting) positions. It is during this time that the upper jaw is retracted in most elasmobranchs

(Liem, 1979, 1980b; Lauder, 1985; Motta and Wilga, 2001; Motta, 2004; Wilga et al.,

2007).

Animals are guided to their food by sensory information. It is generally believed that pre-strike sensory information, gathered on the approach to the prey, contributes to the determination of the motor pattern employed for prey capture. A predator perceives the nature, size, and behavior of its prey, then uses a behavior appropriate to the prey type in order to increase the efficiency of capture (Norton, 1991a; Ferry-Graham, 1997;

Nemeth, 1997a, b; Ferry-Graham, 1998a; Ajemian and Sanford, 2007; Lowry and Motta,

2007). Until recently, the effects of changes in pre-strike sensory information on the kinematics of capture had not been considered. The largemouth bass, Micropterus salmoides, a well-described ram feeder, was found to modulate the kinematics of its prey capture in response to sensory deprivation. This animal switches towards suction-based feeding when visual cues are lacking (Gardiner and Motta, In Press). Blind animals approach prey slowly and strike from a closer proximity, but open their mouths more rapidly, which has been shown to result in greater buccal pressure (Sanford and

Wainwright, 2002; Svanback et al., 2002) and higher peak fluid speeds (Day et al., 2005), causing their prey to move a greater distance at a more rapid velocity as they are being drawn into the predators’ mouths. Prey capture in the absence of vision is mediated by

175

the lateral line, as the combination of lateral line and visual lesions results in the cessation

of feeding behaviors altogether (New, 2002). When deprived of lateral line cues, bass

have higher forward velocities during capture and capture prey earlier in the gape cycle.

Since braking just prior to capture is believed to increase accuracy (Higham et al., 2005;

Higham et al., 2006a), the lateral line appears to add precision to the strikes.

The effects of sensory deprivation on prey capture kinematics have not been

examined in suction feeders. Specialization for suction feeding apparently evolved

independently in conjunction with a benthic lifestyle, and these suction specialists feed

on both elusive and non-elusive prey that live in or on the substrate, are attached to it, or

are associated with the bottom. Suction feeding sharks have been described as having

highly stereotyped capture kinematics that show little or no modulation once the capture

sequence is initiated (Motta and Wilga, 2001; Motta et al., 2002; Robinson and Motta,

2002; Matott et al., 2005), however, some suction feeding teleosts modulate capture in

response to different prey types (Ferry-Graham et al., 2001b; Van Wassenbergh et al.,

2006). Other suction feeding teleosts have been found to modulate in response to missed

prey or prey escapes, presumably as a result of sensory feedback, or a lack thereof, during

capture (Aerts, 1990; Van Wassenbergh and De Rechter, 2011). The feeding modulation

hypothesis states that ram feeding and biting carcharhinid and lamnid sharks have greater

modulatory capabilities than suction feeding sharks (Motta and Wilga, 2001), however,

only a few studies have specifically examined modulation in prey capture kinematics in sharks, by varying prey size, type, or elusivity (Ferry-Graham, 1997; Ferry-Graham,

1998b; Ferry-Graham, 1998a; Matott et al., 2005; Ajemian and Sanford, 2007)

176

Furthermore, sensory involvement in capture kinematics has not yet been considered in

elasmobranchs.

The goal of this study is to examine contributions of sensory information to the

kinematics of prey capture in three species of sharks that use different prey capture

strategies. The blacktip shark, Carcharhinus limbatus, is a fast-swimming epibenthic

shark that ram feeds on highly elusive prey such as bony fish and elasmobranchs from the water column (Frazzetta and Prange, 1987; Castro, 1996). The bonnethead, Sphyrna tiburo, is an epibenthic species that uses ram-biting almost exclusively, capturing crustaceans by scooping them off the substrate (Cortés et al., 1996; Wilga and Motta,

2000; Bethea et al., 2007). The nurse shark, Ginglymostoma cirratum, is a benthic

species that suction feeds on fish, mollusks and crustaceans and has been described as

stalking and ambushing prey, primarily at night (Castro, 2000; Motta et al., 2002; Motta

et al., 2008).

Materials and Methods

Experimental animals

Eighteen young-of-the-year (YOY) blacktip sharks, Carcharhinus limbatus

(Müller and Henle 1839), 51-65cm total length, were collected from Terra Ceia Bay, FL

using rod and reel and gillnet gear. Ten juvenile nurse sharks, Ginglymostoma cirratum

(Bonnaterre 1788), 67-94cm total length, were collected from the waters near Long Key,

FL using rod and reel gear. Sixteen bonnetheads, Sphyrna tiburo (Linnaeus 1758), 69-

95cm total length, were collected from Terra Ceia Bay, FL, and the waters near Sarasota,

FL using gillnet gear. All animals were transported to Mote Marine Laboratory in

177

Sarasota, FL in a transport container equipped with a life support system, and

subsequently held in a 210,000L oval tank operated on a closed recirculating life support

system with sand filtration and heater/chiller units. The tank was maintained at

spring/summer conditions of daylight (12-14 hours) and temperature (24-26°C). Animals

were fed fish, shrimp, and squid, supplemented with Mazuri Vita-Zu Sharks/Rays

Vitamin Supplement Tablets (PMI Nutrition International, St Louis, MO, USA), to

satiation three times per week, except during periods of experimentation, when food was

withheld for 48 hours prior to behavioral trials to ensure that the animals were hungry.

Protocols for animal handling and use were approved by the Institutional Animal Care

and Use Committees at the University of South Florida (W3817) and Mote Marine

Laboratory (11-03-RH1).

Behavioral procedures

Experiments were conducted in a near-laminar flow channel (flume) which was

constructed within the 210,000 liter oval tank: working area was 7.5 m long x 2 m wide,

filled to 120 cm depth, with a flow rate of 2.3 cm/s (Figure 4.1). A 4m x 2m holding area

at the downstream end allowed for animal containment behind a mesh gate. As per

Gardiner and Atema (2007), for each trial, an individual animal was moved into the

flume channel, allowed to acclimate for 30 minutes, then offered a small piece of food to

confirm that it was hungry. The animal was then herded into the holding area. A live,

moving prey item from the natural diet of each species (nurse and blacktip sharks:

pinfish, Lagodon rhomboides (Bigelow and Schroeder, 1948; Castro, 1996, 2000); bonnethead: pink shrimp, Farfantepenaeus duorarum (Cortés et al., 1996)) was tethered,

178 injured from tether attachment, at the upstream end of the flume using a piece of thin, degradable, cotton thread, to restrict it to the area in front of a window in the side of the tank (Figure 4.1). Within each species, the prey items were size-matched to the total length of the predator and prey size was consistent across trials. Prey were suspended midwater (approximately 60cm above the bottom) for the blacktip sharks, approximately

30cm above the bottom for the bonnetheads, and just above the bottom (approximately

10cm for the nurse sharks. The shark was held in the start box for six minutes to allow a plume of sensory cues emanating from the prey to establish along the length of the flume channel. The shark was then released and a trial proceeded for 10 minutes or until the prey was consumed, during which time the shark’s behavior was simultaneously monitored and filmed laterally at 250 frames/s using a Photron FASTCAM-X 1024 PCI

Model 100K camera (Photron USA Inc., San Diego, CA, USA) placed in front of the previously mentioned window in the tank wall (Figure 4.1). During all trials, only the facility’s overhead lights were used, no additional illumination of flume channel was needed for filming. Animals were tested intact, and following blocks of each of the sensory systems (outlined below), alone and in combination.

Sensory deprivation

Basil et al., 2000). To block vision, small pieces of heavy black plastic were glued around the margins of the eyes using cyanoacrylate glue. The sensitivity of the electrosensory

179 system was reduced by painting over the pores of the ampullae of Lorenzini with silicone-rubber paint (bonnetheads; Smooth-On Mold Max Stroke, Smooth-On Inc.,

Easton, PA USA) or with cyanoacrylate glue (blacktip and nurse sharks; The Original

Super Glue, Super Glue Corp., Rancho Cucamonga, CA USA). The location of the pores in all three species has previously been mapped (Kajiura, 2001; Cornett, 2006). Though only the pores were painted over, dripping of the glue or stretching of the silicone-rubber paint may have resulted in occlusion of some of the lateral line canal pores and/or superficial neuromasts on the head. Prior to use on animals, the insulating nature of these two materials was verified by covering one electrode on the prey-simulating electrical stimulus apparatus described in Kajiura and Holland (2002). The pair of electrodes was then immersed in seawater and a current of up to 200mA was applied; no current was detected at the multimeter, indicating that the paint and glue break the electrical circuit and are therefore insulating. All of these blocks were applied while the animal was under anesthesia with tricaine methanesulfonate (MS-222), with a dose of 100mg/L in buffered seawater for induction and 50mg/L for maintenance. Animals were ventilated using a hose attached to a small recirculating pump while the blocks were applied, then revived using fresh seawater. Animals were allowed to recover in approximately 1000L of seawater in a 244cm diameter round tank for three hours, then moved to the flume channel and allowed to acclimatize for a further 30 minutes as above, prior to a behavioral trial.

180

1000L of this solution in the 244cm diameter round tank. For combinations of sensory

blocks, those requiring anesthesia were first applied, then the animal was moved to the

recovery round tank where it was held until it had recovered sufficiently to swim and navigate the tank normally. The streptomycin sulfate was added to the water and the animal was held in this solution for three hours as described above, prior to being moved to the flume channel for behavioral testing. Streptomycin is an ototoxic antibiotic that has been shown to lesion both the surface neuromasts and canal neuromasts in teleosts

(Wersall and Flock, 1964; Kaus, 1987; Montgomery et al., 1997; Faucher et al., 2006a;

which may be caused by interference with the motion of the sensory hairs (Kroese and

van den Bercken, 1982). It does not affect inner ear function unless applied

intraluminally (Matsuura et al., 1971). The duration of the effects of this drug are not

completely understood. Since teleosts treated with this drug return to normal behavior in

20–24 hours (Blaxter and Fuiman, 1989), all lateral line blocked trials were completed

within 6 hours of application of the drug. However, since physical damage to the hair

cells has been found on scanning electron micrographs of streptomycin-treated neuromasts (Faucher et al., 2006a; Faucher et al., 2006b; Faucher et al., 2010), following lateral line lesion treatments, animals were allowed to recover for a minimum of four weeks prior to any other behavioral testing to allow time for the neuromasts to regenerate

(Coombs et al, 2001; Faucher et al, 2006). While hearing with the inner ear is likely contributing to prey localization in sharks, the stimulus field in a closed tank environment

181

is very difficult to control due to echoing off the walls, bottom and surface of the tank,

and thus the sharks’ sense of hearing was not specifically tested in this study.

Video analysis

Only those trials in which capture occurred laterally to the camera were used for analysis. Any feeding sequences in which the prey attempted to escape were excluded from analysis. Videos were digitized using MaxTRAQ Lite+ v.2.2.2.2 software

(Innovision Systems Inc., Columbiaville, MI, USA) and analyzed by multiple readers.

The striking and capture phases of feeding behavior (Chapter 3) were examined. Striking in ram-feeding bony fish is defined as a direct, rapid whole-body acceleration towards the prey, often using an S-start (Nyberg, 1971; New and Kang, 2000; New, 2002). In

ambush predators, such as the Pacific angel shark, Squatina californica, striking begins in

a relatively stationary animal with the raising of the head and a short lunge (Fouts and

Nelson, 1999). In fishes, capture begins with the onset of jaw depression (Ferry-Graham

and Lauder, 2001; Motta and Wilga, 2001; Motta, 2004) and for comparative purposes, in this study was defined as ending when the center of mass of the prey had passed the anterior margin of the mouth. The following kinematic variables were examined. In the final moments of the striking phase, just prior to the onset of capture, (1) the approach velocity = the velocity of the shark in body lengths/s (BL/s) was measured over the last

40-100ms prior to the start of capture. Capture was considered to begin with the start of

jaw depression (t0). The following capture timing variables were measured in ms: (2)

time to peak gape = from t0 to peak gape; (3) duration of peak gape = from peak gape to

the start of jaw closure; (4) time to close jaws = from the start of closure until the gape is

182

closed; (5) total bite time = from t0 to jaw closure; (6) onset of jaw protrusion = from t0 to

the start of jaw protrusion; (7) maximum jaw protrusion = from the onset of jaw

protrusion to peak protrusion; (8) duration of maximum jaw protrusion = from maximum

jaw protrusion to the start of jaw retraction; (9) time to retract upper jaw = from the start

of retraction until the upper jaw is completely retracted; (10) capture time = from t0 until the last portion of the prey crossed the tip of the jaws of the predator (tcapture). The following variables were measured: (11) predator-prey distance at the start of capture = distance between anterior end of the lower jaw of the predator and the estimated center of mass of the prey at t0 ,in cm; (12) capture angle = the angle between the jaws of the

predator (from the corner of the mouth to the the midpoint between the anterior end of the

upper and lower jaw) and the estimated center of mass of the prey at t0 , in °; (13) peak

gape measured between the anterior upper and lower jaw teeth, in cm; (14) gape angle, at

maximum gape, in °; (15) maximum jaw protrusion distance (protrusion of the palatoquadrate), as measured from the underside of the chondrocranium to the tip of the teeth, in cm; (16) ram distance = distance moved by the predator in the earthbound frame

(Dpredator, in cm) from t0 to tcapture; (17) suction distance = the distance moved by the prey

in the earthbound frame (Dprey, in cm) from t0 to tcapture; (18) predator velocity (cm/s)

during capture, from t0 to tcapture; (19) prey velocity (cm/s) during capture, from t0 to

tcapture. These variables were also used to calculate (20) the ram-suction index (RSI),

given by:

RSI = (Dpredator – Dprey)/(Dpredator + Dprey).

This index varies from +1, a pure ram strike, to -1, a pure suction strike (Norton

and Brainerd, 1993). The (21) frequency of bites, as a percentage, was also examined. A

183

bite is considered as any capture event in which the prey is caught in the teeth, rather than

engulfed (Motta and Wilga, 2001; Motta, 2004). In the blacktip shark, (22) onset of

cranial elevation = from t0 to the start of cranial elevation; (23) maximum cranial

elevation = from the start of cranial elevation to the maximum cranial angle, as measured

on the dorsal side of the animal from just anterior to the dorsal fin, to the posterior end of

the chondrocranium, to the tip of the snout; and (24) maximum head elevation = change

in angle from start of cranial elevation to maximum elevation, as measured on the dorsal side of the animal from just anterior to the dorsal fin, to the posterior end of the chondrocranium, to the tip of the snout, in °, were measured. Cranial elevation was not assessed in the nurse shark or bonnethead. Because of the position of the prey above the bottom, the nurse sharks lifted the head while orienting to the prey (Chapter 3), and thus

cranial elevation during capture could not be distinguished from orientation.

Bonnetheads only elevate the head slightly during capture (Wilga and Motta, 2000). The

bonnetheads in this study struck at the prey from below, and thus this slight cranial head

elevation, if present, was difficult to distinguish. For the nurse sharks, (25) start of labial

cartilage protrusion = from t0 to the start of labial cartilage protrusion, in ms, (26) time to maximum labial protrusion = from the start of labial protrusion until the maximum protrusion distance was reached, (27) duration of maximum labial protrusion = from maximum protrusion to the start of retraction, in ms, (28) labial retraction time = from the start of labial retraction to complete retraction, in ms, and (29) maximum labial protrusion distance measure from the corner of the mouth to the anterior margin of the medial labial cartilage, in cm, were measured.

184

Statistical analysis

Rohlf, 1995), respectively. Florida Fish and Wildlife Conservation Commission regulations prohibit the release of any fishes that are held in captivity for more than 30 days or treated with any chemicals and so, in an effort to reduce the number of animals taken from the wild, individual animals were used in more than one, but not all, treatment groups. This design precludes the use of either repeated measures or non-parametric tests, as there are missing cells for each individual animal. Data for the different treatments were therefore compared within each species using one-way Analysis of Variance

(ANOVA). When significant differences were found, Tukey post hoc tests were then used to perform pairwise comparisons of the treatments. Data that failed the normality and/or equality of variance tests were analyzed using non-parametric Kruskal-Wallis one- way ANOVA on ranks and pairwise comparisons were made using Dunn’s Method when significant differences were found. The Benjamini-Hochberg method was used to control

the false discovery rate in multiple statistical tests (Benjamini and Hochberg, 1995). All

analyses were conducted using SigmaPlot 12.0 (Systat Software Inc., San Jose, CA,

USA).

185

Results

Approach Kinematics

In the blacktip sharks, strikes occurred in the control, olfactory, vision, lateral

line, lateral line + olfactory, and electrosensory blocked treatments only. No strikes

occurred in the olfaction + vision, lateral line + vision, and lateral line + olfaction +

vision blocked treatments (see Chapter 3). In the control treatment, during the final

moments of the strike, the blacktip sharks approached the prey at a velocity of 1.77 ±

0.16 BL/s (Table 4.1, Figure 4.2) and began to move the jaws at a distance of 5.59 ±

0.29cm from the prey (Table 4.1, Figure 4.3) and an absolute angle of 8.67 ± 0.64° to the

prey, with respect to the midline of the jaws (Table 4.1). When the lateral line was

blocked, the approach velocity was faster, 3.09 ± 0.75BL/s, and when vision was

blocked, the approach velocity was slower, 0.72 ± 0.16BL/s. When the lateral line +

olfactory systems were blocked, approach velocity was not significantly affected (Table

4.1, Figure 4.2). When the electrosensory system was blocked, capture began at a closer proximity to the prey, 3.52 ± 0.51cm (Table 4.1, Figures 4.3). Predator-prey distance at the start of capture was not significantly different among the other treatments. The capture angle was significantly greater when the lateral line, lateral line + olfaction, or electroreception were blocked, but largest when vision was blocked, such that the prey was positioned above or below the mouth. Blocking olfaction did not significantly affect the capture angle (Table 4.1).

In the nurse shark, strikes occurred in the control, vision, lateral line, lateral line + vision, and electrosensory blocked treatments. In the olfactory, vision + olfaction, lateral

line + olfaction, and lateral line + vision + olfaction blocked treatments either no strikes

186

occurred, or they occurred too rarely to be useful for analysis (see Chapter 3). In the final

moments of the strike, the control nurse sharks approached the prey at a velocity of 0.31

± 0.04BL/s (Table 4.2, Figure 4.2) and began to capture the prey at a distance of 4.03 ±

0.25cm (Table 4.2, Figure 4.3) and from an absolute angle of 31.92 ± 3.56° (Table 4.2).

Blocking vision, alone or in combination with the lateral line, resulted in an approach velocity that was significantly slower than that in the electrosensory or lateral line blocked animals, however, none of these treatments was significantly different from the control treatment. Approach velocity was not affected by any of the other treatments.

Predator-prey distance at the start of capture and capture angle did not vary among any of the treatments.

In the bonnethead, strikes occurred in the control, olfaction, lateral line block treatments, and electrosensory block treatments. No strikes occurred in the vision,

olfaction + vision, or lateral line + vision block treatments. In the electrosensory block

treatment, strikes occurred, but the animals failed to move the jaws and thus capture did

not occur in any of the animals in this treatment group (see Chapter 3). Control

bonnetheads approached the prey with a velocity of 0.70 ± 0.03BL/s and began capturing

prey from an angle of 14.17 ± 3.30° and a distance of 4.74 ± 0.16cm. Since no jaw

movement occurred in the electrosensory blocked treatment, approach velocity and

capture angle were assessed at the distance to the prey at which capture began in the

control treatment. Approach velocity, capture angle and predator-prey distance at the

start of capture did not vary significantly among any of the treatments (Table 4.3, Figures

4.2, 4.3).

187

Capture Kinematics

In the control blacktip sharks, capture began with lower jaw depression which was nearly synchronous with the onset of cranial elevation (-1.87 ± 4.03ms). As the jaw depressed, cranial elevation reached a maximum (47.13 ± 4.20ms). Upper jaw protrusion began (at 60.00 ± 5.02ms) as peak gape was reached (at 69.25 ± 5.65ms). Peak gape was maintained for 4.32 ± 0.28ms (until 73.57ms), then lower jaw began to elevate. The upper jaw reached maximum protrusion (at 114.95 ± 10.29 ms) as the jaws closed (at

116.49 ± 8.92ms). Capture occurred at 75.35 ± 5.41ms, as the jaws began to close (Table

4.1, Figure 4.4).

In the control nurse sharks, capture began with lower jaw depression, followed by the start of protrusion of the labial cartilages (at 21.19 ± 8.86ms). Peak gape was reached

(at 47.30 ± 8.28) as the labial cartilages reached peak protrusion (at 53.11 ± 3.56ms).

Peak gape was maintained for 18.55ms ± 4.56ms (until 71.66ms) and peak labial protrusion for 24.91 ± 3.72ms (until 78.02ms), then the lower jaw began to elevate and the labial cartilages were retracted. The labial cartilages reached full retraction in 27.70 ±

3.69ms (at 105.72ms) and the jaws closed in 56.42 ± 6.37ms (at 128.08ms). Capture occurred at 50.72ms, just after peak gape (Table 4.2, Figure 4.4).

In the control bonnetheads, capture began with lower jaw depression, which reached a maximum (in 113.92 ± 5.87ms), prior to the start of upper jaw protrusion (at

138.68 ± 12.03ms). Peak gape was maintained for 12.86 ± 2.00ms (until 126.79ms), then the lower jaw began to elevate, closing at 220.53 ± 9.02ms, just after peak protrusion was reached (at 200.53ms). Peak protrusion was maintained while the jaws closed (until

226.12ms) and the upper jaw was retracted after the gape had closed (at 329.83ms).

188

Capture occurred at 149.62 ± 6.50ms, as the jaws were closing (Table 4.3, Figure 4.4).

For all three species, the order of events was generally similar in the other treatments,

however, there were differences in the onset and/or duration of the different events, or the

magnitude of displacement, as outlined below (see also Tables 4.1-4.3).

Cranial Elevation

In the control blacktip sharks, cranial elevation began simultaneously with jaw depression, reaching a maximum displacement of 8.51 ± 0.80° after 47.13 ± 4.20ms.

Following blocks of the visual system, cranial elevation began before the onset of lower

jaw depression (-62.89 ± 17.37ms), reaching a greater angle, 15.28 ± 3.37°, over a longer

time, 104.89 ± 21.75ms. When the electrosensory system was blocked, cranial elevation

also began before the onset of lower jaw depression (-64.27 ± 84.40ms), however, these

differences were not significant, likely due to the large variation among animals in this

treatment group (Table 4.1, Figure 4.4). When the lateral line + olfactory systems were

blocked, the timing of the onset of cranial elevation and the maximum angle were not

significantly different, however, the time to maximum elevation was longer, 103.07 ±

24.57ms. The kinematics of cranial elevation were not significantly affected by the

olfactory or lateral line block treatments. The kinematics of cranial depression were not

evaluated because the blacktip sharks tended to turn away from the camera at the end of

capture, prior to complete cranial depression. As noted above (see Materials and

Methods), cranial elevation was not assessed for the nurse sharks and bonnetheads.

189

Gape Cycle

In the control blacktip sharks, the jaws reached their maximum gape, 5.43 ±

0.28cm, and gape angle, 96.71 ± 1.47°, in 69.25 ± 5.65ms, which was maintained for

4.32 ± 0.28ms. Lower jaw elevation lasted 43.9 ± 3.8ms, resulting in a total bite time of

116.49 ± 8.62ms. When the olfactory or lateral line + olfactory systems were blocked,

the time to peak gape was longer, 107.27 ± 8.52ms and 113.47 ± 15.71ms, respectively.

Time to peak gape was not significantly different among the other treatments. The

magnitude of the maximum gape, gape angle, duration of peak gape, time to close, and

total bite time were not significantly different among any of the treatments (Table 4.1,

Figure 4.5).

In the control nurse sharks, the jaws reached their peak gape, 3.24 ± 0.16cm, in

47.30 ± 8.28ms, which was maintained for 14.38 ± 6.72ms. Lower jaw elevation occurred over 56.42 ± 6.37ms, resulting in a total bite time of 121.17 ± 16.07ms. There

were no significant differences in the kinematics of the gape cycle among any of the

treatments (Table 4.2, Figure 4.5).

In the control bonnetheads, peak gape, 4.24 ± 0.17cm, with a gape angle of 4.24 ±

0.17°, was reached in 113.92 ± 5.87ms and maintained for 12.87 ± 2.00ms. Lower jaw

elevation occurred over 93.74 ± 4.57ms, resulting in a total bite time of 220.53 ± 9.02ms.

When the lateral line + olfaction was blocked, the duration of peak gape was significantly

longer, 40.22 ± 13.17ms. The duration of peak gape did not vary among any of the other

treatments. There were no significant differences in the magnitude of the peak gape,

gape angle, time to peak gape, time to close, or total bite time among any of the

treatments (Table 4.3, Figure 4.5).

190

Upper Jaw Protrusion and Labial Cartilage Protrusion

In the control blacktip sharks, as mentioned above, upper jaw protrusion began at

60.00 ± 5.02ms, at approximately the same time as the jaws reached peak gape. The

upper jaw protruded to a maximum displacement of 1.84 ± 0.16cm over 53.70 ± 3.69ms.

When the olfactory or lateral line + olfactory systems were blocked, upper jaw protrusion

began significantly later, at 98.61 ± 9.24ms and 103.73 ± 16.47ms, respectively, but still

in synchrony with peak gape, which was reached later in these treatments (see Gape

Cycle above, Table 4.1, Figure 4.4). The time to maximum jaw protrusion and jaw

protrusion distance were not significantly different among any of the treatments. Jaw

retraction was not assessed because the blacktip sharks tended to turn away from the

camera prior to complete retraction.

In the nurse shark, jaw protrusion was frequently obscured by the anteriorly

moving labial cartilages and was not assessed. The labial cartilages began to protrude

21.19 ± 8.86ms after the start of jaw depression, and reached their maximum displacement, 2.12 ± 0.22cm, over 31.92 ± 3.56ms. The labial cartilages remained at their peak protrusion for 24.91 ± 3.72ms and then retracted over 27.70 ± 3.69ms. When the electrosensory system was blocked, the labial cartilages were retracted significantly more slowly. There were no significant differences in the start of labial protrusion, time to maximum protrusion or duration of maximum among any of the treatments (Table

4.2).

In the bonnethead, upper jaw protrusion began after peak gape had been reached and as the jaws were closing, at 138.68 ± 12.03ms. The upper jaw reached its maximum

191

displacement, 0.92 ± 0.27cm, in 61.43 ± 5.10ms, which was maintained for 26.01 ±

3.61ms, then the upper jaw was retracted over 103.71 ± 17.86ms. There were no significant differences in the kinematics of jaw protrusion among any of the treatments.

Prey Capture and the Ram-Suction Index

In the control blacktip sharks, the RSI value was 0.77 ± 0.03, indicative of a

predominantly ram capture. Blocking the visual system resulted in a lower RSI value,

0.55 ± 0.05, though still predominantly ram capture. The RSI values did not differ

significantly among the other treatments (Table 4.1, Figure 4.6). This index, however,

tends to mask the source of changes in the overall behavior observed, as a higher RSI

value (i.e., more positive, thus indicative of more ram) can be achieved either by the

predator employing more forward motion (more ram) or by the prey moving a smaller

distance (less suction) and vice versa. Since the predator can modulate the amount of

ram and suction independently of one another, by changing the speed of its approach to

the prey or the speed with which it opens its jaws respectively, RSI value alone does not

explain why the observed behavior has changed. Thus, when differences in RSI are

found, it is important to also examine the distances moved by the predator and the prey in

order to clarify how the overall changes have been achieved (Gardiner and Motta, In

Press). The suction distance (distance moved by the prey) was not significantly different

among any of the treatments. The ram distance (distance moved by the predator),

however, was smaller when either the visual or electrosensory systems were blocked

(Table 4.1, Figure 4.7).

192

In the control blacktips, capture occurred at 74.68 ± 5.48ms, as the jaws began to

close. When olfaction was blocked, the time to capture was significantly longer, 114.97

± 8.33ms. There were no significant differences in the time to capture among the other

treatments (Table 4.1, Figure 4.4). The velocity of the prey as it entered the mouth of the

predator was not significantly different among any of the treatments (Table 4.1, Figure

4.8). The velocity of the predator during capture, however, was lower in the vision

blocked animals (63.8 ± 10.0cm/s) and higher in the lateral line blocked animals (189.9 ±

46.0) (Table 4.1, Figure 4.9). The control blacktip sharks typically overtook and

engulfed the prey, swallowing it whole in a ram feeding event. Ram-biting, in which the

prey is caught in the teeth, and subsequently drawn into the mouth in a transport event, was observed in 31.6 ± 7.0% of captures. The frequency of ram-bites was not significantly different among the treatments (Table 4.1, Figure 4.10).

In the control nurse sharks, the RSI value was -0.45 ± 0.10, indicative of

predominantly suction capture. When vision was blocked, the RSI value was

significantly lower, -0.84 ± 0.06, indicating that the blind nurse sharks were using more

suction (Table 4.2, Figure 4.6). The ram distance was significantly smaller in the vision

blocked animals, while the suction distance was significantly higher (Table 4.2, Figure

4.7). There were no significant differences in RSI, ram distance, or suction distance

among any of the other treatments. In the control nurse sharks, capture occurred at 50.72

± 8.55ms, as the jaws reached peak gape. The time to capture did not vary significantly

among any of the treatments (Table 4.2, Figure 4.4). The velocity of the prey as it

entered the nurse sharks’ mouth was not significantly different among any of the

treatments (Table 4.2, Figure 4.8). The velocity of the predator was significantly lower in

193

the vision blocked animals (Table 4.2, Figure 4.9). The control nurse sharks were

typically moving slowly, but opened the mouth rapidly to draw in the prey in a suction

event, engulfing it whole. Suction-biting, in which the prey is caught in the teeth and

subsequently drawn into the mouth in a transport event, occurred only occasionally, in

2.08 ± 2.08% of captures. The frequency of suction-bites did not vary among any of the treatments (Table 4.2, Figure 4.10).

In the control bonnetheads, the RSI value was 0.63 ± 0.02, indicating predominantly ram capture. The RSI values were not significantly different among any of the treatments (Table 4.3, Figure 4.6). Similarly, the ram distance and suction distance were not significantly different among any of the treatments (Table 4.3, Figure 4.7), nor were the velocity of the predator or velocity of the prey (Table 4.3, Figures 4.8, 4.9). In the control bonnetheads, capture occurred at 149.58 ± 6.51ms, while the jaws were closing. When olfaction or lateral line + olfaction were blocked, the time to capture was significantly longer, 184.03 ± 15.27ms and 268.44 ± 39.65ms, respectively, occurring as the jaws were closing and being retracted (Table 4.3, Figure 4.4). The time to capture was not significantly different among the other treatments. The frequency of ram-bites in this species was high, occurring in 75.33 ± 6.09% of captures. This species thus typically captured prey in the bite, rather than engulfing it whole. This behavior was not significantly affected by any of the treatments (Table 4.3, Figure 4.10).

194

Discussion

Approach Kinematics

Ram feeding inherently requires that the predator localize its prey from a distance in order to have sufficient room to accelerate prior to capture. This is especially true for the blacktip shark, which uses a higher approach velocity than the bonnethead, likely

because it hunts for elusive piscivorous prey, which it must rapidly chase down in the water column (Castro, 1996). The bonnethead on the other hand, uses a slower approach velocity, possibly because it typically captures benthic prey, such as lobsters and crabs, by opening the mouth wide and scooping them off the substrate (Cortés et al., 1996;

Wilga and Motta, 2000; Bethea et al., 2007; Mara et al., 2009). In the blacktip shark and

bonnethead, this process of localizing prey from a distance is visually mediated, similar

to ram-feeding bony fishes (New, 2002; Gardiner and Motta, In Press). When visual

information is available, blacktip sharks execute a rapid strike to overtake and engulf the

prey. When visual information is lacking, this species can still successfully capture prey,

but it must rely on lateral line information to orient and strike (Chapter 3). As the lateral

line system only functions over a distance of 0.4-2 body lengths (Denton and Gray, 1983;

Denton and Gray, 1988; Coombs, 1999; Braun and Coombs, 2000; Palmer et al., 2005),

strikes are initiated from a closer proximity (Chapter 3) and, therefore, the blacktip sharks

arrive at the prey at a slower velocity (approach velocity) compared to when visual

information is available. Similarly, in Indo-Pacific tarpon, Megalops cyprinoides, strike

velocity varies inversely with predator-prey distance at the initiation of the strike (Tran et

al., 2010). The bonnethead is unable to strike when visual information is lacking

195

(Chapter 3). The enhanced binocular vision possessed by hammerhead sharks (McComb et al., 2009) may explain the greater reliance on vision demonstrated by this species.

In the blacktip shark, information from the lateral line system appears to regulate the approach speed in these visually guided strikes, as it does in largemouth bass,

(Gardiner and Motta, In Press). When lateral line information is lacking, blacktip sharks use higher velocities at the start of (approach velocity) and during (predator velocity) capture. Thus, similar to largemouth bass, (Gardiner and Motta, In Press), blacktip sharks fail to brake when lateral line information is lacking. This suggests that either the lateral line provides the animals with information on the position of the prey, just prior to capture, which prompts the animals to brake, or that it may aid in regulating swimming speed. The lateral line has been shown to function in regulating swimming speed in several species of schooling fish (Pitcher, 1979; Partridge and Pitcher, 1980; Pitcher et al., 1980). The lateral line detects not only the motion of the prey, but also self-generated noise, such as from swimming or respiratory movements (Watanabe and Anraku, 2007;

Montgomery et al., 2009). The effects of self-generated noise have generally been understudied as most studies on the lateral line system have been conducted on immobilized animals, however, fish are capable of filtering out self-generated mechanosensory noise from respiratory movements in the hindbrain (Bodznick and

Montgomery, 1994) and therefore perhaps those generated by swimming. Swimming movements generate hydrodynamic signals (Bleckmann et al., 1991; Hanke et al., 2000;

Hanke and Bleckmann, 2004), which have primarily been viewed as noise that can mask prey signals (reviewed in Montgomery et al., 2009). Hunting strategies such as thrust and glide (Janssen, 1997) or saltatory (pause and move) search patterns (Bassett et al.,

196

2007), have been proposed to provide the animal with periods of lessened self-generated

mechanosensory noise (pause/glide), allowing them to better detect the prey signal

(Janssen, 1996, 1997; Bassett et al., 2007). However, this self-generated noise from

swimming movements could serve a purpose to the animal, providing it with feedback

regarding its own velocity. The lack of this feedback could explain the higher swimming

velocities observed in several species of fish during tracking and striking after the lateral

line system has been blocked (Hassan et al., 1992; Baker et al., 2002; Gardiner and

Motta, In Press) (Chapter 3).

Whether they are unable to regulate their velocity or unable to detect the prey just

prior to capture, blacktip sharks fail to brake when lateral line information is lacking.

Braking just prior to capture has been suggested to increase the capture accuracy, by

giving the animal more time for steering and positioning (Higham et al., 2005; Higham et

al., 2006a). Intact blacktip sharks position the prey more or less directly in front of the

mouth, but those with the lateral line blocked (alone or in combination with olfaction),

position the prey at greater capture angles, such that the prey is slightly above or below

the head. The failure to brake in lateral line blocked animals may not leave enough time

to make fine adjustments to the position relative to the prey prior to the start of capture,

or these fine adjustments may be lateral line mediated. Furthermore, braking may also decrease the bow wave in front of the head, which has been shown to displace prey,

alerting it to the approach of the predator and prompting an escape response (Viitasalo et

al., 1998; Ferry-Graham et al., 2003). Vision blocked blacktip sharks also demonstrate

greater capture angles, however, because sharks possess a blind area directly in front of

the head that extends beyond the distance at which capture is initiated in the blacktip

197

shark (McComb et al., 2009), it seems unlikely adjustments just prior to capture could be

visually mediated and more likely that the animals simply cannot completely compensate

for the inaccurate alignment of the strike itself that occurs in vision blocked animals

(Chapter 3). When the electrosensory system is blocked, blacktip sharks must make

physical contact with the prey to initiate capture (Chapter 3). As a result, the predator- prey distance at the start of capture is smaller in this treatment. The distance between the

predator’s jaws and the prey is, however, not zero as the prey typically made contact with

the snout, rather than the mouth. The capture angle is also greater, however, this may be

an artifact of the closer proximity of the prey at capture. If the blacktip shark aligns the

capture angle the prey at a given distance, but then fails to move the jaws until it is at a

closer proximity to the prey, if it has moved along the same heading, the capture angle

increases as the distance between the shark and the prey decreases (Figure 4.11)

Alternatively, the electrosensory system and the lateral line may be complementary for fine tuning the blacktip shark’s position prior to capture.

In the bonnethead, there is no difference in the approach velocity among the treatments. This species may rely less on feedback from self-generated mechanosensory noise to regulate its swimming speed, as slower swimming speeds have been suggested to create less noise (Janssen et al., 1990; Janssen, 1996). The approach speed, and therefore any braking, could instead be mediated by vision as the animal perceives the rate of change in the distance between itself and the prey target during the approach.

Hammerhead sharks possess enhanced binocular vision (McComb et al., 2009) and swimming speed during schooling has been shown to be visually regulated in highly visual fish species, such as Pacific bluefin tuna, Thunnus orientalis (Torisawa et al.,

198

2011). Bonnetheads may also need less fine-tuning of the strikes, since slower velocity strikes generate smaller bow waves in front of the head, which decreases the reaction distance in prey (Viitasalo et al., 1998; Ferry-Graham et al., 2003). Since hammerhead sharks, including the bonnethead, have larger blind areas in front of the head than pointed-nosed sharks (McComb et al., 2009), the prey is within the blind area well before capture begins. Thus, this species is likely relying on non-visual cues to line up the jaws with the prey prior to capture. Additionally, this species typically hunts for crustaceans in seagrass beds and thus, the prey may be hidden and visual cues may, therefore, be unavailable at the time of capture (Cortés et al., 1996; Bethea et al., 2007). Although capture does not occur in electrosensory blocked animals, the capture angle (assessed at the distance at which capture occurred in the control treatment) is not significantly affected. Capture angle does not vary significantly among any of the treatments; however, because this species approaches the prey at a slower velocity than the blacktip shark, it may not need to brake, as the slower approach speed may provide the animal with sufficient time to make adjustments to its position just prior to capture. Therefore, the bonnethead may be aligning the jaws with the prey for capture using either the electrosensory or lateral line systems, which are alternative for this task, but the electrosensory system is required in order to initiate jaw movement.

Since suction feeding is only effective over a distance of a few centimeters, suction feeders such as the nurse shark only strike from close proximity (Motta et al.,

2002; Nauwelaerts et al., 2006; Lowry and Motta, 2007; Motta et al., 2008; Wilga and

Sanford, 2008). Suction feeders, therefore, do not need to pinpoint their prey from a distance as ram feeders do. The nurse sharks in this study generally approached the prey

199

slowly, and the approach velocities were slower in vision blocked animals compared to lateral line and electrosensory blocked animals. This suggests that there are slight differences in the striking behaviors based on the sense used to orient for the strike, as this species can orient using vision, the lateral line, or electroreception (Chapter 3).

These differences do not, however, affect the position of the nurse shark at the start of capture, as the predator-prey distance and capture angle does not vary among the treatments. Because suction feeding in the nurse shark is so rapid (Motta et al., 2002;

Matott et al., 2005; Motta et al., 2008), and because suction is effective over only a few centimeters, suction feeding varies over a smaller distance than ram feeding (Ferry-

Graham et al., 2001a), and thus no fine-tuning may be needed just prior to capture.

Furthermore, since suction feeders use little to no forward motion, the prey is primarily alerted to the presence of the predator by the suction flow (Holzman and Wainwright,

2009), which is often too late for a successful escape response.

Capture Kinematics

The capture kinematics observed in the control blacktip sharks in this experiment follow the pattern described previously for this species, as well as other ram feeding carcharhiniform sharks (blacknose sharks Carcharhinus acronotus, lemon sharks

Negaprion brevirostris, and swell sharks Cephaloscyllium ventrosium) (Frazzetta and

Prange, 1987; Ferry-Graham, 1997; Motta et al., 1997). Capture begins with the head being raised while the lower jaw is simultaneously depressed, prey capture occurs while the jaws are at their peak gape, and upper jaw protrusion begins as the lower jaw is being elevated, reaching its maximum just before the jaws close. Cranial elevation begins

200

much earlier in vision blocked animals and also takes longer to reach a larger maximum

angle. This may represent an effort on the part of the animal to partially overcome the inaccurate positioning of the prey that occurs when vision is lacking, by raising the head further before beginning to depress the lower jaw for capture. When the olfactory system

is blocked, alone or in combination with the lateral line, the time to peak gape is longer.

This suggests that the animals are less motivated when there are no food odor cues.

Slower jaw opening, attributed to a lack of motivation, is also found in fish that are approaching satiation (Sass and Motta, 2002). Upper jaw protrusion also begins later in the olfactory-blocked treatments (olfaction alone and lateral line + olfaction), with respect to the start of jaw opening, but still coincides with peak gape and the start of lower jaw elevation, which suggests that protrusion in this species functions to assist in jaw closure as previously hypothesized (Frazzetta and Prange, 1987), by speeding up the closing of the mouth (Wilga et al., 2001; Motta, 2004).

The positive RSI values indicate that the blacktip shark is a ram feeder, as previously described (Frazzetta and Prange, 1987). When visual information is lacking, the RSI value is lower, which can be attributed to changes in the amount of ram, as the ram distance is smaller when vision was blocked, while the suction distance remains unchanged. The slower predator velocity and longer time to capture further reinforce that the blind blacktip sharks are using less ram than when visual cues are available. The

hydrodynamics of ram and suction feeding have not been well studied in sharks, but in

bony fish, the amount of suction generated depends on the rate of buccal expansion,

which is turn influenced by the speed of jaw opening and the magnitude of the gape,

which influence the rate of fluid flow, and the shape of the mouth, as more tubular

201

mouths direct the suction in front of the mouth (Liem, 1993; Carroll et al., 2004; Higham

et al., 2006a; Higham et al., 2006b). Since the speed of jaw opening and magnitude of

the gape are unchanged when vision is blocked, it is unlikely that the blacktip sharks are

modulating the minimal amount of suction. Ram feeding bony fishes primarily modulate

the kinematics of feeding by increasing or decreasing the amount of ram used (Ferry-

Graham and Lauder, 2001; Wainwright et al., 2001; Tran et al., 2010). These studies have primarily examined changes in prey type or size. The largemouth bass, M. salmoides, however, changes the amount of suction used in response to sensory deprivation. Thus, while blind animals use less ram, unlike the largemouth bass

(Gardiner and Motta, In Press), the blacktip shark does not completely change feeding modalities and remains a ram feeder despite changes in the pre-capture sensory information.

The capture kinematics observed in the bonnetheads closely follow the pattern previously described for this species. Cranial elevation and depression, which are slight and highly temporally variable (Wilga and Motta, 2000), were not assessed. Capture begins with lower jaw depression, peak gape is briefly maintained, then lower jaw is elevated, followed by upper jaw protrusion. Prey capture occurs as the jaws are closing and the prey is primarily caught in the teeth, in a ram-bite. Upper jaw retraction occurs well after the jaws have closed. The duration of peak gape is longer in the lateral line + olfaction blocked animals. Some of the strikes in this treatment resulted in a miss, rather than a capture (Chapter 3), which may have influenced the capture kinematics. Asps,

Aspius aspius, maintain peak gape for a longer duration when their prey attempts to escape (Van Wassenbergh and De Rechter, 2011), however, the asp is a suction feeder

202

and thus the overall kinematics of its gape cycle are quite different from those of ram

feeders. In the bonnethead, other changes in capture kinematics are slight and not

significant, however, together these changes result in a change in the time to capture,

such that prey capture occurred earlier in the lateral line blocked animals and later in the

olfaction and lateral line + olfaction blocked animals. In all cases, however, prey capture occurs as the jaws are closing, such that the prey is caught in the teeth, as is typical of ram-biters (Wilga and Motta, 2000; Motta and Wilga, 2001). The positive RSI values indicate that the bonnethead is a ram feeder, as previously described (Wilga and Motta,

2000). The RSI values, ram distance, suction distance, predator velocity, and prey velocity are not affected by any of the treatments. Collectively, these results indicate that little kinematic modulation occurs in response to sensory deprivation in the bonnethead.

These results are consistent with previous work which found no modulation in response to prey type or hardness (Wilga and Motta, 2000). This suggests that the ram-feeding behavior of the bonnethead is highly stereotyped under these experimental conditions.

The kinematics of capture in intact nurse sharks follow the pattern previously described for this species (Motta et al., 2002; Matott et al., 2005; Motta et al., 2008).

Capture begins with lower jaw depression, followed shortly thereafter by labial cartilage protrusion. Maximum labial cartilage protrusion occurs slightly after peak gape. Lower jaw elevation then begins, followed shortly thereafter by labial cartilage retraction. Prey capture occurs while the jaws are at their maximum gape and the labial cartilages maximally protruded. Changes in the various aspects of the gape cycle are not significant among the different sensory deprivation treatments. The negative RSI values indicate that the nurse shark is a suction feeder, as previously described (Motta and Wilga, 1999;

203

Motta et al., 2002; Motta et al., 2008). When vision is blocked, the RSI value is significantly more negative, indicating even greater suction. As the ram distance is significantly smaller and the suction distance significantly greater, blinded nurse sharks are both decreasing the amount of ram used, as well as increasing the amount of suction, though the kinematic pattern remains the same. The predator velocity during capture is also slower, but the prey velocity is unchanged, indicating that the blinded nurse sharks are drawing the prey in over a greater distance, but at the same velocity, compared to when visual cues are available. Since the suction pressure generated relates to the rate of buccopharyngeal expansion (Sanford and Wainwright, 2002; Svanback et al., 2002; Day et al., 2005; Motta et al., 2008), it is likely that the blind nurse sharks are opening the jaws more rapidly than when visual cues are available, but that either these differences were not detected by the sample size of this study. This suggests that when visual cues are unavailable, such as on a dark night or in the case of prey hidden within a crevice in a reef (Motta et al., 2008), the nurse shark may open the jaws more rapidly to generate more suction. This may assist the animal in overcoming a degree of uncertainty regarding the exact location of the prey, as decreasing the ram and increasing the suction results in the ingestion of a rounder parcel of water (Lowry and Motta, 2008). These results agree with previous work that suggested that nurse sharks are obligate suction feeders (Motta and Wilga, 1999; Motta et al., 2002; Matott et al., 2005; Motta et al.,

2008). However, these results are in partial disagreement with previous work on the nurse shark (Matott et al., 2005), which found no kinematic modulation in response to differences in prey size or type, and concluded that capture in this species is highly

204

stereotyped. While the kinematic sequence in this species is highly stereotyped, the rate of movement of the jaw elements is variable, as was found by Motta et al. (2008).

The feeding modulation hypothesis (Motta and Wilga, 2001) states that ram feeding and biting carcharhinid and lamnid sharks are capable of modulating the kinematics of feeding to a greater extent than suction feeding sharks, which have highly

stereotyped capture events. The results of this study disagree in part. While a greater amount of modulation occurred in the ram feeding blacktip shark, as compared to the suction feeding nurse shark, capture in the nurse shark was not as highly stereotyped as previously found (Matott et al., 2005), and modulation in the timing did occur in response to visual deprivation. Furthermore, little to no modulation was found in the ram-biting bonnethead. Only a few studies to date have specifically examined modulation in feeding kinematics in elasmobranchs. The ram feeding swellshark, Cephaloscylluim ventrosium, does not modulate its capture kinematics in response to prey size (Ferry-Graham, 1997).

Similarly, the suction feeding nurse shark, G. cirratum, and chain catshark, Scyliorhinus retifer, show no modulation in response to prey size or type (Matott et al., 2005; Ajemian and Sanford, 2007). Together, these results suggest that while suction feeders are generally more stereotyped than ram feeders, modulatory ability also varies by species.

In comparison to bony fishes, overall less modulation is found in these elasmobranchs in response to sensory deprivation. The largemouth bass, M. salmoides, uses similar amounts of ram (RSI value = 0.73)(Gardiner and Motta, In Press), compared to the blacktip shark (RSI value = 0.77; this study), yet the bass is capable of switching to primarily suction based feeding (Gardiner and Motta, In Press), whereas blacktip sharks modulate their kinematics to a lesser degree, remaining ram feeders. The blacktip shark

205

may be limited in its ability to modulate, based on its anatomy. Largemouth bass possess

the characteristics of both ram and suction feeders, including a fusiform body and

relatively large mouth, suited for ram feeding, and laterally occluded mouth and

protrusible jaws, suited for suction feeding (Norton, 1991b, 1995; Wainwright et al.,

2001; Carroll et al., 2004; Holzman et al., 2008a; Holzman et al., 2008b). The bass can

therefore modulate by decreasing ram, as well as increasing suction (Gardiner and Motta,

In Press). In contrast, the blacktip shark possesses the characteristics of a ram feeding

elasmobranch, including a large gape, body form suited to rapid swimming, and

relatively small labial cartilages (Motta and Wilga, 2001; Motta, 2004). Since the labial

cartilages do not occlude the mouth laterally, any suction generated is likely directed in

all directions, rather than in front of the mouth, as is the case in species that possess large

labial cartilages that aid in forming a more tubular mouth, such as the nurse shark (Motta

and Wilga, 2001; Motta, 2004; Nauwelaerts et al., 2007; Motta et al., 2008; Nauwelaerts

et al., 2008). Thus, it is likely unable to increase the amount of suction, modulating

instead by decreasing the amount of ram. Many carcharhinid and lamnid sharks are

likely similarly anatomically limited in the amount of suction force that can be generated,

as they have similar mouths to the blacktip shark. Similarly, the nurse shark is

morphologically suited for suction feeding with its small tubular mouth, hypertrophied jaw and hyoid adductors, large labial cartilages, and small teeth (Motta, 2004; Motta et al., 2008), and it too remains a suction feeder regardless of sensory deprivation.

Specialized suction feeders, such as the nurse shark, can only suck prey from a distance

of a few centimeters (Lowry and Motta, 2007; Lowry and Motta, 2008; Motta et al.,

2008; Wilga and Sanford, 2008), although this distance can be extended slightly by

206

placing the mouth near the substrate (Nauwelaerts et al., 2007). Since the prey were

suspended off the bottom in this study, the larger suction distance observed in blind nurse sharks is likely nearing the functional limitations of suction feeding. Ram-suction feeding species, such as the spiny dogfish, Squalus acanthias (Wilga and Motta, 1998;

Wilga et al., 2007) and the leopard shark, Triakis semifasciata, (Ferry-Graham, 1998a),

have more intermediate characteristics. Both species have been found to modulate

capture kinematics in response to prey size (Wilga, 1997; Ferry-Graham, 1998a). These

results suggest that capture plasticity is anatomically limited in many elasmobranchs.

Summary and Conclusions

In summary, the blacktip shark typically approaches the prey with rapid velocity

using vision, switches to lateral line cues to fine-tune the strike just prior to capture, such

that it brakes slightly and positions the prey in front of the mouth, then switches to

electrosensory or tactile cues initiate capture by simultaneously depressing the lower jaw

and raising the head. The prey is engulfed using ram capture while the jaws are at their

maximum gape, then the upper jaw begins to protrude as the lower jaw elevates.

Olfactory cues serve to increase motivation, but are not required for capture. When

visual information is lacking, the blacktip shark approaches the prey with a slower

velocity, positions the prey slightly above or below the head, and begins elevating the

head well before the lower jaw depresses, and captures the prey using decreased ram.

When electrosensory cues are lacking, capture begins from a closer proximity, and only

after the animal makes physical contact with the prey. The bonnethead typically

approaches the prey with a slower velocity using vision, switches to lateral line or

207 electrosensory cues to position the prey in front of the mouth, and requires electrosensory cues to initiate capture by depressing the lower jaw. The prey is typically caught in the teeth as the jaws are closing, in ram-bite. Capture does not occur in the absence of vision or electroreception, but otherwise, little to no modulation occurs in response to sensory deprivation. The nurse shark typically approaches the prey very slowly, uses vision, the lateral line, or electroreception to position to prey in front of the mouth, and uses electrosensory or tactile cues to initiate capture by depressing the lower jaw and protruding the labial cartilages. The prey is typically engulfed in a suction capture while the jaws are at their maximum gape and the labial cartilages maximally protruded.

Capture does not occur in the absence of olfaction. When visual cues are unavailable, the nurse shark approaches the prey more slowly and captures prey using decreased ram and increased suction, likely as a result of opening the jaws more rapidly. Therefore, in response to sensory deprivation, the blacktip shark can modulate the kinematic pattern, as well as the rate of movement of the cranial elements, while the nurse shark has a highly stereotyped kinematic pattern, but can modulate the rate of jaw opening, but both the kinematic pattern and the rate of movement are highly stereotyped in the bonnethead.

Literature Cited

Aerts, P. (1990). Variability of the fast suction feeding process in Astatotilapia elegans (Teleostei: Cichlidae): a hypothesis of peripheral feedback control. Journal of Zoology, London 220, 653-678.

Ajemian, M. J. and Sanford, C. P. (2007). Food capture kinematics in the deep-water chain catshark Scyliorhinus retifer. Journal of the Marine Biological Association of the United Kingdom 87, 1277-1286.

208

Atema, J. (1971). Structures and functions of the sense of taste in the catfish, Ictalurus natalis. Brain Behavior and Evolution 4, 273-294.

Basil, J. A., Hanlon, R. T., Sheikh, S. I. and Atema, J. (2000). Three-dimensional analysis of odor tracking by Nautilus pompilius. Journal of Experimental Biology 203, 1409-1414.

Bassett, D. K., Carton, A. G. and Montgomery, J. C. (2007). Saltatory search in a lateral line predator. Journal of Fish Biology 70, 1148-1160.

Bigelow, H. B. and Schroeder, W. C. (1948). Fishes of the Western North Atlantic: Lancelets, cyclostomes, sharks. New Haven, Connecticut.

Bleckmann, H., Breithaupt, T., Blickhan, R. and Tautz, J. (1991). The time course and frequency content of hydrodynamic events caused by moving fish, frogs and crustaceans. Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 168, 749-757.

Bodznick, D. and Montgomery, J. C. (1994). An adaptive filter that cancels self-induced noise in the electrosensory and lateral line mechanosensory systems of fish. Neuroscience Letters 174, 145-148.

209

Carroll, A. M., Wainwright, P. C., Huskey, S. H., Collar, D. C. and Turingan, R. G. (2004). Morphology predicts suction feeding performance in centrarchid fishes. Journal of Experimental Biology 207, 3873-3881.

Castro, J. I. (1996). The biology of the blacktip shark, Carcharhinus limbatus, off the southeastern United States. Bulletin of Marine Science 59, 508-522.

Castro, J. I. (2000). The biology of the nurse shark, Ginglymostoma cirratum, off the Florida east coast and the Bahama Islands. Environmental Biology of Fishes 58, 1-22.

Coombs, S. (1999). Signal detection theory, lateral-line excitation patterns and prey capture behaviour of mottled sculpin. Animal Behaviour 58, 421-430.

Cornett, A. D. (2006) Ecomorphology of Shark Electroreceptors. M.S. Thesis. Florida Atlantic University, Boca Raton, FL, pp. 102.

Day, S. W., Higham, T. E., Cheer, A. Y. and Wainwright, P. C. (2005). Spatial and temporal flow patterns during suction feeding of bluegill sunfish (Lepomis macrochirus) by Particle Image Velocimetry. Journal of Experimental Biology 208, 2661-2671.

Denton, E. J. and Gray, J. (1983). Mechanical factors in the excitation of clupeid lateral lines. Proceedings of the Royal Society of London. Series B: Biological Sciences 218, 1-26.

210

Faucher, K., Parmentier, E., Becco, C., Vandewalle, N. and Vandewalle, P. (2010). Fish lateral system is required for accurate control of shoaling behaviour. Animal Behaviour 79, 679-687.

Ferry-Graham, L. A. (1997). Feeding kinematics of juvenile swellsharks, Cephaloscyllium ventriosum. Journal of Experimental Biology 200, 1255-1269.

Ferry-Graham, L. A. (1998a). Effects of prey size and mobility on prey-capture kinematics in leopard sharks Triakis semifasciata. Journal of Experimental Biology 201, 2433-2444.

Ferry-Graham, L. A. (1998b). Feeding kinematics of hatchling swellsharks, Cephaloscyllium ventriosum (Scyliorhinidae): the importance of predator size. Marine Biology 131, 703-718.

Ferry-Graham, L. A. and Lauder, G. V. (2001). Aquatic prey capture in ray-finned fishes: a century of progress and new directions. Journal of Morphology 248, 99-119.

Ferry-Graham, L. A., Wainwright, P. C. and Bellwood, D. R. (2001a). Prey capture in long-jawed butterflyfishes (Chaetodontidae): the functional basis of novel feeding habits. Journal of Experimental Marine Biology and Ecology 256, 167-184.

Ferry-Graham, L. A., Wainwright, P. C. and Lauder, G. V. (2003). Quantification of flow during suction feeding in bluegill sunfishes. Zoology 106, 159-168.

Ferry-Graham, L. A., Wainwright, P. C., Westneat, M. W. and Bellwood, D. R. (2001b). Modulation of prey capture kinematics in the cheeklined wrasse Oxycheilinus digrammus (Teleostei : Labridae). Journal of Experimental Zoology 290, 88-100.

Fouts, W. R. and Nelson, D. R. (1999). Prey capture by the Pacific angle shark, Squatina californica: visually mediated strikes and ambush-site characteristics. Copeia 1999, 304-312.

Frazzetta, T. H. and Prange, C. D. (1987). Movements of cephalic components during feeding in some requiem sharks (Carcharhiniformes: Carcharhinidae). Copeia 4, 979-993.

Gardiner, J. M. and Motta, P. J. (In Press). Largemouth bass (Micropterus salmoides) switch feeding modalities in response to sensory deprivation. Zoology.

211

Hassan, E. S., Abdel-Latif, H. and Biebricher, R. (1992). Studies on the effects of Ca2+ and Co2+ on the swimming behavior of the blind Mexican cave fish. Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 171, 413-419.

Higham, T. E., Day, S. W. and Wainwright, P. C. (2005). Sucking while swimming: evaluating the effects of ram speed on suction generation in bluegill sunfish Lepomis macrochirus using digital particle image velocimetry. Journal of Experimental Biology 208, 2653-2660.

Higham, T. E., Day, S. W. and Wainwright, P. C. (2006a). Multidimensional analysis of suction feeding performance in fishes: fluid speed, acceleration, strike accuracy, and the ingested volume of water. Journal of Experimental Biology 209, 2713 - 2725.

Higham, T. E., Day, S. W. and Wainwright, P. C. (2006b). The pressures of suction feeding: the relation between buccal pressure and induced fluid speed in centrarchid fishes. Journal of Experimental Biology 209, 3281 - 3287.

Holzman, R., Day, S. W., Mehta, R. S. and Wainwright, P. C. (2008a). Integrating the determinants of suction feeding performance in centrarchid fishes. Journal of Experimental Biology 211, 3296-3305.

Holzman, R., Day, S. W., Mehta, R. S. and Wainwright, P. C. (2008b). Jaw protrusion enhances forces exerted on prey by suction feeding fishes. Journal of The Royal Society Interface 5, 1445-1457.

Janssen, J. (1996). Use of the lateral line and tactile senses in feeding in four antarctic nototheniid fishes. Environmental Biology of Fishes 47, 51-64.

Janssen, J. (1997). Comparison of response distance to prey via the lateral line in the ruffe and yellow perch. Journal of Fish Biology 51, 921-930.

Janssen, J., Coombs, S. and Pride, S. (1990). Feeding and orientation of mottled sculpin to water jets. Environmental Biology of Fishes 29, 43-50.

Kajiura, S. M. (2001). Head morphology and electrosensory pore distribution of carcharhinid and sphyrnid sharks. Environmental Biology of Fishes 61, 125-133. 212

Kajiura, S. M. and Holland, K. N. (2002). Electroreception in juvenile scalloped hammerhead and sandbar sharks. Journal of Experimental Biology 205, 3609- 3621.

Kaus, S. (1987). The effect of aminoglycoside antibiotics on the lateral line organ of Aplocheilus lineatus (Cyprinodontidae). Acta Otolaryngologica 103, 291-298.

Kroese, A. B. A. and van den Bercken, J. (1982). Effects of ototoxic antibiotics on sensory hair cell functioning. Hearing Research 6, 183-197.

Lauder, G. V. (1985). Aquatic feeding in lower vertebrates. In Functional Vertebrate Morphology, eds. M. Hildebrand D. M. Bramble K. F. Liem and D. B. Wake), pp. 210-229. Cambridge: Belknap Press.

Liem, K. F. (1979). Modulatory multiplicity in the feeding mechanism of the cichlids, as exemplified by the invertebrate pickers of Lake Tanganyika. Journal of Zoology (London) 189, 93-125.

Liem, K. F. (1980a). Acquisition of energy by teleosts: adaptive mechanisms and evolutionary patterns. In Environmental Physiology of Fishes, (ed. M. A. Ali), pp. 299-334. New York: Plenum.

Liem, K. F. (1980b). Adaptive significance of intra- and interspecific differences in the feeding repertoires of cichlid fishes. American Zoologist 20, 295-314.

Liem, K. F. (1993). Ecomorphology of the teleost skull. In The Skull. Functional and evolutionary mechanisms, vol. 3 eds. J. Hanken and B. K. Hall), pp. 422-452. Chicago: The University of Chicago Press.

Lowry, D. and Motta, P. J. (2007). Ontogeny of feeding and cranial morphology in the whitespotted bambooshark Chiloscyllium plagiosum. Marine Biology 151, 2013- 2023.

Lowry, D. and Motta, P. J. (2008). Relative importance of growth and behaviour to elasmobranch suction-feeding performance over early ontogeny. Journal of The Royal Society Interface 5, 641-652.

Mara, K., Motta, P. J. and Huber, D. R. (2009). Bite force and performance in the durophagous bonnethead shark, Sphyrna tiburo. Journal of Experimental Zoology 311A, 95-105.

Matott, M. P., Motta, P. J. and Hueter, R. E. (2005). Modulation in feeding kinematics and motor pattern of the nurse shark Ginglymostoma cirratum. Environmental Biology of Fishes 74, 163-174.

213

McComb, D. M., Tricas, T. C. and Kajiura, S. M. (2009). Enhanced visual fields in hammerhead sharks. Journal of Experimental Biology 212, 4010-4018.

Montgomery, J. C., Baker, C. F. and Carton, A. G. (1997). The lateral line can mediate rheotaxis in fish. Nature 389, 960-963.

Montgomery, J. C., Windsor, S. P. and Bassett, D. (2009). Behavior and physiology of mechanoreception: separating signal and noise. Integrative Zoology 4, 3-12.

Motta, P. J., Tricas, T. C., Hueter, R. E. and Summers, A. P. (1997). Feeding mechanism and functional morphology of the jaws of the lemon shark Negaprion brevirostris (Chondrichthyes, Carcharhinidae). Journal of Experimental Biology 200, 2765- 2780.

Motta, P. J. and Wilga, C. D. (1999). Anatomy of the feeding apparatus of the nurse shark, Ginglymostoma cirratum. Journal of Morphology 241, 1-29.

Motta, P. J. and Wilga, C. D. (2001). Advances in the study of feeding behaviors, mechanisms, and mechanics of sharks. Environmental Biology of Fishes 60, 131- 156.

Nauwelaerts, S., Wilga, C., Sanford, C. and Lauder, G. (2006). Hydrodynamics of prey capture in sharks: effects of substrate. Journal of The Royal Society Interface, 1-6.

Nauwelaerts, S., Wilga, C. D., Lauder, G. V. and Sanford, C. P. (2008). Fluid dynamics of feeding behavior in white spotted bamboo sharks. Journal of Experimental Biology 211, 3095-3102.

214

Nauwelaerts, S., Wilga, C. D., Sanford, C. P. and Lauder, G. V. (2007). Substrate passively improves suction feeding in benthic sharks. Journal of The Royal Society Interface 2, 341-345.

Nemeth, D. H. (1997a). Modulation of attack behavior and its effect on feeding performance in a trophic generalist fish, Hexagrammos decagrammus. Journal of Experimental Biology 200, 2155-2164.

Nemeth, D. H. (1997b). Modulation of buccal pressure during prey capture in Hexagrammos decagrammus (Teleostei: Hexagrammidae). Journal of Experimental Biology 200, 2145-2154.

New, J. G. (2002). Multimodal integration in the feeding behaviors of predatory teleost fishes. Brain, Behavior and Evolution 59, 177-189.

Norton, S. F. (1991a). Capture success and diet of cottid fishes: the role of predator morphology and attack kinematics. Ecology 72, 1807-1819.

Norton, S. F. (1991b). Habitat use and community structure in an assemblage of cottid fishes. Ecology 72, 2181-2192.

Norton, S. F. (1995). A functional approach to ecomorphological patterns of feeding in cottid fishes. Environmental Biology of Fishes 44, 61-78.

Norton, S. F. and Brainerd, E. L. (1993). Convergence in the feeding mechanics of ecomorphologically similar species in the Centrarchidae and Cichlidae. Journal of Experimental Biology 176, 11-29.

Nyberg, D. W. (1971). Prey capture in the largemouth bass. American Midland Naturalist 86, 128-144.

Partridge, B. L. and Pitcher, T. J. (1980). The sensory basis of fish schools: relative roles of lateral line and vision. Journal of Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology 135, 315-325.

Pitcher, T. J. (1979). Sensory information and the organization of behaviour in a shoaling cyprinid fish. Animal Behaviour 27, 126-149.

215

Pitcher, T. J., Partridge, B. L. and Wardle, C. S. (1980). A blind fish can school. Science 194, 963-965.

Robinson, M. P. and Motta, P. J. (2002). Patterns of growth and the effects of scale on the feeding kinematics of the nurse shark (Ginglymostoma cirratum). Journal of Zoology 256, 449-462.

Sanford, C. P. and Wainwright, P. C. (2002). Use of sonomicrometry demonstrates the link between prey capture kinematics and suction pressure in largemouth bass. Journal of Experimental Biology 205, 3445-3457.

Sass, G. G. and Motta, P. J. (2002). The effects of satiation on strike mode and prey capture kinematics in largemouth bass, Micropterus salmoides. Environmental Biology of Fishes 65, 441-454.

Sokal, R. R. and Rohlf, F. J. (1995). Biometry: The Principles and Practices of Statistics in Biological Research. New York: W.H. Freeman and Co.

Svanback, R., Wainwright, P. C. and Ferry-Graham, L. A. (2002). Linking cranial kinematics, buccal pressure, and suction feeding performance in largemouth bass. Physiological and Biochemical Zoology 75, 532-543.

Torisawa, S., Fukuda, H., Suzuki, K. and Takagi, T. (2011). Schooling behaviour of juvenile Pacific bluefin tuna Thunnus orientalis depends on their vision development. Journal of Fish Biology 79, 1291-1303.

Tran, H. Q., Mehta, R. S. and Wainwright, P. C. (2010). Effects of ram speed on prey capture kinematics of juvenile Indo-Pacific tarpon, Megalops cyprinoides. Zoology 113, 75-84.

Van Wassenbergh, S. and De Rechter, D. (2011). Piscivorous cyprinid fish modulates suction feeding kinematics to capture elusive prey. Zoology 114, 46-52.

Van Wassenbergh, S., Herrel, A., Adriaens, D. and Aerts, P. (2006). Modulation and variability of prey capture kinematics in clariid catfishes. Journal of Experimental Zoology 305A, 559-569.

Wainwright, P. C., Ferry-Graham, L. A., Waltzek, T. B., Carroll, A. M., Hulsey, C. D. and Grubich, J. R. (2001). Evaluating the use of ram and suction during prey capture by cichlid fishes. Journal of Experimental Biology 204, 3039-3051.

216

Watanabe, K. and Anraku, K. (2007). The activities of the lateral line nerves inducted by swimming motions. Nippon Suisan Gakkaishi 73, 1096-1102.

Wersall, J. and Flock, A. (1964). Suppression and restoration of the microphonic output from the lateral line organ after local application of streptomycin. Life Sciences 3, 1151-1155.

Wilga, C. D. (1997) Evolution of feeding mechanisms in elasmobranchs: a functional morphological approach. Univ. of South Florida, Tampa, pp.

Wilga, C. D., Hueter, R. E., Wainwright, P. C. and Motta, P. J. (2001). Evolution of upper jaw protrusion mechanisms in elasmobranchs. American Zoologist 41, 1248-1257.

Wilga, C. D. and Motta, P. J. (1998). Conservation and variation in the feeding mechanism of the spiny dogfish Squalus acanthias. Journal of Experimental Biology 201, 1345-1358.

Wilga, C. D. and Motta, P. J. (2000). Durophagy in sharks: feeding mechanics of the hammerhead Sphyrna tiburo. Journal of Experimental Biology 203, 2781-2796.

Wilga, C. D., Motta, P. J. and Sanford, C. P. (2007). Evolution and ecology of feeding in elasmobranchs. Integrative and Comparative Biology 47, 55-69.

Wilga, C. D. and Sanford, C. P. (2008). Suction generation in white-spotted bamboo sharks Chiloscyllium plagiosum. Journal of Experimental Biology 211, 3128- 3138.

217

Table 4.1. Kinematic Data Summary – Blacktip Sharks Tukey Test vs. Nose vs. Eye vs. vs. vs. Variable Treatment Mean n p value Block Block LLX LLX + Electrosensory Nose Block Block Control 1.77±0.16 14 N.S. <0.001 N.S. N.S. N.S. Nose Block 1.31±0.12 7 N.S. 0.034 N.S. N.S. Approach Eye Block 0.72±0.16 6 <0.001 N.S. N.S. Velocity <0.001* LLX 3.09±0.75 6 0.016 N.S. (BL/s) LLX + Nose Block 1.14±0.11 5 N.S. Electrosensory Block 1.49±0.4 3 Control 8.67±0.64 14 N.S. <0.001 0.040 0.004 0.001 Capture Nose 11.38±1.74 7 <0.001 N.S. N.S. 0.040 Angle Eyes 36.30±7.53 6 0.021 N.S. N.S. <0.001* (absolute Lateral Line 14.12±2.10 6 N.S. N.S. value, °) LL + Nose 22.06±6.33 5 N.S. Electrosensory 27.66±7.78 3 Control 5.59±0.29 14 N.S. N.S. N.S. N.S. 0.014 Predator-Prey Nose 6.18±0.42 7 N.S. N.S. N.S. 0.037 Distance at Eyes 4.70±0.84 6 N.S. N.S. 0.005 <0.001* Jaw Opening Lateral Line 7.62±0.45 6 N.S. <0.001 (cm) LL + Nose 4.97±0.10 5 0.022 Electrosensory 3.52±0.51 3 Control 69.25±5.65 14 0.024 N.S. N.S. 0.019 N.S. Nose 107.27±8.52 7 N.S. N.S. N.S. N.S. Time to Peak Eyes 92.50±11.69 6 N.S. N.S. N.S. 0.002* Gape (ms) Lateral Line 64.56±10.44 6 0.039 N.S. LL + Nose 113.47±15.71 5 N.S. Electrosensory 67.53±12.75 3 Control 4.32±0.28 14 Nose 3.89±0.07 7 Duration of Eyes 4.00±0.00 6 Peak Gape 0.402 Lateral Line 5.00±1.00 6 (ms) LL + Nose 4.00±0.00 5 Electrosensory 6.50±1.26 3

218

(Table 4.1 continued) Tukey Test vs. Nose vs. Eye vs. vs. vs. Variable Treatment Mean n p value Block Block LLX LLX + Electrosensory Nose Block Block Control 43.9±3.8 14 Nose Block 50.9±6.6 7 Time to Close Eye Block 45.3±6.4 6 0.566 (ms) LLX 43.2±9.9 6 LLX + Nose Block 41.9±3.7 5 Electrosensory Block 66.1±16.3 3 Control 116.49±8.62 14 Nose Block 165.14±14.28 7 Total Bite Eye Block 144.06±15.01 6 0.031 Time (ms) LLX 113.67±18.02 6 LLX + Nose Block 162.33±18.40 5 Electrosensory Block 142.00±9.45 3 Control -1.87±4.03 14 N.S. N.S. N.S. N.S. N.S. Start of Nose Block 14.33±9.48 7 <0.001 N.S. N.S. N.S. Cranial Eye Block -62.89±17.37 6 N.S. N.S. 0.002 <0.001* Elevation LLX 5.22±4.05 6 N.S. N.S. (ms) LLX + Nose Block 5.60±12.22 5 N.S. Electrosensory Block -64.27±84.40 3 Control 47.13±4.20 14 N.S. 0.005 N.S. 0.011 N.S. Time to Nose Block 53.86±2.44 7 N.S. N.S. N.S. N.S. Maximum Eye Block 104.89±21.75 6 0.002 N.S. N.S. Cranial <0.001* LLX 38.39±7.36 6 0.004 N.S. Elevation LLX + Nose Block 103.07±24.57 5 N.S. (ms) Electrosensory Block 53.07±2.93 3 Control 60.00±5.02 14 0.016 N.S. N.S. 0.033 N.S. Nose Block 98.61±9.24 7 N.S. 0.019 N.S. N.S. Start of Jaw Eye Block 75.28±8.81 6 N.S. N.S. N.S. Protrusion 0.002* LLX 54.33±10.27 6 0.029 N.S. (ms) LLX + Nose Block 103.73±16.47 5 N.S. Electrosensory Block 61.00±15.72 3

219

(Table 4.1 continued) Tukey Test vs. Nose vs. Eye vs. vs. vs. Variable Treatment Mean n p value Block Block LLX LLX + Electrosensory Nose Block Block Control 53.70±3.69 14 Time to Nose Block 62.79±9.02 7 Maximum Eye Block 70.56±14.03 6 0.443 Protrusion LLX 50.00±6.51 6 (ms) LLX + Nose Block 52.27±4.23 5 Electrosensory Block 56.73±7.48 3 Control 5.43±0.28 14 Nose Block 5.66±0.43 7 Peak Gape Eye Block 4.70±0.43 6 0.016 (cm) LLX 7.00±0.44 6 LLX + Nose Block 4.78±0.57 5 Electrosensory Block 4.90±0.49 3 Control 96.71±1.47 14 Nose Block 91.36±3.25 7 Gape Angle Eye Block 88.29±4.92 6 0.022 (°) LLX 103.01±2.63 6 LLX + Nose Block 93.66±3.32 5 Electrosensory Block 91.98±1.97 3 Control 1.84±0.16 14 Maximum Nose Block 1.86±0.22 7 Jaw Eye Block 1.43±0.19 6 0.057 Protrusion LLX 2.48±0.25 6 Distance (cm) LLX + Nose Block 1.65±0.20 5 Electrosensory Block 1.34±0.18 3 Control 8.51±0.80 14 N.S. 0.018 N.S. N.S. N.S. Nose Block 7.03±1.17 7 0.013 N.S. N.S. N.S. Maximum Eye Block 15.28±3.37 6 0.010 0.001 0.003 Head 0.001* LLX 6.81±1.39 6 N.S. N.S. Elevation (°) LLX + Nose Block 4.66±0.97 5 N.S. Electrosensory Block 5.20±1.01 3

220

(Table 4.1 Continued) Tukey Test vs. Nose vs. Eye vs. vs. vs. Variable Treatment Mean n p value Block Block LLX LLX + Electrosensory Nose Block Block Control 0.77±0.03 14 N.S. 0.006 N.S. N.S. N.S. Nose Block 0.78±0.04 7 0.011 N.S. N.S. N.S. Eye Block 0.55±0.05 6 0.002 N.S. N.S. RSI 0.002* LLX 0.83±0.03 6 N.S. N.S. LLX + Nose Block 0.75±0.04 5 N.S. Electrosensory Block 0.65±0.09 3 Control 7.25±0.35 14 N.S. N.S. N.S. N.S. N.S. Nose Block 8.63±0.59 7 N.S. N.S. N.S. 0.025 Ram Distance Eye Block 5.84±1.01 6 0.011 N.S. 0.035 <0.001* (cm) LLX 9.15±0.33 6 N.S. 0.009 LLX + Nose Block 6.46±0.34 5 N.S. Electrosensory Block 5.20±1.01 3 Control 1.03±0.13 14 Nose Block 0.98±0.18 7 Suction Eye Block 1.71±0.42 6 0.191 Distance (cm) LLX 0.84±0.15 6 LLX + Nose Block 1.06±0.15 5 Electrosensory Block 1.02±0.14 3 Control 74.68±5.48 14 0.039 N.S. N.S. N.S. N.S. Nose Block 114.97±8.33 7 N.S. N.S. N.S. N.S. Time to Eye Block 110.39±16.46 6 N.S. N.S. N.S. 0.004* Capture (ms) LLX 70.33±12.84 6 N.S. N.S. LLX + Nose Block 112.60±17.58 5 N.S. Electrosensory Block 62.67±19.19 3 Control 17.3±3.1 14 Nose Block 10.2±1.9 7 Prey Velocity Eye Block 18.2±1.3 6 0.461 (cm/s) LLX 17.1±4.2 6 LLX + Nose Block 10.6±2.7 5 Electrosensory Block 22.5±6.4 3

221

(Table 4.1 Continued) Tukey Test vs. Nose vs. Eye vs. vs. vs. Variable Treatment Mean N p value Block Block LLX LLX + Electrosensory Nose Block Block Control 117.1±12.9 14 N.S. N.S. N.S. N.S. N.S. Nose Block 79.0±6.1 7 N.S. 0.028 N.S. N.S. Predator Eye Block 66.4±8.9 6 0.008 N.S. N.S. Velocity 0.003* LLX 198.51±53.3 6 0.010 N.S. (cm/s) LLX + Nose Block 64.3±10.7 5 N.S. Electrosensory Block 103.3±37.9 3 Control 31.6±7.0 14 Nose 33.8±12.2 7 Frequency of Eyes 66.9±15.0 6 Ram-Bites 0.279 Lateral Line 45.8±11.9 6 (%) LLX + Nose 43.3±17.8 5 Electrosensory 25.0±14.4 3

Summary of all of the kinematic variables for the blacktip shark, Carcharhinus limbatus, with all senses intact (control) and after blocks of the sensory systems as indicated. LLX: lateral line blocked. All means are ±s.e.m. The p values are the results of ANOVA or Kruskal-Wallis tests performed on each variable. Value marked (*) are significant after Benjamini-Hochberg corrections. Tukey Test p values reflect the results of pairwise post-hoc comparisons between treatments. N.S.: Not significant at α = 0.05.

222

Table 4.2. Kinematic Data Summary – Nurse Sharks Tukey Test vs. vs. vs. vs. Variable Treatment Mean n p value Eye Block LLX LLX + Eye Electrosensory Block Block Control 0.31±0.04 8 N.S. N.S. N.S. N.S. Approach Eye Block 0.23±0.06 7 0.035 N.S. 0.032 Velocity LLX 0.44±0.06 6 0.004* 0.029 N.S. (BL/s) LLX + Eye Block 0.23±0.03 5 0.027 Electrosensory Block 0.44±0.04 6 Control 4.03±0.25 8 Predator-Prey Eye Block 3.86±0.53 7 Distance at LLX 3.59±0.36 6 0.147 Jaw Opening LLX + Eye Block 5.04±0.40 5 (cm) Electrosensory Block 3.40±0.30 6 Control 16.57±4.67 8 Eye Block 36.86±13.18 7 Capture LLX 25.39±3.83 6 0.349 Angle (°) LLX + Eye Block 35.69±5.05 5 Electrosensory Block 33.21±10.11 6 Control 47.30±8.28 8 0.040 N.S. N.S. N.S. Eye Block 26.95±3.08 7 N.S. N.S. N.S. Time to Peak LLX 35.28±2.85 6 0.046 N.S. N.S. Gape (ms) LLX + Eye Block 34.12±2.55 5 N.S. Electrosensory Block 31.24±2.55 6 Control 14.38±6.72 8 Duration of Eye Block 24.19±4.76 7 Peak Gape LLX 17.89±1.74 6 0.571 (ms) LLX + Eye Block 18.87±2.24 5 Electrosensory Block 14.94±3.47 6 Control 56.42±6.37 8 Eye Block 48.48±7.51 7 Time to Close LLX 47.51±7.14 6 0.280 (ms) LLX + Eye Block 40.47±4.27 5 Electrosensory Block 59.91±6.05 6

223

(Table 4.2 Continued) Tukey Test vs. vs. vs. vs. Variable Treatment Mean n p value Eye Block LLX LLX + Eye Electrosensory Block Block Control 121.17±16.07 8 Eye Block 99.62±8.18 7 Total Bite LLX 100.67±9.47 6 0.390 Time (ms) LLX + Eye Block 92.09±5.72 5 Electrosensory Block 106.27±6.81 6 Control 21.19±8.86 8 Start of Eye Block 14.29±1.19 7 Labial LLX 17.16±2.14 6 0.310 Protrusion LLX + Eye Block 13.51±0.94 5 (ms) Electrosensory Block 12.51±0.98 6 Time to Control 31.92±3.56 8 Maximum Eye Block 22.67±3.80 7 Labial LLX 33.24±2.75 6 0.202 Protrusion LLX + Eye Block 29.49±1.77 5 (ms) Electrosensory Block 27.74±3.70 6 Duration of Control 24.91±3.72 8 Maximum Eye Block 23.62±4.97 7 Labial LLX 25.42±5.97 6 0.987 Protrusion LLX + Eye Block 22.37±3.81 5 (ms) Electrosensory Block 22.06±6.85 6 Control 27.70±3.69 8 N.S. N.S. N.S. <0.05 Labial Eye Block 29.71±2.35 7 N.S. N.S. <0.05 Retraction LLX 34.22±2.09 6 0.005* N.S. N.S. Time (ms) LLX + Eye Block 38.13±1.60 5 N.S. Electrosensory Block 46.01±4.34 6 Control 3.24±0.16 8 0.032 N.S. N.S. N.S. Eye Block 3.98±0.10 7 N.S. N.S. N.S. Peak Gape LLX 3.89±0.15 6 0.023 N.S. N.S. (cm) LLX + Eye Block 3.66±0.26 5 N.S. Electrosensory Block 3.39±0.20 6 N.S.

224

(Table 4.2 Continued) Tukey Test vs. vs. vs. vs. Variable Treatment Mean n p value Eye Block LLX LLX + Eye Electrosensory Block Block Control 2.12±0.22 8 N.S. 0.043 N.S. N.S. Maximum Eye Block 2.83±0.11 7 N.S. N.S. N.S. Labial LLX 2.88±0.10 6 0.028 N.S. N.S. Protrusion LLX + Eye Block 2.76±0.15 5 N.S. Distance (cm) Electrosensory Block 2.67±0.26 6 Control 50.72±8.55 8 Eye Block 46.86±4.09 7 Time to LLX 43.19±4.67 6 0.581 Capture (ms) LLX + Eye Block 37.81±3.64 5 Electrosensory Block 39.88±6.82 6 Control 1.20±0.16 8 0.035 N.S. N.S. N.S. Eye Block 0.48±0.20 7 N.S. 0.006 0.018 Ram Distance LLX 0.83±0.15 6 0.004* N.S. N.S. (cm) LLX + Eye Block 1.58±0.31 5 N.S. Electrosensory Block 1.42±0.33 6 Control 3.35±0.54 8 N.S. N.S. N.S. N.S. Eye Block 4.36±0.62 7 0.018 0.015 0.023 Suction LLX 2.07±0.33 6 0.004* N.S. N.S. Distance (cm) LLX + Eye Block 2.02±0.28 5 N.S. Electrosensory Block 2.14±0.35 6 Control -0.45±0.10 8 0.022 N.S. N.S. N.S. Eye Block -0.84±0.06 7 <0.001 <0.001 <0.001 RSI LLX -0.42±0.113 6 <0.001* N.S. N.S. LLX + Eye Block -0.19±0.07 5 N.S. Electrosensory Block -0.24±0.05 6 Control 31.41±5.27 8 0.043 N.S. N.S. N.S. Predator Eye Block 10.60±4.62 7 N.S. 0.003 0.006 Velocity LLX 22.69±3.32 6 0.002* N.S. N.S. (cm/s) LLX + Eye Block 41.13±5.85 5 N.S. Electrosensory Block 39.08±6.52 6

225

(Table 4.2 Continued) Tukey Test vs. vs. vs. vs. Variable Treatment Mean n p value Eye Block LLX LLX + Eye Electrosensory Block Block Control 99.21±18.89 8 Eye Block 102.68±23.26 7 Prey Velocity LLX 55.22±7.91 6 0.208 (cm/s) LLX + Eye Block 57.41±10.32 5 Electrosensory Block 58.69±2.75 6 Control 2.08±2.08 8 Frequency of Eye Block 4.76±4.76 7 Suction-Bites LLX 10.7±8.20 6 0.483 (%) LLX + Eye Block 0.00±0.00 5 Electrosensory Block 10.8±7.12 6

Summary of all of the kinematic variables for the nurse shark, Ginglymostoma cirratum, with all senses intact (control) and after blocks of the sensory systems as indicated. LLX: lateral line blocked. All means are ±s.e.m. The p values are the results of ANOVA or Kruskal-Wallis tests performed on each variable. Value marked (*) are significant after Benjamini-Hochberg corrections. Tukey Test p values reflect the results of pairwise post-hoc comparisons between treatments. N.S.: Not significant at α = 0.05.

226

Table 4.3. Kinematic Data Summary – Bonnetheads Tukey Test Variable Treatment Mean n p value vs. Nose Block vs. LLX vs. LLX + Nose Block Control 0.70±0.03 18 Approach Nose Block 0.62±0.04 7 Velocity LLX 0.73±0.08 4 0.679 (BL/s) LLX + Nose Block 0.67±0.15 3 Electrosensory Block 0.62±0.07 8 Control 14.17±3.30 18 Capture Nose Block 15.98±3.68 7 Angle LLX 15.81±3.35 4 0.787 (absolute LLX + Nose Block 13.76±5.19 3 value, °) Electrosensory Block 16.98±4.14 8 Predator-Prey Control 4.74±0.16 18 Distance at Nose Block 4.18±0.28 7 0.075 Jaw Opening LLX 4.57±0.37 4 (cm) LLX + Nose Block 5.27±0.33 3 Control 113.91±5.87 18 Time to Peak Nose Block 156.67±18.60 7 0.152 Gape (ms) LLX 102.20±10.95 4 LLX + Nose Block 173.11±54.35 3 Control 12.46±1.93 18 N.S. N.S. 0.004 Duration of Nose Block 26.87±10.89 7 N.S. N.S. Peak Gape 0.002* LLX 8.00±1.08 4 0.002 (ms) LLX + Nose Block 40.22±13.17 3 Control 93.74±4.57 18 N.S. N.S. N.S. Time to Close Nose Block 109.87±15.53 7 0.019 N.S. 0.016 (ms) LLX 69.20±6.52 4 N.S. LLX + Nose Block 73.33±17.49 3 Control 220.53±9.01 18 0.049 N.S. N.S. Total Bite Nose Block 299.52±29.85 7 0.008 N.S. 0.010 Time (ms) LLX 179.40±16.51 4 N.S. LLX + Nose Block 247.33±23.73 3

227

(Table 4.3 Continued) Tukey Test Variable Treatment Mean n p value vs. Nose Block vs. LLX vs. LLX + Nose Block Control 138.68±15.38 18 Start of Jaw Nose Block 127.33±9.34 7 Protrusion 0.340 LLX 111.56±21.83 4 (ms) LLX + Nose Block 181.00±15.00 3 Time to Control 61.43±6.52 18 Max Jaw Nose Block 56.89±10.51 7 0.269 Protrusion LLX 60.44±2.70 4 (ms) LLX + Nose Block 31.33±3.33 3 Duration of Control 26.01±4.62 18 Max Jaw Nose Block 24.44±14.85 7 0.562 Protrusion LLX 19.11±4.38 4 (ms) LLX + Nose Block 40.00±4.00 3 Control 103.71±22.85 18 Time to Nose Block 67.33±6.57 7 Retract Upper 0.723 LLX 88.00±5.55 4 Jaw (ms) LLX + Nose Block 53.33±9.33 3 Control 4.24±0.17 18 Peak Gape Nose Block 3.79±0.36 7 0.621 (cm) LLX 4.58±0.26 4 LLX + Nose Block 4.37±0.34 3 Control 84.58±2.72 18 Gape Angle Eye Block 80.62±6.10 7 0.084 (°) LLX 77.72±1.89 4 LLX + Nose Block 65.71±6.90 3 Maximum Control 0.92±0.04 18 N.S. 0.029 N.S. Jaw Nose Block 0.87±0.11 7 0.049 N.S. 0.029 Protrusion LLX 1.27±0.14 4 N.S. Distance (cm) LLX + Nose Block 1.05±0.25 3

228

(Table 4.3 Continued) Tukey Test Variable Treatment Mean n p value vs. Nose Block vs. LLX vs. LLX + Nose Block Control 7.23±0.36 18 N.S. N.S. 0.007 Ram Distance Nose Block 7.93±0.71 7 N.S. N.S. 0.013 (cm) LLX 7.34±0.36 4 0.042 LLX + Nose Block 10.71±1.45 3 Control 1.66±0.13 18 Suction Nose Block 1.27±0.54 7 0.253 Distance (cm) LLX 1.91±0.22 4 LLX + Nose Block 2.39±0.80 3 Control 0.63±0.02 18 Nose Block 0.74±0.09 7 RSI 0.519 LLX 0.60±0.02 4 LLX + Nose Block 0.65±0.07 3 Control 149.62±6.50 18 N.S. N.S. <0.001 Time to Nose Block 184.03±15.27 7 0.033 N.S. <0.001* Capture (ms) LLX 123.07±13.91 4 <0.001 LLX + Nose Block 268.44±39.65 3 Control 51.01±2.52 18 Predator Nose Block 52.44±7.29 7 Velocity 0.364 LLX 62.13±4.89 4 (cm/s) LLX + Nose Block 46.80±10.57 3 Control 11.90±1.00 18 Prey Velocity Nose Block 9.35±4.17 7 0.348 (cm/s) LLX 16.15±1.78 4 LLX + Nose Block 10.91±4.49 3 Control 75.33±6.09 18 Frequency of Nose Block 75.43±11.87 7 Ram-Bites 0.247 LLX 77.67±8.82 4 (%) LLX + Nose Block 100.00±0.00 3 Summary of all of the kinematic variables for the bonnethead, Sphyrna tiburo, with all senses intact (control) and after blocks of the sensory systems as indicated. LLX: lateral line blocked. All means are ±s.e.m. The p values are the results of ANOVA or Kruskal-Wallis tests performed on each variable. Value marked (*) are significant after Benjamini-Hochberg corrections. Tukey Test p values reflect the results of pairwise post-hoc comparisons between treatments. N.S.: Not significant at α = 0.05.

229

Figures

Figure 4.1. Experimental Flume. Diagram is drawn to scale, with a ~1m TL bonnethead shark in the test arena. The dot represents the location of the prey only and is not representative of size.

230

Figure 4.2. Approach Velocity. The velocity of the strike, just prior to the start of capture (onset of jaw opening) in body lengths/s (BL/s) in the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo, with all senses intact (control) and following blocks of the sensory systems indicated in the figure legend (LL: lateral line). Jaw opening did not occur in the bonnethead, approach velocity in this treatment was assessed at the distance at which capture began in the control treatment for this speices. Error bars are s.e.m. * denotes significant differences between treatments at α = 0.05.

231

Figure 4.3. Predator-Prey Distance at the Start of Capture. The distance between the tips of the predator’s jaws and the center of mass of the prey at t0 (onset of jaw depression) in three species, the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo, with all senses intact (control) and following blocks of the sensory systems indicated in the figure legend (LL: lateral line). Error bars are s.e.m. * denotes values that are significantly different from other treatments at α = 0.05.

232

Figure 4.4. Capture Kinematics. Onset and duration in milliseconds (ms) are shown for each variable measured in three species (left to right): blacktips sharks, Carcharhinus limbatus; nurse sharks, Ginglymostoma cirratum; bonnetheads, Sphryna tiburo, with all senses intact (control) and following blocks of the sensory systems indicated in the figure legend (LL: lateral line). Dotted lines indicate the timing of prey capture. Error bars are s.e.m. For details of statistical analyses, see Tables 4.1 – 4.3. Blacktip shark and bonnethead illustrations from Diane Peebles, with permission. Nurse shark illustration from José Castro, with permission.

233

Figure 4.5. Gape Cycle. Plots of kinematic data for representative prey capture events for three species of sharks (top to bottom): the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo. Data are shown for animals with all senses intact (control) and after blocks of the sensory systems as indicated. For details of statistical analyses, see Tables 4.1 – 4.3. Blacktip shark and bonnethead illustrations from Diane Peebles, with permission. Nurse shark illustration from José Castro, with permission.

234

Figure 4.6. RSI Values. Boxplots of the RSI values for the three species (left to right): blacktip shark, Carcharhinus limbatus; nurse shark, Ginglymostoma cirratum; bonnethead, Sphyrna tiburo, with all senses intact (control) and following blocks of the sensory systems indicated in the figure legend (LL: lateral line). Top and bottom whiskers represent maximum and minimum data points, respectively. RSI value may range from -1 (pure suction) to +1 (pure ram). * denotes significant differences between treatments at α = 0.05. Blacktip shark and bonnethead illustrations from Diane Peebles, with permission. Nurse shark illustration from José Castro, with permission.

235

Figure 4.7. Predator and Prey Movements During Capture. Distances moved by the predator (dark gray) and prey (light gray) during capture for the three species (top to bottom): blacktip shark, Carcharhinus limbatus; nurse shark, Ginglymostoma cirratum; bonnethead, Sphyrna tiburo, with all senses intact (control) and following blocks of the sensory systems indicated in the figure legend (LL: lateral line). The total length of the bars (light gray + dark gray) is centered around the 0cm line to allow for comparisons of the proportion of the distance travelled by predator vs. prey (i.e., the 0cm line represents equal distances travelled by predator and prey). Error bars are s.e.m. * denotes significant differences between treatments at α = 0.05. Blacktip shark, bonnethead, shrimp and pinfish illustrations from Diane Peebles, with permission. Nurse shark illustration from José Castro, with permission.

236

Figure 4.8. Prey Velocity. The velocity of the prey as it is drawn in the mouth of the predator during capture, from the onset of jaw depression in the predator until prey capture. Data are shown for three species of sharks, the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo. Error bars are s.e.m. There were no significant differences among treatments.

237

Figure 4.9. Predator Velocity. The velocity of the predator during capture, from the onset of jaw depression to the capture of the prey in three species, the blacktip shark, Carcharhinus limbatus, the nurse shark, Ginglymostoma cirratum, and the bonnethead, Sphyrna tiburo, with all senses intact (control) and following blocks of the sensory systems indicated in the figure legend (LL: lateral line). Error bars are s.e.m. * denotes significant differences between treatments at α = 0.05.

238

Figure 4.10. Frequency of Bites. The frequency of biting behavior in three species of sharks, the blacktip shark, Carcharhinus limbatus (ram-biting behavior), the nurse shark, Ginglymostoma cirratum (ram-suction behavior), and the bonnethead, Sphyrna tiburo (ram-biting behavior), in animals with all senses intact (control) and following blocks of the sensory systems as indicated. Error bars are s.e.m. There were no significant differences among treatments.

239

Figure 4.11. Hypothetical Capture Angles. An illustration of the change in capture angle, ϴ, that would be expected to occur if an animal were to align the capture angle at a particular position (P1) and distance from the prey (dotted line), but then fail to initiate capture until it was closer to the prey (P2). If the animal continues along the same heading (solid line with arrow), the capture angle from its new position (P2) will be greater (ϴ2 > ϴ1). Diagram is for illustrative purposes only, predator-prey distances, capture angles, and size of the predator and the prey are not representative. Blacktip shark illustration from Diane Peebles, with permission.

240

CHAPTER 5: MULTIMODAL INTEGRATION AND SENSORY PLASTICITY

IN SHARK FEEDING BEHAVIOR

Animals use a variety of sensory signals to accomplish behaviors, such as feeding.

The various senses can play different roles in different phases of these behaviors.

Occasionally, one sensory modality is sufficient to mediate a behavior, but often multiple

cues are combined in the central nervous system, which can result in higher sensitivity, shorter latency, better spatial resolution and better filtering of noise (Stein and Stanford,

2008). Because feeding performance is strongly correlated with fitness (reviewed in

Arnold, 1983), animals will likely use whichever sense(s) provides them with the best performance for a particular phase of feeding to maximize the likelihood of success. This can lead to sensory switching at the interface of different phases of behavior. An animal may change its focus from one cue(s) to another at different distances as it approaches a

target, such as a food item, as new signals become available. For example, hawkmoths

(Sphingidae), use olfaction and vision to approach flowers, vision to place the proboscis,

and vision, touch, and taste to probe for nectar (reviewed in Raguso et al., 2005).

Sensory switching of another kind can occur in animals that use sensory information actively, rather than passively, in that they can switch on the active mode based on sensory cues. The California leaf-nosed bat (Macrotus californicus), for example, switches off echolocation, by ceasing to generate echolocation calls, when sufficient 241

visual information is available (Bell, 1985). Sensory switching may occur as a result of

weighting the cues based on the amount of information that they are likely to provide for

a given task (reviewed in Stein and Stanford, 2008), and it occurs naturally as part of a

behavioral sequence.

There is a degree of flexibility, or plasticity, in the use of the senses, however,

two or more sensory cues can sometimes play alternative roles for a given task, such that

when information from one sense is unavailable, animals can compensate, either partially

or completely, using the remaining senses to accomplish particular behaviors. This

phenomenon, termed “compensatory sensory plasticity” (Lewkowicz and Lickliter,

1994), has generally been used to refer to long-term changes in the CNS, occurring in

individual animals in response to permanent losses of sensory information (i.e., injury or

congenital defects), that allow the animal to compensate to some degree by developing

heightened capabilities with one or more of the remaining senses. For example, cats

deprived of vision at birth develop enhanced auditory localization capabilities

(Rauschecker, 1995). Similar effects have been described in humans (Lessard et al.,

1998). Another form of sensory plasticity involves changes in the tuning of the sensory

systems which can occur through ontogeny. The electrosensory system of skates and

stingrays, for example, is tuned in the embryo for the detection of predators and in the

adult to conspecific and prey cues (Sisneros et al., 1998; Sisneros and Tricas, 2002).

These changes also require long term changes to the sensory system, usually to the

receptors themselves. The photoreceptors in the eye of the lemon shark, Negaprion

brevirostris, change through ontogeny, such that juveniles have pigments that detect yellow-green for maximum sensitivity in their shallow inshore habitat, while adults have

242

pigments that detect blue-green for maximum sensitivity in their oceanic habitat (Cohen

et al., 1990). These changes may be permanent, as in the case of the above

developmental changes, or they may be transient and reversible. The tuning of the

electrosensory system of stingrays (Sisneros and Tricas, 2000) and the auditory systems

of fish (Sisneros and Bass, 2003; Sisneros, 2009b, c, a; Rohmann and Bass, 2011), frogs

(Goense and Feng, 2005) and birds (Lucas et al., 2007) changes seasonally, such that

animals are better tuned to conspecific signals or calls during the breeding season. These

changes are thought to be hormonally induced.

Compensatory sensory plasticity can, however, also occur in the short term or

even instantaneously, with animals compensating for the loss of a particular cue through

sensory switching, by focusing on other, alternative cues to perform a behavior when the

sensory cues which are optimal for this behavior are unavailable. Many fishes, for

example, will switch to lateral line cues to locate and capture prey, when visual

information is unavailable (Janssen et al., 1995; New and Kang, 2000; New et al., 2001;

New, 2002; Gardiner and Motta, In Press). The resulting behavior may, however, be

slightly altered as a result of using this alternative sensory modality, as is the case in

largemouth bass, Micropterus salmoides, which strikes slower and from a closer

proximity, but opens the mouth more rapidly, switching to suction feeding when using the lateral line to locate prey.

Animals may face losses of sensory information due to natural environmental variation as well as changes resulting from human disturbances. Vision can be affected by the natural variation in light levels between day and night, as well as by cloud cover

(McMahon and Holanov, 1995) and increases in turbidity, natural or anthropogenic

243

(Vogel and Beauchamp, 1999). Olfactory receptors are damaged by petroleum oil

(Tierney et al., 2010), pesticides (Tierney et al., 2007; Tierney et al., 2008), detergents

(Bardach et al., 1965), and the byproducts of eutrophication (Yang et al., 2002), which can also mask olfactory cues, impacting feeding and social interactions (Fisher et al.,

2006). Hearing can be affected by damage to the otoliths from petroleum oil (Morales-

Nin et al., 2007) or to the inner ear hair cells by anthropogenic noise (McCauley et al.,

2003). The lateral line system is damaged by ototoxic antibiotics (Kroese and van den

Bercken, 1982; Owens et al., 2009) and metals (Karlsen and Sand, 1987; Baker and

Montgomery, 2001; Faucher et al., 2006; Faucher et al., 2008). The sensitivity of the electrosensory system can by affected by changes in salinity from alteration of freshwater inputs (McGowan and Kajiura, 2009). These losses may be transient, as in the case of the loss of visual information at night, or permanent, as may be the case with damage to the receptors, if the receptors cannot be regenerated.

Over the longer term, focusing on particular cues over others can lead to behavioral changes in individuals or populations. Guppies, Poecilia reticulata, typically focus on visual cues for feeding, but when reared in the absence of visual cues utilize olfactory cues to locate food, even after visual cues have been restored (Chapman et al.,

2010). River and lake populations of mottled sculpin, Cottus bairdi, respond differently to lateral line cues that mimic flowing water and prey vibrations (Coombs and Grossman,

2006). Since these changes occur without changes to the sensory systems themselves, they are likely mediated by experience, as the animals have learned to rely on certain sensory cues over others. Over the longer term, however, this can lead to evolutionary changes to a population or species, as structures that are not being used can be altered or

244

lost. Mexican cavefish, Astyanax fasciatus, exist in two forms, a blind cave-dwelling

form that responds strongly to lateral line cues, and a sighted top-dwelling form that is repelled by lateral line stimuli (Yoshizawa et al., 2010). Although they are considered to

be the same species, the cave-dwelling populations differ in a number of morphological

traits (reviewed in Wilkens, 2010), most notably, the absence of eyes (Wilkens, 1988).

Therefore, sensory switching, when used long term, may lead to behavioral changes and

eventually morphological changes.

The overall goal of this dissertation was to investigate the use of sensory

information in the different phases of feeding behavior in sharks and to elucidate the

complementary and alternative roles of olfaction, vision, mechanoreception by the lateral

line, and electroreception. Understanding which senses are required for particular

behavioral tasks and the sensory switching that occurs naturally at the interface of

different phases of behavior allows us to understand the hierarchy of sensory information

used in typical feeding behavior, as well as alternative strategies that can be used when

particular cues are lacking. Understanding which behavioral tasks absolutely require a

particular sense, that is, for which no alternative cues can be used (e.g. olfaction in nurse

sharks; Chapter 3), allows us to predict the limits of sensory plasticity in these animals, as

the loss of these required cues will result in the cessation of the behavior altogether. This

information is important as the degree of sensory plasticity dictates the ability of the

animals to switch senses, which may not only allow animals to consume a broader diet

and be successful in a greater variety of habitats (Liem, 1980), but may also be crucial for

survival in a changing environment.

245

A prey item might emit an odor, create a hydrodynamic disturbance, such as from

gill movements or swimming, and/or produce a weak electrical field (Kalmijn, 1972).

The threshold for detection of such fields by electrosensitive animals is in the range of

1nV/cm – 0.1μV/cm (Murray, 1962; Kalmijn, 1982; Johnson et al., 1984; Haine et al.,

2001; Kajiura and Holland, 2002; Kajiura, 2003; Jordan et al., 2009; Kajiura and

Fitzgerald, 2009). For fields produced by aquatic animals, this translates to a distance of

less than one-half meter from the source (Kalmijn, 1972). The detection limits of the

visual system of most aquatic animals are not well known and depend on the amount of

available light, the amount of scatter (Duntley, 1963; Mazur and Beauchamp, 2003), and

the background contrast in intensity, polarization and pattern of reflected light (Johnsen

and Sosik, 2004; Johnsen, 2005), but rarely exceed tens of meters. Odor, on the other

hand, can be carried a great distance from the source by the bulk flow. Intermolecular

viscous forces dissipate any turbulent energy until only very patchy odor information is

available (reviewed in Webster, 2007) and sharks are sensitive to amino acids in the

picomolar range (Hodgson and Mathewson, 1978; Zeiske et al., 1986; Meredith and

Kajiura, 2010), similar to teleosts (Hara, 1994). Odor is thus often the first cue

encountered by a shark that is searching for food and it is typically the cue that allows it

to detect that food is present in the environment (Figures 5.1-5.3). Sharks can orient to

these patches of odor, based on differences in the timing of odor arrival at the two nares.

If there is an internarial delay of 0.1-0.5s, the shark will respond with a turn towards the

side that receives the first, not the strongest, odor cue. If there is no delay or a long (1s)

delay, sharks turn to either side with equal frequency. This allows these animals to steer

into oncoming patches and therefore into odor plumes. Once a shark has been alerted to

246

the presence of food and has steered into the plume, it must track the plume to the

vicinity of its source, orient to the food item, strike, and capture the prey (Chapter 2).

Having detected its prey using olfaction, the blacktip shark then typically uses

olfaction in combination with either the lateral line or vision to track, vision to orient and execute a rapid strike from a distance of several body lengths, the lateral line to fine-tune the strike (braking slightly and positioning the prey in front of the mouth) and the electrosensory system to initiate capture, by depressing the lower jaw. The prey is typically engulfed in a ram capture event as the shark overswims the prey. When olfactory information is lacking, no tracking occurs, but visual information, acquired at a distance of a few body lengths, can mediate detection, followed by orientation, rapid striking, and capture. The kinematics of capture are slightly slower, suggesting that olfactory cues are the most motivating for this species. When visual information is lacking, blacktip sharks can track with olfaction in combination with the lateral line and

orient and strike using the lateral line, but only from a distance of less than one body

length. As a result, the strikes are initiated from a greater variety of angles and are

slower, therefore, the animals capture prey using less ram. When information from the

lateral line is lacking, blacktip sharks position the prey slightly above or below the head

and do not brake prior to capture and, therefore, the animals frequently miss; when

successful, the prey is engulfed using increased ram. When information from the

electrosensory system is lacking, physical contact with the prey results in jaw depression,

likely as a result of tactile cues. If olfaction and vision are absent, the blacktip shark does

not detect the prey and feeding behavior does not occur. If vision and the lateral line are

247 blocked, the prey is detected using olfaction but tracking does not occur (Chapters 3 and

4; Figure 5.1).

The nurse shark typically uses olfaction in combination with either vision or the lateral line to track its prey, switches to vision, the lateral line, or electroreception to orient and slowly strike from a close proximity, then to electroreception to initiate capture, from a distance of a few centimeters, by rapidly depressing the lower jaw. The prey is typically engulfed in a suction capture event. When olfaction is blocked, the prey is not detected, feeding generally ceases, and the animals spend more time resting on bottom. These animals will occasionally orient to the prey, but rarely strike when olfaction is blocked. When both vision and the lateral line are blocked, nurse sharks can still perform tracking, possibly using tactile cues, although the process is slower. When vision is blocked, nurse sharks approach the prey even more slowly, and therefore capture prey using less ram. They simultaneously use more suction and therefore are likely opening the jaws more rapidly. When information from the electrosensory system is lacking, physical contact with the prey results in jaw depression, likely as a result of tactile cues (Chapters 3 and 4; Figure 5.2).

Under these experimental conditions the bonnethead typically tracks prey using olfaction in combination with vision or the lateral line, switches to vision to orient from a distance of one to two body lengths and execute a strike, which is slower velocity than that of the blacktip shark. Bonnetheads rely on electroreception to initiate capture by depressing the lower jaw. They prey is typically caught in the teeth as the upper jaw protrudes and the lower jaw is elevating, in a ram-bite. In the absence of olfactory cues, tracking does not occur, but visual information, acquired at a distance of one to two body

248

lengths, can mediate detection, followed by orientation, striking, and capture. When visual information is lacking, tracking occurs but the bonnetheads are unable to orient or strike. This likely applies only to prey in the water column, which may occur when adult bonnetheads consume teleosts (Bethea et al., 2007). When bonnetheads capture crustaceans, they typically approach them from above and scoop them off the substrate

(Cortés et al., 1996; Wilga and Motta, 2000) and are likely capable of detecting them using electroreception. When the electrosensory system is blocked, bonnetheads orient and strike, but fail to depress the lower jaw, even if they make physical contact with the prey, and thus, capture does not occur. In the bonnethead, little to no modulation occurs in response to sensory deprivation (Chapters 3 and 4; Figure 5.3).

Similar sensory hierarchies have been developed for a few other vertebrates, but only for one or two phases of a behavior, such as orientation and striking in fish (New et al., 2001), striking in snakes (Haverly and Kardong, 1996; Young and Morain, 2002), and prey capture in frogs (Monroy and Nishikawa, 2011). This study is the first to reveal a sensory hierarchy that covers a complete behavioral sequence in a vertebrate. All three species in this study rely on a suite of sensory modalities for feeding, however, they differ in the importance of the various senses for some of the behavioral phases. These differences may be linked to differences in habitat, prey type, or hunting strategy or they could be linked to anatomical differences. Differences in the relative volumes of the regions of the brain (telencephalon, diencephalon, mesencephalon, cerebellum, and medulla) (Northcutt, 1978; Demski and Northcutt, 1996), of the four sensory areas

(olfactory bulbs, optic tectum, anterior lateral line lobe, and posterior lateral line lobe)

(Lisney and Collin, 2006; Lisney et al., 2007), or of the sensory organs (olfactory

249

rosettes, eyes) (Theisen et al., 1986; Zeiske et al., 1986; Zeiske et al., 1987; Kajiura et al.,

2005; Lisney and Collin, 2007; Schluessel et al., 2008; Theiss et al., 2009; Schluessel et

al., 2010) have often been suggested to be linked to differences in the importance of the

sensory systems for tasks such as feeding or social behavior. No study to date, however,

has compared differences in behavior with differences in anatomy in elasmobranchs.

The telencephalon, which is primarily olfactory (reviewed in Northcutt, 1978),

comprises the majority of the total brain volume in the carcharinid, sphyrnid, and

ginglymostomid species examined to date, but is relatively largest in sphyrnid sharks,

followed by ginglymostomid sharks, and carcharhinid sharks, at averages of 59.46%,

58.30%, and 56.27% respectively (Yopak et al., 2007). The telencephalon includes the

forebrain, which is multimodal (Cohen et al., 1973; Graeber et al., 1973; Veselkin and

Kovačević, 1973; Bullock and Corwin, 1979; Ebbesson, 1980; Nikaronov and Lukyanov,

1980; Nikaronov, 1983; Bodznick and Northcutt, 1984; Bodznick, 1991; Hoffman and

Northcutt, 2008), but also the olfactory bulbs, which also comprise a relatively larger proportion of the sensory brain volume of adult sphyrnid sharks (61.3%) than of juvenile carcharhinid species (33.2%) (Lisney et al., 2007). Sphyrnid species also have the greatest number of olfactory lamellae, followed by carcharhinid species, and ginglymostomid species (Schluessel et al., 2008). The mesencephalon, on the other hand,

is relatively larger, on average, in carcharhinid sharks than in sphyrnid sharks or ginglymostomid sharks, at 9.76%, 5.45%, 4.39% (Yopak et al., 2007). Similarly, the optic tectum comprises a greater proportion of the sensory brain of juvenile carcharhinid

sharks (45.4%) than adult sphyrnid sharks (19.9%) (Lisney et al., 2007). These results

must be interpreted with caution, however, as the optic tectum is multimodal (Platt et al.,

250

1974; Boord and Northcutt, 1982, 1988; Bodznick, 1991; Hoffman and Northcutt, 2008).

The medulla, to which the lateral line, electrosensory, statoacoustic, gustatory, and tactile systems project, is relatively largest in carcharhinid sharks, followed by ginglymostomid sharks, and sphyrnid sharks at 14.19%, 10.16%, and 9.71%, respectively (Yopak et al.,

2007). The results of Lisney et al. (2007) suggest that the anterior lateral line lobe

(electrosensory) and the posterior lateral line lobe (lateral line and statoacoustic systems), occupy greater proportions of the sensory brain volume in juvenile sphyrnid sharks (6.2% and 22.6%, respectively; but not adults, 3.8% and 15.0%, respectively) than in carcharhinid sharks (4.5% and 17.0%, respectively) (Lisney et al., 2007). These results must be interpreted with caution though, as the adult results are based on a single specimen of one species, the scalloped hammerhead, Sphyrna lewini. In both families and both age classes, however, the posterior lateral line lobe comprises a greater percentage of the sensory brain volume than the anterior lateral line lobe (Lisney et al.,

2007). Together, these results suggest that in juvenile carcharhinid sharks, such as the blacktip sharks in this study, vision is the most important sense, followed by olfaction, then the lateral line, then electroreception. In adult sphyrnid sharks, such as the bonnetheads in this study, olfaction is the most important sense, followed by vision, then the lateral line, then electroreception, while in ginglymostomid sharks, such as the nurse shark, olfaction is the most important sense, followed by the lateral line and electroreception, then vision. However, these sensory systems mediate behaviors other than feeding (reviewed in Hueter et al., 2004) and thus the sizes of the different regions of the brain, sensory areas of the brain, and sensory structures may not directly reflect their importance in feeding behavior.

251

In all three species, long-distance detection is mediated by olfaction. Blacktip

sharks, which hunt for midwater teleost prey in the water column (Castro, 1996) and

bonnetheads, which sometimes hunt for teleost prey and swimming crabs in the water

column above seagrass beds (Cortés et al., 1996; Bethea et al., 2007), can also detect prey

visually. The nurse shark, on the other hand, which often corners fish that are hidden in

reef crevices at night (Hodgson and Mathewson, 1971; Mathewson and Hodgson, 1972;

Motta et al., 2008), relies completely on olfactory cues for prey detection. Tracking in all three species is mediated by olfaction in combination with vision or the lateral line. The nurse shark, however, can also track using olfaction in combination with touch. Since this species typically maintains contact with the substrate (Moss, 1972; Limbaugh, 1975), other benthic species, such as other orectolobiforms and benthic batoids, may also be able to track using tactile cues. In the ram-feeding species, the blacktip shark and bonnethead,

vision mediates orientation. In the blacktip shark, vision appears to play the greatest role

in feeding, as it mediates the long-distance strikes necessary for chasing down highly elusive teleost prey in the water column (Frazzetta and Prange, 1987; Castro, 1996). In

the bonnethead, vision is required for orienting to and striking at prey in the water column, which would be important for capturing teleost prey or swimming crabs (Cortés et al., 1996; Bethea et al., 2007). The blacktip shark can, however, use the lateral line in the absence of vision to orient to and strike at the prey, while the bonnethead cannot. In the nurse shark, vision, the lateral line, and electroreception mediate orientation and striking. Capture in all three species is mediated by electroreception, likely as a result of the visually blind area that exists directly in front of the head in head in sharks (McComb et al., 2009). Bonnetheads require electroreception for capture, which is likely also

252 important for locating benthic prey which might be hidden in the seagrass (Cortés et al.,

1996), but the blacktip shark and nurse shark can rely on tactile cues for capture when electrosensory cues are lacking.

These sensory hierarchies have been developed under a specific set of experimental conditions and they would likely change to some degree under different experimental conditions in captivity or different environmental conditions in the wild.

The animals in these experiments were all tested within the confines of a tank. As such, there is a limited amount of space, which means that there is a far greater likelihood of the shark passing within close proximity of the prey as a result of cruising the tank, as compared to the open ocean. Therefore, some of the alternative strategies observed in these experiments may occur infrequently in the wild as the animal may not get close enough to the prey for detection (e.g. visual detection of prey in an olfactory blocked blacktip). However, this would also suggest that when an animal with some of its senses blocked does not strike or capture prey in the test tank (e.g. olfactory blocked nurse shark), it is incapable of doing so, as it has passed within the range of detection the remaining senses (e.g. electrosensory system). The test tank is also less visually complex than a reef or seagrass bed, but has more visual features than the open ocean. The animals in this study may therefore be more likely to detect prey visually than they would in the wild, as prey may be more difficult to detect visually against a reef relief (Marshall,

2000; Losey, 2003) or seagrass (Cournoyer and Cohen, 2011) background, but less likely to detect prey visually than a pelagic species in the visually homogeneous open ocean

(McFall-Ngai, 1990; Hamner, 1995).

253

Although the species of sharks examined in this study all possess the same suite

of sensory modalities, they differ in terms of which senses they choose to focus on for particular behaviors, likely as a result of differences in the environments that they hunt in, type of prey consumed, and foraging strategies used, as well as anatomical differences in the central nervous system and the sensory organs. Feeding behavior in these animals is generally plastic; in most cases, multiple senses can be used for the same behavioral task and thus the sharks are capable of successfully capturing prey even when the optimal sensory cues are unavailable, by switching to alternative sensory modalities. This suggests that these animals are well adapted to succeed even in the face of a changing environment. This flexibility in behavior and the use of the sensory systems may in part explain the success of this vertebrate group that dates back about 400 million years.

This dissertation represents the first study to examine multimodal integration across a complete behavioral sequence: feeding in sharks. It does not, however, address the mechanisms by which this integration is occurring or the location in the brain where it occurs. As little is known about sensory integration in elasmobranchs, as compared to other vertebrates (e.g., mammals), future research should focus on neurophysiology to determine where and how multimodal integration occurs.

Literature Cited

Arnold, S. J. (1983). Morphology, performance and fitness. American Zoologist 23, 347- 361.

Baker, C. F. and Montgomery, J. C. (2001). Sensory deficit induced by cadmium in banded kokopu, Galaxias fasciatus, juveniles. Environmental Biology of Fishes 62, 455-464. 254

Bardach, J. E., Fujiya, M. and Holl, A. (1965). Detergents: effects on the chemical senses of the fish, Ictalurus natalis (le Sueur). Science 148, 1605-1607.

Bell, G. P. (1985). The sensory basis of prey location by the California leaf-nosed bat Macrotus californicus (Chiroptera: Phyllostomidae). Behavioral Ecology and Sociobiology 16, 343-347.