xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001430588
007 cr mnu|||uuuuu
008 031007s2003 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000106
An examination of modulation of feeding behavior in the nurse shark Ginglymostoma cirratum (Bonaterre 1788)
h [electronic resource] /
by Michael Patrick Matott.
[Tampa, Fla.] :
University of South Florida,
Thesis (M.S.)--University of South Florida, 2003.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 89 pages.
ABSTRACT: The ability of an organism to modulate its feeding behavior is an important focus of feeding ecology studies. Modulation is the ability to distinctly and consistently alter a behavior to accommodate different stimuli. The goal of this study was to examine the ability of the nurse shark Ginglymostoma cirratum to modulate its food capture behavior with different sizes and types of food items. This was carried out through kinematic and electromyographic analysis. Eight sub-adult specimens of G. cirratum were filmed feeding on two different food types (squid and fish) and sizes (gape size and larger than gape size). Filming consisted of high-speed videography utilizing a low-light digital video system. Kinematic variables related to lower jaw movement, mouth width, and head angle were measured from video footage. Up to twelve muscles in each of six specimens were implanted with bipolar electrodes to measure the onset and duration of motor activity. There were no significant differences between food sizes and any of the kinematic variables. Only two muscles showed significant differences in onset time based on food size. In regards to food types, squid bites were significantly faster than fish bites, but when examined proportionately to bite duration only the time to jaw closure remained significantly different. The motor pattern of G. cirratum demonstrates an anterior to posterior sequence, which corresponds to the anterior to posterior kinematic sequence. Little cranial elevation is present during feeding sequences and is not thought to contribute significantly to feeding. Ginglymostoma cirratum is a stereotyped, inertial suction feeder. There is little evidence that there is modulation in feeding behavior based on food size or food type. If modulation does exist in the feeding behavior, it is more likely to occur after prey capture while the prey is being processed and manipulated prior to transport. Initial observations suggested that a novel behavior termed 'spit-suck manipulation' is utilized for larger prey items.
Adviser: PhD, Philip J. Motta
t USF Electronic Theses and Dissertations.
AN EXAMINATION OF MODULATION OF FEEDING BEHAVIOR IN THE NURSE SHARK GINGLYMOSTOMA CIRRATUM (BONATERRE 1788) by MICHAEL MATOTT A thesis submitted in partial fulfillment of the requirements for the degree of M a s t e r o f S c i e n c e Department of Biology College of Arts and Sciences University of South Florida Major Professor: Philip Motta, P h D Robert Hueter, Ph.D. Florence Thomas, Ph.D. Date of Approval: April 10, 2003 Keywords: feeding behavior, kinematics, electromyography, elasmobranch, modulation, Ginglymostoma cirratum Copyright 2003 Michael Matott
Acknowledgements This work was supported by a Mote Marine Laboratory USF Summer Fellowship and a Lerner-Gray Marine Fund. This project could not have been completed without the assistance of my fellow graduate students, particularly Desiree Sasko, Heather Porter, Gaddy Bergmann, Dayv Lowry, Dan Huber, Alpa Patel, Lisa Whitenack, and Mason Dean. In addition, numerous undergraduate students contributed their time and assistance to this project, most especially Greg Sass, Ross Meyer, and Amelia Griffiths. Finally I am indebted to my family for their support and encouragement throughout the process of this project.
i T able of Contents List of Tables ii List of Figures iv Abstract vi Chapter One: Introduction 1 Functional Morphology 1 Functional Morphology and Feeding Behavior 2 Modulation and Variability 9 Elasmobranch Feeding Mechanism 12 Experimental Animal 14 Research Goals 15 Chapter Two: Materials and Methods 16 Kinematics 16 Electromyography 19 Statistical Analysis 22 Chapter Three: Results 24 Description of Feeding Behavior 24 Kinematic Variables 30 Electromyography 41 Chapter Four: Discussion 56 Feeding Behavior 56 Modulation 62 Variability 65 Motor Pattern 66 Post capture manipulation 69 Summary 70 References 72
ii List of Tables Table 1 Component loadings of kinematic variables for kinematic variables for principal components analysis. 31 Table 2 Mean kinematic values (plus or minus standard error) from prey capture events of eight nurse sharks Ginglymostoma cirratum. 36 Table 3 Results of Kruskal-Wallis ANOVA on Ranks for kinematic duration variables for different food sizes (gape sized and larger than gape). 37 Table 4 Results of Kruskal-Wallis ANOVA on Ranks for kinematic duration variables for different food types (fish and squid). 38 Table 5 Results of one-way ANOVA and Kruskal-Wallis ANOVA on Ranks for the nurse shark Ginglymostoma cirratum for food types (squid and fish), scaled to bite duration. 39 Table 6 Results of Kruskal-Wallis ANOVA on Ranks for kinematic duration variables for eight nurse sharks Ginglymostoma cirratum 40 Table 7 Mean angular values and gape distance (plus or minus standard error) from food capture events of four nurse sharks Ginglymostoma cirratum 42 Table 8 Results of one-way ANOVA for distance and angular variables for the nurse shark Ginglymostoma cirratum for different food sizes. 43 Table 9 Results of one-way ANOVA for distance and angular variables for the nurse shark Ginglymostoma cirratum for different food types. 44 Table 10 Results of one-way ANOVA for distance and angular variables for the nurse shark Ginglymostoma cirratum 45 Table 11 Component loadings of EMG variables for principal components analysis. 47
iii Table 12 Mean values for onset and duration of muscle firing in the nurse shark Ginglymostoma cirratum for the overall data set, gape-sized food items, and larger than gape-sized food items. 52 Table 13 Results of one-way ANOVA and Kruskal-Wallis ANOVA on Ranks for EMG variables based on food size. 53 Table 14 Results of one-way ANOVA and Kruskal-Wallis ANOVA on Ranks for EMG variables based on food type. 54 Table 15 Results of one-way ANOVA and Kruskal-Wallis ANOVA on Ranks for individual nurse sharks for EMG variables. 55
iv List of Figures Figure 1. Landmark variables for the nurse shark Ginglymostoma cirratum. 18 Figure 2. Ventral musculature of the nurse shark Ginglymostoma cirratum with implanted muscles shaded red 20 Figure 3. Dorsal musculature of the nurse shark Ginglymostoma cirratum with implanted muscles shaded red 21 Figure 4. Kinematic profile of feeding in a nurse shark Ginglymostoma cirratum. 25 Figure 5. Kinematic profile for feeding in the nurse shark Ginglymostoma cirratum. 26 Figure 6. Kinematic sequence of spit-suck manipulation in the nurse shark Ginglymostoma cirratum 28 Figure 7. Second example of spit-suck manipulation in the nurse shark Ginglymostoma cirratum 29 Figure 8. PCA of kinematic variables by individual shark. 32 Figure 9. PCA of kinematic variables indicating food type 33 Figure 10. PCA of kinematic data with food size indicated. 34 Figure 11. PCA of all EMG variables with individual shark indicated. 48 Figure 12. PCA of all EMG variables by food type. 49 Figure 13. PCA of all EMG variables by food size. 50 Figure 14. Compilation of muscle firing for the nurse shark Ginglymostoma cirratum with error bars representing standard error. 57 Figure 15. Compilation of muscle firing for the nurse shark Ginglymostoma cirratum for food that is gape-sized, with error bars representing standard error. 58
v Figure 16. Compilation of muscle firing for the nurse shark Ginglymostoma cirratum for food that is larger than gape-sized, with error bars representing standard error. 59
vi An examination of modulation of feeding behavior in the nurse shark Ginglymostoma cirratum (Bonaterre 1788) Michael Matott ABSTRACT The ability of an organism to modulate its feeding behavior is an important focus of feeding ecology studies. Modulation is the ability to distinctly and consistently alter a behavior to accommodate different stimuli. The goal of this study was to examine the ability of the nurse shark Ginglymostoma cirratum to modulate its food capture behavior with different sizes and types of food items. This was carried out through kinematic and electromyographic analysis. Eight sub-adult specimens of G. cirratum were filmed feeding on two different food types (squid and fish) and sizes (gape size and larger than gape size). Filming consisted of high-speed videography utilizing a low-light digital video system. Kinematic variables related to lower jaw movement, mouth width, and head angle were measured from video footage. Up to twelve muscles in each of six specimens were implanted with bipolar electrodes to measure the onset and duration of motor activity. There were no significant differences between food sizes and any of the kinematic variables. Only two muscles showed significant differences in onset time based on food size. In regards to food types, squid bites were significantly faster than fish bites, but
vii when examined proportionately to bite duration only the time to jaw closure remained significantly different. The motor pattern of G. cirratum demonstrates an anterior to posterior sequence, which corresponds to the anterior to posterior kinematic sequence. Little cranial elevation is present during feeding sequences and is not thought to contribute significantly to feeding. Ginglymostoma cirratum is a stereotyped, inertial suction feeder. There is little evidence that there is modulation in feeding behavior based on food size or food type. If modulation does exist in the feeding behavior, it is more likely to occur after prey capture while the prey is being processed and manipulated prior to transport. Initial observations suggested that a novel behavior termed 'spit-suck manipulation' is utilized for larger prey items.
Introduction Functional Morphology Studies in morphology often focus on the evolutionary development of structures and their function. Functional morphology examines physical characteristics of structures and their use within an organism. There are wide varieties of applications of functional morphology. For instance, numerous studies have linked body shape to locomotory ability in fish (Webb, 1982; 1984; Arreola and Westneat, 1996; Fish and Shannahan, 2000). Other studies in fish functional morphology include studies on reproduction (Okuda et al., 2002), ontogeny (Osse, 1995; Hjelm et al., 2000; Koumoundouros et al., 2001) and vision (Andison and Sivak, 1994; Vandermeer et al., 1995; Higgs and Fuiman, 1996; Wagner et al., 1998; Bozzano et al., 2001) The common focus of these studies is that the form of an organism affects the way in which it can perform a function. An extension of functional morphology that has gained interest over the past thirty years is ecomorphology (Dullemeijer, 1980; Goldschmid and Kotrschal, 1989; Bock, 1990; Motta and Kotrschal, 1992). Ecomorphology addresses how form and function interact within the framework of an organism's environment (Karr and James, 1975; Goldschmid and Kotrschal, 1989; Bock, 1990; Wake, 1992). The combination of modern morphological techniques and ecological studies has resulted in the rapid growth of this field.
2 Studies of morphological characteristics have allowed scientists to infer dietary breadth (Liem, 1979; Wainwright and Lauder, 1992), locomotory capabilities (Webb, 1984; Jayne, 1988; Gibb et al., 1991; Arreola and Westneat, 1996), habitat partitioning (Ehlinger, 1990; Douglas and Matthews, 1992; Kaicounis and Brigham, 1995), and convergence of morphological and ecological characters in a variety of species or groups (Wiens, 1991; Norton and Brainerd, 1993; Hugueny and Pouhly, 1999). Functional Morphology and Feeding Behavior Morphology and its relationship to feeding is a frequently studied topic (e.g. Liem, 1979; Gatz Jr., 1979; Lauder and Liem, 1982; Sanford and Lauder, 1988; 1989; Wainwright et al., 1988; Jayne, 1988; Reilly and Lauder, 1990a; Witte et al., 1990; Wiens, 1991; Herrel et al., 1999; Ferry-Graham et al., 2001a). One reason for this is its universality (Schwenk, 2000) All animals must eat to survive. The comparison of feeding structures, strategies, and performance provides insight into the evolution and diversity of animal life (Schwenk, 2000). The feeding system of the gnathostomes has a basic mechanical design and behavioral pattern that has been conserved across evolutionary time (Wainwright and Lauder, 1986; Reilly and Lauder, 1990b; Lauder and Shaffer, 1993). The numerous variations that do exist within this basic design allow for a wide range of specializations in feeding. Variation in the feeding system can occur within the morphology of the system, the behavior as controlled by the motor pattern, or both (Wainwright et al., 1989; Wainwright and Lauder, 1992). Motor activity of the muscles precedes mechanical activity and can be used to describe behavioral patterns. Some studies of fish, in which
3 morphological change has occurred across taxa over evolutionary time, have shown the motor patterns in feeding are conserved (Wainwright and Lauder, 1986; Wainwright, 1989). Other studies have shown that the feeding behavior in fish can be altered by changing the motor pattern while the morphological structures remain the same (Liem, 1980; Lauder, 1983a). Frequently differences in the feeding mechanism involve an addition to the basic design (Lauder and Shaffer, 1993). Tetraodontiform fishes, for example, have highly derived adductor mandibulae complexes that likely arose from duplication and subsequent divisions of the existing muscle, resulting in an increase in functional complexity (Friel and Wainwright, 1998; 1999). The development of the pharyngeal jaw apparatus for crushing of durophagous prey in some teleost fish is another example of a modification that has greatly expanded feeding ability (Liem, 1978; 1980; Meyer, 1987; Norton and Brainerd, 1993). The general kinematic pattern of feeding in fishes can be characterized by four phases (Liem, 1978; 1979). The preparatory phase, when present, consists of compression of the buccopharyngeal space, reducing its volume prior to mouth opening. It is thought that this preparatory phase increases the amount of suction that can be generated by increasing the overall volume change that occurs (Lauder, 1980b; 1983c) The expansive phase is from the time when the mouth begins to open to the maximum gape. In suction feeding it is during this phase that the prey item is drawn into the buccal cavity. The compressive phase is from the time of maximum gape to mouth closure. In fish, water flow continues unidirectionally in this phase by flowing out the gills (Lauder and Shaffer, 1993). The final phase is the recovery phase, in which the components of the feeding apparatus return to their initial resting state. Within this general kinematic
4 pattern there are numerous opportunities for variation of feeding behavior (Lauder, 1978; 1983; 1985; Liem, 1980; Wainwright and Lauder, 1986; 1992; DeVree and Gans, 1989; Wainwright et al., 1989; Ehlinger, 1990; Norton and Brainerd, 1993; Clifton and Motta, 1998; Wainwright and Shaw, 1999; Wainwright, 1999; Alfaro et al., 2001; Sanford, 2001). The complete behavioral sequence of feeding in an animal is characterized by three sequences: capture, manipulation, and transport (Lauder, 1983a). Capture is the acquisition of the prey item by the predator. Manipulation is the processing of the prey item to facilitate transport for digestion. In some cases, there is no manipulation and the prey item is transported whole. Transport is the process of moving the prey item from the buccal cavity into the digestive tract. Movements of the feeding mechanism during any of these four sequences may include preparatory, expansive, compressive, and recovery phases. For instance, during manipulation, the jaws may be opened and closed repeatedly in order to ease transport of a prey item. Repeated opening and closing behaviors would involve repeated expansive and compressive phases. In aquatic prey capture the prey may be acquired through inertial suction feeding, ram feeding, biting, or some combination of these behaviors (Liem 1980). During inertial suction feeding, the prey item is drawn into the mouth of the animal due to the sub-ambient pressure generated by the opening of the mouth. Suction feeding is believed to be the most common mode of feeding for aquatic vertebrates (Lauder, 1985) In ram feeding the predator overtakes the prey (Norton, 1995). Biting is a behavior distinct from either ram or suction feeding. Biting is when the jaws close down on a prey item, applying force to the surrounding material. Biting is used to secure moving prey, remove
5 portions of material from large prey items, or to remove attached prey items, such as barnacles, from a substrate (Liem, 1980; Norton, 1995). It should be noted that ram, suction, and biting are not mutually exclusive behaviors. Some species utilize all three behaviors, including a combination of ram and suction (Liem, 1993; Wainwright, 1999). However, there are species that utilize only one type of feeding behavior. Barracuda, needlefish, and gar are exclusively ram feeders (Porter, 2002). Among sharks, the Orectolobiformes contain multiple species that appear to be obligate suction feeders (Moss, 1977; Wu, 1994; Motta et al., 2002). The morphological and behavioral characteristics of these specialized species are of great interest for the insights they provide on constructional constraints and morphological specialization. Among the elasmobranchs, ram feeding has been described in Lamniformes (Tricas, 1985), Carcharhiniformes (Hobson, 1963; Moss, 1972; Frazzetta and Prange, 1987; Frazzetta, 1994; Motta et al., 1997; Wilga and Motta, 2000), Orectolobiformes (Gudger, 1941a; 1941b; Sanderson and Wasserug, 1993), Squatiniformes (Fouts and Nelson, 1999), and the Batoidea (Bigelow and Schroeder, 1953; Sanderson and Wasserug, 1993). Biting is observed in teleosts such as the parrotfish, cleaner wrasse, and various cichlids (Barel, 1983; Clifton and Motta, 1998; Norton, 1995; Bouton et al., 1999; Alfaro et al., 2001; Devaere et al., 2001). For elasmobranchs biting has been described principally among Carchariniformes and Lamniformes, often as part of ram feeding (Hobson, 1963; Tricas and McCosker, 1984; Tricas, 1985; Motta et al., 1997). Biting is seen in elasmobranchs typically with prey items that are too large to be
6 swallowed whole. Morphological structures associated with biting in elasmobranchs include multiple rows of serrated teeth and protrusibility of the upper jaw (Gilbert, 1970). The cookie-cutter shark Isistius brasiliensis has a unique feeding mechanism that allows it to remove smooth plugs of flesh from prey (Shirai and Nakaya, 1992) A common pattern seen with biting in sharks is for the lower jaw to be used to secure the prey prior to depression of the upper jaw and subsequent cutting of the prey item (Hobson, 1963; Gilbert, 1970; Frazzetta and Prange, 1987). This is found in sharks with dignathic heterodonty, in which the upper and lower jaws have different types of teeth (Compagno, 1990). A behavior that is frequently seen among elasmobranchs during biting is lateral head shaking. This behavior is thought to improve the cutting efficiency of the teeth by producing a shearing force on the prey item (Frazzetta and Prange, 1987) Vigorous lateral head shaking has been observed in white sharks Carcharodon carcharias lemon sharks Negaprion brevirostris tiger sharks Galeocerdo cuvier and Caribbean reef sharks Carcharhinus perezi among others (Springer, 1961; Moss, 1977; Tricas, 1985; Frazzetta and Prange, 1987; Frazzetta, 1994). The functional morphology of the suction feeding mechanism has been studied in a wide variety of teleost families including the Cichlidae (Liem, 1979; 1980; Barel, 1983; Wainwright et al., 2001), Labridae (Clifton and Motta, 1998; Wainwright, 1988; Westneat and Wainwright, 1989; Ferry-Graham et al., 2001b), and Centrarchidae (Lauder, 1983b; 1983c; Wainwright and Lauder, 1992). These studies have resulted in the identification of a general pattern of feeding sequences consisting of preparatory, expansive, compressive, and recovery phases, anterior to posterior movement of water (and entrained prey), and the anterior to posterior activation of muscles during feeding.
7 These same phases have been used to describe suction feeding in elasmobranchs (e.g. Ferry-Graham, 1997; 1998b; Wilga and Motta, 1998b; Motta et al., 2002; Sasko et al., Unpublished Manuscript) Morphological features that have been suggested to be beneficial for suction feeding fishes include a small, circular gape, protrusible upper jaw, and hypertrophied jaw abduction musculature (Liem, 1993; Norton, 1995). Welldeveloped labial cartilages, such as those found in orectolobiform sharks, have been proposed as an additional morphological feature that improves suction feeding in elasmobranchs (Wu, 1994; Motta et al., 2002). These cartilages are extended anteriorly during expansion of the buccal cavity and help form a narrow, circular gape. Kinematically, suction feeding is faster than ram feeding. In some suction feeding aquatic vertebrates there is a fast opening phase in which the expansive phase is much faster than the compressive phase (Lauder, 1985; Lauder and Reilly, 1990; 1994; Reilly and Lauder, 1992; Summers et al., 1998; Motta et al., 2002). Compared to teleosts, there have been far fewer studies on suction feeding in elasmobranchs. Suction feeding in elasmobranchs has been studied in the Batoidea (Belbenoit, 1986; Wilga and Motta, 1998b; Sasko et al., Unpublished Manuscript), Squaliformes (Wilga and Motta, 1998a), Carcharhiniformes (Ferry-Graham, 1998a), Heterodontiformes (Edmonds et al., 2001), and Orectolobiformes (Tanaka, 1973; Wu, 1994; Motta et al., 2002; Robinson and Motta, 2002). Suction feeding in batoids differs from that in other elasmobranchs due to a derived jaw depression mechanism that results in a highly kinetic jaw that can be protruded ventrally (Wilga and Motta, 1998b; Sasko et al., Unpublished Manuscript). Despite the differences in anatomy, the kinematic pattern is somewhat similar to that of
8 other elasmobranchs. Belbenoit (1986) found that the feeding behavior of Torpedo marmorata appeared to be stereotyped, although this was not specifically tested. A study of suction feeding in the Atlantic guitarfish Rhinobatos lentiginosus revealed a novel compressive transport phase not described in other elasmobranchs (Wilga and Motta, 1998b). Furthermore, palatoquadrate protrusion in R. lentiginosus occurs during the compressive phase, which is similar to other elasmobranchs, but protrusion in the suction-feeding cownose ray Rhinoptera bonasus occurs during the expansive phase due to tight ligamentous connections of the upper and lower jaw (Wilga and Motta, 1998b; Sasko et al., Unpublished Manuscript). Palatoquadrate protrusion may also occur in the expansive phase for the lesser electric ray Narcine brasiliensis and in orectolobids, including the nurse shark Ginglymostoma cirratum (Dean, personal communication; Wu, 1994). The spiny dogfish Squalus acanthias and leopard shark Triakis semifasciata are benthic sharks that utilize suction feeding to some degree in feeding (Ferry-Graham, 1998a; 1998b; Wilga and Motta, 1998a). Squalus acanthias uses primarily suction in 69% of its capture bites compared to just 24% of bites being suction dominated in T. semifasciata (Ferry-Graham, 1998a; Wilga and Motta, 1998a). Feeding mechanisms of these sharks support the suggestion that suction and ram feeding are not mutually exclusive behaviors. Heterodontiform and orectolobiform sharks are well suited for suction feeding (Tanaka, 1973; Wu, 1994; Edmonds et al., 2001; Motta et al., 2002; Robinson and Motta, 2002). These sharks are primarily benthic and feed on invertebrates, marine gastropods, and small fish (Castro, 2000; Compagno, 2001). Negative pressures of up to 102 kPa
9 have been recorded in suction-feeding nurse sharks (Tanaka, 1973) Modulation and Variability One of the most interesting components of feeding behavior is the degree of modulation a species is capable of performing during feeding. Modulation is the ability of the organism to distinctly and consistently alter its feeding behavior in order to accommodate different stimuli (Liem, 1978; 1979). This differs from variation, in that individuals may vary their feeding behaviors for prey types or sizes without following a consistent pattern. To be defined as modulation, a feeding strategy must be distinct for a specific prey type or size and consistently generated. Liem (1978) defined modulation based on elusivity of prey types and the different strategies employed by cichlids to capture these prey items. He found that piscivorous cichlids consistently had shorter bite times for elusive prey over sluggish prey (Liem, 1978). Modulation has been examined in both teleosts and elasmobranchs (Liem, 1978; Lauder, 1981; Muller and Osse, 1984; Nemeth, 1997; Friel and Wainwright, 1999; Ferry-Graham et al., 2001b; for a review of elasmobranch studies see Motta and Wilga, 2001). These studies and others have examined modulation based on different prey types (Norton, 1995; Nemeth, 1997) and sizes (Lauder, 1981; Nemeth, 1997; Ferry-Graham, 1998a) as well as predator capture strategies (Norton and Brainerd, 1993). Consequently, we can predict that prey that is larger than gape size may require more biting and manipulation than prey that is smaller than gape size. This should result in differences in motor pattern to accommodate for the differences in size. It is expected that larger prey items will result in shorter duration of kinematic variables because faster mouth opening can generate
10 greater suction force to accommodate the greater mass of the prey item (Lauder, 1980a; Muller et al., 1982; Muller and Osse, 1984). Kinematic studies of fishes that have taken into account prey size have not found modulation in feeding between prey items of different size (Lauder, 1981; Nemeth, 1997; Ferry-Graham, 1998a). Furthermore, acquisition of larger prey items by suction force requires a greater amount of suction than smaller prey items (Lauder, 1983c), therefore, we can predict that the elusivity and size of the prey should influence the amount of ram or suction feeding a predator uses in prey capture. In teleosts, a frequent focus of modulation studies has examined whether the morphological structure of the feeding apparatus predisposes certain fishes to specific capture strategies (Lauder, 1980b; 1983b; Norton and Brainerd, 1993; Norton, 1995; Friel and Wainwright, 1999). It has been proposed that fishes with small mouths should make use of suction capture more than fishes with large mouths (Norton, 1991; Norton, 1995). Nemeth (1997) proposed that fish with an intermediate mouth size might be able to utilize a greater array of feeding strategies by having a "generalized" mouth capable of utilizing both suction and ram feeding strategies. Kinematics of feeding in Heterodontus francisci and Ginglymostoma cirratum have suggested that feeding may be stereotyped within these species, although the motor pattern of either species has yet to be examined (Edmonds et al., 2001; Motta et al., 2002; Robinson and Motta, 2002). Based on food items that were smaller than gape, Motta et al (2002) suggested that G. cirratum was stereotyped in its use of inertial suction feeding, as there were no significant differences in kinematics between bites for individual sharks, although there was variation among individuals. An ontogenetic study of kinematics in
11 G. cirratum supported the suggestion that nurse sharks are stereotyped in their feeding behavior (Robinson and Motta, 2001). Wu (1994) studied protrusion in three species of orectolobiform sharks: epaulette shark Hemiscyllium ocellatum nurse shark Ginglymostoma cirratum and spotted wobbegong Orectolobus maculatus He found that the feeding kinematics were stereotyped for each species, with amount of palatoquadrate protrusion being the one variable to differ among species. In elasmobranchs, modulation has been examined in relatively few species (FerryGraham, 1997; 1998a; 1998b; Wilga and Motta, 1998b; Edmonds et al., 2001). A study of juvenile leopard sharks Triakis semifasciata found no evidence of modulation based on prey size or elusivity, although it was suggested that the sharks may be able to modulate the amount of suction used in prey capture to compensate for prey size (Ferry-Graham, 1998a). The results of this study also suggested that the prey items offered might not have differed significantly in elusivity to be a valid test of modulatory capability. Studies on hatchling and juvenile swellsharks failed to detect modulation based on prey size (Ferry-Graham, 1997; Ferry-Graham, 1998b). A study of horn sharks Heterodontus francisci did not detect modulation based on prey accessibility, but this study utilized only one prey type (Edmonds et al., 2001). Wilga and Motta (1998b) reported modulation in motor activity between feeding phases in the Atlantic guitarfish Rhinobatos lentiginosus but modulation was not examined for different prey types, sizes, or elusivity. Other feeding studies on sharks have examined variation within a species without focusing on modulation. For instance, variation in the timing and degree of cranial elevation has been found for Negaprion brevirostris (Frazzetta and Prange, 1987) Moss
12 (1972) noted differences in N. brevirostris and Galeocerdo cuvier feeding based on position in the water column. Sharks feeding on items on the bottom picked them up by swimming over them and impaling them on the teeth of the lower jaw. Prey items in midwater were often swallowed whole by simply opening the mouth and engulfing the prey item. Surface bites resulted in cranial expansion and arching of the back. Items that could not be swallowed whole elicited biting and lateral head-shaking behaviors. A common result of these studies, and other studies on feeding behavior in elasmobranchs, is that inter-individual variation within a species tends to be high, which may mask the presence of modulation. The low sample sizes typical of most elasmobranch studies make it difficult to demonstrate statistically significant differences between feeding behaviors. Elasmobranch Feeding Mechanism The feeding mechanism of elasmobranchs has been the focus of a number of recent functional morphological studies (Moss, 1972; 1977; Shirai and Nakaya, 1992; Frazzetta, 1994; Wu, 1994; Motta et al., 1997; Ferry-Graham, 1997; 1998b; Wilga, 1997; 1998b; Edmonds et al., 2001). The jaws of elasmobranchs consist of cartilaginous structures with a prominent palatoquadrate cartilage forming the upper jaw and Meckels cartilage forming the lower jaw. Most sharks have a hyostylic form of jaw suspension, where the palatoquadrate articulates with the neurocranium primarily via the hyoid arch (Moss, 1972; 1977; Maisey, 1984; Wilga et al., 2000a; Wilga et al., 2001). The hyoid arch consists of paired hyomandibulae, paired ceratohyals, and a single basihyal. Articulation of the palatoquadrate with the hyoid occurs at the hyomandibula. There is a
13 loose articulation between the palatoquadrate and ethmoid region of the cranium in hyostylic sharks. The most studied feeding mechanism of elasmobranchs has been that of carcharhiniform sharks. In these sharks the jaw is opened by contraction of a number of muscles including the coracomandibularis, coracohyoideus, and coracoarcualis (Motta et al., 1997; Wilga et al., 2000b). Contraction of the coracomandibularis results in depression of the mandible, and consequently mouth opening (Wilga et al., 2000b). The contraction of this muscle pulls the hyoid in a posteroventral direction, pulling both the palatoquadrate and mandible anteroventrally (Wilga et al., 2000b). The coracoarcualis forms a complex with the coracohyoideus that functions to depress the hyoid. The major jaw adductor is the quadratomandibularis. This muscle is a complex of multiple heads (Motta and Wilga, 1999) The palatoquadrate is raised by the levator palatoquadrati. In carcharhiniform sharks the levator palatoquadrati is believed to contribute to protrusion of the jaws in association with the preorbitalis muscle (Motta et al., 1997; Wilga and Motta, 1998a; Wilga et al., 2000a). In squaliform sharks the preorbitalis muscle is the primary muscle responsible for palatoquadrate protrusion (Wilga and Motta, 1998a). Wu (1994) proposed a mechanism for protrusion in Orectolobiformes in which the interhyoideus and intermandibularis mediate protrusion through action on the ceratohyals.
14 Experimental Animal In order to examine modulation in feeding with elasmobranchs it is important to choose a species in which the presence or absence of modulation is liable to be clearly demonstrated. Motta et al. (2002) have suggested that specialized suction feeders, such as the nurse shark, are the least likely to demonstrate modulation. The nurse shark, Ginglymostoma cirratum (Bonaterre 1788) (Galea; Orectolobiformes; Ginglymostomidae) is widely distributed throughout tropical and sub-tropical coastal waters on both sides of the Atlantic and in the Eastern Pacific (Compagno, 2001). Nurse sharks are benthic and are often found beneath overhanging coral heads or other structures on reefs. Numerous characteristics of the nurse shark are suited for suction feeding. The mouth is sub-terminal when closed, but becomes terminal when opened. The labial cartilages frame the mouth when open, creating a small, nearly circular opening (Wu, 1994; Motta et al., 2002; Robinson and Motta 2002). Nurse sharks have small, unserrated homodont dentition, with each tooth consisting of a large central conical cusp and two smaller lateral cusps (Goto, 2001; Compagno, 2001). This dentition type is suited for piercing prey, but not for cutting or tearing (Cappetta, 1987; Williams, 2001). Nurse sharks have a robust jaw and hyoid apparatus that accommodates the large quadratomandibularis, coracomandibularis, and coracohyoideus muscles (Moss, 1965; Wu, 1994). Nurse sharks do well in captivity and are therefore a popular research subject (Hamasaki and Gruber, 1965; Gelsleichter et al., 1998; Carrier and Luer, 1990). The anatomy of G. cirratum is well studied (Wu, 1994; Motta and Wilga, 1999), as is prey capture behavior (Tanaka, 1973; Wu, 1994; Motta et al., 2002; Robinson and Motta,
15 2002). Previous reports stated that durophagous invertebrate prey made up a large portion of the nurse sharks diet (Gudger, 1921; Compagno, 2001). However, Castro (2000) found that small fish, primarily grunts (Haemulidae), were more common than invertebrates in a sample of nurse sharks from Florida and the Bahamas. Research Goals The goal of this study is to examine the modulatory ability of G. cirratum feeding behavior in response to different sizes and types of prey. Based on the results from Motta et al. (2002), it is expected that the kinematic pattern for prey capture will be stereotyped among individuals, with differences primarily in duration of kinematic variables among individuals. Variation among individuals has been a common finding in studies on elasmobranch feeding, (Ferry-Graham, 1997; 1998a; 1998b; Wilga and Motta, 1998a; Motta et al., 1991; Motta et al., 2002) and is expected in this study. Modulation between food types and sizes is not expected for either kinematic or motor patterns, based on the stereotyped response seen previously for nurse sharks with smaller than gape-sized food items (Motta et al., 2002; Robinson and Motta, 2002).
16 Materials and Methods Kinematics Eight sub-adult Ginglymostoma cirratum (TL 61.0 -110.0 cm; mean TL 85.3 cm) were trained to feed in a 2.4 m diameter, 1,400 l semicircular tank with a 0.5 by 1.7 m acrylic window at Mote Marine Laboratory, Sarasota, Florida and in a similar tank at the University of South Florida, Tampa, Florida. Food types consisted of fish (Atlantic thread herring Opisthonema oglinum ) and squid ( Loglio spp.) in two size classes: approximately gape width, and approximately twice gape width. High-speed video was recorded with a Redlake PCI 1000 camera at 250 fields/sec and a NAC HSV200 at 200 fields/sec. The first two animals were filmed using the NAC camera and the remaining animals were filmed using the Redlake camera system. Lateral and ventral views of the shark feeding were acquired with a mirror platform set at 45 beneath the animal. Twelve kinematic variables of prey capture were measured and analyzed for each prey type and size. Events following prey capture were not analyzed as post-capture events often lasted longer than the Redlake camera recorded and the shark often swam off-field with the prey item before performing manipulation, transport, and recovery behaviors. The kinematic variables that were measured were (1) bite duration, (2) time to mandible depression, (3) time to mandible elevation, (4) total time of mandible
17 depression and elevation, (5) gape duration, and (6) time from mandible depression to maximum gape. Hyoid depression and palatoquadrate protrusion could not be measured because these events were not typically visible. The palatoquadrate was obscured by extension of the labial cartilages and hyoid depression was masked by contraction of the ventral musculature. To measure distance data, selected video images from each animal were captured and analyzed using SigmaScan Pro 4.01. Variables that were measured using landmarks were (1) gape width, (2) resting jaw angle, (3) jaw angle at maximum gape, (4) resting cranial angle, (5) maximum cranial angle, and (6) total cranial elevation angle (Figure 1). Bite duration was measured from the field prior to lower jaw depression to the field in which the lower jaw stopped moving upwards. Time to mandible depression was measured starting from the field prior to lower jaw movement to the field in which the lower jaw stopped moving downward. Time to mandible elevation was measured from the field prior to upward movement of the lower jaw to the field in which the lower jaw stopped moving upward. Total time of mandible depression and elevation is simply the addition of the previous two duration variables. The latter measure differs from bite duration in that bite duration encompasses lower jaw depression and elevation but does not provide an indicator of whether or not the animal held its jaws agape for any period of time. The difference between bite duration and the total time for mandible depression and elevation indicates the amount of time the jaw was held agape during a prey capture sequence, and this short time period was defined as gape duration. The time from mandible depression to maximum gape was measured from the field prior to lower jaw
19 movement to the field in which the mouth was at its maximum gape. Gape width was measured from the anteriormost point of the mandible to the anteriormost point of the palatoquadrate. Jaw angle was defined by the anteriormost points of the palatoquadrate and mandible, using the bottom of the medial labial cartilage as the vertex (Figure 1). For resting jaw angle the field prior to mandible depression was chosen. Gape angle was measured using the field in which maximum gape occurred. Cranial angle was defined using a point on the dorsal surface above the first gill slit and the tip of the rostrum, with a point on the dorsal surface above the eye serving as the vertex (Figure 1). Resting cranial angle was measured from the same field as resting jaw angle. Maximum cranial angle was measured from the field in which maximum gape occurred. The difference between maximum cranial angle and resting cranial angle represents total cranial elevation angle. Electromyography Six of the G. cirratum were used for electromyography (EMG) studies. Specimens were anesthetized with 0.1-g/L tricaine methanesulfonate (MS-222). Up to twelve cranial muscles were implanted with bipolar electrodes inserted using 23 gauge hypodermic needles similar to the method of Motta et al. (1997). Electrodes were implanted in the superficial head of the superior division of the preorbitalis, anterior division of the preorbitalis, quadratomandibularis, interhyoideus, intermandibularis, levator hyomandibularis, coracobranchialis, coracoarcualis, coracomandibularis, coracohyoideus, and epaxialis (Figures 2 and 3; nomenclature follows Motta and Wilga, 1999). Electrodes were differential amplified at 1000X (AM Systems Inc., Model 1 700),
22 bandpass (100-3000 Hz) and notch (60 Hz) filtered. Signals were recorded simultaneously on a Yokogowa DL416 oscilloscope and pulse code modulator (AR Vetter MV 520) and stored on VHS tape. One channel recorded a digital pulse synchronized with a pulsing LED recorded by the high-speed camera. Feeding experiments commenced approximately one hour following recovery from surgery after the animal resumed normal swimming and respiratory behaviors. Food was offered on wooden tongs with the longest dimension of the food item parallel to the mouth of the animal. The food item was held loosely in the tongs and released prior to the prey capture in most cases. In some cases the animal removed the food item directly from the tongs before it could be released. Following the experiment all six animals were sacrificed by an overdose of MS-222 in order for electrode placement to be verified via dissection, following University of South Florida and Mote Marine Laboratory Institutional Animal Care and Use guidelines (USF IACUC #1734). EMG data were converted from analog to digital using a Cambridge Electronics Design (CED) 1400 converter and downloaded to computer. The prey capture sequences were then analyzed using Spike2 and custom EMG analysis program (Cambridge Electronics Design, Ltd.). All sequences were referenced with respect to jaw opening as determined from the video. Statistical Analysis Mean, standard error, standard deviation, and variance were determined for each kinematic variable. All variables were regressed against total lengths to investigate for size dependent difference. As there were no significant relationship between size and any
23 of the kinematic variables the original data was further analyzed. A principal components analysis (PCA) was performed separately on both kinematic and EMG variables using a correlation matrix with SPSS 11.0 software. Missing data for each PCA was filled using a linear trend point. The first three components of each PCA were plotted on a scatter plot. Additional scatter plots were made by grouping data by individual, food type, and food size. Normality and equality of variance assumptions were tested respectively by the Kolmogrov-Smirnov normality test and the Levene Median test using SigmaStat 2.03. Comparison of kinematic and motor events for prey capture events were made using separate one-way ANOVAs for each kinematic and EMG variable. The factors that were examined were individual, prey type, and prey size. A Tukeys multiple comparison test was performed a posteriori to test all pairwise comparisons (Zar 1996). Non-parametric Kruskal-Wallis ANOVA on ranks was performed in cases in which assumptions of normality and heterogeneity of variance were not met. Dunnets multiple comparison test was performed on significant ANOVA on ranks when there were equal sample sizes. Dunns multiple comparison test was performed on significant ANOVA on ranks when the sample sizes were unequal. Two-way or three-way ANOVA was not possible due to sample size and the unbalanced design due to missing data in some cells, so interaction effects could not be directly determined.
24 Results Description of Feeding Behavior The nurse shark Ginglymostoma cirratum captured food in all cases through the use of inertial suction. The food item was visibly drawn to the mouth of the animal following rapid buccal expansion. In most cases, the animal stopped moving forward before capture. In other cases, the animal captured the food item while swimming towards it. Capture bites followed the same basic pattern (Figure 4). The mandible was depressed and expansion of the buccopharyngeal cavity occurred. The expansion occurred in an anterior to posterior direction. Mandible depression lasted on average 46 msec. Maximum gape was reached shortly after mandible depression, on average 56 msec after the start of mandible depression, however in 26% of the bites the times to mandible depression and maximum gape were simultaneous. The mouth was then held agape for an average of 66 msec. During this period, the food item was drawn to the mouth as a result of water flow into the buccopharyngeal cavity. Once the mandible began to elevate, it took an average of 53 msec for closure of the jaws. The average bite duration was 164 msec, but ranged from 48 to 628 msec. The kinematic profiles for the overall data set, by food size, and food type are given in Figure 5. The differences in these kinematic profiles with respect to food type and size are discussed below.
26 Kinematic Profile 0 50 100 150 200 250 Duration (msec) Overall Gape Size Plus Gape Size Fish Squid Figure 5: Kinematic profile for feeding in the nurse shark Ginglymostoma cirratum Profiles are shown for the overall data set, capture sequences with gape-sized food, capture sequences with larger than gape-sized food, capture sequences with fish as the food type, and capture sequences with squid as the food type. The first bar represents the duration of mandible depression. The represents the time at maximum gape. The second bar represents the duration of mandible elevation. The space between the two bars represents the time the mouth is held agape. During this period cranial elevation, if present, occurs.
27 In some cases, typically with gape-sized prey, the prey item was transported completely into the buccal cavity during the initial acquisition sequence and immediately transported. Dissection of the stomach of three subjects following EMG experiments showed no or very little evidence of cutting or tearing of the food by the teeth. Smaller sized pieces of fish were apparently swallowed whole without processing by the teeth. When the food item was larger than gape-sized it would often be held between the jaws and then manipulated. This manipulation consisted of one or both of the following behaviors: reorientation of the food item along its long axis, and repeated spit and suction behaviors termed 'spit-suck manipulation'. Reorientation of the food item consisted of either the animal dropping the food item and reacquiring it, or the animal actively expelling the item from its mouth and then sucking it back in. In some cases the reoriented food item was then transported in a single suction event. Spit-suck manipulation was observed with some large food items and consisted of the repeated alteration of suction and spitting behaviors with the food item located in the buccal cavity (Figures 6 and 7). In some cases the food item was seen to come partially out of the mouth and then be sucked back in. Flesh and viscera were at times visibly expelled from the gill slits during the spit-suck manipulation. Repeated compression and flaring of the branchial region was observed during these manipulation events. These post-capture manipulation events were not quantified due to the fact that they often exceeded the video recording time and/or were carried out off-frame. From dissection of the stomach, fish that were manipulated in this way were shredded, but there was no way to determine if the teeth were used to shred the fish or if the damage to the food item was
30 due to suction force alone. Squid that were manipulated in this way were not shredded and showed no sign of deep cutting by the teeth, although there were occasionally shallow tooth marks. Kinematic Variables Principal components analysis of all kinematic variables for all food types and sizes resulted in five components that explained 79.98% of the variance. The majority of the duration variables loaded high and positively on the first component (Table 1). Resting jaw angle and gape jaw angle loaded positively on the second component, while resting cranial angle and maximum cranial angle loaded negatively. Change in gape angle and resting jaw angle loaded heavily on principal component three. Total cranial elevation angle and gape size loaded on principal component four. Figure 8 shows the scatter plot of the first three components with the eight individual sharks, indicating the loadings of each bite. There is some clustering by individual, but there is no distinct separation of the individuals. When food type (fish versus squid) is plotted it indicates that there is overlap between food types (Figure 9). Similarly, when all bites are plotted based on large versus small food size there is overlap between the two food sizes with no evident separation (Figure 10). All of the duration variables (bite duration, mandible depression, mandible elevation, mandible depression and elevation, gape duration, time to maximum gape) had non-normal data distributions that resisted transformation. All statistical analyses of these variables employed non-parametric statistics. Table 2 shows the mean values for all duration variables overall, by food size, and
31 PC1 PC2 PC3 PC4 PC5 Duration of Mandible Depression .824 .238 -.144 -.072 -.186 Duration of Mandible Elevation .763 .180 -.068 .148 -.200 Duration of Mandible Depression and Elevation .922 .241 -.123 .047 -.225 Time to Max Gape .867 .199 -.075 -.038 -.075 Gape Duration .372 .053 .265 -.042 .854 Gape Jaw Angle -.352 .691 -.132 .278 .245 Resting Jaw Angle -.231 .693 .574 -.068 -.040 Gape Change Angle -.104 -.081 -.817 .400 .299 Maximum Cranial Angle .417 -.664 .353 .357 -.028 Resting Cranial Angle .406 -.599 -.004 -.319 .127 Total Cranial Elevation Angle -.071 -.154 .456 .781 -.098 Gape Size -.026 -.071 -.129 .533 -.056 Bite Duration .890 .201 -.079 -.017 .351 Cumulative Percent Variance 33.29 48.85 60.28 71.03 79.98 Table 1: Component loadings of kinematic variables for principal components analysis. Loadings above 0.5 are indicated by shading. The last row indicates the cumulative percentage of the variance explained by the principal components.
32 -6 -4 -2 0 2 4 6 -2 -1 0 1 2 3 4 5 -3 -2 -1 0 1 2 3 4 PC 3 P C1 P C2 1 2 3 4 5 6 7 8 Du rat i o n Vari ab l es Ja w A ngl e s C ra ni a l A ngl e s Figure 8: PCA of kinematic variables by individual shark. Individual sharks are indicated by the number in the legend. Duration variables that loaded positively on PC1 were bite duration; duration of mandible depression, duration of mandible elevation, and duration of mandible depression plus elevation. Jaw angles that loaded positively on PC2 were resting jaw angle and gape jaw angle. Cranial angles that loaded negatively on PC2 were maximum cranial angle and resting cranial angle. For PC3 the variables that loaded were GC, gape change angle; and RJA, resting jaw angle. Component loadings are given in Table 1. RJA GC
33 -6 -4 -2 0 2 4 6 -2 -1 0 1 2 3 4 5 -3 -2 -1 0 1 2 3 4 PC 3 P C 1 P C2 S q u i d Fi sh D uration V ariables J a w A n g l e s C r a n i a l A n g l e s Figure 9: PCA of kinematic variables indicating food type. There is no distinct clustering between food types, but there is substantial variation in the data set. Duration variables that loaded positively on PC1 were bite duration; duration of mandible depression, duration of mandible elevation, and duration of mandible depression plus elevation. Jaw angles that loaded positively on PC2 were resting jaw angle and gape jaw angle. Cranial angles that loaded negatively on PC2 were maximum cranial angle and resting cranial angle. For PC3 the variables that loaded were GC, gape change angle; and RJA, resting jaw angle. Component loadings are given in Table 1. RJA GC
34 -6 -4 -2 0 2 4 6 -2 -1 0 1 2 3 4 5 -3 -2 -1 0 1 2 3 4 PC 3 P C 1 PC2 Ga p e L a rg e r Th a n Ga p e D ur ation V ar iables Jaw An g les C ran ial An g les Figure 10: PCA of kinematic data with food size indicated. There is no distinct clustering between food sizes, but there is substantial variation for in the data set. Duration variables that loaded positively on PC1 were bite duration; duration of mandible depression, duration of mandible elevation, and duration of mandible depression plus elevation. Jaw angles that loaded positively on PC2 were resting jaw angle and gape jaw angle. Cranial angles that loaded negatively on PC2 were maximum cranial angle and resting cranial angle. For PC3 the variables that loaded were GC, gape change angle; and RJA, resting jaw angle. Component loadings are given in Table 1. RJA GC
35 by food type. There was no significant difference in the durations for larger than gapesized food as compared to gape-sized food (Table 3). For the six duration variables there were significant differences between food type (Table 4). In all cases, squid bites were faster than fish bites (Table 2). In order to determine if squid bites differed proportionately from fish bites, all bites were rescaled with respect to bite duration, by dividing all variables for each bite by their respective bite duration. One-way ANOVAs were run for each of the remaining kinematic duration variables. Only the duration of mandible elevation was significantly different between fish and squid (Table 5). With the rescaled values, it took significantly longer for mandible elevation with squid than with fish. There were significant differences among individual sharks for each duration variable (Table 6). Two animals (#7 and #8) were generally faster for duration variables than the other individuals (Table 6). All of the angular variables (resting jaw angle, gape angle, change in gape angle, resting cranial angle, maximum cranial angle, and total cranial elevation angle) and gape distance had a normal distribution and were therefore analyzed with parametric statistics. Table 7 shows the mean values for all of the angular variables overall, by food size, and food type. Overall, the resting jaw angle ranged from 8 to 104 indicating that the mouth was being held open prior to prey capture in some sequences. The jaw angle at maximum gape ranged from 58 to 137 The resting cranial angle had a narrower range than resting jaw angle, ranging from 159 to 176 The cranial angle at maximum gape had a similar range to resting cranial angle, ranging from 154 to 180 Total cranial
36 Food Item Bite Duration (ms) Duration of Mandible Depression (ms) Duration of Mandible Elevation (ms) Duration of Mandible Depression and Elevation (ms) Gape Duration (ms) Time to Maximum Gape (ms) Overall 164.18 (.11) 45.17 (.58) 53.32 (.58) 102.37 (.72) 66.03 (.07) 56.33 (.92) Gape Size 156.84 (.23) 44.83 (.05) 51.76 (.26) 101.02 (.97) 61.51 (.91) 60.17 (.10) Plus Gape Size 168.24 (.28) 45.36 (.34) 54.06 (.49) 103.10 (.24) 68.60 (.15) 54.23 (.67) Fish 178.35 (.17) 48.50 (.08) 55.45 (.31) 106.97 (.42) 75.57 (.77) 61.59 (.19) Squid 133.31 (.07) 37.69 (.60) 54.30 (.51) 92.28 (.15) 44.60 (.09) 44.77 (.66) Table 2: Mean kinematic values (plus or minus standard error) from food capture events of eight nurse sharks Ginglymostoma cirratum Overall refers to mean values for all data. Gape size and plus gape size refer to means by food size, regardless of food type. Fish and squid refer to means by food type, regardless of food size.
37 H df p Bite Duration 0.102 1 0.749 Duration of Mandible Depression 3.89 x 10 -4 1 0.984 Duration of Mandible Elevation 0.0655 1 0.798 Duration of Mandible Depression and Elevation 0.0105 1 0.919 Time to Maximum Gape 0.971 1 0.324 Gape Duration 0.874 1 0.350 Table 3: Results of Kruskal-Wallis ANOVA on Ranks for kinematic duration variables for different food sizes (gape sized and larger than gape). There were no significant differences between food sizes for any of the kinematic duration variables.
38 H df p Bite Duration 11.951 1 <0.001 Duration of Mandible Depression 9.268 1 =0.002 Duration of Mandible Elevation 1.701 1 =0.192 Duration of Mandible Depression and Elevation 7.803 1 =0.005 Time to Maximum Gape 8.341 1 =0.004 Gape Duration 6.859 1 =0.009 Table 4: Results of Kruskal-Wallis ANOVA on Ranks for kinematic duration variables for different food types (fish and squid). In all cases squid bites were faster than fish bites. All of the kinematic duration variables were significant with the exception of duration of mandible elevation.
39 df p power Duration of Mandible Depression F=0.0306 1 0.861 0.047 Duration of Mandible Elevation F=4.924 1 0.028 0.492 Duration of Mandible Depression and Elevation F=2.936 1 0.089 0.265 Time to Maximum Gape H=0.0391 1 0.843 na Gape Duration F=2.936 1 0.089 0.265 Table 5: Results of one-way ANOVA and Kruskal-Wallis one-way ANOVA on Ranks for the nurse shark Ginglymostoma cirratum for food types (squid and fish), scaled to bite duration. Each kinematic variable was divided by bite duration prior to analysis. Duration of mandible elevation was the only variable that showed significant differences, with mandible elevation taking longer in squid bites than fish bites.
40 H df p post hoc Bite Duration 72.882 7 <0.001 7,8 < 3,4,5,6 Duration of Mandible Depression 49.925 7 <0.001 7 < 3,4,5,6 8 < 4,5 Duration of Mandible Elevation 24.276 7 =0.001 7,8 < 4 Duration of Mandible Depression and Elevation 39.773 7 <0.001 7,8 < 4,5,6 Time to Maximum Gape 46.364 7 <0.001 7 < 1,3,4,5 8 < 4,5 Gape Duration 62.743 7 <0.001 7,8 < 3,4,5,6 Table 6: Results of Kruskal-Wallis ANOVA on Ranks for kinematic duration variables for eight nurse sharks Ginglymosoma cirratum Post hoc column provides comparison of medians for individuals as determined by Dunnet's test.
41 elevation angle ranged from -8.5 to 14 indicating that in some bites there was actually cranial depression at maximum gape. Cranial elevation, although minor, was present in 74% of the bites and was the reason that the end of mandible depression and maximum gape were not coincident. Gape distance was significantly greater for larger than gape-sized food (Table 8). There were no significant differences among the angular variables for the two food sizes (Table 8). There were no significant differences among the angular variables or gape distance for the two different food types (Table 9). Significant differences among individual sharks were detected for gape angle, resting cranial angle, and gape distance, but no pattern was apparent (Table 10). Electromyography Principal components analysis of the electromyographic data resulted in seven components that explained 79% of the variance (Table 11). There was no pattern apparent with any of the components. More variables loaded on principal component one than any of the other components, but there was no pattern as far as which functional or anatomical muscle groups accounted for the variability. As with the kinematic data, there is some separation of points by individual (Figure 11), but no apparent separation for food type or size (Figures 12 and 13). The mean onset times and durations for each implanted muscle were calculated (Table 12). These values were used to determine a mean motor pattern for all bites (Figure 14) as well as the motor patterns for gape-sized food (Figure 15) and larger than
42 Food Item Resting Jaw Angle Gape Angle Change in Gape Angle Resting Cranial Angle Maximum Cranial Angle Total Cranial Elevation Angle Gape Distance (cm) Overall 40.26 ( 2.99) 85.06 ( 2.54 ) 45.53 ( 2.63) 167.96 ( 0.63) 169.64 ( 0.74) 1.56 ( 0.65) 2.63 ( 0.04) Gape Size 31.90 ( 3.75) 81.88 ( 5.22 ) 49.99 ( 6.06) 167.53 ( 0.96) 168.49 ( 1.11) 0.95 ( 0.88) 2.48 ( 0.069) Plus Gape Size 44.45 ( 3.93) 86.61 ( 2.82 ) 43.36 ( 2.57) 168.176 ( 0.83) 170.20 ( 0.95) 1.86 ( 0.87) 2.73 ( 0.05) Fish 42.75 ( 3.82) 85.66 ( 3.24 ) 44.04 ( 3.52) 168.02 ( 0.77) 170.41 ( 0.90) 2.24 ( 0.74) 2.59 ( 0.05) Squid 34.85 ( 4.48) 83.72 ( 4.02 ) 48.86 (3.26) 167.84 ( 1.15) 167.92 ( 1.21) 0.074 ( 1.26) 2.75 ( 0.07) Table 7: Mean angular values and gape distance (plus or minus standard error) from food capture events of four nurse sharks Ginglymostoma cirratum Resting jaw angle refers to angle of the jaw in the frame prior to mandible depression. Resting cranial angle refers to the angle of the head in the frame prior to mandible depression. Overall refers to mean values for all data. Gape size and plus gape size refer to means by food size, regardless of food type. Fish and squid refer to means by food type, regardless of food size.
43 F df p post hoc power Gape Jaw Angle 0.0331 1 0.856 na 0.047 Resting Jaw Angle 2.200 1 0.145 na 0.176 Change in Gape Angle 1.655 1 0.205 na 0.116 Resting Cranial Angle 0.0373 1 0.848 na 0.047 Cranial Angle 1.393 1 0.244 na 0.089 Total Cranial Elevation Angle 0.0696 1 0.793 na 0.047 Gape Distance 9.476 1 0.003 1 < 2 0.841 Table 8: Results of one-way ANOVA for distance and angle variables for the nurse shark Ginglymostoma cirratum for different food sizes. Angular tests based on data from four individuals. Gape distance based on data from eight individuals. Post hoc column provides comparison of means as determined by Tukey test. The number in the post hoc column indicates the food size, with 1 indicating gape-sized food and 2 indicating larger than gape-sized food.
44 F df p power Gape Jaw Angle 0.739 1 0.394 0.047 Resting Jaw Angle 3.062 1 0.087 0.274 Change in Gape Angle 0.784 1 0.380 0.047 Resting Cranial Angle 0.0373 1 0.848 0.047 Cranial Angle 1.901 1 0.174 0.143 Total Cranial Elevation Angle 2.652 1 0.110 0.227 Gape Distance 3.155 1 0.078 0.290 Table 9: Results of one-way ANOVA for distance and angle variables for the nurse shark Ginglymostoma cirratum for different food types. Angular tests based on data from four individuals. Gape distance based on data from eight individuals. No significant differences were found for angular data or gape distance between food types.
45 F df p post hoc power Gape Jaw Angle 3.577 3 0.021 7 < 5,6,8 0.591 Resting Jaw Angle 0.797 3 0.502 na 0.049 Change in Gape Angle 0.882 3 0.457 na 0.049 Resting Cranial Angle 4.398 3 0.018 8 < 5,6,7 0.730 Cranial Angle 1.754 3 0.169 na 0.193 Total Cranial Elevation Angle 1.007 3 0.399 na 0.050 Gape Distance 3.384 7 0.003 6 < 1,2,3,4,5,7,8 0.844 Table 10: Results of one way ANOVA for distance and angle variables for the nurse shark Ginglymostoma cirratum Angular tests based on data from four individuals. Gape distance based on data from eight individuals. Post hoc column provides comparison of means for individuals as determined by Tukey test. The number in the post hoc column indicates the individual shark.
46 gape-sized food (Figure 16). Values for the coracoarcualis and superficial head of the posterior division of the preorbitalis are only given for the mean motor pattern due to small sample sizes. In all cases the first muscle to fire in the motor sequence was the coracomandibularis, which mediates jaw opening (Figures 14, 15 and 16). Based on the combined data set, the next muscle to fire was the superficial head of the posterior division of the preorbitalis, closely followed by the coracohyoideus (Figures 14). For gape-sized food the coracohyoideus preceded the start of jaw opening (Figure 15). The epaxialis and levator hyomandibularis began activity near the end of mandible depression (Figures 14, 15, and 16). Contraction of the coracobranchialis occurred around the time of maximum gape and continued until about the end of jaw closure (Figures 14, 15, and 16). The duration of activity of the coracoarcualis was very long, in some cases extending well past jaw closure (Figure 14). The quadratomandibularis, interhyoideus, intermandibularis, and anterior division of the preorbitalis all fired after maximum gape, but prior to the start of jaw closure (Figures 14, 15, and 16). Two onsets of muscle firing differed between gape-sized and larger than gapesized food items (Table 13). These were onset of firing in the levator hyomandibularis and epaxialis, with onset being earlier in gape-sized food items for the levator hyomandibularis and earlier in larger than gape-sized food items for the epaxialis. The only significant differences between food types were found for onset of firing in the levator hyomandibularis and onset of the anterior division of the preorbitalis (Table 14). The levator hyomandibularis fired sooner with fish bites than squid bites. The anterior division of the preorbitalis fired sooner for squid bites than fish bites. Significant individual differences were found for six EMG variables (Tables 15).
47 PC1 PC2 PC3 PC4 PC5 PC6 CB Onset .373 .357 .494 .065 -.285 .443 CB Duration .080 .342 .247 -.025 .622 .038 CC Onset .609 .062 -.280 -.520 -.049 .002 CC Duration .914 -.219 -.053 -.250 .101 -.067 CH Onset -.050 .452 .578 -.129 .138 -.328 CH Duration -.037 .160 -.092 .447 -.043 -.436 CM Onset -.599 .222 .403 -.350 .427 -.113 CM Duration .706 -.194 -.302 .298 -.354 -.089 EP Onset .781 .201 -.002 .138 .175 -.162 EP Duration .146 .597 -.449 .282 .254 .007 IH Onset -.043 .146 .030 .699 .100 -.108 IH Duration .616 -.062 .105 .447 .196 -.139 IM Onset .625 .347 .197 -.415 -.329 -.159 IM Duration .118 -.661 .376 .147 .373 -.249 LH Onset .037 .561 .570 .192 -.352 .069 LH Duration .164 -.184 .182 .196 .233 .784 POAD Onset .806 .139 -.008 -.104 -.013 .013 POAD Duration .153 .389 -.607 -.140 .403 .227 POSP Onset -.863 .231 -.164 .106 -.119 -.048 POSP Duration -.910 .274 -.153 .142 .090 .040 QM Onset .781 .384 -.055 .265 .060 .083 QM Duration .135 -.374 .328 .298 .065 .195 Cumulative Percent Variance 29.69 41.18 51.46 60.74 67.89 74.17 Table 11: Component loadings of EMG variables for principal components analysis. Loadings above 0.5 are indicated by shading. The last row indicates the cumulative percentage of the variance explained by the principal components. Muscle abbreviations are as follows: CB, coracobranchialis; CC, coracoarcualis; CH, coracohyoideus; CM, coracomandibularis; EP, epaxialis; IH, interhyoideus; IM, intermandibularis; LH, levator hyomandibularis; POAD, anterior divison of the preorbitalis; POSP, superficial head of the superior division of the preorbitalis; QM, qaudratomandibularis.
48 -4 -3 -2 -1 0 1 2 3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 P C3 P C 1 P C2 3 4 5 6 7 8 Figure 11: PCA of all EMG variables with individual shark indicated. No pattern was detected in the component loadings. There is some separation by individual evident from the scatter plot. Component loadings are given in Table 11.
49 -4 -3 -2 -1 0 1 2 3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 PC3 PC 1 PC2 Fish Squid Figure 12: PCA of all EMG variables by food type. No pattern was detected in the component loadings. Despite large variation there is no distinct clustering between food types. Component loadings are given in Table 11.
50 -4 -3 -2 -1 0 1 2 3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 P C3 PC1 PC 2 Gape Larger Than Gape Figure 13: PCA of all EMG variables by food size. No pattern was detected in the component loadings. Despite large variation there is no distinct clustering between food sizes. Component loadings are given in Table 11.
51 Most of these differences were in onset of muscle firing for the coracobranchialis, epaxialis, quadratomandibularis, and interhyoideus (Table 15). The vast majority of EMG variables, both onset times and durations of activity, were not significantly different among individuals.
52 Overall Gape sized Larger than Gape Onset (ms) Duration (ms) Onset (ms) Duration (ms) Onset (ms) Duration (ms) CB 65.528 103.165 75.927 101.298 52.530 105.499 CC 73.991 142.858 114.143 175.368 43.876 118.475 CH 11.551 88.284 -14.798 86.034 35.873 90.360 CM -28.838 103.428 -31.313 130.258 -26.363 76.598 EP 34.955 60.557 38.726 61.204 30.713 59.829 IH 78.926 61.489 64.024 53.787 85.312 64.789 IM 95.545 52.533 102.692 57.952 89.980 50.884 LH 42.293 76.057 34.134 98.532 47.732 64.611 POAD 87.084 83.164 103.068 85.524 55.115 78.444 POSP 9.875 56.5125 4.985 57.225 19.655 55.088 QM 80.585 58.093 98.092 52.900 61.732 63.686 Table 12: Mean values for onset and duration of muscle firing in the nurse shark Ginglymostoma cirratum for the overall data set, gape-sized food items, and larger than gape-sized food items. Onset is calibrated to start of mandible depression. See Table 11 for muscle abbreviations.
53 df p post hoc power CB Onset F=0.920 1 0.352 na 0.048 CB Duration F=0.0161 1 0.901 na 0.048 CH Onset F=0.268 1 0.610 na 0.048 CH Duration F=2.867 1 0.103 na 0.245 CM Onset H=1.103 1 0.328 na na CM Duration F=1.219 1 0.288 na 0.069 EP Onset F=5.698 1 0.033 2 < 1 0.513 EP Duration F=0.660 1 0.431 na 0.048 IH Onset F=0.362 1 0.556 na 0.048 IH Duration F=0.0737 1 0.790 na 0.048 IM Onset F=0.214 1 0.651 na 0.048 IM Duration F=0.0303 1 0.864 na 0.048 LH Onset F=4.989 1 0.041 1 < 2 0.454 LH Duration F=0.0354 1 0.853 na 0.048 POAD Onset F=0.920 1 0.360 na 0.048 POAD Duration F=0.0298 1 0.866 na 0.048 QM Onset F=0.473 1 0.499 na 0.048 QM Duration F=2.702 1 0.114 na 0.225 Table 13: Results of one-way ANOVA and Kruskal-Wallis ANOVA on Ranks for EMG variables based on food size. Post hoc column provides comparison of means as determined by Tukey test. The number in the post hoc column indicates the food size, with 1 being gape-sized and 2 being larger than gape. See Table 11 for muscle abbreviations.
54 df p post hoc power CB Onset F=2.442 1 0.138 na 0.193 CB Duration F=0.425 1 0.524 na 0.048 CH Onset F=0.704 1 0.410 na 0.048 CH Duration F=0.0971 1 0.758 na 0.048 EP Onset F=0.826 1 0.380 na 0.048 EP Duration F=1.911 1 0.190 na 0.136 IH Onset F=0.529 1 0.478 na 0.048 IH Duration F=0.214 1 0.650 na 0.048 IM Onset F=4.344 1 0.056 na 0.387 IM Duration F=1.134 1 0.305 na 0.061 LH Onset F=5.460 1 0.034 1 < 2 0.499 LH Duration F=1.183 1 0.294 na 0.066 POAD Onset F=4.976 1 0.050 2 < 1 0.429 POAD Duration F=0.370 1 0.557 na 0.048 QM Onset F=1.294 1 0.268 na 0.077 QM Duration F=0.917 1 0.349 na 0.048 Table 14: Results of one-way ANOVA for EMG variables based on food type. Post hoc column provides comparison of means as determined by Tukey test. The number in the post hoc column indicates the food type, with 1 being fish and 2 being squid. See Table 11 for muscle abbreviations.
55 df p post hoc power CB Onset F= 5.107 3 0.025 8 < 4 0.609 CB Duration F=0.341 3 0.796 na 0.050 CH Onset F=1.335 3 0.289 na 0.103 CH Duration F=0.512 3 0.678 na 0.049 CM Onset H=6.751 2 0.034 5 < 7 na CM Duration F=2.060 2 0.167 na 0.190 EP Onset F= 6.224 2 0.014 7,8 < 5 0.724 EP Duration F= 3.942 2 0.048 7 < 5 0.460 IH Onset H= 7.745 2 0.021 7 < 5 na IH Duration F=0.888 2 0.433 na 0.049 IM Onset F=0.942 3 0.451 na 0.050 IM Duration F=1.848 3 0.192 na 0.178 LH Onset F=0.639 3 0.603 na 0.050 LH Duration F=0.140 3 0.935 na 0.050 POAD Onset F=4.026 2 0.056 na 0.439 POAD Duration F=1.627 2 0.249 na 0.122 QM Onset F= 21.979 3 <0.001 7,8 < 6 < 5 1.000 QM Duration F=1.672 3 0.205 na 0.162 Table 15: Results of one-way ANOVA and Kruskal-Wallis ANOVA on Ranks for individual nurse sharks for EMG variables. Post hoc column provides comparison of means for individuals as determined by Tukey test. For the Kruskal-Wallis ANOVA the post hoc test provides comparison of medians for individuals as determined by Dunnet's T test. The number in the post hoc column indicates the individual shark. See Table 11 for muscle abbreviations.
56 Discussion Feeding Behavior The nurse shark Ginglymostoma cirratum is an inertial suction feeder with a stereotyped pattern of prey capture. Like other suction feeding animals it possesses a number of morphological features that are advantageous for suction feeding (Liem, 1993; Norton, 1995). These features include a small, anteriorly directed mouth, short jaw length, robust jaw abductors, and a fast initial buccal expansion (Tanaka, 1973; Wu, 1994; Summers et al., 1998; Motta and Wilga, 1999; Motta et al., 2002; Robinson and Motta, 2002). Feeding in G. cirratum is characterized by a fast expansive sequence. The duration of mandible depression and total bite duration for suction feeding sharks has been shown to be shorter than for most ram feeding sharks (Tricas and McCosker, 1984; Tricas, 1985; Ferry-Graham, 1998a ; Motta et al., 1997; Wilga and Motta 1998a; 2000; Edmonds et al., 2001; Motta et al., 2002). The average time for mandible depression in this study (46 msec) is slower than that reported for G. cirratum for small food items (26 msec) (Motta et al. 2002). Likewise, the average bite duration in this study (164 msec) is slower than Motta et al. (2002) reported (100 msec). The suction feeding horn shark Heterodontus francisci is similarly fast with an average of 54 msec for mandible depression and 131 msec for total bite duration (Edmonds et al., 2001). Times for mandibular depression in ram feeding sharks include 75 msec for Negaprion brevirostris
60 103 msec for Sphyrna tiburo 115 msec for Carcharhinus perezi 140 msec for Carcharodon carcharias and 353 msec for Cephaloscyllium ventriosum (Tricas, 1985; Ferry-Graham, 1997; Motta et al., 1997; Wilga and Motta, 2000; Motta and Wilga, 2002). Bite durations for primarily ram feeding sharks are often twice as slow as those of the nurse shark, with 250 msec for S. tiburo 309 msec for N. brevirostris 383 msec for C. perezi 405 msec for C. carcharias and 419 msec for C. ventriosum (Tricas, 1985; Ferry-Graham, 1997; Motta et al., 1997; Wilga and Motta, 2000; Motta and Wilga, 2002). Sharks that utilize a combination of ram and suction, such as Squalus acanthias and Triakis semifasciata have intermediate values for both time to mandible depression and total bite duration as compared to elasmobranchs that feed using either primarily suction or ram (Wilga and Motta, 1998a; Ferry-Graham 1998a). Nurse sharks have a small mouth that forms a circular opening during the expansive phase of feeding, which is advantageous for suction feeding (Barel, 1983; Norton, 1995). Previous studies using mathematical modeling and performance testing have shown that a small circular opening and expansive buccal cavity are advantageous for generating larger sub-ambient pressures (Lauder, 1980b; Barel, 1983; Muller and Osse, 1984; Van Leeuwen and Muller, 1984; Wainwright et al., 2001). The formation of a circular opening in orectolobids is assisted by the protrusion of the labial cartilages (Wu, 1994; Motta et al., 2002). This has also been observed in the leopard shark Triakis semifasciata spiny dogfish Squalus acanthias and horn shark Heterodontus francisci (Ferry-Graham, 1998a; Wilga and Motta, 1998a; Edmonds et al., 2001). The mouth of the nurse shark at rest is sub-terminal, but protrusion of the labial cartilages and slight cranial elevation result in a terminally directed mouth, allowing the animal to capture
61 prey in front of or beneath it (Lauder and Clark, 1984) Strong jaw abductors are another feature of suction feeding fish (Liem, 1993) The coracomandibularis and coracohyoideus muscles in nurse sharks are apparently hypertrophied, compared to other species of shark (Moss 1963). This may allow the nurse shark to generate large negative pressures via a relatively stronger contractive force which may expand the buccal cavity further and faster than other sharks (Moss, 1965; Wu, 1994). Nurse sharks have been measured to generate suction up to 102 kPa (Tanaka, 1973) and it has been proposed that the nurse shark may be able to generate enough suction to cause cavitation of the water (Motta et al., 2002). Cranial elevation does not appear to be an integral component of suction feeding in nurse sharks. In 35% of the bites, the total cranial angle at maximum gape was negative, indicating that at maximum gape the head of the animal was depressed relative to its position prior to jaw opening. In only one case was cranial elevation angle in the nurse shark greater than 10 as compared to an average cranial elevation angle of 19 for the sandtiger shark Carcharias taurus (Lowry, Matott, and Huber; unpublished data). Ram feeding sharks such as Carcharodon carcharias C. taurus and N. brevirostris all show larger cranial elevations than suction feeding sharks (Tricas and McCosker, 1984; Tricas, 1985; Motta et al., 1991; Klimley et al., 1996; Motta et al., 1997). Cranial elevation is thought to be useful for increasing the gape during feeding, repositioning sub-terminal mouths towards the prey, and possibly assisting in palatoquadrate protrusion (Moss, 1972; Tricas and McCosker, 1984; Frazzetta and Prange, 1987; Frazzetta, 1994). While this may be of importance for predominantly ram feeding sharks, cranial elevation would be a disadvantage in suction feeding sharks as a larger gape is less effective in
62 suction feeding. The nurse shark maintains a small mouth opening that maximizes suction generation. The small amount of head depression and elevation that is present in feeding events in nurse sharks may also be related to the terminal position of the mouth which does not require cranial elevation for positioning of the prey. Modulation The nurse shark Ginglymostoma cirratum does not appear to modulate its capture behavior based on food size as presented in this study. Slight modulation may be present based on food type, however the results are not conclusive. Kinematic and electromyographic results support the hypothesis that G. cirratum is a stereotyped inertial suction feeder. Squid bites were significantly faster for all six duration variables, but when these bites were analyzed with respect to their individual bite durations the only difference between bites was the time required to close the mandible (Tables 4 and 5). These results suggest that squid bites, although on the whole faster than fish bites, are very similar kinematically. There were no differences for any of the six landmark variables with respect to food type (Table 9). Only two muscles showed significant differences in activity between food types: the onset of activity for both the levator hyomandibularis and ventral division of the preorbitalis (Table 14). If the levator hyomandibularis is used to aid in closing the jaw or to assist in protrusion, then based on the kinematics it would have been expected to fire sooner in squid bites rather than in fish bites (Table 14). Due to small sample sizes with respect to squid bites, these electromyographic results should be taken cautiously. As the difference in durations between squid bites and fish bites is a
63 consistent result, it suggests however that some modulation based on food type is occurring. Based on the landmark kinematic variables, the proportional duration variables, and the electromyographic data available for food type, modulation, if present, is only occurring in the kinematic durations. The differences in the durations may be due to the relative densities of the food items, with fish having a higher density than squid. No significant differences in kinematic and motor activity duration variables could be found between food items that were gape-sized and those that were twice gape size (Tables 3 and 13). As would be expected, gape distance was significantly larger for twice gape-sized food items as opposed to gape-sized food (Table 8). In only two muscles were activities significantly different based on prey size (Table 13). Earlier onset times were found for firings of the epaxialis and levator hyomandibularis for gapesized food. The reason for this is not apparent. Based on studies in other sharks, the epaxialis is used to raise the head during the expansive phase and the levator hyomandibularis is used to retract the hyomandibulae during the recovery phase (Motta et al., 1997; Wilga and Motta, 1998a). The few studies in elasmobranchs that have examined modulation have failed to demonstrate it during food capture (Ferry-Graham, 1997; 1998a; 1998b; Wilga and Motta, 1998b; Edmonds et al., 2001). There are, however, reasons to be cautious about suggesting that modulation is absent in elasmobranchs. First, most of these studies have been on benthic or epi-benthic species (Ferry-Graham, 1997; Wilga and Motta, 1998a; Edmonds et al. 2001). It is possible that the few species studied do not modulate, but that other elasmobranchs do. There have been numerous studies on teleost fish that have demonstrated modulation, but there have also been studies on teleosts that have failed to
64 demonstrate modulation (Norton, 1995; Nemeth 1997). Secondly, some of these studies may not have provided food items that were different enough to require modulation of the feeding behavior. Ferry-Graham (1998a) suggested that the live prey items offered to juvenile Triakis semifasciata may not have been truly elusive, given the fact that all 'elusive' prey items were captured. The study on Heterodontus francisci utilized only one food type and examined only differences in presentation (Edmonds et al., 2001). The description of modulation in motor activity between feeding phases in the Rhinobatos lentiginosus is based on a single prey item and does not strictly follow the definition of modulation (Wilga and Motta, 1998b). In the strict sense of the term, differences between feeding phases are not an example of modulation as these are separate behaviors and modulation is a distinct pattern within a single behavior (Liem, 1978; 1979). Ferry-Graham (1998a) suggested that T. fasciata may modulate the amount of suction generated based on prey size. Modulation was not present in Cephaloscyllium ventriosum based on prey size (Ferry-Graham, 1997; 1998b). For suction feeding, prey size would presumably influence modulation of feeding behavior. However, food size had no discernible effect on the kinematics of feeding in the G. cirratum with the exception of gape size. It is possible that the food sizes offered did not differ sufficiently to result in modulation. Castro (2000) found that most prey items found in the stomachs of G. cirratum were of small size compared to the predator, which suggests that the food sizes used in this study were as large or larger than those naturally found for nurse sharks. The kinematic results of the two size classes that were offered are similar to those previously reported for nurse sharks with smaller than gape-sized food (Motta et al., 2002; Robinson and Motta, 2002).
65 Although there were no differences detected in the initial capture sequence, it is possible that modulation occurs in the manipulation sequence based on prey size. Spitsuck manipulation occurred only with prey items that were larger than gape size. However, spit-suck manipulation only occurred in some of the larger than gape-size bites. Variability Variation among individuals is a common finding of feeding studies in lower vertebrates such as fish and salamanders (Shaffer and Lauder, 1985; 1986; Wainwright and Lauder, 1986; Reilly and Lauder, 1989a; 1989b; Lauder and Shaffer, 1993; Gillis and Lauder, 1994; Gillis and Lauder, 1995; Reilly, 1996). For example, Wainwright and Lauder (1986) found significant variation among individual motor patterns within species in an examination of four species of Centrarchidae. Other studies on fishes that have demonstrated high inter-individual variability include studies on three species of Labridae and the blue-gill sunfish Lepomis macrochirus (Sanderson, 1990; Gillis and Lauder, 1995). Differences in feeding behavior between populations of the same species have been found for populations of Centrarchidae and Balistidae, among others (Turingan et al., 1995; Cutwa and Turingan, 2000; Durie and Turingan, 2001; Huskey and Turingan, 2001). Inter-individual variation has been found in other elasmobranch studies including studies on Triakis fasciata Negaprion brevirostris and Squalus acanthias (FerryGraham, 1997; Motta et al., 1997; Wilga and Motta, 1998). As with many other studies on elasmobranchs, significant inter-individual variation was found for most of the variables (Tables 6, 10, and 15). Some of the individual differences are to be expected. Differences in resting cranial angle between
66 individuals are understandable, as different animals are likely to have different head profiles at the beginning of prey capture depending on how they approach the food. The principal component analysis of the kinematic variables separated the duration variables from the angular variables, but there was no clear separation between either food types or sizes. Similar to the kinematic data, the principal component analysis of the electromyographic variables also showed no clear separation of onset or duration of muscle firing based on either food type or size. Motor Pattern The coracomandibularis was always the first muscle to fire and the only muscle to consistently fire before mandible depression (Figures 14, 15, and 16; Table 12). This strongly suggests that the coracomandibularis is the primary mandible depressor. This is consistent with previous studies on Negaprion brevirostris Squalus acanthias Sphyrna tiburo and Rhinobatos lentiginosus in which the coracomandibularis fired before jaw opening (Motta et al., 1997; Wilga and Motta, 1998a; 1998b; Wilga and Motta, 2000). The superficial head of the posterior division of the preorbitalis and coracohyoideus fire at approximately the same time, with the coracohyoideus having a longer duration of activity. The coracohyoideus functions to depress the basihyal cartilage of the hyoid arch in a posteroventral direction, expanding the buccopharyngeal cavity. Expansion of the buccopharyngeal cavity generates the suction force. The superficial head of the posterior division of the preorbitalis inserts on the ventral edge of the mandible (Motta and Wilga, 1999). Wu (1994) suggests that the preorbitalis in orectolobids is not involved in palatoquadrate protrusion as it is in squaliform and
67 carcharhiniform sharks (Motta et al., 1997; Wilga and Motta, 1998a). The contraction of the superficial head of the posterior division of the preorbitalis in G. cirratum would likely adduct the mandible. Activity of this muscle during jaw opening may be an indicator of reciprocal activation, in which an antagonist muscle fires increasing the tension and output force of the resultant action, in this case mandible depression (Levin et al., 1992; Gribble and Ostry, 1998). In G. cirratum the anterior to posterior expansion of the feeding mechanism is reflected in the anterior to posterior contraction of the coracomandibularis, coracohyoideus, coracobranchialis and coracoarcualis. This pattern is not found in Negaprion brevirostris or Squalus acanthias both of which have nearly simultaneous firings of these hypobranchial muscles (Motta et al., 1991; Motta et al., 1997; Wilga and Motta, 1998a) In G. cirratum the coracoarcualis inserts onto the coracohyoideus (Motta and Wilga, 1999), but does not begin to fire until the coracohyoideus is nearly finished firing. It is possible that the action of the coracoarcualis is to continue the power stroke of the coracohyoideus and increase the posterior expansion of the buccal cavity. Due to low sample size with the coracoarcualis, its placement in the motor pattern must be taken cautiously. Activity in the coracoarcualis had the greatest variation in onset or duration of any muscle examined. Before the end of mandible depression, the epaxialis begins firing. The epaxialis serves to raise the head and this muscle is likely firing in the bites in which cranial elevation contributes to maximum gape. Food items that were larger than gape-size elicited a faster onset time of this muscle (Table 13). It is possible that this earlier onset of activity corresponds with the larger gape distance required for larger than gape-sized
68 food. Previous studies on Negaprion brevirostris Sphyrna tiburo and Squalus acanthias have all shown action in the epaxialis with concomitant cranial elevation prior to mandible depression (Motta et al., 1991; Motta et al., 1997; Wilga and Motta, 2000). The levator hyomandibularis begins activity around the time the mandible ceases lowering and before maximum gape is reached. Activity in this muscle at this stage of the kinematic sequence is unexpected, as the levator hyomandibularis presumably elevates the hyoid. This would be expected in the compressive phase of the kinematic sequence as seen in other elasmobranchs examined to date, rather than at the end of the expansive phase (Motta et al., 1997; Wilga and Motta, 2000). In 25% of the bites there was a second burst of the levator hyomandibularis that does occur during the compressive phase, prior to the end of mandible elevation. The quadratomandibularis, anterior division of the preorbitalis, interhyoideus, and intermandibularis all fire just prior to the start of mandible elevation. The quadratomandibularis, based on its size, location and motor pattern, is believed to be the primary adductor of the lower jaw, as demonstrated in other studies of elasmobranchs (Motta et al., 1991; Wilga and Motta, 1998a). The anterior division of the preorbitalis likely assists in raising the mandible and may assist in retraction of the labial cartilages, given its insertion onto the ventral labial cartilage and labial fold (Motta and Wilga, 1999). The actions of the intermandibularis and interhyoideus are not apparent, but it is possible that they are involved in palatoquadrate protrusion as proposed by Wu (1994). In Wus (1994) model contraction of the interhyoideus pulls the ceratohyals medially, resulting in an anterior rotation of the hyomandibulae and by association the upper and lower jaws are displaced anteriorly. The intermandibularis contributes to jaw protrusion
69 by laterally compressing the two halves of the mandible, forming a more acute symphyseal angle and extending the length of the lower jaw anteriorly (Wu, 1994). Both the interhyoideus and intermandibularis fire after maximum gape but prior to complete jaw closure Post capture manipulation Spit-suck manipulation is a behavior that varied between feeding events. It was observed in all but one of the individuals used in this study. Spit-suck manipulation was found only in bites with larger than gape-sized food, but was not present in all of these bites. In some cases, the food item was completely expelled from the buccal cavity, and then reacquired, but typically the food item was partially retained within the buccal cavity during the alteration of suction and spitting. Spit-suck manipulation often resulted in the viscera and some flesh of the food item being expelled from the external gills and mouth. It cannot be determined with certainty whether or not this was caused by the teeth tearing the food item or by the suction force alone. Examination of stomach contents of nurse sharks often resulted in intact fish and squid with tooth indentations but no evidence of tearing. Shredded fish in the stomach of G. cirratum did not show clear tooth marks. The teeth in nurse sharks are small and unserrated (Goto, 2001; Compagno, 2001). It seems possible that the teeth are used in spit-suck manipulation to retain the food item in the buccal cavity or that in some cases the back and forth movement of the food item across the teeth results in shearing or rasping of the flesh.
70 Summary Based on the results of this study Ginglymostoma cirratum is an inertial suction feeder that acquires prey in a stereotyped manner. Prey capture is characterized by a fast expansive phase, slight or no cranial elevation, and an anterior-to-posterior expansion of the buccopharyngeal cavity. Closure of the jaws begins while posterior expansion continues in the pharyngeal cavity. Modulation in G. cirratum is not apparent based on either food type or food size. Durations for squid bites are faster than for fish bites, but are otherwise kinematically similar. There were few significant differences between onset of muscle activity for different food types. Other than differences in gape size corresponding to differences in food size, there were no significant differences between food sizes and feeding kinematics or motor patterns. Modulation may be present in later phases of the feeding sequence. Spit-suck manipulation is one behavioral pattern that the nurse shark may utilize to process larger prey items for easier transport. High inter-individual variability was found for nearly all of the kinematic variables and for some of the electromyographic variables. This is an indicator of the range of response possible within a population and serves to underscore the importance of sampling multiple individuals to characterize accurately their behavior. Jaw opening in G. cirratum is mediated by action in the coracomandibularis, and then is assisted by contraction of the coracohyoideus and coracoarcualis that serve to depress the hyoid apparatus, resulting in expansion of the buccopharyngeal cavity. In
71 some sequences the epaxialis fires prior to maximum gape, likely indicating that the slight cranial elevation present in G. cirratum contributes to the gape. Adduction of the jaw is accomplished primarily via activity in the quadratomandibularis, with additional adduction likely provided by the anterior division of the preorbitalis. Although palatoquadrate protrusion could not be visibly correlated with muscle activity, the onset times for the interhyoideus and intermandibularis are consistent with the reported sequence and timing for palatoquadrate protrusion in orectolobids (Wu, 1994).
72 Literature Cited Alfaro, M. E., J. Janovetz, and M. W. Westneat. 2001. Motor control across trophic strategies: muscle activity of biting and suction feeding fishes. American Zoologist. 41:1266-1279. Andison, M. E., and J. G. Sivak. 1994. The functional morphology of the retractor lentis muscle of a teleost fish, Astronotus ocellatus Canadian Journal of Zoology. 72:1880-1886. Arreola, V. I., and M. W. Westneat. 1996. Mechanics of propulsion by multiple fins: kinematics of aquatic locomotion in the burrfish ( Chilomycterus schoepfi ). Proceedings of the Royal Society of London B. 263:1689-1696. Barel, C. D. N. 1983. Towards a constructional morphology of cichlid fishes (Teleostei, Perciformes). Netherlands Journal of Zoology. 33:357-424. Belbenoit, P. 1986. Fine analysis of predatory and defensive motor events in Torpedo marmorata (Pisces). Journal of Experimental Biology. 121:197-226. Bigelow, H. B., and W. C. Schroeder. 1953. Fishes of the Western North Atlantic, p. 563. In: Memoir: Sears Foundation for Marine Research, Yale University 1. Vol. 2: Sawfishes, guitarfishes, skates and rays, New Haven. Bock, W. J. 1990. From biologische anatomie to ecomorphology. Netherlands Journal of Zoology. 40:254-277. Bouton, N., F. Witte, J. J. M. van Alphen, A. Schenk, and O. Seehausen. 1999. Local adaptations in populations of rock-dwelling haplochromines (Pisces: Cichlidae) from southern Lake Victoria. Proceeding of the Royal Society of London B. 266:355-360. Bozzano, A., R. Murgia, S. Vallerga, J. Hirano, and S. Archer. 2001. The photoreceptor system in the retinae of two dogfishes, Scyliorhinus canicula and Galeus melastomus : possible relationship with depth and predatory lifestyle. Journal of Fish Biology. 59:1258-1278. Cappetta, H. 1987. Chondricthyes II: Mesozoic and Cenozoic Elasmobranchii, p. 1-193. In: Handbook of Paleoichthyology. Vol. 3B. H. P. Schultze (ed.). Gustav Fischer Verlag, Stuttgart. Carrier, J. C., and C. A. Luer. 1990. Growth rates in the nurse shark, Ginglymostoma cirratum Copeia. 1990:686-692. 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. Clifton, K. B., and P. J. Motta. 1998. Feeding morphology, diet, and ecomorphological relationships among five Caribbean labrids (Teleostei, Labridae). Copeia:953966.
73 Compagno, L. J. V. 1990. Relationships of the megamouth shark, Megachasma pelagios (Lamniformes: Megachasmidae), with comments on its feeding habits. NOAA Technical Report, NMFS. 90:357-379. 2001. Sharks of the world: An annotated and illustrated catalogue of shark species known to date. Cutwa, M. M., and R. G. Turingan. 2000. Intralocality variation in feeding biomechanics and prey use in Archsargus probatocephalus (Teleostei, Sparidae), with implications for the ecomorphology of fishes. Environmental Biology of Fishes. 59:191-198. Devaere, S., D. Adriaens, W. Verraes, and G. G. Teugels. 2001. Cranial morphology of the anguilliform clariid Channallabes apus (Gunther, 1873) (Teleostei: Siluriformes): are adaptations related to powerful biting? Journal of Zoology, London. 255:235-250. DeVree, F., and C. Gans. 1989. Functional morphology of the feeding mechanisms in lower tetrapods. Progress in Zoology. 35:115-127. Douglas, M. E., and W. J. Matthews. 1992. Does morphology predict ecology? Hypothesis testing within a freshwater stream fish assemblage. Oikos. 65:213234. Dullemeijer, P. 1980. Animal ecology and morphology. Netherlands Journal of Zoology. 30:161-178. Durie, C. J., and R. G. Turingan. 2001. Relationship between durophagy and feeding biomechanics in gray triggerfish, Balistes capriscus : intraspecific variation in ecological morphology. Florida Scientist. 64:20-28. Edmonds, M. A., P. J. Motta, and R. E. Hueter. 2001. Food capture kinematics of the suction feeding horn shark, Heterodontus francisci. Environmental Biology of Fishes. 62:415-427. Ehlinger, T. J. 1990. Habitat choice and phenotype-limited feeding efficiency in bluegill: Individual differences and trophic polymorphism. Ecology. 7:886-896. Ferry-Graham, L. A. 1997. Feeding kinematics of juvenile swellsharks, Cephaloscyllium ventriosum Journal or Experimental Biology. 200:1255-1269. 1998a. Effects of prey size and mobility on prey-capture kinematics in leopard sharks, Triakis semifasciata Journal of Experimental Biology. 201:2433-2444. 1998b. Feeding kinematics of hatchling swellsharks, Cephaloscyllium ventriosum (Scyliorhinidae): The importance of predator size. Marine Biology. 131:703-718. Ferry-Graham, L. A., P. C. Wainwright, and G. V. Lauder. 2001a. Sources of fluid flow during feeding in bluegill sunfish as identified using DPIV. American Zoologist. 41:1442-1442. Ferry-Graham, L. A., P. C. Wainwright, M. W. Westneat, and D. R. Bellwood. 2001b. Modulation of prey capture kinematics in the cheeklined wrasse Oxycheilinus digrammus (Teleostei : labridae). Journal of Experimental Zoology. 290:88-100. Fish, F. E., and L. D. Shannahan. 2000. The role of the pectoral fins in body trim of sharks. Journal of Fish Biology. 56:1062-1073. Fouts, W. R., and D. R. Nelson. 1999. Prey capture by the Pacific angel shark, Squatina californica Visually mediated strikes and ambush-site characteristics. Copeia:304-312.
74 Frazzetta, T. H. 1994. Feeding mechanisms in sharks and other elasmobranchs. Advances in Comparative and Environmental Physiology. 18:31-57. Frazzetta, T. H., and C. D. Prange. 1987. Movements of cephalic components during feeding in some requiem sharks (Carchariniformes: Carcharinidae). Copeia:979993. Friel, J. P., and P. C. Wainwright. 1998. Evolution of motor patterns in Tetraodontiform fishes: Does muscle duplication lead to functional diversification. Brain, Behaviour, and Evolution. 52:159-170. 1999. Evolution of complexity in motor patterns and jaw musculature of Tetraodontiform fishes. Journal of Experimental Biology. 202:867-880. Gatz Jr., A. J. 1979. Ecological morphology of freshwater stream fishes. Tulane Studies in Zoology and Botany:91-124. Gelsleichter, J., E. Cortes, C. Manire, R. Hueter, and J. Musick. 1998. Evaluation of toxicity of oxytetracycline on growth of captive nurse sharks, Ginglymostoma cirratum. Gibb, A. C., B. C. Jayne, and G. V. Lauder. 1991. Pectoral Fin Locomotion in the Bluegill Sunfish. American Zoologist. 31:A97-A97. Gilbert, P. W. 1970. Studies on the anatomy, physiology, and behavior of sharks, p. 45. Office of Naval Research, Ithaca, New York. Gillis, G. B., and G. V. Lauder. 1994. Aquatic prey transport and the comparative kinematics of Ambystoma tigrinum feeding behaviors. Journal of Experimental Biology. 187:159-179. 1995. Kinematics of Feeding in Bluegill Sunfish Is There a General Distinction between Aquatic Capture and Transport Behaviors. Journal of Experimental Biology. 198:709-720. Goldschmid, A., and K. Kotrschal. 1989. Ecomorphology: development and concepts. Progress in Zoology. 35:501-512. Goto, T. 2001. Comparative anatomy, phylogeny, and cladistic classification of the order Orectolobiformes (Chondrichthyes, Elasmobranchii). Memoirs of the Graduate School of Fisheries Sciences, Hokkaido University. 48:1-100. Gribble, P. L., and D. J. Ostry. 1998. Independent coactivation of shoulder and elbow muscles. Experimental Brain Research. 123:355-360. Gudger, E. W. 1921. Note on the morphology and habits of the nurse shark, Ginglymostoma cirratum Copeia. 1921:54-59. 1941a. The feeding organs of the whale shark, Rhineodon typus Journal of Morphology. 68:81-99. 1941b. The food and feeding habits of the whale shark Rhineodon typus Journal of the Elisha Mitchell Scientific Society. 57:57-72. Hamasaki, D. I., and S. H. Gruber. 1965. The photoreceptors of the nurse shark, Ginglymostoma cirratum and the sting ray, Dasyatis sayi. Herrel, A., P. Aerts, J. Fret, and F. DeVree. 1999. Morphology of the feeding system in Agamid lizards: Ecological correlates. Anatomical Record. 254:496-507. Higgs, D. M., and L. A. Fuiman. 1996. Ontogeny of visual and mechanosensory structure and function in Atlantic menhaden Brevoortia tyrannus Journal of Experimental Biology. 199:2619-2629.
75 Hjelm, J., L. Persson, and B. Christensen. 2000. Growth, morphological variation and ontogenetic niche shifts in perch (Perca fluviatilis) in relation to resource availability. Oecologia. 122:190-199. Hobson, E. S. 1963. Feeding behavior in three species of sharks. Pacific Science. 17:171193. Hugueny, B., and M. Pouhly. 1999. Morphological correlates of diet in an assemblage of West African freshwater fishes. Journal of Fish Biology. 54:1310-1325. Huskey, S. H., and R. G. Turingan. 2001. Variation in prey-resource utilization and oral jaw gape between two populations of largemouth bass, Micropterus salmoides Environmental Biology of Fishes. 61:185-194. Jayne, B. C. 1988. Muscular mechanisms of snake locomotion: An electromyographic study of the sidewinding and concertina modes of Crotalus cerastes Neroda fasciata and Elaphe obsoleta Journal of Experimental Biology. 140:1-33. Johnston, I. A. 1983. Dynamic properties of fish muscle. In: Fish Biomechanics. P. W. Webb and D. Weihs (eds.). Praeger Publishers, New York. Kaicounis, M. C., and R. M. Brigham. 1995. Intraspecific variation in wing loading affects habitat use by little brown bats ( Myotis lucifugus ). Canadian Journal of Zoology. 73:89-95. Karr, J. R., and F. C. James. 1975. Eco-morphological configurations and convergent evolution in species and communities., p. 258-291. In: Ecology and evolution of communities. J. M. Diamond (ed.). Harvard University Press, Cambridge, MA. Klimley, A. P., P. Pyle, and S. D. Anderson. 1996. The behavior of white sharks and their pinniped prey during predatory attacks. In: Great Whtie Sharks: The biology of Carcharodon carcharias A. P. a. A. Klimley, D.G. (ed.). Academic Press, San Diego. Koumoundouros, G., P. Divanach, L. Anezaki, and M. Kentouri. 2001. Temperatureinduced ontogenetic plasticity in sea bass (Dicentrarchus labrax). Marine Biology. 139:817-830. Lauder, G. V. 1978. Functional anatomy of feeding in primitive Actinopterygian fishes. American Zoologist. 18:622-622. 1980a. Prey capture by suction feeding an experimental analysis of mouth cavity pressures in sunfishes. American Zoologist. 20:931-931. 1980b. The suction feeding mechanism in sunfishes (Lepomis): An experimental analysis. Journal of Experimental Biology. 88:49-72. 1981. Intraspecific functional repertoires in the feeding mechanism of the Characoid fishes, Lebiasina Hoplias and Chalceus Copeia. 1:154-168. 1983a. Food capture., p. 280-311. In: Fish Biomechanics. P. W. Webb and D. Weihs (eds.). Praeger Publishers, New York, New York. 1983b. Functional and morphological bases of trophic specialization in sunfishes (Teleostei, Centrarchidae). Journal of Morphology. 178:1-21. 1983c. Prey capture hydrodynamics in fishes: Experimental tests of two models. Journal or Experimental Biology. 104:1-13. 1985. Functional morphology of the feeding mechanism in lower vertebrates. Fortschritte Der Zoologie. 30:179-188. Lauder, G. V., and B. D. Clark. 1984. Water flow patterns during prey capture by teleost fishes. Journal of Experimental Biology. 113:143-&.
76 Lauder, G. V., and K. F. Liem. 1982. Symposium Summary Evolutionary patterns in Actinopterygian fishes. American Zoologist. 22:343-345. Lauder, G. V., and S. M. Reilly. 1990. Metamorphosis of the feeding mechanism in tiger salamanders ( Ambystoma tigrinum ), the ontogeny of cranial muscle mass. Journal of Zoology. 222:59-74. 1994. Amphibian feeding behavior: Comparative biomechanics and evolution, p. 163195. In: Biomechanics of feeding in vertebrates. Bels (ed.). Lauder, G. V., and H. B. Shaffer. 1993. Aquatic vertebrate skull design, p. 113-149. In: The Skull. Vol. 3: Functional and Evolutionary Mechanisms. Levin, M. F., A. G. Feldman, T. E. Milner, and Y. Lamarre. 1992. Reciprocal and coactivation commands for fast wrist movements. Experimental Brain Research. 89:669-677. Liem, K. F. 1978. Modulatory multiplicity in the functional repertoire of the feeding mechanism in cichlid fishes. Journal of Morphology. 158:323-360. 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. 1980. Adaptive significance of intraand interspecific differences in the feeding repertoires of cichlid fishes. American Zoologist. 20:295-314. 1993. Ecomorphology of the Teleost Skull, p. 422-452. In: The Skull. Vol. 3: Functional and Evolutionary Mechanisms. J. Hanken and B. K. Hall (eds.). The University of Chicago Press, Chicago. Maisey, J. G. 1984. Higher elasmobranch phylogeny and biostratigraphy. Zoological Journal of the Linnean Society. 82:33-54. Meyer, A. 1987. Phenotypic plasticity and heterochrony in Cichlasoma managuense (Pisces: Cichlidae) and their implications for speciation in cichlid fishes. Evolution. 41:1357-1369. Moss, S. A. 1965. The feeding mechanisms of three sharks: Galecerdo cuvier (Peron & LeSuer), Negaprion brevirostris (Poey), and Ginglymostoma cirratum (Bonaterre). p. 139. In: Biology. Cornell University, Ithaca. 1972. The feeding mechanism of sharks of the family Carcharhinidae. Journal of Zoology (London). 167:423-436. 1977. Feeding mechanisms in sharks. American Zoologist. 17:355-364. Motta, P. J., R. E. Hueter, and T. C. Tricas. 1991. An electromyographic analysis of the biting mechanism of the lemon shark, Negaprion brevirostris : Functional and evolutionary implications. Journal of Morphology. 201:55-69. Motta, P. J., R. E. Hueter, T. C. Tricas, and A. P. Summers. 2002. Kinematic analysis of suction feeding in the nurse shark, Ginglymostoma cirratum (Orectolobiformes, Ginglymostomatidae). Copeia:24-38. Motta, P. J., and K. Kotrschal. 1992. Correlative, experimental, and comparative evolutionary approaches in ecomorphology. Netherlands Journal of Zoology. 42:400-415. Motta, P. J., T. C. Tricas, R. E. Hueter, and A. P. Summers. 1997. Feeding mechanism and functional morphology of the jaws of the lemon shark Negaprion brevirostris (Chondrichthyes, Carcharhinidae). Journal of Experimental Biology. 200:27652780.
77 Motta, P. J., and C. A. Wilga. 2001. Advances in the study of feeding behaviors, mechanisms, and mechanics of sharks. Environmental Biology of Fishes. 60:131156. Motta, P. J., and C. D. Wilga. 1999. Anatomy of the feeding apparatus of the nurse shark, Ginglymostoma cirratum Journal of Morphology. 241:1-29. Muller, M., and J. W. M. Osse. 1984. Hydrodynamics of suction feeding in fish. Transactions of the Zoological Society of London. 37:51-135. Muller, M., J. W. M. Osse, and J. H. G. Verhagen. 1982. A quantitative hydrodynamical model of suction feeding in fish. Journal of Theoretical Biology. 95:49-79. Nemeth, D. H. 1997. Modulation of buccal pressure during prey capture in Hexagrammos decagrammus (Teleostei: Hexagrammidae). Journal of Experimental Biology. 200:2145-2154. Norton, S. F. 1991. Capture success and diet of cottid fishes the role of predator morphology and attack kinematics. Ecology. 72:1807-1819. 1995. A functional approach to ecomorphological patterns of feeding in cottid fishes. Environmental Biology of Fishes. 44:61-78. Norton, S. F., and E. L. Brainerd. 1993. Convergence in the feeding mechanics of ecomorphologically similar species in the Centrarchidae and Cichlidae. Journal of Experimental Biology. 176:11-29. Okuda, N., M. Miyazaki, and Y. Yanagisawa. 2002. Sexual difference in buccal morphology of the paternal mouthbrooding cardinalfish Apogon doederleini Zoological Science. 19:801-807. Osse, J. W. M. 1995. Comparative functional morphology of fish larvae during metamorphosis. European Journal of Morphology. 33:188-189. Reilly, S. M. 1996. The metamorphosis of feeding kinematics in Salamandra salamandra and the evolution of terrestrial feeding behavior. Journal of Experimental Biology. 199:1219-1227. Reilly, S. M., and G. V. Lauder. 1989a. Evolution of Tetrapod Prey Transport Behavior Quantitative Kinematics and Muscle Function in a Primitive Tetrapod. American Zoologist. 29:A91-A91. 1989b. Physiological Bases of Feeding-Behavior in Salamanders Do Motor Patterns Vary with Prey Type. Journal of Experimental Biology. 141:343-358. 1990a. The evolution of tetrapod feeding behavior: Kinematic homologies in prey transport. Evolution. 44:1542-1557. 1990b. Metamorphosis of cranial design in tiger salamanders ( Ambystoma tigrinum ) a morphometric analysis of ontogenic change. Journal of Morphology. 204:121137. 1992. Morphology, behavior, and evolution: Comparative kinematics of aquatic feeding in salamanders. Brain, Behavior, and Evolution. 40:182-196. Robinson, M. P., and P. J. Motta. 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. Sanderson, S. L. 1990. Versatility and specialization in labrid fishes: ecomorphological implications. Oecologia. 84:272-279.
78 Sanderson, S. L., and R. Wasserug. 1993. Vertebrate Suspension Feeding, p. 37-112. In: The Skull. Vol. 3: Functional and Evolutionary Mechanisms. J. Hanken and B. K. Hall (eds.). The University of Chicago Press, Chicago. Sanford, C. P. 2001. The novel 'tongue-bite apparatus' in the knifefish family Notopteridae (Teleostei: Osteoglossomorpha): are kinematic patterns conserved within a clade? Zoological Journal of the Linnean Society. 132:259-275. Sanford, C. P. J., and G. V. Lauder. 1988. Functional morphology of the tongue bite in Notopterus (Osteoglossomorpha). American Zoologist. 28:A115-A115. 1989. Functional morphology of the tongue bite in Osteoglossomorph fishes. American Zoologist. 29:A14-A14. Sasko, D. E., R. E. Hueter, and P. J. Motta. Unpublished Manuscript. Prey capture behavior and kinematics of the Atlantic cownose ray, Rhinoptera bonasus Schwenk, K. 2000. Tetrapod feeding in the context of vertebrate morphology, p. 3-20. In: Feeding: Form, function, and evolution in tetrapod vertebrates. K. Schwenk (ed.). Academic Press, San Deigo. Shaffer, H. B., and G. V. Lauder. 1985. Aquatic Prey Capture in Ambystomatid Salamanders Patterns of Variation in Muscle-Activity. Journal of Morphology. 183:273-284. 1986. Functional design of the feeding mechanism in lower vertebrates: unidirectional and bidirectional flow systems in the tiger salamander. Zoological Journal of the Linnean Society. 88:277-290. Shirai, S., and K. Nakaya. 1992. Functional morphology of feeding apparatus of the cookie-cutter shark, Isistius brasiliensis (Elasmobranchii, Dalatiinae). Zoological Science. 9:811-821. Springer, S. 1961. Dynamics of the feeding mechanism of large galeoid sharks. American Zoologist. 1:83-185. Summers, A. P., K. F. Darouian, A. M. Richmond, and E. L. Brainerd. 1998. Kinematics of aquatic and terrestrial prey capture in Terrapene carolina with implications for the evolution of feeding in Cryptodire turtles. Journal of Experimental Zoology. 281:280-287. Tanaka, S. K. 1973. Suction feeding by the nurse shark. Copeia:606-608. Tricas, T. C. 1985. Feeding ethology of the white shark, Carcharodon carcharias Memoirs of the Southern California Academy of Sciences. 9:81-91. Tricas, T. C., and J. E. McCosker. 1984. Predatory behavior of the white shark ( Charcaradon carcharias ) with notes on its biology. Proceedings of the California Academy of Science. 43:221-238. Turingan, R. G., P. C. Wainwright, and D. A. Hensley. 1995. Interpopulation variation in prey use and feeding biomechanics in Caribbean triggerfishes. Oecologia. 102:296-304. Van Leeuwen, J. L., and M. Muller. 1984. Optimum sucking techniques for predatory fish. Transactions of the Zoological Society of London. 37:137-169. Vandermeer, H. J., G. C. Anker, and C. D. N. Barel. 1995. Ecomorphology of retinal structures in zooplanktivorous haplochromine cichlids (Pisces) from Lake Victoria. Environmental Biology of Fishes. 44:115-132.
79 Wagner, H. J., E. Frohlich, K. Negishi, and S. P. Collin. 1998. The eyes of deep-sea fish II. Functional morphology of the retina. Progress in Retinal and Eye Research. 17:637-685. Wainwright, P. C. 1988. Morphology and ecology: Functional basis of feeding constraints in Caribbean labrid fishes. Ecology. 69:635-645. 1989. Prey processing in haemulid fishes: Patterns of variation in pharyngeal jaw muscle activity. Journal of Experimental Biology. 141:359-375. 1999. Ecomorphology of prey capture in fishes, p. 375-387. In: Advances in Ichthyological Research. E. Saksena (ed.). Jiwaji University Press, Gwalior, India. Wainwright, P. C., L. A. Ferry-Graham, T. B. Waltzek, A. M. Carroll, C. D. Hulsey, and J. R. Grubich. 2001. Evaluating the use of ram and suction during prey capture by cichlid fishes. Journal of Experimental Biology. 204:3039-3051. Wainwright, P. C., and G. V. Lauder. 1986. Feeding biology of sunfishes Patterns of variation in the feeding mechanism. Zoological Journal of the Linnean Society. 88:217-228. 1992. The evolution of feeding biology in sunfishes (Centrarchidae), p. 470-491. In: Systematics, historical ecology and North American Freshwater Fishes. R. Mayden (ed.). Stanford U. Press,, Stanford. Wainwright, P. C., C. P. Sanford, S. M. Reilly, and G. V. Lauder. 1989. Evolution of motor patterns: aquatic feeding in salamanders and ray-finned fishes. Brain, Behavior, and Evolution. 34:329-341. Wainwright, P. C., C. P. J. Sanford, S. M. Reilly, and G. V. Lauder. 1988. The evolution of motor patterns used during prey capture in lower vertebrates. American Zoologist. 28:A115-A115. Wainwright, P. C., and S. S. Shaw. 1999. Morphological basis of kinematic divesity in feeding sunfishes. Journal of Experimental Biology. 202:3101-3110. Wake, M. H. 1992. Morphology, the study of form and function, in modern evolutionary biology, p. 289-346. In: Functional Morphology and Evolution. Oxford Publishers Ltd. Webb, P. W. 1982. Locomotor patterns in the evolution of actinopterygian fishes. American Zoologist. 22:329-342. 1984. Body, form, locomotion and foraging in aquatic vertebrates. American Zoologist. 24:107-120. Westneat, M. W., and P. C. Wainwright. 1989. Feeding mechanism of Epibulus insidiator (Labridae: Teleostei): Evolution of a novel functional system. Journal of Morphology. 202:129-150. Wiens, J. A. 1991. Ecomorphological comparisons of the shrub-desert avifaunas of Australia and North America. Oikos. 60:55-63. Wilga, C. A., and P. J. Motta. 2000. Durophagy in sharks:Feeding mechanics of the hammerhead Sphyrna tiburo Journal of Experimental Biology. 203:2781-2796. Wilga, C. A. D., and P. J. Motta. 1998a. Conservation and variation in the feeding mechanism of the spiny dogfish, Squalus acanthias Journal of Experimental Biology. 201:1345-1358. 1998b. Feeding mechanism of the Atlantic guitarfish, Rhinobatos lentiginosus : modulation of kinematic and motor activity. Journal of Experimental Biology. 201:3167-3184.
80 Wilga, C. D. 1997. Evolution of feeding mechanisms in elasmobranchs, p. 261. In: Biology. University of South Florida, Tampa. Wilga, C. D., R. E. Hueter, P. C. Wainwright, and P. J. Motta. 2000a. Function and evolution of upper jaw protrusion mechanisms in elasmobranchs. American Zoologist. 40:1259-1259. 2001. Evolution of upper jaw protrusion mechanisms in elasmobranchs. American Zoologist. 41:1248-1257. Wilga, C. D., P. C. Wainwright, and P. J. Motta. 2000b. Evolution of jaw depression mechanics in aquatic vertebrates: insights from Chondrichthyes. Biological Journal of the Linnean Society. 71:165-185. Williams, M. E. 2001. Tooth retention in cladodont sharks: with a comparison between primitive grasping and swallowing, and modern cutting and gouging feeding mechanisms. Journal of Vertebrate Paleontology. 21:214-226. Witte, F., C. D. N. Barel, and R. J. C. Hoogerhoud. 1990. Phenotypic plasticity of anatomical structures and its ecomorphological significance. Netherlands Journal of Zoology. 40:278-298. Wu, E. H. 1994. Kinematic analysis of jaw protrusion in orectolobiform sharks: A new mechanism for jaw protrusion in elasmobranchs. Journal of Morphology. 222:175-190.