TY - JOUR
AU - Nishikawa, Kiisa, C.
AB - Abstract Based on studies of a few model taxa, amphibians have been considered stereotyped in their feeding movements relative to other vertebrates. However, recent studies on a wide variety of amphibian species have revealed great diversity in feeding mechanics and kinematics, and illustrate that stereotypy is the exception rather than the rule in amphibian feeding. Apparent stereotypy in some taxa may be an artifact of unnatural laboratory conditions. The common ancestor of lissamphibians was probably capable of some modulation of feeding movements, and descendants have evolved along two trajectories with regard to motor control: (1) an increase in modulation via feedback or feed-forward mechanisms, as exemplified by ballistic-tongued plethodontid salamanders and hydrostatic-tongued frogs, and (2) a decrease in variation dictated by biomechanics that require tight coordination between different body parts, such as the tongue and jaws in toads and other frogs with ballistic tongue projection. Multi-joint coordination of rapid movements may hamper accurate tongue placement in ballistic-tongued frogs as compared to both short-tongued frogs and ballistic tongued-salamanders that face simpler motor control tasks. Decoupling of tongue and jaw movements is associated with increased accuracy in both hydrostatic-tongued frogs and ballistic-tongued salamanders. INTRODUCTION Until recently, study of the motor control of amphibian feeding was limited to a few model taxa, for example the genera Bufo and Rana representing the Anura, and salamanders of the genus Ambystoma representing the Caudata. Kinematic studies of these taxa gave the impression that amphibian feeding in general is highly stereotyped, that is, it is performed in much the same way every time, with little variation in the timing or extent of movements. However, kinematic studies of a great diversity of salamanders, frogs and caecilians have revealed, in the last 15 yr or so, that these taxa are probably exceptions among amphibians, and were unfortunate models on which to base generalizations about feeding in amphibians. Indeed, Bufo is now known to be among the most stereotyped of frogs in its feeding movements (Nishikawa et al., 1992; Nishikawa and Gans, 1996), and Ambystoma lies at the low end of variation for salamanders (Larsen and Guthrie, 1975; Reilly and Lauder, 1989, 1990, 1992; Beneski et al., 1995). The perspective gained over this time, and the rapid growth of understanding in the field have motivated the current survey of motor control in amphibians. Now is a good time to reassess some early conclusions that were drawn, particularly that of stereotypy of movement, and to highlight some instructive examples and general evolutionary patterns to present a more balanced view of amphibian feeding. A major conclusion that can be drawn from a survey across living amphibians is that feeding mechanisms are extremely varied, and that kinematics of feeding are accordingly diverse. Part of this diversity stems from the biphasic life cycle that is ancestral for amphibians and which characterizes most living taxa. Most amphibians make a transition from an aquatic larval stage in which suction-based feeding mechanics and lateral line and chemosensory cues dominate (Himstedt et al., 1982; Bartels et al., 1990), to a more terrestrial adult stage in which feeding is accomplished with the tongue and jaws, and visual stimuli become more important in guiding feeding movements (Ewert, 1987; Roth, 1987). There are clear exceptions to this ancestral developmental pattern, such as direct development and viviparity in which the larval stage is skipped, as well as paedomorphosis in which the ancestral larval characteristics are retained in the adult stage. Direct development and viviparity have evolved in all three groups of Lissamphibia: frogs, salamanders and caecilians. Perennibranchiation (a form of paedomorphosis in which external gills persist in adults) has evolved repeatedly only in salamanders, and the perennibranchiate adults use larval biomechanics and sensorimotor control of feeding movements. Feeding movements can be extremely rapid (i.e., five msec for tongue projection in bolitoglossine salamanders; Thexton et al., 1977; Larsen et al., 1989) and controlled by feed-forward mechanisms in which sensory feedback plays no role, or movements can be relatively slow and deliberate (i.e., about 150 msec for tongue protraction in the frog Hemisus, which has a hydrostatic tongue; Ritter and Nishikawa, 1995), relying heavily on feedback to adjust movements as they are performed. Movements can be triggered and guided by visual stimuli alone, as in many frogs and salamanders (Ewert, 1987; Roth, 1987), or by a combination of mechanical and chemical cues as in most caecilians and some frogs (Himstedt and Simon, 1995; Lettvin et al., 1959; Comer and Grobstein, 1981). Secondarily aquatic adult amphibians provide another source of diversity. Although the common ancestor of amphibians most likely had a terrestrial adult stage, many adult amphibians are now either facultative or obligate aquatic feeders. Adult frogs that feed in water generally use modifications of terrestrial mechanisms, such as tongue protraction and jaw prehension, and only one taxon (Hymenochirus) has re-evolved suction feeding (O'Reilly et al., 2002). All caecilians and many salamanders that feed in water as adults use jaw prehension, although the retention of suction feeding in adult salamanders is widespread and probably related to their generalized morphology as compared to frogs and caecilians. Those salamanders that lose all ability to suction feed are specialized for tongue projection. Biomechanical diversity has been accompanied by diversity of motor control strategies. For example, the fundamentally different biomechanics of ballistic tongue projection in frogs and salamanders call for different characteristics of motor planning. Frogs use a mechanism that requires tight coordination between jaw and tongue movements via feedback, while salamanders have decoupled the tongue from the jaws and project the tongue using feed-forward mechanisms in which feedback is not used to coordinate movements. Because motor control mechanisms (i.e., feedback vs. feed-forward control) have only been examined directly in a few taxa of frogs and salamanders, we must infer mechanisms for most taxa based on behavioral and kinematic evidence. The lack of direct evidence for many taxa and behavioral data on only a fraction of the extant species makes reconstruction of the evolution of motor control in a phylogenetic context difficult and tenuous. Although a formal phylogenetic analysis of characters should be conducted when more data become available, such an analysis using current data would include many unknown character states. Therefore, instead of a formal phylogenetic analysis, we discuss the trends observed in each of the three groups of extant amphibians with examples of representative taxa, as well as the patterns observed in the Lissamphibia as a whole. Definitions For the purposes of this discussion, it is necessary to lay down some operational definitions of terms that appear in the literature that relate to amphibian feeding and the neural control of movement. Amphibians capture prey using a variety of behaviors that can involve movements of the jaws, hyobranchial apparatus, tongue, limbs, and the entire body. Several modes of prey capture have been discussed in the literature; we describe our understanding of these terms here. Jaw prehension is the grasping of prey between the jaws, and is performed both in water and on land; this is also called biting. Tongue prehension is the grasping of prey with the tongue, and involves tongue protraction or projection followed by tongue retraction. Suction feeding is drawing a single, relatively large prey item into the mouth by a single expansion of the buccal cavity, and is performed only in water. Filter feeding is removing multiple small food particles from the water either by rhythmically repeated buccal expansions that draw water into the mouth, or by moving forward over the particles. Lunging is forward movement of the entire body and can be combined with the other behaviors. A given behavior such as suction feeding in a salamander can be highly variable in its duration and in the extent of the involved movements. A behavior that shows variation is one in which the movements are not identical each time the behavior is performed. All movements show some variation; movements that exhibit relatively low variation are generally termed stereotyped. Variation that would not be considered modulation is variation caused solely by differences in the physical circumstances in which the behavior is performed, such as the medium (water or air), the mass of the prey, or body temperature. For this discussion, modulation is produced by active adjustment of movements by the central nervous system. Variation and modulation in behavior are produced by motor pattern variation. We define motor pattern as the pattern of depolarization of motor neurons that innervate the muscles of interest, in response to a sensory trigger or releaser. A muscle activity pattern is the pattern of muscle fiber depolarization that results from the execution of a motor pattern. We understand the anatomical and functional distinction between motor pattern and muscle activity pattern, but for the purposes of this discussion, we consider electromyograms to be approximations of motor neuron activity in addition to recordings of muscle activity. Movement begins with a motor program, or a representation of the plan for movement in the central nervous system at levels higher than the motor neurons in the brainstem and spinal cord (Fig. 1) (Grobstein et al., 1983, 1985; Ghez and Krakauer, 2000). Regions of the brain that participate in the formation of the motor program activate motor neurons, a mechanism called feed-forward control. Sensory information can influence the ultimate generation of a motor pattern (activity pattern of the motor neurons) by altering the motor program. Information obtained during the movement can influence the motor pattern through feedback circuits that are of three general types (Shepherd, 1994): central feedback, which occurs within the central nervous system as information is conveyed from lower to higher centers; reflex feedback, in which information is conveyed from the effector organs (i.e., muscles, joints) to the central nervous system (we use the term proprioceptive feedback if the feedback information is processed in higher brain centers than the spinal cord); finally, sensory feedback, in which information from the external environment is transmitted to the central nervous system via the sense organs. The term sensorimotor feedback is often used to encompass some or all of these types of feedback. Motor control Variation and modulation of behavior have obvious advantages, such as allowing an animal to adjust its behavior in response to changing circumstances. Somewhat less obvious are the relative advantages and disadvantages of feed-forward control versus feedback control of behavior. In general, slower behaviors are those more likely to be modified by feedback, simply because the time required to transmit signals through neurons and across synapses is proportional to the length of axons and the number of synapses. This places an absolute lower limit on the time needed to make an adjustment to a movement, so the shorter the duration of movement, the fewer adjustments can be made (Matsushima et al., 1988). Therefore, feed-forward modulation allows short-duration movements to be altered to fit changing circumstances (e.g., prey position or evasiveness), but movements cannot be altered by feed-forward modulation once they have begun. Feedback modulation provides a mechanism to adjust movements while they are being performed (e.g., if the prey attempts to escape), but it is only possible with movements of relatively long duration. A minimum time for a feedback to alter subsequent movements is about 16 msec in toads (Matsushima et al., 1988) but this value will vary in proportion to axon diameter as well as body size (e.g., approximately 12 msec in the diminutive (<2 g) Bolitoglossa occidentalis, Thexton et al., 1977). The temporal limits of effective feed-back modulation allow us to make some inferences about motor control from kinematic data alone. For example, if the behavior performed has a duration of less than 15 msec in a 50 g toad, sensory-feedback control is highly unlikely. In contrast, if the movement is of long duration and the animal alters its behavior mid-strike, we can be fairly certain that feedback modulation is occurring. However, the presence or absence of variation and even modulation in feeding movements is not sufficient to determine if feedback control is occurring. The example discussed below of tongue projection in Bufo illustrates that proprioceptive feedback can be used to reduce the variation among feedings, and the example of Ensatina shows that modulation can be accomplished using feed-forward mechanisms without feedback. SALAMANDERS Aquatic feeding in salamanders All larval salamanders that have been studied capture aquatic prey using suction feeding (Deban and Wake, 2000). In suction feeding, the hyobranchial apparatus is rapidly depressed while the mouth is opened, expanding the buccal cavity and drawing water and prey into the mouth. Labial lobes restrict water flow to the front of the mouth. Water exits through the gill slits at the back of the head. Lunging may or may not be used. The facial and hypoglossal nerves serve the muscles used to generate flow, the branchiohyoideus and rectus cervicis respectively, which together unfold the hyobranchial apparatus ventrally. The geniohyoideus, intermandibularis and interhyoideus muscles, innervated by the hypoglossal, trigeminal and facial nerves, respectively, elevate the hyobranchial apparatus and squeeze water out the gill slits while prey are retained in the mouth. Prey capture in larvae is triggered by electrical, mechanical, visual or olfactory cues (Himstedt et al., 1982; Bartels et al., 1990). Any modulation of movement that occurs in response to prey position and/or evasiveness, such as modulation of lunge length or the strength of hyobranchial depression, is probably controlled by feed-forward mechanisms. The short duration of movement in suction feeding may preclude sensory feedback from influencing movements during the strike. Cave-dwelling salamanders usually retain larval morphology and suction-feeding behavior throughout life. Proteus of the family Proteidae and several plethodontids have evolved an obligate cave existence (Duellman and Trueb, 1986). Some taxa (Eurycea, Gyrinophilus) retain vision and probably use it when light is available, while others (Proteus, Typhlomolge, Haideotriton) have lost the use of vision and must therefore rely on chemical, mechanical, or electrical sensory modalities to locate and capture prey (Peck, 1973; Uiblein et al., 1992). Viviparity occurs in Salamandra and Mertensiella of the family Salamandridae, in which fetuses feed, using an unknown mechanism, on unfertilized eggs or smaller siblings in the oviducts of the mother and young are born as larvae or metamorphosed juveniles (Muskhelishvili, 1964; Özeti, 1979; Alcobendas et al., 1996; Greven, 1998). Aquatic prey capture in generalized adult salamanders is by either jaw prehension or suction feeding. Many adult salamanders are terrestrial and return to water only to breed, and are not specialized for aquatic feeding. An example is Ambystoma, which feeds using hyobranchial depression suggestive of suction feeding, but also raises the tongue from the floor of the mouth and grasps the prey with the jaws (Lauder and Shaffer, 1986; Lauder and Reilly, 1988). Salamanders with derived tongue protraction mechanisms generally do not suction feed, because their hyobranchial specializations for tongue protraction are at odds with the requirements of suction feeding (Wake, 1982; Roth and Wake, 1985; Deban, 1997a). Examples of this group are all metamorphosing plethodontids, the salamandrids Salamandrina and Chioglossa, and the hynobiid Salamandrella. These taxa use jaw prehension typically, however a few plethodontids are known to use tongue prehension under water (Deban and Marks, 1992, 2002; Schwenk and Wake, 1993; Deban, 1997a). Newts of the family Salamandridae are specialized as adults for aquatic prey capture; they possess robust hyobranchial skeletons suited to suction feeding and grow labial lobes during the breeding season that facilitate directional water flow during suction feeding (Matthes, 1934). Permanently aquatic adults, such as Pachytriton, are the most proficient suction feeders and have reduced tongues (Özeti and Wake, 1969; Miller and Larsen, 1989). Hynobiids of the genus Batrachuperus are also powerful suction feeders as metamorphosed adults (Deban, personal observation). Taxa that undergo less complete metamorphosis and retain larval morphology and can be concluded to be suction feeders, but consequently lack tongue protraction. These include cryptobranchids, amphiumids, and ambystomatids formerly placed in the genus Rhyacosiredon. Truly perennibranchiate forms such as sirenids and proteids suction feed as their larvae do (Reilly and Lauder, 1992; Reilly and Altig, 1996). Adult salamanders that feed in water possess both ampullary organs and neuromasts, and probably use both to direct their feeding strikes, in addition to visual, olfactory, and tactile cues (Himstedt et al., 1982; Fritzsch and Wahnschaffe, 1983). Terrestrial feeding in salamanders Generalized post-metamorphic salamanders are terrestrial or semi-aquatic, and capture prey on land using tongue prehension with limited tongue protraction distance (derived tongue protraction is discussed below). The tongue is protracted by folding and forward thrusting of the hyobranchial apparatus by the subarcualis rectus muscle and the tongue pad is flipped by the genioglossus muscle (Wake and Deban, 2000). Tongue protraction and flipping are controlled through the glossopharyngeal, vagus, and hypoglossal nerves; tongue retraction is accomplished by the rectus cervicis muscles, innervated by the hypoglossal nerve (Wake et al., 1983; Roth et al., 1990; Deban and Wake, 2000). Vision predominates as a trigger of the feeding strike in terrestrial adults (Martin et al., 1974; Jaeger et al., 1982; Roth, 1987), and tongue and jaw movements may be modulated via feed-forward and feedback control, depending on the speed of prey capture. Chemical and mechanical triggers predominate in the dark where vision fails (Roth, 1976). Examples of living salamanders with a generalized prey-capture mechanism are members of the families Ambystomatidae, Dicamptodontidae, and Rhyacotritonidae. These salamanders have a suction feeding aquatic larva and a tongue-flipping terrestrial adult with a short, massive tongue. The tongue is protracted using a combination of hyobranchial thrusting by the subarcualis rectus muscles pulling the epibranchials forward, and action of the genioglossus muscle that attaches the tongue pad to the mandibular symphysis. Kinematic studies of Ambystoma indicate that feeding is rapid, but relatively imprecise, with little capacity for modulation of tongue and jaw movements (Larsen and Guthrie, 1975), either by feedback or feed-forward mechanisms. Study of motor control in more species from these families is needed to determine if the results from Ambystoma apply to other salamanders with generalized morphology. Derived tongue protraction mechanisms in adult salamanders have evolved along three different trajectories, along with increased reliance on vision to guide feeding movements. The Salamandridae possess a specialized mechanism in which the ceratohyals, which anchor the main protractor muscles (subarcualis rectus), are moved forward during tongue protraction and increase the reach of the tongue. The ceratohyals are moved forward by action of the subhyoideus muscles (Findeis and Bemis, 1990), which are innervated by a branch of the facial nerve. An elaboration of this system has evolved twice within the Salamandridae, in the ancestors of Chioglossa and Salamandrina. These taxa possess elongated radial elements that support the large tongue pad and are rotated forward during tongue protraction by the basiradialis muscles to flip the pad a greater distance out of the mouth (Özeti and Wake, 1969). This flipping, in combination with forward thrusting of the tongue skeleton as well as the ceratohyals, contributes to long-distance protraction in these taxa (Miller and Larsen, 1990). Ambystomatids and Rhyacotritonids, although they possess limited tongue protraction (Larsen and Guthrie, 1975; Larsen, personal communication), possess a genioglossus lateralis muscle (innervated by a branch of the hypoglossal nerve) which they may use to move the ceratohyals forward in a manner similar to salamandrids during tongue protraction to increase tongue reach. The second type of derived tongue protraction has evolved in the Plethodontidae, the lungless salamanders (Regal, 1966; Lombard and Wake, 1976, 1977). Unlike the situation in salamandrids, in plethodontids the ceratohyals remain stable during protraction and only the tongue skeleton is thrust forward by action of the subarcualis rectus muscles, innervated by branches of the glossopharyngeal and vagus nerves. The rectus abdominis is continuous with the rectus cervicis to form an elongate retractor muscle innervated by the ramus hypoglossus of the first and second spinal nerves. Plethodontids possess tongue skeletons that fold to form compact projectiles, and at least twice within the family, ballistic projection has evolved in which the tongue skeleton leaves the mouth completely and travels partway to the target under its own momentum. These species generally do not lunge forward with the whole body as do shorter-tongued species. Specializations for extended projection distance in plethodontids include elongated posterior hyobranchial elements (epibranchials) and associated protractor muscles, elongated retractor muscles that originate on the pelvis and are slack when at rest (forming loops in front of the heart), slender and flexible cartilaginous elements that form the tongue skeleton, and a compact and lightweight tongue pad that is free from the lower jaw (i.e., no genioglossus muscle is present). Vision dominates the control of tongue protraction in plethodontids, and the ballistic-tongued species have the greatest binocularity and the most ipsilateral retinotectal projections among salamanders, which are important in prey localization. The details of the evolution of the diverse tongue mechanics in plethodontids are complex, and are discussed in detail by other authors (Lombard and Wake, 1977; Roth and Wake, 1985). The third type of derived tongue protraction occurs in the Hynobiidae. Although tongue protraction is not as extreme in this group as in either the Salamandridae or the Plethodontidae, the hynobiids possess unusual specializations related to tongue protraction. Radial loops of cartilage connect the anterior tips of the ceratohyals (and sometimes the basibranchial) and may form an elongate figure-eight lying between the ceratohyals. The exact biomechanical function of these loops is unknown, but they may extend and support the tongue pad (Larsen et al., 1996). Two pairs of epibranchials are present in adults, one bony and one cartilaginous, compared to a single pair in most other families, and the protractor muscle encompasses both epibranchials. Some taxa (e.g.,Onychodactylus) possess elongate cornua projecting forward from the basibranchial that presumably extend the reach of the tongue pad. Lateral tongue aiming has been observed in Salamandrella (Fig. 2). Tongue aiming may be a specialization of the Hynobiidae, although data are obviously needed on more taxa to test this generalization. The closest relatives of the Hynobiidae, the Cryptobranchidae, display asymmetrical hyobranchial depression during suction feeding (Cundall et al., 1987; Elwood and Cundall, 1994). If asymmetrical movement is found to be widespread in the Hynobiidae, it may constitute a synapomorphy for the Cryptobranchoidea. Tongue aiming has not been observed in any other group except the Plethodontidae. No examinations of motor control have been conducted on the Hynobiidae, although there is no reason to expect they are unusual. Vision seems to guide prey capture, at least terrestrially (Roth, 1987). Prey capture is sufficiently slow (e.g., >200 msec) for feedback to play a role in tongue or jaw movements (Matsushima et al., 1988). Another departure from the presumed ancestral feeding in salamanders includes the increased importance of the jaws in subduing prey in the Ambystomatidae and Dicamptodontidae. These taxa possess robust jaws and large heads, and some are known to feed on small mammals. They also possess massive tongues, particularly the Dicamptodontidae, which are limited in their reach but have a large surface area that would be helpful in grasping large or heavy prey. These taxa, which often feed in mammal burrows, may rely heavily on tactile cues in capturing prey. Terrestrial salamanders have evolved a cave-dwelling lifestyle several times independently. One species, the plethodontid Typhlotriton spelaeus, metamorphoses from a “normal” surface-dwelling larva into a blind cave form that is largely terrestrial and is capable of significant tongue projection (it presumably uses tongue prehension as an adult). It likely relies on olfactory and tactile cues to locate and capture prey, but its feeding behavior remains undescribed. Adult salamanders that are facultative cave-dwellers (e.g., plethodontids Hydromantes, Eurycea) have large eyes that are extremely sensitive to light, and use vision when possible to capture prey. In complete darkness they rely on chemical and mechanical modalities (Roth, 1987). Ancestral feeding in salamanders All salamander larvae and all paedomorphic species suction feed using the same basic mechanism (i.e., hyobranchial depression), therefore it is likely that the common ancestor of salamanders had larvae that suction fed as modern larvae do. Given that all salamanders possess a fundamentally similar tongue protraction mechanism based on hyobranchial movements, it is likely that the common ancestor of salamanders used some form of tongue prehension to capture prey on land, with similar motor control as extant forms (Thexton et al., 1977; Reilly and Lauder, 1990; Roth et al., 1990; Deban and Dicke, 1999). Tongue reach was probably limited, and the jaws were probably emphasized as in Ambystoma (Larsen and Guthrie, 1975). Prey capture was guided visually, although olfactory and tactile cues might have been used as well (Roth, 1976, 1987). In water, suction feeding by adults was probable, because it is unlikely the ancestral adult was as highly specialized for tongue protraction as some modern taxa. An intermediate between suction feeding and jaw prehension, such as that displayed by adult Ambystoma is another possibility. Modulation and variation Although most studies of salamander feeding do not directly examine modulation, variation in kinematics is typically noted or recorded. In addition, the extent to which kinematics vary between successful captures and misses have been recorded for some taxa, and can give us an idea of the modulability of feeding movements. Among larval salamanders, there is little observed variation in hyobranchial and jaw movements (Lauder and Shaffer, 1985), which may reflect the need to coordinate jaw and hyobranchial movements precisely to generate maximum flow of water into the mouth. Lunging distance is highly variable in some taxa and is modulated in response prey distance (Deban and Marks, 2002). Miller and Larsen (1990) examined terrestrial prey capture in several species of salamandrids and found that although kinematics were highly variable within individuals, there was no effect of capture success on feeding kinematics. These results suggests that the salamandrids can modulate, and that sensory feedback does not play a role in modulation. If they do modulate tongue and jaw movements, salamandrids probably do so by feed-forward control. However, their relatively long gape cycles (up to 200 msec) do not preclude modulation via feedback. Plethodontids display considerable feed-forward modulation of tongue movements (e.g., distance, elevation, lateral angle relative to the head) (Maglia and Pyles, 1995) and of the timing and amplitude of muscle activity (Deban and Dicke, 1999). Sensory feedback control is unlikely due to the extremely short duration of tongue protraction, particularly in ballistic-tongued species such as Bolitoglossa, that appear to be specialized for speed (e.g., 5 msec for tongue projection) (Thexton et al., 1977; Deban, personal observation). Denervation of the tongue tip in Ensatina has no effect on feeding kinematics or on the capacity to modulate tongue and jaw movements (Deban, 1997b), further indicating that modulation is via feed-forward control rather than feedback. Among aquatically feeding species Cryptobranchus and Amphiuma are known to modulate their hyobranchial and jaw movements depending on prey type (Erdman and Cundall, 1984; Elwood and Cundall, 1994), and cryptobranchids are capable of asymmetrical jaw and hyoid movements (Cundall et al., 1987). Whether or not this modulation is under feedback control has not been investigated, however it may be possible, given the duration of the behavior of up to 300 msec for the entire gape cycle in large individuals. In contrast, muscle activity patterns of several feeding muscles were found to remain constant in larval Ambystoma when they were feeding on prey that differed in elusiveness (Reilly and Lauder, 1989). An earlier study found that larval Ambystoma show variable muscle activity pattern among and within individuals (Shaffer and Lauder, 1985). Across metamorphosis, muscle activity patterns in aquatically-feeding Ambystoma were found to remain unchanged despite profound morphological alterations (Lauder and Shaffer, 1988). However, a major hyobranchial depressor, the branchiohyoideus, is lost at metamorphosis, so activity of this muscle in larvae could not be compared that of adults. When metamorphosed animals were fed on land and used tongue prehension rather than suction, they displayed a new muscle activity pattern (Lauder and Shaffer, 1988), principally due to the enlargement and increased importance of the main tongue protractor muscle, the subarcualis rectus. FROGS Aquatic feeding in frogs Tadpoles have diversified tremendously in habit and morphology (Orton, 1953; McDiarmid and Altig, 1999), yet most share the basic filter-feeding mechanism that consists of scraping mouth parts (jaw sheaths) that are used to generate suspensions of organic particles, and a rhythmic buccal pump and branchial filters to filter these particles from the water. Buccal pumping is accomplished by depression of the ceratohyals by action of the orbitohyoideus muscles, innervated by the facial nerve (Wassersug and Hoff, 1979; Schlosser and Roth, 1995), followed by elevation by the intermandibularis and interhyoideus muscles. The branchial arches are modified to form elaborate filter baskets through which water is pumped. Food is entrapped in mucous and transported by ciliary action to the esophagus (Wassersug, 1972). All microhylids and pipoids (e.g.,Xenopus and Rhinophrynus) lack hard mouth parts and most are obligate suspension feeders, typically hovering off the substrate to feed. Chemical and mechanical cues are used to locate food; vision is probably only used by generalized tadpoles to avoid predation. Modulation of buccal pumping frequency and stroke volume is extensive in tadpoles, performed in response to changes in the density of food particles (Wassersug and Hoff, 1979; Feder and Wassersug, 1984). Some tadpoles have evolved away from the herbivorous, filter-feeding lifestyle that characterizes most taxa. Larvae of the African pipid Hymenochirus are macrophagous carnivores that capture prey by suction feeding (Sokol, 1962). These tadpoles have large frontal eyes and probably direct their prey capture strikes visually. The speed and extent of mouth and hyobranchial movements are varied, and perhaps modulated in response to prey distance or evasiveness (unpublished data, S.M.D.). Pseudhymenochirus larvae are similar morphologically and probably have similar feeding behavior. Larvae of the leptodactylid Lepidobatrachus llanensis are also macrophagous carnivores that have lost scraping mouth parts, and use a sit-and-wait strategy to capture relatively enormous prey by suction (Ruibal and Thomas, 1988). They have well-developed eyes and probably also direct their strikes visually. Egg-eating has evolved numerous times in tadpoles, particularly arboreal, bromeliad-dwelling species; little is known of the ingestion behavior among these diverse taxa, but suction feeding is likely in some species that engulf eggs whole. Carnivorous tadpoles of the pelobatid genus Scaphiopus consume conspecifics, and appear to have morphological specializations such as enlarged jaw muscles and modified jaw sheaths, suggesting that they may bite off and consume larger pieces of food than their herbivorous conspecifics. Many adult frogs capture prey in water and most use the jaws to grasp or engulf prey. Jaw prehension is common in aquatic frogs, often in combination with abortive tongue protraction, but suction feeding is present only in pipids (Fig. 3). Tongue reduction has occurred repeatedly among aquatic frogs, and extreme reduction is a characteristic of the entire family Pipidae. Adult pipids orient to disturbances in the water using lateral line mechanoreceptors that are visible as “stitches” along the body (Dijkgraaf, 1956; Russell, 1976), however electroreceptors (ampullary organs) are absent in frogs (Fritzsch, 1990). Terrestrial feeding in frogs Many adult frogs capture prey on land using a forward lunge of the entire body in combination with protraction of a short tongue that reaches only slightly beyond the margins of the jaws (Regal and Gans, 1976; Nishikawa and Cannatella, 1991; Nishikawa and Roth, 1991; Deban and Nishikawa, 1992). Prey are often grasped between the jaws upon tongue retraction, and the head is flexed ventrally as the mouth closes. Tongue protraction is accomplished via mechanical pulling (sensuNishikawa, 2000) in which the genioglossus muscle pulls the tongue pad toward the lower jaw tip. As a result of this shortening, the tongue pad bulges dorsally and may extend slightly beyond the jaw tip; ventral bending of the lower jaw rotates the tongue pad toward the prey (Nishikawa and Cannatella, 1991; Nishikawa and Roth, 1991; Deban and Nishikawa, 1992; Nishikawa, 2000). Prey capture is triggered mainly by visual but also by olfactory and tactile cues (Shinn and Dole, 1978; Dole et al., 1981; Comer and Grobstein, 1981). Frogs with this generalized feeding behavior include many of the basal taxa such as the archeobatrachians Ascaphus, Discoglossus and Bombina, as well as mesobatrachians Pelobates, and neobatrachians Hyla, Limnodynastes and Leptodactylus. While many taxa retain this generalized, ancestral feeding mode of tongue prehension with a relatively short tongue, novelties have arisen repeatedly. Long-distance tongue protraction has evolved several times along two major biomechanical trajectories. The first, ballistic tongue projection or inertial elongation, has arisen at least five times independently, in megophryids, phylomedusine hylids, bufonids, ranoids, and leptodactylids. Ballistic projection in frogs involves the tongue shortening under action of the genioglossus muscle, and then lengthening and traveling to the prey under its own momentum (i.e., ballistically). Inertial elongation requires tight coordination between jaw and tongue movements to be successful, because much of the momentum that the tongue acquires is kinetic energy transferred from the lower jaw being accelerated and then decelerated. Tongue reach in taxa that use inertial elongation is much greater than those that use mechanical pulling, usually more than a jaw length. Lunging is usually reduced because the tongue covers more distance to the prey, although the arboreal phylomedusines still lunge dramatically (Gray and Nishikawa, 1995). While salamanders with ballistic tongue projection rarely miss their target (Wiggers et al., 1995), frogs that use ballistic projection can be highly inaccurate (e.g., 30% in Bufo, Gray, 1997). Perhaps the latter are inaccurate due to the fundamentally different mechanism of projection and the need to coordinate the movements of multiple joints. The second type of long-distance tongue protraction, hydrostatic protraction or hydrostatic elongation, has probably evolved three times, in microhylids, hemisotids, and rhinophrynids (Trueb and Gans, 1983; Ritter and Nishikawa, 1995; Nishikawa, 2000). Hydrostatic elongation of the tongue involves narrowing of the tongue vertically and consequent lengthening and protraction from the mouth. Some taxa with this mechanism of protraction, such as Hemisus, have the ability to aim the tongue in all three dimensions and with great precision (Ritter and Nishikawa, 1995). Tongue protraction is modulated to a high degree by either feed forward or feedback mechanisms, and the relatively slow tongue movements are completely decoupled from jaw movements, unlike in inertial elongation. In myrmecophagous anurans, many of which use hydrostatic tongue protraction, olfactory and tactile cues have become especially important for detecting and localizing prey in the darkness of ant and termite colonies. The use of the forelimbs is important in many frogs for positioning large prey in the mouth for swallowing, but some derived uses have evolved. From the ancestral behaviors of wiping and scooping, the derived behaviors of grasping and grasping with rotation of the wrist have evolved repeatedly in arboreal clades (Gray et al., 1997). Wiping involves re-orienting prey in the mouth using the palm of the hand, scooping involves pushing prey into the mouth with the back of the hand, grasping involves wrapping the fingers around the prey, and grasping with wrist rotation is similar to grasping, but the palms are rotated to face the mouth. Many arboreal species have forelimbs that are modified for grasping twigs in a variety of orientations; the forelimbs have supplemented the tongue as organs of prey capture in these taxa. Observations have been made of arboreal species simply grasping prey with the forelimbs after unsuccessful attempts at tongue prehension (Gray et al., 1997). The motor control of the forelimbs as used in feeding has undergone extensive evolution, but remains unstudied. Ancestral feeding in frogs Based on the distribution of feeding characters in extant anurans, particularly the basal clades (Nishikawa and Cannatella, 1991; Nishikawa and Roth, 1991), the ancestral condition of the Anura was most likely terrestrial or semi-aquatic as an adult with an aquatic larva. Prey were captured in water using jaw prehension, triggered by visual or tactile information. Because electroreceptors are absent from the lateral line of all adult frogs, including fully aquatic species, and mechanoreceptors are rare (found only in pipids), the ancestral anuran probably did not use lateral line senses in prey capture. The larvae of ancestral anurans was most likely an herbivorous filter feeder that relied on chemical cues to locate food. The larva probably possessed the scraping mouth parts and the branchial filter seen in most extant anuran larvae, and likewise probably also pumped water rhythmically. They probably could modulate pumping rate and/or pumping volume in response to changes in food density via sensory feedback (Wassersug and Hoff, 1979; Feder and Wassersug, 1984). Modulation and variation Modulation in the ancestral adult anuran appears to have included varying the extent and duration of body, tongue and jaw movements. The short-tongued Cyclorana displays a predictable and unique combination of these movements when presented with each of five prey types (Valdez and Nishikawa, 1997) and at least some anurans with mechanical-pulling type tongues (e.g.,Litoria and Hyla) are capable of multiple tongue protractions in a single gape cycle (Deban and Nishikawa, 1992). Modulation via sensory feedback has been accentuated in frogs that have hydrostatic tongue protraction, especially Hemisus, which probes for prey with the tongue and grasps it with the prehensile tongue tip (Ritter and Nishikawa, 1995). At the other extreme are frogs that rely on inertial elongation of the tongue, such as Bufo, which have reduced variation to a minimum as a consequence of the required tight coordination between tongue and lower jaw movements. They accomplish this coordination with proprioceptive feedback through the hypoglossal nerve that disinhibits activation of the jaw depressors when the tongue protractor muscle is activated. Experimental transection in Bufo of the hypoglossal nerve innervating the main tongue protractor muscle disrupts this coordination and the mouth fails to open during feeding attempts (Nishikawa et al., 1992; Nishikawa and Gans, 1996). In Rana, transection of the hypoglossal nerve may or may not inhibit mouth opening, depending on prey dimensions (Anderson and Nishikawa, 1996). CAECILIANS Aquatic feeding in caecilians As far as is known, all basal caecilians (Rhinatrematidae, Ichthyophiidae and Uraeotyphlidae) have a life history that includes a free-living aquatic larval stage and a terrestrial, post-metamorphic phase (Wake, 1977a, b, 1992; Nussbaum, 1979; Wilkinson, 1992; Wilkinson and Nussbaum, 1996). A free-living aquatic larval life stage is also retained in some deeply-nested (i.e., non-basal) taxa (Grandisonia, Parker, 1958; Sylvacaecilia, Largen et al., 1972; Praslinia, Nussbaum, 1992). In larvae, prey are located using electrical cues (Himstedt and Fritzsch, 1990) probably in combination with chemical and tactile cues. The visual system cannot form images and is apparently not used to localize prey items (Fritzsch et al., 1985). The larvae capture prey using suction feeding with little or no forward movement of the head (O'Reilly, 2000). Prey items are moved to the esophagus using hydraulic transport. In larvae, muscles served by the trigeminal (jaw adduction), facial (interhyoideus, levator hyoideus), vagus and hypoglossal (hyobranchial musculature) nerves all function to move the jaws and hyobranchial apparatus during feeding. Among adult caecilians, only the South American typhlonectine caeciliids (sensuHedges et al., 1993) and the Seychellian caeciliid Hypogeophis are known for certain to forage regularly in water. Hypogeophis, Chthonerpeton and Nectocaecilia are semiaquatic, while Typhlonectes, Potomotyphlus and Atretochoana are completely aquatic in habit (Nussbaum, 1992; Wilkinson and Nussbaum, 1999). Other caecilians may facultatively forage in water as adults (e.g.,Ichthyophis kohtaoensis, Crapon De Crapona and Himstedt, 1985). A combination of behavioral observations and anatomical evidence suggests that aquatically-foraging adult caecilians use a combination of electrical, chemical, and mechanical cues to initiate feeding (Fritzsch and Wake, 1986; Schmidt and Wake, 1990; Himstedt and Simon, 1995; O'Reilly, 2000). After prey capture is initiated, jaw prehension is used to ingest the prey item (Wilkinson, 1991; O'Reilly, 2000). Although marked hyoid depression can be seen during ingestion in Typhlonectes natans, there is no evidence that sufficient flow into the mouth is generated to help procure prey items (O'Reilly, 2000). Prey capture involves a coordinated effort of virtually all of the muscles in the body including those involved in jaw, hyoid, and trunk movements. Swallowing is achieved by a combination of lingual, hydraulic, and inertial transport (Bemis et al., 1983; O'Reilly, 2000). While smaller prey items are swallowed whole, larger items can be dismembered by a combination of long axis body rolling and rubbing the side of the head against the substrate. The feeding movements of Typhlonectes are more rapid than those of terrestrial caecilians of similar body size. It is not clear whether this is the result of an evolutionary shift in the rate of feeding movements, or a facultative effect of feeding in water versus feeding on land. Terrestrial feeding in caecilians Terrestrial caecilians use jaw prehension to capture prey. Although their tongues are large, they do not protrude them beyond the threshold of the jaws (Bemis et al., 1983; O'Reilly, 2000). The hyobranchial apparatus no longer functions in prey capture, only cranial muscles innervated by the trigeminal and facial nerves function during the apprehension of food items (Bemis et al., 1983; Nussbaum, 1983). However, prey capture on land requires the coordinated movement of the body to place the jaws on the prey item, requiring motor control of the trunk musculature. As in aquatic prey capture, caecilians can dismember larger prey on land by spinning on the long axis of the body (Tanner, 1971; Bemis et al., 1983). Ancestral feeding in caecilians Given that larva are present in the three most basal major clades, the common ancestor of living caecilians had a free-living aquatic larval stage that used suction feeding to capture prey (O'Reilly, 2000). Prey capture was triggered by electrical, chemical and tactile cues. Small prey were eaten whole, but larger prey could be dismembered if too large to swallow. After metamorphosis, jaw prehension was used to capture prey, as in extant forms. Although capable of feeding in water, post-metamorphic individuals were primarily terrestrial. Chemical and tactile cues initiated prey capture, which was highly variable within individuals. As in the larval stage, very large prey could be attacked and dismembered (as described above), while smaller prey tended to be eaten whole. Although there is currently a consensus that the interhyoideus contributed to mouth closing ancestrally in caecilians, there is some debate about whether or not it played a dominant role in jaw closing. Nussbaum (1983) argues that the interhyoideus played a relatively minor role in jaw adduction, as is seen in living rhinatrematids where the levators mandibulae are the primary jaw closing muscles. However, Carroll (2000) argues that the condition in rhinatrematids is secondary and that the ancestral caecilian displayed an arrangement more similar to that seen in living Ichthyophis, where the interhyoideus is the dominant jaw closing muscle. From a motor control perspective, this distinction is a minor one, because both scenarios require simultaneous activation of the interhyoideus and levators mandibulae as a component of the ancestral feeding pattern. Modulation and variation The data available suggest that larval caecilian feeding behavior is highly stereotyped and not modulated. However, all data at this point come from a single species from which no attempt was made to elicit modulation by varying prey type or other conditions (O'Reilly, 2000). If the prey capture behavior of Epicrionops turns out to be typical for caecilian larvae (Fig. 4), the sheer speed of movement would preclude feedback modulation, however, feed-forward modulation remains a possibility. The gape cycle of post-metamorphic caecilians is extremely variable, whether feeding on land or in water. The available evidence suggests that this variation is in fact modulation, probably driven primarily by tactile feedback, but perhaps also by olfactory information via a feed-forward mechanism. PATTERNS OF MOTOR CONTROL We see immediately from examining a variety of amphibians that they are remarkably diverse in their feeding morphology, function and motor control. Given this diversity, generalizations derived from a few taxa early in the study of amphibian feeding are bound to give a distorted picture of amphibian feeding. It is unfortunate that a few taxa with stereotyped feeding movements (e.g.,Bufo and Ambystoma) served as models. Variation and modulation of feeding movements seem to occur almost everywhere one looks among amphibians, and a high degree of variation is probably the ancestral condition. Much evolutionary diversification of feeding mechanics and motor control has occurred and it is likely that stereotypy is a derived feature of amphibian feeding. The example of feedback coordination of tongue and jaws in Bufo (in which activation of the tongue protractor muscle disinhibits the jaw levators, permitting the mouth to open; Nishikawa and Gans, 1992, 1996) now makes extreme stereotypy an interesting novelty that may lead to the discovery of particular biomechanical or neurobiological constraints. Stereotypy observed in the laboratory may not accurately reflect an animal's abilities to modulate its feeding movements. Relatively few studies have examined modulation directly by altering the conditions in which the animals are feeding and determining the effect on kinematics or muscle activity patterns. In most cases, modulation has been revealed (Salamanders: Erdman and Cundall, 1984; Miller and Larsen, 1990; Elwood and Cundall, 1994; Maglia and Pyles, 1995; Deban, 1997b; Deban and Dicke, 1999; Frogs: Anderson, 1993; Anderson and Nishikawa, 1996; Ritter and Nishikawa, 1995; Caecilians: O'Reilly, 2000), and only rarely, modulation was absent (e.g.,Reilly and Lauder, 1989). Apparently stereotyped movements in animals feeding on a single prey type that is presented in a uniform manner are likely to represent only a small portion of the animal's repertoire of feeding movements, and should be viewed with caution. Stereotypy in such cases may be an artifact of laboratory conditions. Another pattern to emerge from this survey is the difference between salamanders and frogs in the relationship and coordination of tongue and jaw movements, particularly in species with inertia-based feeding mechanisms, such as bolitoglossine plethodontids and bufonids. Frogs with inertial elongation of the tongue rely on jaw movements to project the tongue, and tongue and jaw movements must be tightly coordinated. In Bufo, proprioceptive feedback from the tongue through the hypoglossal nerve delays the onset of activity in the jaw levators, allowing the mouth to open at precisely the right moment for tongue protraction (Nishikawa and Gans, 1996). This coordination of tongue and jaw movements is necessary in frogs because of the attachment of the tongue at the mandibular symphysis; the movements are biomechanically linked, especially in frogs with derived inertial elongation. Similar feedback systems have evolved repeatedly in ballistic-tongued frogs, among ranoids, phyllomedusine hylids, and leptodactylids. In Bufo, tongue accuracy is very poor (Gray, 1997), compared to that of some short-tongued frogs, such as Hyla, and compared to projectile-tongued salamanders. In salamanders, the evolutionary process is reversed. Salamanders with generalized tongue protraction have an attachment of the tongue pad to the mandible via the genioglossus muscle (as in all amphibians), and the tongue pad is flipped over the mandibular symphysis to some extent during tongue protraction. In plethodontid salamanders with derived tongue projection, the tongue pad has been freed of the lower jaw (the genioglossus is elongated or lost) (Lombard and Wake, 1977) and tongue projection follows a linear trajectory (Deban, 1997a, b). Thus, in salamanders, jaw and tongue movements have become decoupled and coordination of the two movements is less important in tongue projecting salamanders than in frogs with derived inertial feeding mechanics. This decoupling of tongue and jaws avoids the complex multi-joint coordination that ballistic-tongued frogs are faced with, and may account for the high degree of accuracy with no cost in speed observed in bolitoglossine salamanders (Wiggers et al., 1995). Studies of amphibian feeding provide examples of some general principles of the motor control of goal-directed movements. (1) The duration of movements appears to be directly related to level of variation present. For example, the extraordinarily slow feeding movements of adult caecilians are associated with the highest degree of variation and modulation. In contrast, the extremely rapid movements of suction feeding amphibians seem to display the least variation. (2) Feedback does not always play a role in the modulation of feeding movements. Feedback control may be absent either because the duration of movement is extremely short or because of biomechanical constraints which prohibit the alteration of a movement after it has begun. (3) Stereotypy does not preclude the use of feedback control. As seen in toads, stereotypy may indicate that proprioceptive or reflex feedback is needed to control a particular biomechanical system. For example, ballistic systems may not be under continuous muscular control, limiting the opportunity for adjustments. (4) Rapid ballistic movements, as seen in bolitoglossine salamanders, can still be modulated to a high degree via feed-forward control and can be very accurate (Deban and Dicke, 1999); that is, there need not always be a trade-off between speed and accuracy of movement. (5) Finally, decoupling of tongue and jaw movements may help increase accuracy of tongue placement because fewer joints are involved and coordination is simplified, as exemplified by the mechanisms of bolitoglossine salamanders and hydrostatic-tongued frogs like Hemisus. Open in new tabDownload slide Fig. 1. Scenario of the motor control of movement. Sensory information can influence movement at different times in the process, as shown by the vertical arrows Open in new tabDownload slide Fig. 1. Scenario of the motor control of movement. Sensory information can influence movement at different times in the process, as shown by the vertical arrows Open in new tabDownload slide Fig. 2. Video sequence of the hynobiid salamander Salamandrella keyserlingii feeding on a termite. This relatively slow feeding event illustrates tongue prehension as well as the precise control of lateral tongue aiming. Times are in milliseconds from the start of mouth opening at time zero Open in new tabDownload slide Fig. 2. Video sequence of the hynobiid salamander Salamandrella keyserlingii feeding on a termite. This relatively slow feeding event illustrates tongue prehension as well as the precise control of lateral tongue aiming. Times are in milliseconds from the start of mouth opening at time zero Open in new tabDownload slide Fig. 3. Video sequence of an adult Hymenochirus, an aquatic frog in the family Pipidae, capturing a worm using suction feeding. Only members of this genus of frogs are known to suction feed. Note the bending of the lower jaw which reduces the lateral gape, the movement of the worm towards the frog, and the substantial hyobranchial depression. Times are in milliseconds from the start of mouth opening at time zero Open in new tabDownload slide Fig. 3. Video sequence of an adult Hymenochirus, an aquatic frog in the family Pipidae, capturing a worm using suction feeding. Only members of this genus of frogs are known to suction feed. Note the bending of the lower jaw which reduces the lateral gape, the movement of the worm towards the frog, and the substantial hyobranchial depression. Times are in milliseconds from the start of mouth opening at time zero Open in new tabDownload slide Fig. 4. Ciné sequence of the larval caecilian Epicrionops suction feeding on a piece of earthworm. Note the extremely rapid movement of the worm, which blurs in the second image and is engulfed in the third. Times are in milliseconds from the start of mouth opening at time zero Open in new tabDownload slide Fig. 4. Ciné sequence of the larval caecilian Epicrionops suction feeding on a piece of earthworm. Note the extremely rapid movement of the worm, which blurs in the second image and is engulfed in the third. Times are in milliseconds from the start of mouth opening at time zero 1 From the Symposium Motor Control of Vertebrate Feeding: Function and Evolution presented at the Annual Meeting of the Society for Integrative and Comparative Biology, 3–7 January 2001, at Chicago, Illinois. 2 E-mail: sdeban@usa.net 3 Present address: Department of Biology, University of Miami, Coral Gables FL 33124 We thank Michael Alfaro and Anthony Herrel for organizing the symposium in which this paper was presented. Paul Grobstein and Curt Anderson provided helpful insight into the origin and proper use of the term “motor program.” J.C.O. was supported by a Darwin Postdoctoral Fellowship from the University of Massachusetts. Two anonymous reviewers provided helpful suggestions for improving the manuscript. Some equipment used to obtain feeding sequences was provided by the Gompertz professorship, University of California Berkeley, courtesy of D. B. Wake. References Alcobendas , M. , H. Dopazo, and P. Alberch. 1996 . Geographic variation in allozymes of populations of Salamandra salamandra (Amphibia: Urodela) exhibiting distinct reproductive modes. J. Evol. Biol , 9 83 -102. Crossref Search ADS Anderson , C. W. 1993 . The modulation of feeding behavior in response to prey type in the frog Rana pipiens. J. Exp. Biol , 179 1 -12. Anderson , C. W. , and K. C. Nishikawa. 1996 . The roles of visual and proprioceptive information during motor program choice in frogs. J. Comp. Physiol. A , 179 753 -762. Crossref Search ADS PubMed Bartels , M. , H. Münz, and B. Claas. 1990 . Representation of lateral line and electrosensory systems in the midbrain of the axolotl, Ambystoma mexicanum. J. Comp. Physiol. A , 167 347 -356. Crossref Search ADS Bemis , W. E. , K. Schwenk, and M. H. Wake. 1983 . Morphology and function of the feeding apparatus of Dermophis mexicanus (Amphibia: Gymnophiona). Zool. J. Linn. Soc , 77 75 -96. Crossref Search ADS Beneski , J. T. Jr., , J. H. Larsen Jr.,, and B. T. Miller. 1995 . Variation in the feeding kinematics of mole salamanders (Ambystomatidae: Ambystoma). Can. J. Zool , 73 353 -366. Crossref Search ADS Comer , C. , and P. Grobstein. 1981 . Tactually elicited prey acquisition behavior in the frog, Rana pipiens, and a comparison with visually elicited behavior. J. Comp. Physiol , 142 141 -150. Crossref Search ADS Carroll , R. L. 2000 . Eocaecilia and the origin of caecilians. In H. Heatwole and R. L. Carroll (eds.), Palaeontology: The evolutionary history of the Amphibia. Chapter 17, pp. 1402–1411. Amphibian biology, 4, pp. 973–1494. Beatty Press, Surrey. Crapon De Crapona , M. , and W. Himstedt. 1985 . Das aquatische Verhalten der Blindwühle Ichthyophis kohtaoensis Taylor 1960 (Gymnophiona: Ichthyophiidae). [The aquatic behavior of the caecilian Ichthyophis kohtaoensis (Gymnophiona: Ichthyophidae)]. Salamandra , 21 192 -196. Cundall , D. , J. Lorenz-Elwood, and J. D. Groves. 1987 . Asymmetric suction feeding in primitive salamanders. Experientia , 43 1229 -1231. Crossref Search ADS Deban , S. M. 1997 . Development and evolution of feeding behavior and functional morphology in salamanders of the family Plethodontidae. Ph.D. Diss., Univ. of California, Berkeley. Deban , S. M. 1997 . Modulation of prey-capture behavior in the plethodontid salamander Ensatina eschscholtzii. J. Exp. Biol , 200 1951 -1964. PubMed Deban , S. M. , and U. Dicke. 1999 . Motor control of tongue movement during prey capture in plethodontid salamanders. J. Exp. Biol , 202 3699 -3714. PubMed Deban , S. M. , and S. B. Marks. 1992 . Aquatic prey capture in plethodontid salamanders. Amer. Zool , 32 140A . Deban , S. M. , and S. B. Marks. 2002 . Metamorphosis and evolution of feeding behaviour in salamanders of the family Plethodontidae. Zool. J. Linn. Soc. (In press). Deban , S. M. , and K. C. Nishikawa. 1992 . The kinematics of prey capture and the mechanism of tongue protraction in the green tree frog Hyla cinerea. J. Exp. Biol , 170 235 -256. Deban , S. M. , and D. B. Wake. 2000 . Aquatic feeding in salamanders. In K. Schwenk (ed.), Feeding: Form, function and evolution in tetrapod vertebrates, pp. 65–94. Academic Press, San Diego. Dijkgraaf , S. 1956 . Elektrophysiologische Untersuchungen an der Seitenlinie von Xenopus laevis. Experientia , 12 276 -278. Crossref Search ADS PubMed Dole , J. W. , B. B. Rose, and K. H. Tachiki. 1981 . Western toads (Bufo boreas) learn odor of prey insects. Herpetologica , 37 63 -68. Duellman , W. E. , and L. Trueb. 1986 . Biology of amphibians. McGraw Hill, New York. Elwood , J. R. L. , and D. Cundall. 1994 . Morphology and behavior of the feeding apparatus in Cryptobranchus alleganiensis (Amphibia: Caudata). J. Morph , 220 47 -70. Crossref Search ADS Ewert , J.-P. 1987 . Neuroethology of releasing mechanisms: Prey catching in toads. Behav. Brain Sci , 10 337 -405. Crossref Search ADS Erdman , S. , and D. Cundall. 1984 . The feeding apparatus of the salamander Amphiuma tridactylum: Morphology and behavior. J. Morph , 181 175 -204. Crossref Search ADS Feder , M. E. , and R. J. Wassersug. 1984 . Aerial versus aquatic oxygem consumption in the larvae of the clawed frog, Xenopus laevis. J. Exp. Biol , 108 231 -245. Findeis , E. K. , and W. E. Bemis. 1990 . Functional morphology of tongue projection in Taricha torosa (Urodela: Salamandridae). Zool. J. Linn. Soc , 99 129 -158. Crossref Search ADS Fritzsch , B. 1990 . The evolution of metamorphosis in amphibians. J. Neurobiol , 21 1011 -1021. Crossref Search ADS PubMed Fritzsch , B. , and U. Wahnschaffe. 1983 . The electroreceptive ampullary organs of urodeles. Cell Tissue Res , 229 483 -503. Crossref Search ADS PubMed Fritzsch , B. , W. Himstedt, and M.-D. Crapon de Caprona. 1985 . Visual projections in larval Ichthyophis kohtaoensis (Amphibia: Gymnophiona). Dev. Brain. Res , 23 201 -210. Crossref Search ADS Fritzsch , B. , and M. H. Wake. 1986 . The distribution of ampullary organs in Gymnophiona. J. Herp , 20 90 -93. Crossref Search ADS Ghez , C. , and J. Krakauer. 2000 . The organization of movement. In E. Kandel, J. H. Schwartz, and T. M. Jessell (eds.), Principles of neural science, pp. 653–673. McGraw-Hill, New York. Gray , L. A. 1997 . Tongue morphology, feeding behavior and feeding ecology in anurans. Ph.D. Diss., Northern Arizona University, Flagstaff. Gray , L. A. , and K. C. Nishikawa. 1995 . Feeding kinematics of phyllomedusine tree frogs. J. Exp. Biol , 198 457 -463. PubMed Gray , L. A. , J. C. O'Reilly, and K. Nishikawa. 1997 . Evolution of forelimb movement patterns for prey manipulation in anurans. J. Exp. Zool , 277 417 -424. Crossref Search ADS PubMed Greven , H. 1998 . Survey of the oviduct of salamandrids with special reference to the viviparous species. J. Exp. Zool , 282 507 -525. Crossref Search ADS PubMed Grobstein , P. , C. Comer, and S. K. Kostyk. 1983 . Frog prey capture behavior: Between sensory maps and directed motor output. In J.-P. Ewert, R. Capranica, and D. Ingle (eds.), Advances in vertebrate neuroethology, pp. 331–347. Plenum Press, New York, New York. Grobstein , P. , A. Reyes, L. Zwanziger, and S. K. Kostyk. 1985 . Prey orienting in frogs: Accounting for variations in output with stimulus distance. J. Comp. Physiol. A , 156 775 -785. Crossref Search ADS Hedges , S. B. , R. A. Nussbaum, and L. R. Maxson. 1993 . Caecilian phylogeny and biogeography inferred from mitochondrial DNA sequences of the 12S rRNA and 16S rRNA genes (Amphibia: Gymnophiona). Herp. Monograph , 7 64 -76. Crossref Search ADS Himstedt , W. , J. Kopp, and W. Schmidt. 1982 . Electroreception guides feeding behavior in amphibians. Naturwissenschaften 69. Himstedt , W. , and B. Fritzsch. 1990 . Behavioural evidence for electroreception in larvae of the caecilian Ichthyophis kohtaoensis (Amphibia, Gymnophiona). Zoologische Jahrbücher, Abteilung für allgemeine Zool. Physiol. Tiere , 94 486 -492. Himstedt , W. , and D. Simon. 1995 . Sensory basis of foraging behaviour in caecilians (Amphibia: Gymnophiona). Herpetol. J , 5 266 -270. Jaeger , R. G. , D. E. Barnard, and R. G. Joseph. 1982 . Foraging tactics of a terrestrial salamander: Assessing prey density. Am. Natur , 119 885 -890. Crossref Search ADS Largen , M. J. , P. A. Morris, and D. W. Yalden. 1972 . Observations on the caecilian Geotrypetes grandisonae Taylor (Amphibia; Gymnophiona) from Ethiopia. Mon. Zool. Ital. N S Suppl. IV , 8 185 -205. Larsen , J. H. , J. T. Beneski, and B. T. Miller. 1996 . Structure and function of the hyolingual system in Hynobius and its bearing on the evolution of prey capture in terrestrial salamanders. J. Morph , 227 235 -248. Crossref Search ADS Larsen , J. H. Jr., , J. T. Beneski Jr.,, and D. B. Wake. 1989 . Hyolingual feeding systems of the Plethodontidae: Comparative kinematics of prey capture by salamanders with free and attached tongues. J. Exp. Zool , 252 25 -33. Crossref Search ADS Larsen , J. H. Jr. , and D. J. Guthrie. 1975 . The feeding system of terrestrial tiger salamanders (Ambystoma tigrinum melanostictum Baird). J. Morph , 147 137 -154. Crossref Search ADS Lauder , G. V. , and S. M. Reilly. 1988 . Functional design of the feeding mechanism in salamanders causal bases of ontogenetic changes in function. J. Exp. Biol , 134 219 -234. Lauder , G. V. , and H. B. Shaffer. 1985 . Functional morphology of the feeding mechanism in aquatic ambystomatid salamanders. J. Morph , 185 297 -326. Crossref Search ADS Lauder , G. V. , and H. B. Shaffer. 1986 . Functional design of the feeding mechanism in lower vertebrates: Unidirectional and bidirectional flow systems in the tiger salamander (Ambystoma tigrinum). Zool. J. Linn. Soc , 88 277 -290. Crossref Search ADS Lauder , G. V. , and H. B. Shaffer. 1988 . Ontogeny of functional design in tiger salamanders (Ambystoma tigrinum): Are motor patterns conserved during major morphological transformations? J. Morph , 197 249 -268. Crossref Search ADS Lettvin , J. Y. , H. R. Maturana, W. S. McCulloch, and W. H. Pitts. 1959 . What the frog's eye tells the frog's brain. Proc. Inst. Radio Engrs , 47 1940 -1951. Lombard , R. E. , and D. B. Wake. 1976 . Tongue evolution in the lungless salamanders, Family Plethodontidae. I. Introduction, theory and a general model of dynamics. J. Morph , 148 265 -286. Crossref Search ADS Lombard , R. E. , and D. B. Wake. 1977 . Tongue evolution in the lungless salamanders, Family Plethodontidae. II. Function and evolutionary diversity. J. Morph , 153 39 -80. Crossref Search ADS Maglia , A. M. , and R. A. Pyles. 1995 . Modulation of prey-capture behavior in Plethodon cinereus (Green) (Amphibia: Caudata). J. Exp. Zool , 272 167 -183. Crossref Search ADS Martin , J. B. , N. B. Witherspoon, and M. H. Keenleyside. 1974 . Analysis of feeding behavior in the newt Notophthalmus viridescens. Can. J. Zool , 52 277 -281. Crossref Search ADS Matsushima , T. , T. Satou, and K. Ueda. 1988 . Neuronal pathways for the lingual reflex in the Japanese toad. J. comp. Physiol. A , 164 173 -193. Crossref Search ADS PubMed Matthes , E. 1934 . Bau und Funktion der Lippensäume wasserlebender Urodelen. Zeitschrift für Morphologie und Ökologie der Tiere , 28 155 -169. Crossref Search ADS McDiarmid , R. W. , and R. Altig. 1999 . Tadpoles: The biology of anuran larvae. University of Chicago Press, Chicago. Miller , B. T. , and J. H. Larsen Jr. 1989 . Feeding performance in aquatic postmetamorphic newts (Urodela: Salamandridae): Are bidirectional flow systems necessarily inefficient? Can. J. Zool , 67 2414 -2421. Crossref Search ADS Miller , B. T. , and J. H. Larsen Jr. 1990 . Comparative kinematics of terrestrial prey capture in salamanders and newts (Amphibia: Urodela: Salamandridae). J. Exp. Zool , 256 135 -153. Crossref Search ADS Muskhelishvili , T. A. 1964 . New findings about reproduction of the Caucasus salamander (Mertensiella caucasica). Soobshch. Akad. Nauk. Gruz. S.S.R , 36 183 -185. Nishikawa , K. 2000 . Feeding in frogs. In K. Schwenk (ed.), Feeding: Form, function, and evolution in tetrapod vertebrates, pp. 117–147. Academic Press, San Diego. Nishikawa , K. C. , C. W. Anderson, S. M. Deban, and J. C. O'Reilly. 1992 . The evolution of neural circuits controlling feeding behavior in frogs. Brain Behav. Evol , 40 125 -140. Crossref Search ADS PubMed Nishikawa , K. C. , and D. Cannatella. 1991 . Kinematics of prey capture in the tailed frog Ascaphus truei (Anura: Ascaphidae). Zool. J. Linn. Soc , 103 289 -307. Crossref Search ADS Nishikawa , K. C. , and C. Gans. 1992 . The role of hypoglossal sensory feedback during feeding in the marine toad, Bufo marinus. J. Exp. Biol , 264 245 -252. Nishikawa , K. C. , and C. Gans. 1996 . Mechanisms of tongue protraction and narial closure in the marine toad Bufo marinus. J. Exp. Biol , 199 2511 -2529. PubMed Nishikawa , K. C. , and G. Roth. 1991 . The mechanism of tongue protraction during prey capture in the frog Discoglossus pictus. J. Exp. Biol , 159 217 -234. Nussbaum , R. A. 1979 . The taxonomic status of the caecilian genus Uraeotyphlus Peters. Occasional Papers of the Museum of Zoology, University of Michigan , 687 1 -20. Nussbaum , R. A. 1983 . The evolution of a unique dual jaw-closing mechanism in caecilians (Amphibia: Gymnophiona) and its bearing on caecilian ancestry. J. Zool , 199 545 -554. Crossref Search ADS Nussbaum , R. A. 1992 . Caecilians. In H. G. Cogger and R. G. Zweifel (eds.), Reptiles and amphibians, pp. 52–59. Smithmark, New York. O'Reilly , J. C. 2000 . Feeding in Caecilians. In K. Schwenk (ed.), Feeding: Form, function and evolution in tetrapod vertebrates, pp. 149–166. Academic Press, San Diego. O'Reilly , J. C. , S. M. Deban, and K. C. Nishikawa. 2002 . Derived life history characteristics constrain the evolution of aquatic feeding behavior in amphibians. Zoology. (In press). Orton , G. L. 1953 . The systematics of vertebrate larvae. Syst. Zool , 2 63 -75. Crossref Search ADS Özeti , N. 1979 . Reproductive biology of the salamander Mertensiella luschani antalyana. Herpetologica , 35 193 -197. Özeti , N. , and D. B. Wake. 1969 . The morphology and evolution of the tongue and associated structures in salamanders and newts (Family Salamandridae). Copeia 1969:91–123. Parker , H. W. 1958 . Caecilians of the Seychelles Islands with description of a new subspecies. Copeia 1958:71–76. Peck , S. B. 1973 . Feeding efficiency in the cave salamander Haideotriton wallacei. Int. J. Speleol , 5 15 -19. Crossref Search ADS Regal , P. J. 1966 . Feeding specializations and the classification of terrestrial salamanders. Evolution , 20 392 -407. Crossref Search ADS Regal , P. J. , and C. Gans. 1976 . Functional aspects of the evolution of frog tongues. Evolution , 20 392 -407. Crossref Search ADS Reilly , S. M. , and R. Altig. 1996 . Cranial ontogeny in Siren intermedia (Caudata: Sirenidae): Paedomorphic, metamorphic, and novel patterns of heterochrony. Copeia 1996:29–41. Reilly , S. M. , and G. V. Lauder. 1989 . Physiological bases of feeding behavior in salamanders: Do motor patterns vary with prey type? J. Exp. Biol , 141 343 -358. Reilly , S. M. , and G. V. Lauder. 1990 . The strike of the tiger salamander: Quantitative electromyography and muscle function during prey capture. J. Comp. Physiol , 167 827 -839. Crossref Search ADS Reilly , S. M. , and G. V. Lauder. 1992 . Morphology behavior and evolution: Comparative kinematics of aquatic feeding in salamanders. Brain Beh. Evol , 40 182 -196. Crossref Search ADS Ritter , D. , and K. Nishikawa. 1995 . The kinematics and mechanism of prey capture in the African pig-nosed frog (Hemisus marmoratum): Description of a radically divergent anuran tongue. J. Exp. Biol , 198 2025 -2040. PubMed Roth , G. 1976 . Experimental analysis of the prey catching behavior of Hydromantes italicus Dunn (Amphibia, Plethodontidae). J. Comp. Physiol , 109 47 -58. Crossref Search ADS Roth , G. 1987 . Visual behavior in salamanders. Springer, Berlin. Roth , G. , K. C. Nishikawa, D. B. Wake, U. Dicke, and T. Matsushima. 1990 . Mechanics and neuromorphology of feeding in amphibians. Neth. J. Zool , 40 115 -135. Crossref Search ADS Roth , G. , and D. B. Wake. 1985 . Trends in the functional morphology and sensorimotor control of feeding behavior in salamanders: An example of the role of internal dynamics in evolution. Acta Biotheoretica , 34 175 -192. Crossref Search ADS PubMed Ruibal , R. , and E. Thomas. 1988 . The obligate carnivorous larvae of the frog, Lepidobatrachus laevis (Leptodactylidae). Copeia 1988:591–604. Russell , I. J. 1976 . Amphibian lateral line receptors. In R. Llinás and W. Precht (eds.), Frog neurobiology, pp. 513–550. Springer, Berlin. Shaffer , H. B. , and G. V. Lauder. 1985 . Aquatic prey capture in ambystomatid salamanders: Patterns of variation in muscle activity. J. Morph , 183 273 -284. Crossref Search ADS Schlosser , G. , and G. Roth. 1995 . Distribution of cranial and rostral spinal nerves in tadpoles of the frog Discoglossus pictus (Discoglossidae). J. Morph , 226 189 -212. Crossref Search ADS Schmidt , A. , and M. H. Wake. 1990 . Olfactory and vomeronasal systems of caecilians (Amphibia: Gymnophiona). J. Morph , 205 255 -268. Crossref Search ADS Schwenk , K. , and D. B. Wake. 1993 . Prey processing in Leurognathus marmoratus and the evolution of form and function in desmognathine salamanders (Plethodontidae). Biol. J. Linn. Soc , 49 141 -162. Crossref Search ADS Shepherd , G. M. 1994 . Neurobiology. Oxford University Press, New York, Oxford. Shinn , E. A. , and J. W. Dole. 1978 . Evidence for a role for olfactory cues in the feeding response of leopard frogs, Rana pipiens. Herpetologica , 34 167 -172. Sokol , O. M. 1962 . The tadpole of Hymenochirus boettgeri. Copeia 1962:272–284. Tanner , K. 1971 . Notizen zur Pflege und zum Verhalten einiger Blindwühlen (Amphibia: Gymnophiona). Salamandra , 7 91 -100. Thexton , A. J. , D. B. Wake, and M. H. Wake. 1977 . Tongue function in the salamander Bolitoglossa occidentalis. Arch. Oral Biol , 22 361 -366. Crossref Search ADS PubMed Trueb , L. , and C. Gans. 1983 . Feeding specializations of the Mexican burrowing toad, Rhinophrinus dorsalis (Anura: Rhinophrynidae). J. Zool , 199 189 -208. Crossref Search ADS Uiblein , F. , J. P. Durand, C. Juberthie, and J. Parzefall. 1992 . Predation in caves: The effects of prey immobility and darkness on the foraging behaviour of two salamanders, Euproctus asper and Proteus anguinus. Behav. Process , 28 33 -40. Crossref Search ADS Valdez , C. M. , and K. C. Nishikawa. 1997 . Sensory modulation and behavioral choice during feeding in the Australian frog, Cyclorana novaehollandiae. J. Comp. Physiol. A , 180 187 -202. Crossref Search ADS PubMed Wake , D. B. 1982 . Functional and developmental constraints and opportunities in the evolution of feeding systems in urodeles. In D. Mossakowski and G. Roth (eds.), Environmental adaptation and evolution, pp. 51–66. Gustav Fischer, Stuttgart, New York. Wake , D. B. , and S. M. Deban. 2000 . Terrestrial feeding in salamanders. In K. Schwenk (ed.), Feeding: Form, function and evolution in tetrapod vertebrates, pp. 95–116. Academic Press, San Diego. Wake , D. B. , G. Roth, and M. H. Wake. 1983 . Tongue evolution in lungless salamanders, family Plethodontidae. III. Patterns of peripheral innervation. J. Morph , 178 207 -224. Crossref Search ADS Wake , M. H. 1977 . Fetal maintenance and its evolutionary significance in the amphibia: Gymnophiona. J. Herp , 11 379 -386. Crossref Search ADS Wake , M. H. 1977 . The reproductive biology of caecilians: an evolutionary perspective. In E. H. Taylor and S. I. Guttman (eds.), Reproductive biology of amphibians, pp. 73–101. Plenum, New York. Wake , M. H. 1992 . Reproduction in caecilians. In W. C. Hamlett (ed.), Reproductive biology of South American vertebrates, pp. 112–120. Springer, New York. Wassersug , R. 1972 . The mechanism of ultraplanktonic entrapment in anuran larvae. J. Morph , 137 279 -287. Crossref Search ADS Wassersug , R. J. , and K. Hoff. 1979 . A Comparative study of the buccal pumping mechanism of tadpoles. Biol. J. Linn. Soc , 12 225 -259. Crossref Search ADS Wiggers , W. , G. Roth, C. Eurich, and A. Straub. 1995 . Binocular depth perception mechanisms in tongue-projecting salamanders. J. Comp. Physiol. (A) , 176 365 -377. Crossref Search ADS Wilkinson , M. 1991 . Adult tooth crown morphology in the Typhlonectidae (Amphibia: Gymnophiona): A reinterpretation of variation and its significance. Z. Zool. Syst. Evolut.—Forsch , 29 304 -311. Crossref Search ADS Wilkinson , M. 1992 . On the life history of the caecilian genus Uraeotyphlus (Amphibia: Gymnophiona). Herpetol. J , 2 121 -124. Wilkinson , M. , and R. A. Nussbaum. 1996 . On the phylogenetic position of the uraeotyphlidae (Amphibia: Gymnophiona). Copeia 1996:550–562. Wilkinson , M. , and R. A. Nussbaum. 1999 . Evolutionary relationships of the lungless caecilian Atretochoana eiselti (Amphibia: Gymnophiona: Typhlonectidae). Zool. J. Linn. Soc , 126 191 -223. Crossref Search ADS The Society for Integrative and Comparative Biology
TI - The Evolution of the Motor Control of Feeding in Amphibians
JF - Integrative and Comparative Biology
DO - 10.1093/icb/41.6.1280
DA - 2001-12-01
UR - https://www.deepdyve.com/lp/oxford-university-press/the-evolution-of-the-motor-control-of-feeding-in-amphibians-gGf5JiHWxv
SP - 1280
VL - 41
IS - 6
DP - DeepDyve
ER -