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 001469375
007 cr mnu|||uuuuu
008 040524s2004 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000277
Egner, Sarah A.
Auditory sensitivity of the sergeant majors (Abudefduf saxatilis) from post-settlement juvenile to adult
h [electronic resource] /
by Sarah A. Egner.
[Tampa, Fla.] :
University of South Florida,
Thesis (M.S.)--University of South Florida, 2004.
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 70 pages.
ABSTRACT: There is much evidence supporting the idea that pelagic larvae of coral reef fishes are active participants in their dispersal and return to a reef, however, the mechanisms used to navigate are still uncertain. It has been proposed that sensory cues, such as hearing, play a role. Sound is a potentially important cue for organisms in marine environments, especially in noisy environments like coral reefs. Sensory organs, including otolithic organs, of most coral reef fish form within the first few days of life. The auditory brainstem response (ABR) technique was used to measure hearing on a wide size range of sergeant majors (Abudefduf saxatilis). Complete audiograms were measured for 32 fish ranging in size from 11-121 mm. Significant effects of standard length on hearing thresholds at 100 and 200 Hz were detected. At these lower frequencies, thresholds increased with an increase in size. All fish were most sensitive to the lower frequencies (100-400 Hz). The frequency range that fish could detect sounds was dependent upon the size of the fish; the larger fish (> 50mm) were more likely to respond to higher frequencies (1000-1600 Hz). A. saxatilis have poor hearing sensitivity in comparison to audiograms of other hearing generalists including other species of Pomacentrids. Due to the high hearing thresholds found in this study in comparison to recorded ambient reef noise, it is unlikely that sound plays a significant role in the navigation of the pelagic larvae of sergeant majors to the return of the reef from large distances.
Adviser: Mann, David A.
auditory brainstem response (ABR).
x Marine Science
t USF Electronic Theses and Dissertations.
Auditory Sensitivity of Sergeant Majors ( Abudefduf saxatilis) from Post-settlement Juvenile to Adult by Sarah A. Egner A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science College of Marine Science University of South Florida Major Professor: David A. Mann, Ph.D. John C. Ogden, Ph.D. Joseph J. Torres, Ph.D. Date of Approval: April 9, 2004 Keywords: damselfish, auditory brainstem response (ABR), hearing sensitivity, larval settlement, coral reef Copyright 2004, Sarah A. Egner
ACKNOWLEDGEMENTS This project could not have been comp leted without the help of numerous individuals. My advisor, Dr. David Mann, has been supportive throughout the whole process, offering countless hours of assistance and guidance. I want to thank the other members of my committee, Dr. Jose Torres and Dr. John Ogden, for their much valued input on my project. Others who have helped include the staff at the Keys Marine Lab who provided me with lab space and assisted in catching the fish, Henry Feddern who caught the larger fish and Peter Boumwa who assisted in collecting the juvenile fish. Additionally, I must thank th e other members of the Mari ne Sensory Lab especially Mandy Hill, Brandon Casper and Jim Locascio who have been with me since the beginning. Most importantly, much thanks to all of my family a nd friends and to the person I love most in the world, all of whom have provided me with much needed love and support throughout my graduate years.
i TABLE OF CONTENTS LIST OF TABLES ii LIST OF FIGURES iii ABSTRACT iv INTRODUCTION 1 Fish hearing 2 Ontogenetic changes in audition 5 Settlement of pelagic fish 6 Auditory brainstem response 8 Purpose 8 METHODS 9 Fish acquisition and maintenance 9 Experimental Setup 9 Sound Generation and ABR Acquisition 12 Data Analysis 15 RESULTS 16 DISCUSSION 37 Auditory sensitivity of sergeant majors 37 Effect of size on the auditory sensitivity of the sergeant major 43 Use of reef sound as a navigational cue for Pelagic larvae 47 Future directions 52 LITERATURE CITED 54
ii LIST OF TABLES Table 1 Standard length and weight of each fish. 11 Table 2 Threshold sound levels for each fish. 20 Table 3 Results of ANOVA analyzing the overall goodness of fit the regression line for each frequency. 31 Table 4 Probabilities of st atistical differences among proportion of responses detected by each size group at each frequency. 36
iii LIST OF FIGURES Figure 1 Schematic drawings of the se nsory hair cell orientation patterns on the otolithic organs. 4 Figure 2 Diagram of ABR recording setup. 14 Figure 3 Input-output functions for each frequency for two fish. 18 Figure 4 Example of ABR waveform in time and frequency domain. 22 Figure 5 Audiogram of average threshold levels for all fish at each tested frequency. 24 Figure 6 Average audiograms for the three size groups of fishes. 26 Figure 7 Regression lines for all freq uencies tested examining standard length versus threshold level. 28-29 Figure 8 Percent of individuals in eac h size group that detected sound at each frequency. 34 Figure 9 Sergeant major audiogram in comparison to audiograms of two hearing specialists. 39 Figure 10 Audiogram of sergeant majo rs and other hearing generalists. 41 Figure 11 Average audiograms of four size groups of juveniles and adult Stegastes partitus (redrawn from Kenyon 1996). 46 Figure 12 Audiogram of smallest size group of sergeant majors in comparison to ambient reef noises. 51 Figure 13 Estimated distance from r eef that smallest size group of sergeant major can detect chorus spectra. 53
iv AUDITORY SENSITIVITY IN SERGEANT MAJORS ( Abudefduf saxatilis) FROM POST-SETTLEMENT JUVENILE TO ADULT Sarah A. Egner ABSTRACT There is much evidence supporting the idea th at pelagic larvae of coral reef fishes are active participants in their dispersal a nd return to a reef, however, the mechanisms used to navigate are still uncertain. It ha s been proposed that sensory cues, such as hearing, play a role. Sound is a potentially important cue for organisms in marine environments, especially in noisy environm ents like coral reefs. Sensory organs, including otolithic organs, of most coral reef fish form within the first few days of life. The auditory brainstem response (ABR) tec hnique was used to measure hearing on a wide size range of sergeant majors ( Abudefduf saxatilis ). Complete audiograms were measured for 32 fish ranging in size from 11-121 mm. Significant effects of standard length on hearing thresholds at 100 and 200 Hz were de tected. At these lower frequencies, thresholds increased with an increas e in size. All fish were most sensitive to the lower frequencies (100-400 Hz). The fre quency range that fish could detect sounds was dependent upon the size of the fish; the la rger fish (>50mm) were more likely to respond to higher frequencies (1000-1600 Hz). A. saxatilis have poor hearing sensitivity in comparison to audiograms of other hear ing generalists includi ng other species of Pomacentrids. Due to the high hearing thre sholds found in this study in comparison to recorded ambient reef noise, it is unlikely that sound pl ays a significant role in the navigation of the pelagic larvae of sergeant ma jors to the return of the reef from large distances.
1 INTRODUCTION Most of the ambient noise in the ocea n falls between 50-5000 Hz (Cato 1992). One of the noisiest habitats in the ocean is the coral reef. Sounds on or nearby a reef can be abiotic in origin, such as waves crashing on the reef, as well as biotic, short duration clicks and snaps produced predominantly by sn apping shrimp and other invertebrates and fish (Leis et al. 2002). Fish produce ep iphenomenal sounds during sudden movement and when feeding (Colson et al. 1998) and many fish use sound production voluntarily during spawning, schooling and aggressive terr itorial behaviors (Myrberg and Fuiman 2002). Nocturnal activity by snapping shrim p, fish and urchins creates an Â‘evening chorusÂ’ on the reef (Cato 1980; McCauley 1994, 1995). Numerous species in the large percif orm family Pomacentridae (damselfish), represented on reefs worldwide by approxi mately 320 species, are well-known sound producers (Allen 1975). One of the most comm on Pomacentrid species is the sergeant major ( Abudefduf saxatilis ) (Alshuth et al. 1998). A. saxatilis is an Atlantic species found along the western Atlantic coast, on most of the reefs in the Cari bbean, around islands of the mid-Atlantic, Cape Verde, and along the tropical coast of western Africa south to Angola (Allen 1991). Due to their abundance, and the lack of h earing data on this common species, sergeant majors were chosen for this study. Like most coral reef fishes, sergeant majors have a planktonic larval stage (Alshuth et al. 1998). They depos it demersal adhesive eggs on hard substrates (Alshuth et al. 1998) and the hatc hing of the larvae occurs four to seven days after fertilization (Alshuth 1998, Foster 1987). The larvae are pe lagic for an average of 17-20 days before settling on a reef (Wellington and Victor 1989). Coral reef fish larvae are usually found between 50-100 meters with the highest dive rsity of taxa in the upper 50 m (Cowen 2002). Pomacentrid larvae are concentrated in the mid depths (20-60m). Newly settled sergeant majors inhabit inshor e and offshore coral or rocky reefs and are found at a depth
2 range of 1-12 m (Allen 1991). Sergeant majo rs grow to a maximum standard length of 150 mm. Sergeant majors are not the most vocal of the damselfish but they do produce sound and live in a noisy environment (Santi ago and Castro 1997, Tolimieri et al. 2000, Fishelson 1970). There is little published data on sound produced by sergeant majors though damselfish are among the best stud ied sound-producing fishes (Lobel and Kerr 1999, Spanier 1979, Chen and Mok 1988, Myrb erg et al. 1993, Lobel and Mann 1995, Santiago and Castro 1997). A study on an Eastern Atlantic species, Abudefduf luridus found sound production by this fish during some aggressive interactions (Santiago and Castro 1997). Sound was in the frequency ra nge of 50-500 Hz with most of the energy concentrated in the low end of the spectrum. Courtship associated sounds studied in a Pacific species, Abudefduf sordidus, were found to be highly variable and of low amplitude in comparison to other Pomacentrids (Lobel and Kerr 1999). Due to the prevalence of sound production in the family, se rgeant majors most lik ely utilize sound in some fashion. Sound propagates well in water, but the di stance it travels whil e still detectable depends on five main factors: frequency, intensity, water depth, background noise and receiver sensitivity (Kingsford et al. 2002). A ll other things being equal, lower frequency sounds such as the crash of breaking waves tr avels further than hi gher frequencies like a snapping shrimp Â‘click.Â” The strength of a sound signal must exceed background noise to be detectable. In coral reef waters, b ackground noise levels ar e high, and wind can add considerably to the noise (K ingsford 2002). Some biolog ical sound from reefs still exceeds background noise levels at 4-20 km from its source (McCauley 1997). Fish hearing In fish, sound is detected by three otolithic organs, the saccule, utricle and lagena (Popper and Fay 1997). Each of these organs is a fluid filled sac with a sensory epithelium that has numerous sensory hair cell s arranged in specific patterns (Figure 1). The hair cells make up a ciliary bundle which consists of a cluster of
3 Figure 1 Schematic drawings of the sensory hair cell orientation patter ns on the otolithic organs. The arrows approximate the orientati on of the major portion of the hair cells in each epithelial region. The regions are se parated by the dotted lines. The patterns represent the majority of fish species. (D:dorsal, M: mediol ateral, R: Rostral) (Redrawn from Popper and Platt 1993).
4 M R D R Utricle Lagena D R Saccule
5 stereocilia and an eccentrically placed si ngle long kinocilium (Popper and Fay 1999). The base of each ciliary bundle synapses w ith neurons of the eighth cranial nerve (Myrberg and Fuiman 2002). Depolarization occurs when stereocilia are bent in the direction of the kinocilium. The displacement of the direct ionally sensitive hair cells relative to the otolith s is the effective auditory stimu lus for fish (Montgomery et al. 2001). Fish are categorized as hearing gene ralists or specialists depending upon the connection or the proximity of the swim bl adder to the ear (Popper and Platt 1993, Yan et al. 2000). Fish with the best hearing se nsitivity are those with swim bladder specializations that fa cilitate the transducti on of the pressure com ponent of the receptive field to particle displacement (Montgomery et al 2001). Without some specialization, such as the Weberian ossicle system f ound in ostariophysans (series of four bones connecting the swim bladder to the ears) ( von Frisch 1936), the swimbladder diverticulae in holocentrids (Coombs and Popper 1979), or th e gas filled bullae in mormyrids (Stipetic 1939), particle motion from the incident sound wave is thought to be the only stimulus that can be sensed by the fish auditory sy stem (Yan et al. 2000, Kenyon et al. 1998). Reef fish typically do not belong to hearing specialist groups and relatively little is known about their hearing capabilit ies (Montgomery et al 2001). Ontogenetic changes in audition Ontogenetic changes in hearing ar e well known in mammals (Rbsamen 1992, Zimmerman 1993, Lecanuet and Schaal 1996, McFadden et al. 1996, Reimer 1996) and birds (Gray and Rubel 1985, Golubeva a nd Tikhonov 1985), but have not been well studied in fish. Most of these mammalian a udition studies have had similar results; an increase in size was positively correlated w ith increased auditory sensitivity (Lecanuet and Schaal 1996, McFadden et al. 1996, Re imer 1996), broadening of hearing range (Rbsamen 1992), especially at higher freque ncies, and a shift in the most sensitive frequency of hearing (Golube va and Tikhonov 1985, Reimer 1996). A few studies have been conducted on the ontogenetic changes in hearing sensitivity for fishes (Popper 1971, Kenyon 1996, Wysocki and Ladich 2001, Higgs et al.
6 2003). Popper studied two different size gr oups of goldfish, usi ng a shock conditioning technique, and found that hearing threshol ds were not dependent on size (Popper 1971). The two size groups used had similar audiograms. Kenyon, however, conducted psychophysical experiments, util izing electric shock, on four size groups of the bicolor damselfish ( Stegastes partitus ). Hearing thresholds decr eased exponentially with an increase in size (Kenyon 1996). A study on the croaking gourami ( Trichopsis vittata ), a hearing specialist, measured evoked poten tials and had findings similar to Kenyon (Wysocki and Ladich 2001). The authors of the latter two studies suggest the use of only two size groups in PopperÂ’s study swayed the results (Kenyon 1996, Wysocki and Ladich 2001). In addition, goldfish are ostariophys an fishes, which have a fundamentally different peripheral ear morphology in co mparison to damselfish and gouramis. There is physiological evidence to suggest that ontogenetic auditory changes occur in fish. Previous studies indicate fish lacking swim bl adders have relatively poor hearing in both frequency range and sensitivit y (Yan et al. 2000). The closer the swim bladder is to the ear, the more sensitive th e audition. The swim bl adder is thought to act as an amplifier for some fish by transferring s ound pressure into displacement (Yan et al 2000). Though the role of the sw im bladder in hearing gene ralists is not well studied (Yan et al. 2000), as the fish grow, the swim bladder presumably increases in size, which could affect audition (Kenyon 1996). A second physiological process that coul d affect hearing ab ility is the ongoing addition of inner ear sensory hair cells (Lanford et al. 1996). Post-embryonic proliferation of hair cells in otolithic endorgans has been seen in various teleosts and elasmobranches (Corwin 1983, Popper and Hoxter 1984, Lombarte and Popper 1994), but the effect this increase in hair cells ha s on hearing is still unclear (Popper and Fay 1999). Settlement of pelagic larval fish Most coral reef fishes have a bipartite lif e cycle where they spend days to months in the open ocean as developing larvae until th e fish eventually settle on the reef as juveniles (Leis 1991). Sergeant majors have a larval stage approximately 17-20 days in
7 duration (Wellington and Victor 1989). One que stion in coral reef fish ecology that has become increasingly important is how pelagic la rval coral reef fish navigate their return to the reef. There is much recent evidence supporting th e idea that the pelagic larvae of coral reef fish are active participants in their di spersal and return to a reef (Stobutzki and Bellwood 1997, 1998, Armsworth 2000). They are strong swimmers, capable of swimming tens of kilometers to greater than 90 kilometers no n-stop at speeds fast enough to overcome currents (Stobutzki and Bellwood 1997). The methods used to navigate, however, are still uncertain. Fish form their sensory organs early in development, usually within the first few days of life (Leis and McCormick 2002, Myrberg and Fuiman 2002); the use of senses such as vision, olfaction, and hearing have been propos ed in literature. Because the coral reef is a noisy envir onment and because sound is used by many coral reef fish in communication, it is pl ausible that larval fish use hearing as a navigational tool (Tolimieri et al. 2000). Otolithic organs are often present at a very early stage in larval deve lopment (Cato 1978, Leis et al 1996, Leis and McCormick 2002). In sergeant majors, otoliths have b een seen in the auditory vesi cles in the eggs (Alshuth et al. 1998). Sound can travel long distances underwater, is highly directional, and has little attenuation (Rogers and Cox 1988). Larval fi sh are found tens of meters to hundreds of kilometers from a reef (Leis and McCormic k 2002) but high ambient noise levels from fish and invertebrates near reefs may exceed background noise for tens of kilometers from the source (McCauley 1995). The noise le vels are greatest at night, when snapping shrimp are most active, whic h is the time reef fishes te nd to settle (McCauley 1995). Studies using light traps th at broadcast reef sounds have shown some, but not all, species of larval reef fish to be more attract ed to the Â“noisy trapsÂ” than Â“quiet trapsÂ” (Tolimieri et al. 2000, Leis et al. 2003). In addi tion, the behavior of reef fish larvae differ in response to broadcasted reef sounds rather than broadcasted random sounds (Leis et al. 2002). These studies show that the larval fishes can detect reef sounds and can differentiate between reef sounds and random noise, but the frequency range detected, the
8 distance the sounds can be perceived and th e ability to actually localize the sound all require further study. Auditory brainstem response The auditory brainstem response (ABR) is an electrophysiological technique for measuring hearing thresholds in fishes and other vertebrates (Kenyon et al. 1998). Electrodes placed cutaneously or inserted s ubdermally in proximity to the organismÂ’s brainstem, directly measure nerve impulses created in the eighth nerve and brain in response to sounds (Corwin et al. 1982). Si gnal averaging is used to pull the evoked potential signal out of background noise. The benefits of ABR in comparison to traditional behavioral tests include prompt eval uation of hearing, ability to repeatedly test animals and a minimization of overall physiologi cal strain on the test subjects (Kenyon et al. 1998; Yan and Popper 1991). Purpose The purpose of this study was to measure audiograms for sergeant majors ranging in age from newly settled juveniles to a dults using the audito ry brainstem response technique. Though the reef is one of the noisiest habitats in th e sea, coral reef fishes are not well studied in regards to hearing (Montgom ery et al. 2001). Audiograms have yet to be measured for sergeant majors and few studi es have succeeded in determining hearing thresholds for juvenile fish (Kenyon 1996, Popper 1971, Wysocki and Ladich 2001). The audiograms were used to provide information on three aspects of sergeant major audition. First, the audiograms were analyzed to determine the frequency range each fish could detect and auditory thresholds at each of the frequencies for which there was a response. The audiograms were also exam ined to ascertain whether size is a factor in the hearing sensitivity of the sergeant major. The thresholds were compared to determine how hearing changed as the fish in creased in length. Las tly, the audiograms of the smallest fish were analyzed to determin e whether it is feasible for pelagic larval sergeant majors to use sound as a navigational tool.
9 METHODS Fish acquisition and maintenance Audiograms were measured for 32 sergean t majors ranging in size from 11.1 mm to 121 mm (Table 1). The majority of the fish in this study were ca ught with nets in the Florida Keys by divers on SCUBA. The 12 sm allest fish (<30 mm) were collected from lobster traps set up in water off of Long Ke y, FL. The traps were collected every 1-3 days so the fish collected were likely newly settled. All fish were collected, held for 1-5 days, te sted and either euthanized or released. Fish #1-28 were collected near Long Key a nd were housed in flow-through holding tanks at the Keys Marine Lab (KML) in Long Key, FL. Auditory brainstem response hearing tests (ABRs) were performed on these fish on-s ite at the Keys Marine Lab. The larger fish were collected off of Tavernier, Flor ida Keys, maintained together in a 275 gallon cylindrical tank (S=35, T=26C) at the University of South Florida and fed a few pinches of Tropical Fish Flakes twice a day. The identica l ABR setup, including all instrumentation and test tank, was used to pe rform ABRs on the six largest fish at the Marine Sensory Lab at USF. All procedures were approved by the University of South Florida Institutional Animal Care and Use Committee. Experimental setup Hearing thresholds were determined for each fish using ABR. An individual fish was secured in a harness constructed from Nitex mesh fastened with clamps and suspended from laboratory stands. Harnesses we re custom made for each animal so that movement was restricted while allo wing respiration to occur normally. The apparatus consisted of a PVC pipe ( 1.2 m high, 30 cm in diameter), closed at the bottom, and oriented upright. At KML, th e test tank was set up in a separate room in which only hearing tests were conducted. At USF, the test tank was set up in an
10 Table 1 Standard length and weight for each fish Group 1 (n=12) included fishes less than 30 mm, Group 2 (n=10) included fish es 30-50 mm and Group 3 (n=10) included fishes greater than 50 mm.
11 GROUP 1 GROUP 2 GROUP 3 fish # SL (mm) weight (g) fish # SL (mm) weight (g) fish # SL (mm) weight (g) 1 11.1 0.08 13 31.7 1.73 23 51.4 8.53 2 11.7 0.08 14 34.8 2.35 24 52.6 9.38 3 12.7 0.16 15 37.5 2.37 25 57.4 10.92 4 12.8 0.13 16 37.8 2.82 26 89.6 47.37 5 13.2 0.15 17 38.8 2.68 27 105 63.23 6 19.1 0.76 18 38.9 3.34 28 108 72.5 7 19.8 0.40 19 39.1 3.57 29 115 80.48 8 21.4 0.70 20 46.1 5.73 30 116 75.32 9 21.6 0.69 21 47.3 6.29 31 120 77.34 10 24.0 0.72 22 48.7 6.41 32 121 75.93 11 27.3 1.11 12 28.0 1.23
12 audiology booth. The PVC container was f illed with sea water (S=35, T=26C) to a height of 1.12 m. The fish was suspende d 46 cm below the surface and a speaker was placed at the bottom of the PVC pipe. Subdermal stainless steel needle electr odes (Rochester Electro-Medical) were used for recording the ABR signal. An elect rode was inserted about 1 mm into the head, over the medulla region. The reference electrode was placed within the fishÂ’s dorsal musculature and a ground electrode was placed directly in the water in close proximity to the animal. After a fish was tested it was weighed a nd measured before being returned to the tank. Data from any fish that died or escaped during ABR testing were not used. Sound generation and ABR acquisition Sound stimuli and ABR waveform recordings were produced with a Tucker-Davis Technologies (TDT) ABR workstation (Figure 2) TDT SigGen and BioSig software was used to generate the sound stimuli, with an RP2.1 Enhanced Real-T ime Processor (digital to analog converter), a PA5 Programmable Attenuator and a power amplifier (Hafler Trans.Ana P1000 110 Watt Professional Power Am plifier) before being sent to the TST 229 AQUA (Clark Synthesis) speaker where so und was emitted. Stimuli consisted of 20 ms pulsed tones gated with a Hanning window to improve resolution of peaks. The phase of the tone was alternated be tween presentations to minimize electrical artifacts from the recordings Acoustic stimuli were calibrated with a Reson hydrophone (sensitivity -212 dB V/1 Pa) connected to the RP2. During calibration, the hydrophone was positioned in the experimental setup in place of the fis h, and the sound levels were measured with BioSig, without phase alternation. Eight different frequencies were pr esented during each trial: 100, 200, 400, 800, 1000, 1200, 1400, 1600 Hz. Sound level at each frequency was presented at up to 150 dB and decreased in 6 dB steps until a threshold level was determined. Evoked potentials recorded by the elec trode were fed through the RA16 Medusa Amplifier to the RA16 Medusa Base Station (anal og to digital converter), routed into the
13 Figure 2 Diagram of ABR-recording setup. (RP 2.1 Enhanced Real-Time Processor, PA4 Programmable Attenuator, P1000 110 Watt Professional Power Amplifier, RE: Recording Electrode, RFE: Reference El ectrode, GE: Ground Electrode, RA16 Medusa Amplifier, RA16 Medusa Base Station).
14 P1000 PA5 RP2.1 RE RFE GE speaker RA16 Amp RA16 Base
15 computer and averaged by BioSig software A total of 2000 signal presentations were averaged to measure the evoked respons e at each level of each frequency. Data analysis Hearing thresholds were determined by using power spectra which were calculated with an 8192-point FFT (Fast Four ier Transform) for all ABR waveforms and analyzed for the presence of significant peaks (peaks at twice the frequency of the stimulus that were at least 3 dB above bac kground levels). The FFT converts a signal in the time domain into the frequency domain a llowing the viewer to determine if there is a signal in response to th e frequency tested. Analyses of significant peaks was done using visual inspection, which is the traditional means of determining thresholds in ABR audiometry (Kenyon et al. 1998, Jacobsen 1985, Kileny and Shea 1986, Gorga et al. 1988; Hall, 1992; Song and Schacht 1996). ABR th resholds were defined as the lowest sound level where significant FFT peaks for th e dominant frequency were apparent. Thresholds were determined for each fish at each frequency and were plotted as a linear regression comparing threshold versus standard length. At each frequency, the R values were calculated and sl opes of the regression lines were determined. An ANOVA analyzing the overall goodness of fit of the regression line at each frequency was then performed. The highest frequency tested, 1600 Hz, was not included in the regression analysis due to the small number of fish for which a response was detected (n=6). Due to repetition of statistical test s on the same subjects (ANOVA pe rformed for regression line at seven different frequencies) a Bonferroni correction was us ed to determine significant values. The alpha level was set at 0.05 so ta king the Bonferroni co rrection into account (0.05/7), values were consid ered significant when p<0.007. Thresholds of every fish were averaged to form one audiogram as representative for all 32 fish. Fish were then separated into three size groups (11-30mm, 31-50mm, >50mm), thresholds were averaged for each group and an audiogram was produced for each group.
16 RESULTS ABR waveforms in the time (nV/ms) and frequency domain (FFT) show that as sound level of the stimulus decreased, the amplitude of the ABR waveforms decreased, as well as the ABR voltages measured by th e electrode (Figure 3). The dominant frequency in the power spectra was approxi mately twice the stimulus frequency. Figure 4 shows the ABR output response from one individual fish representative of Group 1 (A) and one fish from Group 3 (B). For both fish, the greatest output resulted from tones played at 100 and 400 Hz. The ABR voltage is slightly gr eater for the larger fish. Thresholds and frequency range detected were determined for all fish (Table 2). Mean thresholds for all fish suggest th e fish are most sensitive at the lower frequencies tested (100, 200, 400 Hz) and requ ire loud sound levels (>140 dB SPL) to detect tones at higher frequencies (800 -1600 Hz) (Figure 5). The most sensitive frequency was 100 Hz (118 dB, SD=10.9). Audiograms were produced for three size gr oups (Figure 6). A ll size groups had the lowest thresholds at the lower frequencies (100-400 Hz) ranging from 112-133 dB. The most sensitive frequency was 100 Hz for all three size groups. Threshold levels greatly increased for every size group at 800 Hz. The frequency step from 400 Hz to 800 Hz caused about a 15 dB change in threshold for all size groups. The degree of variability be tween individual fish was large at each frequency (Figure 7). The R2 values are very low for all of the regressions but the slopes for all frequencies that were detected by more than one size group are positive, indicating threshold levels increase as the fish increa ses in size. The regr ession lines for 100 Hz (p=0.001) and 200 Hz (0.006) were found to be significant, indicating a true effect between size and threshold at these fre quencies (Table 3). For 800 Hz, 1000 Hz and 1200 Hz the p values were much larger (p=0.451, 0.458, 0.530, respectively). The p
17 Figure 3 Example of an ABR response from fish #1 (11.1 mm) when played tone at 800 Hz. A) Response in time domain. B) Response in frequency domain (FFT).
18 13 26 39 52 Time (ms) 1600 2500 3400 4300 Frequency (Hz) nV 147 dB 141 dB 135 dB A dB V 147 dB 141 dB 135 dB B
19 Table 2 Threshold sound levels (dB re 1 Pa) for each fish. NR denotes no response from the fish at the loud est sound level played.
20 GROUP 1 (<30 mm), n=12 fish # SL (mm) 100 200 400 800 1000 1200 1400 1600 1 11.1 106 107 109 129 142 NR NR NR 2 11.7 136 137 121 147 NR NR NR NR 3 12.7 106 113 115 147 NR NR NR NR 4 12.8 112 143 139 147 NR NR NR NR 5 13.2 100 107 109 135 NR NR NR NR 6 19.1 124 125 121 141 NR NR NR NR 7 19.8 112 107 121 141 130 148 NR NR 8 21.4 106 119 115 141 148 NR NR NR 9 21.6 112 107 115 135 148 NR NR NR 10 24 106 107 115 141 NR NR NR NR 11 27.3 112 125 127 129 142 148 NR NR 12 28 112 113 127 147 NR NR NR NR MEAN 112 117 119 140 142 148 --SD 10 13 9 7 7 0 --GROUP 2 (30-50 mm), n=10 fish # SL (mm) 100 200 400 800 1000 1200 1400 1600 13 31.7 112 119 121 147 148 NR NR NR 14 34.8 112 125 109 129 142 NR NR NR 15 37.5 130 149 121 135 NR NR NR NR 16 37.8 118 137 127 135 136 142 NR NR 17 38.8 130 107 109 129 136 148 NR NR 18 38.9 142 137 151 NR NR NR NR NR 19 39.1 112 131 121 141 NR NR NR NR 20 46.1 112 125 127 141 148 NR NR NR 21 47.3 124 113 121 117 130 136 142 NR 22 48.7 118 137 133 141 148 NR NR NR MEAN 121 128 124 135 141 142 142 -SD 10 13 12 9 7 6 --GROUP 3 (>50mm), n=10 fish # SL (mm) 100 200 400 800 1000 1200 1400 1600 23 51.4 106 125 121 141 148 148 148 NR 24 52.6 118 137 115 147 136 148 136 135 25 57.4 112 125 121 141 124 130 136 129 26 89.6 124 125 121 141 136 148 NR NR 27 105 118 149 133 141 148 148 148 147 28 108 124 143 133 141 136 142 148 147 29 115 130 131 139 141 NR NR NR NR 30 116 136 137 133 147 148 148 NR NR 31 120 136 131 121 147 148 148 142 147 32 121 130 131 127 135 148 148 148 147 MEAN 124 133 126 143 141 145 144 142 SD 10 8 8 4 9 6 6 8
21 Figure 4 Example of ABR response from two different individuals. A) fish #8 (21.4 mm, Group 1) B) fish #24 (52.6 mm, Group 3)
22 0 0.001 0.002 0.003 0.004 0.005 0.006 100110120130140150160170 SPL dB re 1 uPaABR V 100 Hz 200 Hz 400 Hz 800 Hz 1000 Hz 0 0.001 0.002 0.003 0.004 0.005 0.006 0.007 0.008 0.009 100110120130140150160170 SPL dB re 1 uPaABR V 100 Hz 200 Hz 400 Hz 800 Hz 1000 Hz 1200 Hz 1400 Hz 1600 Hz A B
23 Figure 5 Average audiogram for all fish at each te sted frequency with standard deviation of thresholds indicated by the bars (n=32).
24 80 90 100 110 120 130 140 150 160 0500100015002000 Frequency (Hz)threshold (dB SPL)
25 Figure 6 Average audiograms for the three size gr oups of fish. The threshold sound level for each fish at each frequency for which there was a response was averaged, within size groups, to determine the threshold level at each frequency.
26 100 110 120 130 140 150 160 0500100015002000 Frequency (Hz)Threshold (dB SPL) <30 mm, n=12 30-50 mm, n=10 > 50 mm, n=10
27 Figure 7 (A-G) Regression lines for frequencies test ed with equations and r values indicated. 1600 Hz is not pictured due to the small percentage of fish from which a response was detected.
28 100 Hzy = 0.1645x + 110.44 R2 = 0.2936 80 90 100 110 120 130 140 150 160 020406080100120140 Standard Length (mm)dB SPL 200 Hzy = 0.1693x + 117.14 R2 = 0.219 80 90 100 110 120 130 140 150 160 020406080100120140 Standard Length (mm)dB SPL 400 Hzy = 0.1093x + 117.34 R2 = 0.163680 90 100 110 120 130 140 150 160 020406080100120140 Standard Length (mm)dB SPL A B C
29 800 Hzy = 0.0274x + 138.4 R2 = 0.0197 80 90 100 110 120 130 140 150 160 020406080100120140 Standard Length (mm)dB SPL 1000 Hzy = 0.0371x + 138.92 R2 = 0.03280 90 100 110 120 130 140 150 160 020406080100120140 Standard Length (mm)dB SPL 1200 Hzy = 0.0279x + 142.81 R2 = 0.0337 80 90 100 110 120 130 140 150 160 020406080100120140 Standard Length (mm)dB SPL 1400 Hzy = 0.0847x + 136.28 R2 = 0.2816 80 90 100 110 120 130 140 150 160 020406080100120140 Standard Length (mm)dB SPL D E F G
30 Table 3 Results of ANOVA analyzing the ove rall goodness of fit on the regression equation at each frequency. Responses at 1600 Hz were too few for statistical analysis (df=4). Bonferroni correction was used to determine significant values (0.05/7). Asterisks denote values considered significantly different (P<0.007).
31 Sums of Squares Degrees of freedom Mean Squares P value 100 Hz Regression 1083 1 1083 0.001* Residual 2605 30 86.83 Total 3688 200Hz Regression 1147 1 1147 0.006* Residual 4090 30 136.4 Total 5238 400 Hz Regression 478.4 1 478.4 0.022 Residual 2445 30 81.52 Total 2924 800 Hz Regression 30.07 1 30.07 0.451 Residual 1496 29 51.58 Total 1526 1000 Hz Regression 36.65 1 36.65 0.438 Residual 1107 19 58.30 Total 1144 1200 Hz Regression 13.96 1 13.96 0.530 Residual 400.0 12 33.34 Total 414.0 1400 Hz Regression 55.75 1 55.75 0.176 Residual 142.25 6 23.71 Total 198.0
32 value at 1400 Hz was lower (p=0.176), though no t significant. It s hould be noted that there was a small number of responses detected at this frequency (df=7). All of these fish except for one (SL=47.3 mm) are in the la rgest size group. Eff ects of size on hearing ability are much more apparent at th e lower frequencies tested (100-400 Hz). In addition, the frequency ra nge detected by the fish a ppears to be a function of the fish length (Figure 8). The larger fi sh were more likely to respond to higher frequency sounds. Almost 100% of the test ed fish responded to sound presented from 100-800 Hz. As the frequency increased above 8 00 Hz, the number of smaller fish with a response decreased. A two-sided difference te st looking at the diffe rence of proportions of responses detected between the three gr oups at each frequency at the highest sound level, found significant differences at frequencies 1000-1600 Hz (Table 4).
33 Figure 8 Percent of individuals in each size gr oup that detected sound at each frequency at the maximum sound level that could be generated.
34 0 20 40 60 80 100 120 1002004008001000120014001600 Frequency (Hz)% individuals detecting sound <30 mm, n=12 31-50 mm, n=10 >50 mm, n=10
35 Table 4 Two-sided difference test for significance of differe nces between proportion of responses at the loudest sound le vel presented at each frequenc y. Asterisk denotes values considered significantly different (p<0.05).
36 1000 Hz1200 Hz 1400 Hz 1600 Hz Groups 1 and 2 0.204 0.478 0.350 Groups 2 and 3 0.278 0.014* 0.014* 0.010* Groups 1 and 3 0.030* 0.003* 0.003* 0.006*
37 DISCUSSION Auditory sensitivity of sergeant majors The data from this study suggests that sergeant majors hear lower frequencies (100-400 Hz) at lower threshol ds than higher frequencies (800-1600 Hz). Overall, however, they have poor and would be classified as hearing generalists (Figure 9 and 10). This classification is expected because, though this fish has a swim bladder, a connection between the bladder and the auditory endorgans or any other accessory auditory structure has not been found in Pomacentrids (Myrberg 1980, Myrberg et al. 1986). Figure 9 shows the sergeant major audiogram with a udiograms of two known hearing specialists, the goldfish (Cyprinidae, Carassius auratus ) and the soldierfish (Holocentridae, Myripristis kuntee ). The most sensitive frequency for the sergeant major is 100 Hz at 118 dB while the thresholds for the most sensit ive frequency for the soldierfish is 50 dB (1000-1600 Hz), and 51 dB for the goldfish (500 Hz) The frequency range detected is greater in the hearing sp ecialists as well. Only two reef fish families, Chaetodontidae and Holocentridae, have been classified as hearing speciali sts. Chaetodontids have a late rophysic connection, a unique connection between the lateral line and the ea r, which is believed to enhance hearing sensitivity (Webb 1998; Webb and Smith 2000) The Holocentridae have a range of morphologies from close apposition of the ante rior end of the swim bladder to the inner ear to essentially no apparent specializat ion (Coombs and Popper 1979). Perhaps in a loud environment such as a reef, hearing spec ializations are not as necessary, or hearing specializations in untested families are yet to be found. Figure 10 shows the audiogram produced fo r the sergeant majors along with other teleost species of the coral reef classified as hearing generalists. ABR is considered more conservative in judging threshold levels than the classical behavioral approaches that
38 Figure 9 Sergeant major audiogram with audiogra ms of two hearing specialists, the goldfish ( Carassius auratus ) (Popper, 1971) and soldierfish ( Myripristis kuntee ) (Coombs and Popper 1971).
39 0 20 40 60 80 100 120 140 160 100100010000 Frequency (Hz)SPL (dB re 1 uPa) sergeant major (A. saxatilis) soldierfish (M. kuntee) (Coombs and Popper 1979) goldfish (C. auratus) (Popper 1971)
40 Figure 10 Audiogram of sergeant majors (thr esholds adjusted for comparisons by subtracting 10 dB) and other hearing generalists St: average for 6 species of Stegastes (Myrberg and Spires 1980); Av: Adioryx vexillarius Hs: Haemulon sciurus Eg: Epinephalus guttatus Tb: Thalassoma bifasciatum La: Lutjanus apodus (Tavolga and Wodinsky 1963).
41 60 70 80 90 100 110 120 130 140 150 100100010000 Frequency (Hz)dB re 1 uPa A saxatilis St Av Hs Eg Tb La
42 were used to create the other audiograms (K enyon et al. 1998). ABR is a relatively new technique for determining thresholds fo r fish, however, work with humans has demonstrated that auditory thresholds dete rmined using tone-burst ABR are generally higher, by 10-20 dB, than those obtained us ing behavioral methods (Kenyon et al. 1998, Gorga et al. 1988). Kenyon et al. (1998) found no significant difference in thresholds determined using ABR and those found with be havioral methods but they did see that ABR thresholds were generally higher than behavioral values below 1500 Hz. This was taken into account for comparisons in this study by subtracting 10 dB from the threshold levels found for the sergeant major. Thresholds for the sergeant majors are still comparatively high. The adjusted threshold at the most sensitivity frequency of the sergeant major (108 dB, 100 Hz) is above the majority of the other speciesÂ’ most sensitive frequency ( Stegastes [average of six species]: 82 dB, 500 Hz; Adioryx vexillarius : 90 dB, 600 Hz; Haemulon sciurus : 80 dB, 100 Hz; Epinephalus guttatus : 90 dB, 200Hz; Thalassoma bifasciatum : 107 dB, 500 Hz; Lutjanus apodus : 110 dB, 300 Hz). It should be noted that the frequency range detected by the sergeant major is greater than for most of the other species represented, however, this is most likely a limitation of the maximum sound levels used in the other studies. In comparison to other Pomacentrids, se rgeant major thresholds are especially high. Despite no known auditory specializatio ns, many Pomacentrid species have been found to have low hearing thresholds. Serg eant majors are not as vocal as many other members of the Pomacentrid family. This sp ecies is also not as territorial as other damselfish such as the bicolor damselfish ( Stegastes partitus ) (Myrberg 1997). Territoriality, a behavior which incor porates the production of sound in many Pomacentrid species (Myrberg 1997), is usua lly seen in sergeant majors only during reproductive periods (Fishelson 1970 ). Sergeant majors are like ly to be soniferous during the spawning season, not only due to the territorial behavior that occurs, but even more because they have been observed performing the Â“signal jumpÂ” during courtship (Prappas et al. 1991). The male courtship display in damselfishes is known as the signal jump, where the male rises in the water column and swims down rapidly while producing a
43 pulsed sound anywhere in the frequency ra nge of 100-1000 Hz, depending on the species (Fishelson 1964, Myrberg 1972, Spanier 1979, Lobel and Mann 1995). It is thought that Abudefduf sordidus do not rely on sound during courtship as much as other Pomacentrids such as Dascyllus and Stegastes spp. (Lobel and Kerr 1999). The courtship associated sounds produced by the A. sordidus were highly variable in pulse and number length, making courtship s ounds appear rudimentary in comparison to the repetitious acoustic display exhibited by other Pomacentrids. Studies on the propagation of fish sounds in shallow water suggest that they attenuate greatly over short distance s (Mann and Lobel 1997, Rogers and Cox 1988, Forrest et al. 1993). Due to attenuation and background noise levels, damselfish sounds are thought to be used over distance less than 11-12 m (Mann and Lobel 1997). The study on A. sordidus found that courtship associated be havior of this species could be detected only within two meters (Lobel and Kerr 1999). The threshold levels of this species are probably low enough to detect sounds produced by neighboring fish. The most sensitive frequency for Stegastes is 500 Hz (Figure 10). This has adaptive value because the peak frequency of Stegastes sound falls within a frequency band centered around 500 Hz (Kenyon 1996). The most sensitive frequency for the sergeant major was determined to be 100 Hz. Further study on sergeant major sound production could determine whether the most sensitive frequency of 100 Hz has adaptive significance as well. A study on Abudefduf luridus found the most common frequencies produced to be in the frequency range of 50-500 Hz, with most of the energy concentrated on the lower frequencies (Santia go and Castro 1997). This data correlates well with what was found in this study. Effect of size on the auditory sensitivity of the sergeant major Size of fish significantly affected auditory sensitivity and the frequency range the fish was able to detect. At the lower fre quencies (100-200 Hz), thresholds increased with an increase in length. In a ddition, the larger fish more readily responded to the higher frequencies (800-1600 Hz) at a significant leve l (Table 4). However, the most sensitive frequency was 100 Hz for each size group.
44 These results are in contrast to a study on the bicolour damselfish ( Stegastes partitus ), which found an exponential decrease in threshold with an increase in size (Kenyon 1996) (Figure 11). All size groups re sponded to sounds in the frequency range of 300-1500 Hz. The most sensitive frequency at 500 Hz remained the same in each size group. Popper (1971) found that there was no di fference in audition between two size groups of goldfish ( Carassius auratus ). Kenyon (1996) notes th at the difference in results may be due to the physiological di fferences between th e hearing specialist goldfish and the hearing genera list bicolour damselfish ( S. partitus ). Another explanation may be that only two size groups of gol dfish were tested (Kenyon 1996, Wysochi and Ladich 2001). It is important to note that the met hods used by Kenyon, Popper and in this study are different. Rather than the auditory brainstem response technique, Kenyon (1996) conducted psychophysical experiments where the fishes were trained to associate sound with an electric shock. The fish eventually showed an avoidance re sponse with the onset of detectable sound. Popper (1971) used a similar behavioral shock conditioning technique where the fish were shocked if they did not swim across a ba rrier in response to a sound. As mentioned previously, ABR is thought to be more conservative in determining thresholds (Kenyon 1998). It is interesting that at the lower frequenc ies, where length appeared to be a factor in the sound level the fish was able to detect threshold levels actua lly increased with an increase in size. The closer proximity of the swim bladder to the otolith in a smaller fish may allow the bladder to act as a more pronoun ced amplifier at certain frequencies. The lower thresholds exhibited by the juve nile fish could al so be explained by variation in electrode placement. In the attempt to be as consistent as possible with every trial, the electrode was placed about 1 mm belo w the surface of the head of every fish. In smaller fish, therefore, the electrode was likel y closer to the brain which may have caused a larger ABR response in some of the sm aller fish compared to larger fish.
45 Figure 11 Average audiograms of four size gr oups of juveniles (12-14 mm: n=5, 1517mm: n=6, 19-25 mm: n=7, 2938 mm: n=6) and adult Stegastes partitus (n=8). (Redrawn from Kenyon 1996).
46 60 70 80 90 100 110 120 130 140 150 100100010000 Frequency (hz)Sound pressure level (dB re 1 uPa) 12-14 mm 15-17 mm 19-25 mm 29-38 mm >45 mm
47 Though Kenyon (1996) and Popper (1971) did no t see an increase in frequency range with an increase in size, a similar re sult was found in an ontogenetic study utilizing ABR on the audition of the hear ing specialist croaking gourami ( Trichopsis vittata ) (Wysocki and Ladich 2001). The increased fr equency range for this species, however, is attributed to the resonance of the air-filled suprabranchial chamber (SBC), which is utilized for air breathing and also acts as an accessory hearing structure. The sergeant major has no known accessory hearing structure. The role of the swim bladder in hearing generalists is not well understood, but due to the fact that there is no connection between the swim bladder and the otolith in the sergeant major, it is unlikely that any increase in size of the swimbladder would affect audition (Yan et al. 2000). Hearing generalists rely on particle ve locity detected by sensory hair cells (kinocilium and stereocilia) on the otolith. Kinocilium length generally corresponds with the different frequencies at which hair cel ls are stimulated by incoming sound (Platt and Popper 1984). Regions of the sensory epitheliu m in goldfish with l onger kinocilia were considered responsive to lower frequencie s while those with shorter kinocilia were considered responsive to higher frequencies. Perhaps the ongoing a ddition of the sensory hair cells on the sensory epithelium of the ot olithic organs and the exact placement of the new sensory hair bundles plays a role in the frequencies the fish can detect. It should be noted that the differences in hearing range could partly be a reflection of the maximum sound level that could be generated by the system. Perhaps if the speaker were capable of broadcasting the hi gher frequencies at a higher sound level, a response could have been detected from the smaller fish at the higher frequencies. Use of reef sound as a navigational cue for pelagic larvae In order for larvae to detect a reef from a few kilometers away they need to hear reef sound levels below approximately 80 dB re: 1 Pa/ Hz, and to recognize these sounds against broadband background noise (Mont gomery et al. 2001). The majority of sound from the reef is the result of snapping shrimp (Leis et al. 2002) and fish choruses (Cato 1992). The data in this study suggest that it is highly unlikely that larval sergeant majors use coral reef noise as a navigational cue (Figure 12).
48 To determine the possibility of the use of sound as a navigational cue, it is important to consider the distance a fish mu st be able to detect the sound. McCauley (1997) measured the chorus spectra from ins hore fish calling that were recorded at 4.3 km from Feather Reef (Queensland, Australia) This spectra is redrawn in figure 12 in comparison to the audiogram of smallest size group of sergeant majors. Recordings done off of Lizard Island, Great Barrier Reef show th at at 75 meters from the reef some of the clicks of the broadcast sounds are evident a bove the background noise (Leis et al. 2002). At 160 meters, only an occasional click is evident above the background noise. Pelagic larvae are tens to hundreds of kilometers fr om the reef (Cowen 2002). Judging from the data collected from this study in comparison to recorded reef sounds from other sources, it is more likely the fish use sound and their a uditory sense in closerange orientation. Though KenyonÂ’s (1996) study on the hearing of damselfish juveniles supports the idea that the larvae cannot use sound as a navigational cue, studies have been conducted which suggest a role for audition in settlement (Tolimieri et al. 2000, Leis et al. 2002, 2003) Tolimieri and Leis both found th at significantly more larvae of specific taxa (Tolimieri: Tripterygiidae, Leis: Apogoni dae, Mullidae, Pomacentridae, Serranidae and Sphyraenidae) were caught in light traps broadcasting re corded reef sounds than in Â“silentÂ” traps (Tolimieri et al. 2000, Leis et al. 2003). It is important to consider, however, that the distance the larvae were coming from was unknown. All of the trapped larvae could have arrived from within a few meters, where particle velocities would be higher than at greater distances. McCauleyÂ’s (1997) data on a reef chorus measured 4.3 km from the Great Barrier Reef was used to determine the farthest distan ce from which juvenile sergeant majors can detect reef sounds (Figure 13). The equati on Transmission Lost= 20 log r (r=distance), which assumes spherical spreaduing, was used to determine that 6 dB was lost per distance doubled. The farthest distance from which sergeant major juveniles can detect a reef chorus was approximated to be about .54 km from the reef. The farthest distance from which a juvenile fish in this study would be able to detect this reef chorus is 2.15 km.
49 Another consideration when studying sound detection as a navigational cue is the ability of the fish to localize the sound. This study looked only at th e ability of the fish to detect certain frequencies and sound levels Even if the fish do have the ability to detect the sound from the pelagic environmen t, they must be able to determine the direction the sound is coming from. A study on the black axil chromis ( Chromis atripectoralis ) was conducted in which reefs sounds were broadcasted to larv ae released 50-100m from the speaker (Leis et al. 2002). It was found that the larval fish responded to the sound and had a different behavioral response when played random noise as opposed to recorded reef sounds. The response of the larval fish to reef sounds however, was to swim faster and in random directions so there was no i ndication that the fish had the ability to local ize the sound. Due to the high speed of sound in water and the small distance between the ears of fish, fish are presumably unable to use in teraural time, phase and intensity differences of sound pressure to localize a sound source li ke vertebrates in air (Fay and Feng 1987). However, localization has been demonstr ated in some adult fishes including Pomacentrids (Myrberg and Spires 1980, Myrber g et al. 1986). Different ciliary bundle types are found in different regions of the m aculae (Myrberg and Fuiman 2002). There is evidence for the ability of a fish to determ ine the direction of a sound source depending on which part of the macula is most stimulated (Platt and Popper 1984, Enger 1976, Popper 1977, Lu and Popper 1998). The theoreti cal problem with th is is that a 180 ambiguity results from the oscillations of th e hair cells. In order to use directional information from reef sound, larvae need to be able to hear direc tionally and resolve the 180 ambiguity. Although sound may be unlikely as the sole navigational cue for the reef fish larvae, it is probable that the return to the reef from the pelagic environment is not by chance (Leis and McCormick, 2002). Most pelagi c larvae do not settle on the first reef they come upon. If reefs were difficult to lo cate from open water, most larvae would not leave a reef once it was found (Leis and Ca rson-Ewart 1998, 1999). In addition, it has been observed that settlement stage larvae in open water tend to have roughly linear
50 Figure 12 Audiogram of smallest group size grou p of sergeant majors (11-30mm), with threshold range at each of th e tested frequencies indicated by the bars, in comparison to spectrum level ambient reef noise; morning and night chor uses (Cato 1980), snapping shrimp (Tavolga 1974), chorus spectra from inshore reef calling r ecorded 4.3 km from Feather Reef (Queensland, Australia) (redra wn from McCauley 1997). Sergeant major audiogram was altered for comparison purposes, taking into account a frequency bandwidth of 10% of test frequency.
51 50 60 70 80 90 100 110 120 130 140 150 100100010000 Frequency (Hz)dB SPL snapping shrimp morning chorus night chorus sergeant major (<30mm) chorus spectra (4.3 km from reef)
52 Figure 13 Sounds of a reef chorus spectra measured 4.3 km from the reef (McCauley 1997) and the audiogram of Gr oup 1 (<30mm) were used to approximate the distance the juvenile fish in the study would be able to detect the sounds. Sergeant major audiogram was altered for comparison purposes, taking into account a frequency bandwidth of 10% of test frequency. The bars indicate threshol d range for the 12 fish in the smallest size group of sergeant majors. The equation Tran smission Lost=20 log r (r=distance), which assumes spherical spreading, was used to de termine that 6 dB is lost per distance doubled.
53 50 60 70 80 90 100 110 120 130 140 100100010000 Frequency (Hz)dB SPL4.3 km 0.27 km 0.54 km 2.15 km 1.08 km
54 horizontal trajectories, or at least trajectorie s that are significantly different from random (Leis et al 1996, Leis and Carson-Ewart 1999, 2001). Armsworth (2000) modeled the swimming responses of weak and strong swimming larvae reacting to current dependent and current independent cues emanating from a reef. Current dependent cues incl ude odor, while current independent would include sound. He concluded that larval swimming and sens ory abilities are far more important than circulation in determining enhancement of settlement. Rather than audition and sound from the reef being the sole cue for pelagic larval fish to follow, it is more likely that the larvae use a combination of their sensory organs to navigate their return to the reef. Vision is obviously im portant over small scales (5-15 m) (Leis and Carson-Ewart 1998, 1999, 2002) and olfaction has been shown to play a role in some Pomacentrid and Apogonid species at distances of a few to tens of meters (Sweatman 1988, Elliot et al. 1995, Arvedlund et al. 1999, Atem a et al. 2002). Other possible cues include differences in wind or wave induced turbulence, gradients in abundance of fish, plankton or reef detritus and di fferences in temperature of lagoonal or reef flat water flowing from a reef (Leis and McCormick 2002) Cues may vary with ontogeny and with distance from the reef. The use of cues is most likely different among species depending on their sensory abilities. Though most likely quite complex, unders tanding the naviga tional process of larval reef fish on their return to the reef is important for the ma ny groups exploiting reef fish populations and for managing reef fisherie s. The coral reef is a highly exploited ecosystem and an understanding of the supply rate of larvae to coral r eefs is critical in determining the structure of coral reef populations (Doherty and Williams 1988, Doherty and Fowler 1994, Armsworth 2000). Studies such as this one which provide information on the sensory capabilities of larval reef fi sh offer insight on popul ation connectivity and are important to management decisions such as the design of marine protected areas. Future directions This study provided important hearing data on one of the less vocal and less studied, though very abundant, damselfishes The differences between the hearing
55 abilities of this particular Pomacentrid specie s in comparison to other species in the same family are interesting due to the large variation. More audiograms should be produced for other Pomacentrids and more Abudefduf species to see if a si milar pattern of higher thresholds with a greater frequency range exis ts for large versus small fish. The higher sensitivity to low frequency sounds of juvenile s in comparison to the adults is different than what has been found in previous ont ogenetic studies. Th e production of more audiograms for coral reef fish in general is important because the reef is such a noisy habitat and a highly exploited environment. Determining the frequency range and sound levels a newly settled juvenile fish can detect is the first step in understanding the use of audition as a guide for larval fish. The auditory brainstem respons e technique is a rapid method for determining the hearing abilities of a fish. A better understandi ng of how ABR threshol ds are related to behavioral thresholds would be beneficial. In addition, other factors must be considered such as the distance from which the fish must be able to detect the sounds and the ability of a larval fish to localize the sound. More measurements of reef sounds as a function of distance from the reef, particularly in the Fl orida Keys where the fish in this study were collected, would be beneficial for comparisons with fish hearing ability. In addition, more recordings need to focus on lower fre quency sounds from the reef such as breaking waves. According to the data from this study it is not likely that reef noise is used by pelagic larval sergeant majors as the only source for navigating their way to a reef for settlement. Further study on auditory capabilit ies of newly settled juveniles, or possibly larval fish, of many more sp ecies is important in fully understanding ontogenetic changes in audition and the possibility of using sound as a guide.
56 LITERATURE CITED Allen GR (1975) Damselfishes of the South S eas. TFH Publications, Neptune City, New Jersey Allen GR (1991) Damselfishes of the wo rld. Mergus Publishers, Melle, Germany Alshuth SR, Tucker JW, Hateley J (1998) E gg and larval development of laboratoryreared sergeant major, Abudefduf saxatilis (Pisces, Pomacentridae). Bull Mar Sci 62(1): 121-133 Armsworth PR (2000) Modelling the swimming res ponse of late stage larval reef fish to different stimuli. Mar Ecol Prog Ser 195: 231-247 Arvedlund M, McCormick MI, Fautina DG, B ildsoe M (1999) Host recognition and possible imprinting in the anemonefish Amphiprion melanopus (Pisces: Pomacentridae) Mar Ecol Prog Ser 188: 207-218 Atema J, Kingsford MJ, Gerlach G (2002) Larval reef fish could use odour for detection, retention and orientation to reefs. Mar Ecol Prog Ser 241: 151-160 Cato DH (1978) Marine biological choruses in tropical waters near Australia. J Acoust. Soc. Am 64: 736-743 Cato DH (1980) Some unusual sounds of appa rent biological origin responsible for sustained noise in the Timor Sea. J Acoust Soc Am 68: 1056-1060 Cato DH (1992) The biological contribution to the ambient noise in waters near Australia. Acoust Aust 20: 76-80 Chen KC, Mok HK (1988) Sound production in the anenomefishes, Amphiprion clarki and A. frenatus (Pomacentridae), in captivity. Japan J. Ichtyol 35: 90-97 Colson DJ, Patek SN, Brainerd EL, Lewis SM (1998) Sound production during feeding in Hippocampus seahorses (Syngnathidae). Envi ron Biol Fishes 51: 221-229 Coombs S, Popper AN (1979) Hearing differe nces among Hawaiian squirrelfish (family Holocentridae) related to differences in the peripheral auditory system. J Comp Physiol A 132: 203-207
57 Corwin JT (1983) Postembryonic growth of the macula neglecta auditory detector in the ray, Raja clavata : Continual increases in hair ce ll number, neural convergence, and physiological sensitivity. J Comp Neurol 217: 345-356 Corwin JT, Bullock TH, Schweitzer J (1982) Th e auditory brainstem response in five vertebrate classes. Electroen cephalogr Clin Neurophysiol 54: 629-641 Cowen RK (2002) Larval dispersal and rete ntion and consequences for population connectivity. In: Sale PF (ed) Coral r eef fishes: dynamics and diversity in a complex ecosystem. Academic Press, San Diego, p 149-170 Doherty PJ, Fowler AJ (1994) An empirical test of recruitmen t limitation in a coral reef fish. Science 263: 935-939 Doherty PJ, Williams D (1988) The replenishm ent of coral reef fish populations. Oceanogr Mar Biol 26: 487-551 Elliott JK, Elliott JM, Mariscal RN (1995) Ho st selection, location and association behaviors of anenomefishes in field sett lement experiements. Mar Biol 122: 370390 Enger PS (1976) On the orientation of hair cells in the labyrinth of the perch ( Perca fluviatilus ). In: Schuijf A, Hawkins AD (eds ) Â“Sound reception in fishÂ” Elsevier Scientific Publishing, New York, p49-61 Fay RR, Feng AS (1987) Directional heari ng among nonmammalian vertebrates. In: Yost WA, Gourevitch G (eds) Directi onal Hearing, Springer-Verlag, New York New York, p 179-213 Fishelson L (1964) Observation on the biology an d behavior of Red Sea coral fishes. Contributions to the knowledge of the Red Sea. 30: 11-26 Fishelson L (1970) Behaviour and ecology of a population of Abudefduf saxatilis (Pomacentridae, Teleosttei) at Eilat (Red Sea). Anim Behav 18: 225-237 Forrest TG, Miller GL, Za gar JR (1993) Sound propagation in shallow water: implications for acoustic communication by aquatic animals. Bioacoustics 4: 259-270 Foster SA (1987) Diel and l unar patterns of repr oduction in the Caribbean and Pacific sergeant major damselfishes Abudefduf saxatilis and A. troschelii Mar Biol 95: 333-343 Frisch K von (1936) ber den Gehrsinn der Fische. Biol Rev 11:210-246
58 Gorga MP, Kaminski JR, Beauchaine KA, Jesteadt W (1988) Auditory brainstem response to tone bursts in normally hear ing subjects. J Speech Hear Res 31: 8797 Golubeva TB, Tikhonov AV (1985) The voice and h earing of birds in ontogeny. In: Acta XVII Internat Orthinol C ongr Nauka, Moscow, pp 259-274 Gray L, Rubel EW (1985) Development of aud itory thresholds and frequency difference limens in chickens. In: Gottlieb G, Krosnegor N (eds) Measurement of audition and vision in the first year of postnat al life. Ablex Publ Corp, Norwood, NJ, pp145-165 Higgs DM, Rollo AK, Souza MJ, Popper AN ( 2003) Development of form and function in peripheral auditory stru ctures of the zebrafish ( Danio rerio ). J Acoust Soc Am 113(2): 1145-54 Kenyon TN (1994) The significance of sound in terception to males of the bicolor damselfish, Pomacentrus partitus during courtship. Environ Biol Fishes 40: 391405 Kenyon TN (1996) Ontogenetic ch anges in the auditory sensitivity of damselfishes (Pomacentridae). J Comp Physiol A 179: 553-561 Kenyon TN, Ladich F, Yan HY (1998) A compara tive study of hearing ability in fishes: the auditory brainstem response appr oach. J Comp Physiol 182: 307-318 Kingsford MJ, Leis JM, Shanks A, Lindeman KC, Morgan, SG, Pineda J (2002) Sensory environments, larval abilities and local self-recruitment. Bull Mar Sci 70(1): 309340 Lanford PJ, Presson JC, Popper AN (1996) Cell pr oliferation and hair cell addition in the ear of the goldfish, Carrassius auratus Hear Res 100: 1-9 Lecanuet JP, Schaal B (1996) Fetal sensory co mpetencies. Eur J Obstet Gynecol Reprod Biol 68: 1-23 Leis JM (1991) The pelagic stage of reef fishes: The larval bi ology of coral reef fishes. In: Sale PF (ed) The ecology of fishes on co ral reefs. Academic Press, San Diego, CA, p193-230 Leis JM, Carson-Ewart BM (1998) Complex behavior by coral-r eef fish larvae in openwater and near-reef pelagic environmen ts. Environ Biol Fishes 53: 259-266
59 Leis JM, Carson-Ewart BM (1999) In situ swimming and settlement behavior of larvae of an Indo-Pacific coral-reef fish, th e Coral Trout (Pisces, Serranidae, Plectropomus leopardus) Leis JM, Carson-Ewart BM (2001) Behavior of pelagic larvae of four coral-reef fish species in the ocean and an atol l lagoon. Coral Reefs 19: 247257 Leis JM, Carson-Ewart, BM (2002) In situ settlement behavior of damselfish larvae (Pisces: Pomacentridae). J Fish Biol 61 : 325-346 Leis JM, McCormick MI (2002) The Biology, beha vior and ecology of the pelagic, larval stage of coral reef fishes. In: Sale PF (ed) Coral reef fishes: dynamics and diversity in a complex ecosystem. A cademic Press, San Diego, CA, p171-198 Leis JM, Carson-Ewart BM, Cato DH (2002) Sound detection in situ by the larvae of a coral-reef damselfish (Pomacentridae ). Mar Ecol Prog Ser 232: 259-268 Leis JM, Sweatman HPA, Reader SE (1996) What the pelagic stages of coral reef fishes are doing out in the blue water: daytime field observations of larval behavioral capabilities. Mar Freshwater Res 47: 401-411 Leis JM, Carson-Ewart BM, Hay AC, Cato DH (2003) Coral-reef sounds enable nocturnal navigation by some reef-fish larvae in some places and at some times. J Fish Biol 63: 724-737 Lobel JM, Kerr LM (1999) Courtship sounds of the Pacific damselfish, Abudefduf sordidus (Pomacentridae). Biol Bull 197: 242-244. Lobel PS, Mann DA (1995) Spawning sounds of the domino damselfish ( Dascyllus albisella (Pomacentridae), and the relationship to male size. Bioacoustics 6: 187198 Lombarte A, Popper AN (1994) Quantitative anal yses of postembryonic hair cell addition in the otolithic endorgans of the inner ear of the European hake, Merluccius merluccius (Gadiformes, Teleostei). J Comp Neurol 345: 419-428 Lu Z, Popper AN (1998) Morphological polarizati ons of sensory hair cells in the three otolithic organs of a teleost fish: fluore scent imaging of cili ary bundles. Hear Res 126: 47-57 Mann DA, Lobel PS (1997) Propagation of damsel fish (Pomacentridae) courtship sounds. J Acoust Soc Am 101(6): 3783-3791
60 McCauley RD (1994) Distributions and levels of snapping shrimp noise in northern Australia. Report for the Defen ce Science and Technology Organization McCauley RD (1995) Aspects of marine biologi cal sound in Northern Au stralia III: reef associated fish choruses. Report for the Defence Science and Technology Organization McCauley RD (1997) Aspects of marine biol ogical sound in northern Australia IV: reef associated fish choruses. Defen ce Science and Technology Organization McFadden SL, Walsh EJ, McGee J (1996) Onset and development of auditory brainstem response in the Mongolian gerbil ( Meriones unguiculatus ). Hear Res 100: 68-79 Montgomery JC, Tolimieri N, Haine OS (2001) Active habitat selection by presettlement reef fishes. Fish and Fisheries 2: 261-277 Myrberg AA (1972) Ethology of the bicolor damselfish, Eupomacentrus partitus (Pisces: Pomacentridae): a comparative analysis of laboratory and field behavior. Anim Behav Monogr 5: 199-283 Myrberg AA (1997) Sound production by a coral reef fish ( Pomacentrus partitus ): evidence for a vocal, territorial Â“keepoutÂ” signal. Bull Mar Sci 60(3): 1017-1025 Myrberg AA, Fuiman LA (2002) The sensory world of coral reef fishes. In: Sale PF (ed) Coral reef fishes: dynamics and diversity in a complex ecosystem. Academic Press, San Diego, p 123-148 Myrberg AA, Spires JY (1980) Hearing in Dams elfishes: An analysis of Signal Detection among closely related species. J Comp Physiol 140: 135-144 Myrberg AA, Ha SJ, Shamblott MJ (1993) The sounds of bicolor damselfish ( Pomacentrus partitus) : predictors of body size and a spectral basis fpr individual recognition and assessment. J Acoust Soc Amer 94: 3067-3070 Myrberg AA, Kramer E, Heinecke P (1965) Sound production by three cichlid fishes. Science 149: 555-558 Myrberg AA, Mohler M, Catala J (1986) Sound production by ma les of a coral reef fish ( Pomacentrus partitus) : Its significance to females. Anim Behav 34: 923-933 Platt C, Popper AN (1984) Variation in length s of ciliary bundles on hair cells along the macula of the sacculus in two species of teleost fishes. Scanning Electron Microsc 4: 1915-1924
61 Popper AN (1971) The effects of size on auditory capacities of the goldfish. J Aud Res 11: 239-247 Popper AN (1977) A scanning electron microscopi c study of the sacculus and lagena in the ears of 15 teleost fishes. J Morphol 153: 397-417 Popper AN, Fay RR (1997) Evolution of the ear and hearing: issues and questions. Brain, Behav Evol 50:213-221 Popper AN, Fay RR (1999) The auditory periphery in fishes. In: Fay RR, Popper AN (eds) Comparative Hearing: Fish and Amphibians. Springer, New York, New York Popper AN, Hoxter B (1981) The fine structures of the sacculus and lagena of a teleost fish. Hear Res 5: 245-263 Popper AN, Platt C (1993) Inne r ear and lateral line. In: Evans DH (ed) The Physiology of Fishes, CRC Press, Boca Raton, FL p99-136 Prappas JM, Greene LE, White R (1991) Reproductive behavior of the sergeant major, Abudefduf saxatilis within a closed system aquari um. Environ Biol Fish 31: 3340 Reimer K (1996) Ontogeny of hearing in the marsupial, Monodelphis domestica as revealed by brainstem auditory evoked potentials. Hear Res 92: 143-150 Rogers PH, Cox M (1988) Underwater sound as a biological stimulus. In: Atema J, Fay RR, Popper AN, Tavolga WN (eds) Sensory biology of aquatic animals. Springer-Verlag, New York, New York, p131-149 Rbsamen R (1992) Postnatal development of cen tral auditory frequency maps. J Comp Physiol A 170: 129-143 Santiago JA, Castro JJ (1997) Acoustic behavior of Abudefduf luridus J Fish Biol 51: 952-959 Spanier (1979) Aspects of species recognition by sound in four species of Damselfishes, Genus Eupomacentrus (Pisces: Pomacentridae). Z. Tierpsychol 51: 301-316 Stipetic E (1939) ber das Gehrorgan der Mormyriden. Z Vergl Physiol 26: 740-752 Stobutzki IC, Bellwood DR (1997) Sustained swimming abilities of the late pelagic stages of coral reef fishes. Mar Ecol Prog Ser 149: 35-41 Stobutzki IC, Bellwood DR (1998) Nocturnal orie ntation to reefs by la te pelagic stage coral reef fishes. Coral Reefs 17: 103-110
62 Sweatman H (1988) Field evidence that settling coral reef fish larvae detect resident fishes using dissolved chemical cues J Exp Mar Biol Ecol 124: 163-174 Tavolga WN (1974) Signal/noise ratio and the cr itical band in fishes. J Acoust Soc Am 55: 1323-1333 Tavolga WN, Wodinsky J (1963) Auditory capacitie s in fishes. Pure tone thresholds in nice species of marine teleosts. Bull Amer Mus nat Hist 126: 177-240 Tavolga WN, Wodinsky J (1965) Auditory capacit ies in fishes: threshold variability in the bluestriped grunt ( Haemulon sciurus) Anim Behav 13: 301-311 Tolimieri N, Jeffs A, Montgomery JC (2000) Ambient sound as a cu e for navigation by the pelagic larvae of reef fish es. Mar Ecol Prog Ser 207: 219-224 Webb JF (1998) Photic behavior and vertical migration in herring larvae. Mar Beh and Physiol 11: 139-156. Webb JF, Smith WL (2000) The latero-physic connection: a unique link between the swim bladder and the lateral line system in Chaetodon (Perciformes: Chaetodontidae). Copeia 1032-1036 Wellington GM, Victor BC (1989) Planktonic larval durati on of one hundred species of Pacific and Atlantic damselfi shes. Mar Biol 101: 557-567 Wysochi LE, Ladich F (2001) The ontogenetic development of auditory sensitivity, vocalization and acoustic communi cation in the labyrinth fish Trichopsis vittata J Comp Physiol A 187: 177-187 Yan HY, Popper AN (1991) An automated positive reward method for measuring acoustic sensitivity in fish. Be hav Res Methods Instrum 23: 351-356 Yan HY, Fine ML, Horn NS, Coln WE (2000) Va riability in the role of the gasbladder in fish audition. J Comp Physiol A 186: 435-445 Zimmerman (1993) Behavioral meas ures of auditory thresholds in developing tree shrews ( Tupia belangeri ). J Acoust Soc Am 94: 3071-3075