Life history and bioenergetics of the bonnethead shark, sphyrna tiburo (Linnaeus): a comparison of two populations

Citation
Life history and bioenergetics of the bonnethead shark, sphyrna tiburo (Linnaeus): a comparison of two populations

Material Information

Title:
Life history and bioenergetics of the bonnethead shark, sphyrna tiburo (Linnaeus): a comparison of two populations
Creator:
Parsons, Glenn R.
Place of Publication:
Tampa, Florida
Publisher:
University of South Florida
Publication Date:
Language:
English
Physical Description:
xiv, 170 leaves : ill. ; 29 cm

Subjects

Subjects / Keywords:
Hammerhead sharks -- Growth ( lcsh )
Bonnethead shark ( lcsh )
Dissertations, Academic -- Marine science -- Doctoral -- USF ( FTS )

Notes

General Note:
Thesis (Ph. D.)--University of South Florida, 1987. Bibliography: leaves 158-167.

Record Information

Source Institution:
University of South Florida
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
020904733 ( ALEPH )
18823384 ( OCLC )
F51-00171 ( USFLDC DOI )
f51.171 ( USFLDC Handle )

Postcard Information

Format:
Book

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

LIFE HISTORY AND BIOENERGETICS OF THE BONNETHEAD SHARK, Sphyrna tiburo (Linnaeus): A COMPARISON OF TWO POPULATIONS by Glenn R. Parsons A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Marine Science in the University of South Florida August, 1987 Major Professor: John C. Briggs

PAGE 2

Graduate Council University of South Florida Florida CERTIFICATE OF APPROVAL Ph.D. Dissertation This is to certify that the Ph.D. Dissertation of Glenn R. Parsons with a major in Marine Science has been approved by the Examining Committee on 6 July 1987 as satisfactory for the dissertation requirement for the Ph.D. degree. Examining Committee: Major Dr. Joffn Briggs Member; Dr. Thomasd Hopkins Member: Dr. Robert Muller

PAGE 3

This dissertation is dedicated to my wife Cheryl B. Parsons for her help even when waist deep in sharks. ii

PAGE 4

ACKNOWLEDGMENTS During the course of this research work many people graciously donated their time and effort. I would like to extend a most sincere thanks to Dr. John Briggs, my major professor, for his unswerving support. My committee members, Dr. Harold Grier, Dr. Thomas Hopkins, Dr. Robert Muller and Dr. Joseph Torres, each provided valuable assistance and to them I am grateful. Likewise, I thank Dr. Behzad Mahmoudi of the Florida Department of Natural Resources for serving as chairperson of my oral defense. I am deeply indebted to USF Marine Science students Jeff Brown, Randy Hochberg, Kristie Killam, Mike Mitchell, Ernst Peebles, Iliana Quintero and Gregg Tolley. I also thank Dr. John Shively, of Smith-Kline Laboratory, for his enthusiastic support. A large part ofthis dissertation was possible only through the assistance of the expert staff of the Sea World Marine Science and Conservation Center. Special thanks are due to Don Chiddick, Bill Gibbs, Jim Kepley, Randy Mickney, Ed Olsen:.;Moore, and John Swanson. Frank Murru, Curator of Fishes, Sea World of Orlando, donated spec imens for which I am grateful. I thank the employees of the Florida Department of Natural Resources especially Lew Bullock, John Darovek, Jack Gartner, Mark Godcharles, Mark Leiby, Lori Marshall, Bob McMichael, Mike Murphy, and Roy Williams. Finally, I would like to thank Dr. Karen Steidinger for her assistance. This work was funded in part by a fellowship awarded to the author by the Gulf Oceanographic Research Foundation, a grant from the USF Marine Science Department and a facility support grant from Sea World of Florida. iii

PAGE 5

TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES ABSTRACT CHAPTER ONE: INTRODUCTION CHAPTER TWO: CAPTIVE MAINTENANCE AND REPRODUCTIVE BIOLOGY Intmduction Materials and Methods Results and Discussion Conclusions CHAPTER THREE: PATTERNS OF ACTIVITY AND SWIMMING VELOCITY Introduction Materials and Methods Results Discussion Conclusions CHAPTER FOUR: AN EXAMINATION OF THE GROWTH INHIBITORY EFFECT OF TAGGING AND TETRACYCLINE TREATING BONNETHEAD SHARKS, Sphyrna tiburo Introduction Materials and Methods Results Discussion Conclusions CHAPTER FIVE: AGE DETERMINATION AND GROWTH Introduction Materials and Methods Results Discussion Conclusions iv vi viii xii 3 3 4 7 33 35 35 36 38 43 49 52 52 54 56 60 62 64 64 66 70 105 115

PAGE 6

CHAPTER SIX: THE BIOENERGETICS OF THE BONNETHEAD SHARK, Sphyrna tiburo: DESIGN AND APPLICATION OF A RESPIROMETER FOR DETERMINATION OF METABOLIC RATE Introduction Materials and Methods Results Discussion Conclusions LITERATURE CITED APPENDIXES APPENDIX 1 STANDARD DEVIATIONS OF MEAN LENGTHS CALCULATED.USING THE VON BERTALANFFY GROWTH EQUATION APPENDIX 2 STANDARD DEVIATIONS OF MEAN WEIGHTS CALCULATEDUSING THE LOGISTIC GROWTH EQUATION y 117 117 120 124 142 156 158 168 169 170

PAGE 7

LIST OF TABLES Table 1 The environmental parameters monitored during the study. The values shown are the extremes recorded over the year 1983. 9 Table 2. Prey items found in gut content examination of bonnethead sharks from Florida Bay and Tampa Bay. 13 Table 3. Tetracycline experiment I. This experiment examined theeffect of 10, 50 and 75 mg/kg injections of tetracycline on growth of the bonnethead shark. Unless otherwise indicated, numbers in parentheses are standard deviations. 57 Table 4. Tetracycline experiment II. This experiment examined the effect of O, 12.5 and 25 mg/kg injections of tetracycline on growth of the bonnethead shark. Unless otherwise indicated, numbers in parentheses are standard deviations. 59 Table 5. The results of exam1n1n g the vertebral rings of sharks held in captivity for periods of from 0.3 to 2 3 years. 73 Table 6 The results of the study to validate the vertebral ring aging technique in the bonnethead shark. These animals were injected with tetracycline, held in captivity for 0 .08 to 1.45 years and then examined for ring formation. (Tetra.= tetracycline). 75 Table 7. Ford-Walford and nonlinear regression estimates of the von Bertalanffy growth equation. The coefficients of the equation are presented for sharks from Tampa Bay, Florida Bay and captive animals. 81 Table 8. Nonlinear regression estimates of the logistic growth equation. The coefficients of the equation are presented for sharks from Tampa Bay, Florida Bay and captive an imals. 87 Table 9. Back-calculated total lengths (em) determined from vertebral rings for sharks from Tampa Bay, Florida Bay and those held in captivity. (B = birth). 104 Table 10. Summer (May through August) energy budget for c aptive female bonnethead sharks of age class 0 to 3 years. v i

PAGE 8

All values are in kilocalories per day and numbers in parentheses are percentages of the total budget that each component represents. (I -ingestion, P and Pr = production of somatic and reproductive respectively, Mr c routine metabolism and E = energy excreted as waste). 129 Table 11. Summer (May through August) energy budget of male bonnethead sharks from Florida Bay of age class 0 to 3 years. All values are in kilocalories per day and numbers in parentheses are percentages of the total budget that each component represents. (I = ingestion, P and P =production of somatic and reproductive tissue, Fespectively, Mr -routine metabolism and E energy excreted as waste). 130 Table 12. Summer (May through August) energy budget for female bonnethead sharks from Tampa Bay of age class 0 to 3 years. All values are in kilocalories per day and numbers in parentheses are percentages of the total budget that each component represents. (I = ingestion, Ps and Pr =production of somatic and reproductive tissue, respectively, M = routine metabolism and E =energy excreted waste). 131 Table 13. Summer (May through August) energy budget for male bonnethead sharks from Tampa Bay of age class 0 to 3 years. All values are in kilocalories per day and numbers in parentheses are percentages of the total budget that each component represents. (I = ingestion, Ps and Pr =production of somatic and reproductive t1ssue respectively, Mr = routine metabolism and E = energy excreted as waste). 132 Table 14. Metabolic rates (in kcal/day) for sharks from Florida and Tampa Bays. The values in parentheses represent the 95% confidence intervals. 133 vii

PAGE 9

LIST OF FIGURES Figure 1.Collection sites for the bonnethead shark study. These sites were located in a warm temperate area (Tampa Bay) and a tropical area (Florida Bay). 5 Figure 2.-A comparison of the average monthly temperatures reco rded for the Sea World system, for Florida Bay and for Tampa Bay during the year 1983. 10 Figure Average monthly dissolved oxygen concentration for the Sea World system and the Florida Bay site during 1983. 11 Figure 4.-The claspers of (A) an immature (49.1 em TL) and (B) a mature (87.4 em TL) bonnethead shark. 15 Figure 5.-Development of the claspers of the bonne thead shark. This figure indicates maturation at about 68 to 70 cmTL in Florida Bay and about 80 em TL in Tampa Bay. 16 Fi gure O varian e g g develo pment in the femal e b o nnethead shark. The plot indicates maturation o ccurs at about 80 cmTL. 18 Fi gure 7.-The distribution of adult females collected in Florida Bay. Most individuals were within the 8 5 to 95 em size class. 2 1 Figure 8.The distribution of adult females collected in Tampa Bay. Most individuals were within the 95 to 105 em sizeclass. 2 2 Figure 9.Embryonic development of the bonnethe a d shark from fertilization to parturition. Embryonic growth through gestation is estimated at about 8.0 em/month in Tampa Bay, while in Florida Bay the g rowth rate is 5.5 em/month. 2 4 Figure 10.Bonnethead shark embryos demonstrating the difference in growth rate between the two populations. Embryo A from Florida Bay and Embryo B from Tampa Bay are both approximately 2 months old. 2 7 Figure 11.-A section of the umbilical cord of the bonnethead S. tiburo. 2 9 vii i

PAGE 10

Figure 12.-Scanning electron micrograph of a single umbilical appendiculae of the bonnethead shark, S. tiburo (54x magnification). 31 Figure 13.-Scanning electron micrograph of the surface of an umbilical appendiculae of the bonnethead shark, S. tiburo (600x magnification). 32 Figure 14. Changes in the swimming velocity (em/sec) of the bonnethead shark, Sphyrna tiburo over a 24 hour period. 39 Figure 15. Changes in the swimming velocity (bl/sec) of the bonnethead shark Sphyrna tiburo over a 24 hour period. 40 Figure 16. Average catch rates of bonnethead sharks taken in Florida Bay over a 24 hour period. 42 Figure 17. Average catch rates of bonnethead sharks taken in TampaBay during the period 0800 to 1700. 44 Figure 18. The relationship between voluntary swimming speed (U0 ) and total length for the bonnethead 45 Figure 19. A photomicrograph of a sectioned vertebral centrum from a bonnethead shark previously injected with tetracycline. The fluorescent label is clearly visible. 68 Figure 20. A picture of a bonnethead shark vertebrae in which -the vertebral rings have been accentuated using the penciling technique. 71 Figure 21. Von Bertalanffy growth curve of male bonnethead sharks collected from Tampa Bay (B = birth, R = vertebral ring number, L =length). 77 Figure 22. Von Bertalanffy growth curve of female bonnethead shark s collected from Tampa Bay (B = birth, R vertebral ring number, L = length). 78 Figure 23. Von Bertalanffy growth curve of male bonnethead sharks collected from Florida Bay and sharks held at least part of their lives in captivity (B = birth, R =vertebral ring number, L = length). 79 Figure 24. Von Bertalanffy growth curve of female bonnethead sharks collected from Florida Bay and sharks held at least part of their lives in captivity (B = birth, R =vertebral ring number, L length). 80 Figure 25. Growth in weight of female bonnethead sharks collected from Tampa Bay (B = birth, R = vertebral ring number, W =weight). 83 Figure 26. Growth in weight of male bonnethead sharks ix

PAGE 11

collected from Tampa Bay (B birth, R vertebral ring number, W weight). 84 Figure 27. Growth in weight of female bonnethead sharks collected from Florida Bay and held in captivity (B birth, R-vertebral ring number, W weight). 85 Figure 28. Growth in weight of male bonnethead sharks collected from Florida Bay. and held in captivity (B = birth, R vertebral ring number, W weight). 86 Figure 29. Individual growth records of sharks held in captivity. 89 Figure 30. Growth in length of bonnethead sharks born in captivity in August 1983. The mean and 95% confidence interval are shown. 91 Figure 31. Growth in weight of bonnethead sharks born in captivity in August 1983. The mean and 95% confidence interval are shown. 92 Figure 32. Growth in length of bonnethead sharks born in captivity in August 1985. The mean and 95% confidence interval are shown. 93 Figure 33. Growth in weight of bonnethead shark s born in captivity in August 1985. The mean and 95% confidence interval are shown. 9 4 Figure 34. Growth in length of bonnethead shark s born in captivity in August 1984. Also shown is a plot of water temperature data for the environment in which the sharks were held captive. The arrows indicate the time of annulus (-ring) formation in these animals. Means and 95% confidence intervals are shown. 95 Figure 35. Growth in weight of bonnethead shark s born in captivity in August 1984. Also shown is a plot of water temperature data for the environment in which the sharks were held captive. The arrows indicate the time of annulus (-ring) formation in these animals. M eans and 95% confidence intervals are shown. 96 Figure 36. A composite growth curve of female bonnethead sharks held in captivity. This plot was constructed using known age animals born in captivity and by assigning ages to animals that were introduced into captivity. (B birth, A = age in years, L a length). 98 Figure 37. A composite growth curve of female bonnethead sharks held in captivity. This plot was constructed as in Figure 19. (B A age in years, W weight). 99 X

PAGE 12

Figure 38. The relationship between vertebral radius and total length for Tampa Bay sharks (L = length, V = vertebral radius). 101 Figure 39. The relationship between vertebral radius and total length for Florida Bay sharks (L = length, V vertebral radius). 102 Figure 40. A diagram of the respirometer used for determining metabolic rate in the bonnethead shark 121 Figure 41. A linear regression of absolute oxygen consumption (C) on body weight (W). The dotted lines are the 95% confidence limits. 126 Figure 42. An exponential regression equation of oxygen consumption (C) on body weight (W). The dotted lines are the 9 5 % confidence intervals. 127 Figure 43. A power performance curve for bonnethead sharks of 1 to 1.6 kg in weight 135 Figure 44. A linear regression of gut content weight (expressed as percent of total body weight) on total body weight. 137 Figure 45. An exponential regression of gut content weight on total weight. (I equals total weight of ingested food and W equals body weight). 138 xi

PAGE 13

LIFE HISTORY AND BIOENERGETICS OF THE BONNETHEAD SHARK, Sphyrna tiburo (Linnaeus): A COMPARISON OF TWO POPULATIONS by Glenn R. Parsons An Abstract Of a dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Marine Science in the University of South Florida August, 1987 Major Professor: John C. Briggs xii

PAGE 14

A study of two populations of the bonnethead shark, Sphyrna tiburo was conducted in the Florida Keys and in Tampa Bay Florida from September, 1982 to December, 1986. In addition, a captive shark colony ranging in size from 5 to 100 indi victuals, was held for study in a semi-natural environment at the Sea World Marine Science and Conservation Center located in the Florida Keys. The maintenance of animals in captivity and the collection of animals from widely separated geographical areas allowed the examination of latitudinal variation. Differences in embryonic deve lopment, birth size, and size at maturation were noted. The data suggest an inverse relationship between embryonic development and average environmental temperature. Embryonic development is faster, and size at birth and size at maturation is larger in Tampa Bay. Validation of the vertebral ring method of aging allowed examination of age and growth Growth rates are in general faster and size at age larger in the Tampa Bay population. The general energy budget equation, I = P + M + E where I = consumption, P = production, M = metabolism and E = excretory products, was used to examine bioenergetics in the bonnethead shark. The design and application of a respirometer for use with large active fish species allowed an estimate of metabolic rate. This and the measurements of swimming velocities and daily activity patterns allowed an estimate of energy consumed during routine activity. The summer energy budgets of male and female sharks from both Tampa and Florida Bays are compared. xiii

PAGE 15

Abstract approved :3' , Mafr Professor Titfe and Department Date of Approval xiv

PAGE 16

CHAPTER ONE: INTRODUCTION The bonnethead shark, Sphyrna tiburo is an excellent candidate for basic research investigations due to its abundance, relatively high resistance to capture stress, and propensity to captivity. This dissertation reports the results of an investigation into the basic biology of the bonnethead shark The objectives of this research were : (1) to examine various characteristics of two bonnethead shark populations from widely separated geographic locales, (2) to identify those population parameters which show the greatest latitudinal variability (3) to construct energy budgets for sharks from these different regions and use these energy budgets as a framework such that the life histories of the two populations could be compared. In some cases, researchers of the marine environment are without the luxury of a body of literature on which to draw. When investigating shark biology, it is most often the responsibility of the investigator to do the basic "grotmdwork" on which more sophisticated studies may be erected. In this investigation it was necessary to delineate the captive requirements of the bonnethead shark (Chapter Two) then compare the captive and field environment. This was necessary such that the validity of extrapolations from captivity to the wild could be addressed In Chapter Two geographic variability in reproduction was also examined. In Chapter Three the activity patterns and swimming velocity of the bonnethead shark were

PAGE 17

2 examined. Activity can represent a very large percent of the total energy budget in some animals and must be taken into account In constructing an energy budget (Chapter Six) estimates of the production of reproductive tissue (Chapter Two) and somatic tissue (Chapter Five) are necessary. Before determining somatic tissue production it was necessary to validate the vertebral ring method of aging and prior to this the effect of the validation technique on growth had to be ascertained (Chapter Four) Finally, in constructing an energy budget an estimate of metabolism must be made. This required the construction of a respirometer large enough to accomodate these highly active animals (Chapter Six). The completed energy budgets provided a common framework on which the life history strategies of the two populations of sharks could be compared.

PAGE 18

3 CHAPTER TWO: CAPTIVE MAINTENANCE AND REPRODUCTIVE BIOLOGY Introduction Our knowledge of shark biology has been greatly enhanced by their maintenance in captivity. Their generally uncooperative nature, combined with the inherent difficulties of making field observations on free-ranging sharks have in some cases, been circumvented by captive maintenance. Clark ( 1963) lists the species of sharks that have been maintained in captivity and in the years since that report, much progress has been made in capture techniques, handling, and in delineating the captive requirements of sharks. Several recent research programs have utilized sharks held in captivity. Weihs (1981) was able to examine swimming speeds of two species of carcharhinids using animals in captivity. Cleaning behavior by a species of labrid on several species of captive sharks was observed by Keyes ( 1982) Gilmore et al. (1983) used data from captive sand tiger sharks, Odontaspis taurus, to examine the oviphagous reproductive strategy. The bonnethead shark, Sphyrna tiburo is a small species of hammerhead shark that does well in captivity and is thus a good candidate for certain research programs. Clark (1963) reported that bonnethead sharks had been maintained in captivity for only a few months At present, bonnethead sharks have been held in captivity for at least 3 years(this study) Despite the bonnethead shark's apparent

PAGE 19

4 propensity to captivity, few studies have been devoted to its biology. Myrberg and Gruber (1974) utilized a captive colony of bonnethead sharks to examine patterns of intra-specific behavior. The only other recent study that deals exclusively with bonnethead shark biology examined placentation and gestation (Schlernitzauer and Gilbert 1966). In this chapter the physical parameters under which a captive bonnethead shark colony was maintained is examined. This is compared with the same parameters that free-ranging bonnethead sharks in Florida Bay are likely to experience. In addition, reproductive information from captive animals was incorporated with the same from wild-caught sharks to obtain a complete picture of the bonnethead shark reproductive cycle. Latitudinal variation in reproduction was examined by sampling sharks from a tropical (Florida Bay) and warm temperate (Tampa Bay) area. Materials and Methods From July 1982 until December 1986 bonnethead sharks were collected from Tampa Bay at 2 7'N, 82'W and from Florida Bay at 2 4'N, 81'W (Figure 1). The majority of sharks were collected over grass flats using gill nets fished continuously for as long as 48 hours. The nets ranged from 175 to 350 meters in length and were 7 em stretch mesh. The nets were c hecked at 1 hour intervals for live spe c imens, and up to 6 hour intervals when specimens were needed for examination of reproductive condition. Bonnethead sharks collected from Florida Bay that were in good condition were immediately placed in a simulated natural environment

PAGE 20

5 Figure 1 -Collection sites for the bonnethead shark study. These sites were located in a warm temperate area (Tampa Bay) and a tropical area (Florida Bay).

PAGE 21

6 condition were immediately placed in a simulated natural environment at the Sea World Marine Science and Conservation Center (formerly the Sea World Shark Institute) located on Long Key in the Florida Keys. The shark colony was maintained in a large, shallow water enclosure, approximately 30 meters long meters wide and 1 meter deep. Water was supplied to this enclosure by aerating and filtering water pumped directly from Florida Bay. Water quality parameters were monitored on a daily basis throughout the length of the study. The parameters examined were temperature, dissolved oxygen, and salinity using a Hydro Lab meter. Ammonia and pH were monitored using a simple test kit. Sharks were fed to satiation a daily diet of squid, shrimp and fish (Osmerus mordax and Mallotus villosus). Sharks from both Florida Bay and Tampa Bay were examined for determination of reproductive condition. All sharks were immediately weighed and sexed and total, fork and standard lengths were measured to the nearest 0.1 em. Maturity in males was evidenced by the length and calcified nature of the claspers and presence of sperm in the seminal vesicles. Maturity in females was determined by the diameters of the largest ovarian eggs. When gravid females were examined the number, sex, weight and total length of all embryos were recorded. To examine the umbilical appendiculae of the bonnethead shark samples of umbilical cord were obtained from a 91 em TL, gravid bonnethead shark collected from Florida Bay on 7/21/82. The samples were immediately preserved in 2.5% glutaraldehyde in filtered seawater (Corwin 1978). The samples remained in fixative for periods ranging from 1 day to 1 month. For light microscopy, the samples were examined with no further treatment. For scanning electron microscopy (SEM), the

PAGE 22

samples were taken through a seawater to distilled water series (30%, 50%, 70%, and 100% deionized water) to remove any salt, and then from distilled water into an ethanol series (30%, 50%, 70% and 100% ethanol) and finally into a 100% absolute ethanol bath. The samples were left in each bath for 15 minutes. The specimens were critical point dried using a Denton DCP-1 Critical Point Dryer, mounted on aluminum stubs with colloidal silver paste, and sputter coated with gold for about 3 minutes. The specimens were examined with an ISI-DS 130. scanning electron microscope. The following scope specifications were used; standard function, 2 mode, 15KV, 15 mm working distance and 70 uA emission current. All micrographs were taken using a 35 mm camera and 32 ASA, Pan-X film. Results and Discussion Captive Maintenance From July 1982 to December 1986 as few as five and as many as 100 bonnethead sharks of all sizes and both sexes were maintained in captivity at the Sea World Marine Science and Conservation Center. Water quality was monitored on a daily basis. For comparison with the bonnethead sharks' natural environment, the same water quality parameters were obtained for an area within Florida Bay (24'N, 81'W). Physical data were recorded at a site within a grass flat approximately two meters deep and about 1 mile west of Marathon Key. The range for each water quality parameter examined, over the year 7

PAGE 23

8 1983 is shown in Table 1. Temperature values are very similar for both Sea World and Florida Bay ranging from about 19C to 32C. Dissolved oxygen stayed within a fairly narrow range in the Sea World system (5.7 to 7.8 ppm) while Florida Bay fluctuated more widely (4.4 to 8.4 ppm). Salinity values for Sea ,World and Florida Bay show interesting differences. The Sea World system was in general less saline than the Florida Bay site. The maintenance facility, covered only by screening for shade, is subject to salinity fluctuations during rainstorms. The occasional condition in Florida Bay may be a result of evaporation. The values for pH were practically identical between the two areas and, ammonia, not measured at the Florida Bay site, was undetectable at Sea World. Figure 2 shows the average monthly temperature for the year 1983 for both Sea World and for Florida Bay. The seasonal fluctuations in temperature are very similar. The summer months are warmest with temperature in both environments averaging about 30C and then dropping to about 20C during the winter. The average monthly dissolved oxygen concentration is seen in Figure 3. The inverse relationship between water temperature and dissolved oxygen is readily apparent. During the summer months when water temperature is highest, the dissolved oxygen concentration is lowest. Florida Bay averaged a very low 5 ppm dissolved oxygen (about 80% of saturation) during the summer, up to saturation (ca. 7 ppm) in winter. These dissolved oxygen and temperature readings were made during daylight hours over the grass flats in Florida Bay. During the summer months, collection records show that bonnethead sharks utilize the shallow grass flats almost exclusively at night and move out of the area during the day

PAGE 24

Table 1. The environmental parameters monitored during the study. The values shown are the extremes recorded over the year 1983. Temp. D.O. Salinity Ph Ammonia SEA WORLD 19-32 C ppm 32-'36 ppt 7.8-7.9 undetectable FLORIDA BAY 19-31 C 4.4-"8.4 ppm 34-39 ppt 7 .6...;8. 1 9

PAGE 25

10 35 T 30 p oi0-a-( 0c) 25 0 A /. T u 20 0/;/ R "" E 0 e SEA WORLD 15 I 0 FLORIM BAY 0 TAMPA BAY 0 J F M A M J J A S 0 N D Figure 2.-A comparison of the average monthly for the Sea World system, for Florida Bay and for Tampa Bay during the 1983.

PAGE 26

7.0 I 6.5 6.0 5.5 5.0 e SEA WORLD 0 FLORIDA BAY 4.5 J F M A M J J A S 0 N D Figure 3.-Average monthly dissolved oxygen con centration for the Sea World system and the Florida Bay site during 1983. 11

PAGE 27

12 (see Chapter 3) There is evidence which suggests that other shark species likewise move to shallower water at night. Sciarrotta and Nelson ( 1 977) used telemetry devices to follow blue shark, Prionace glauca, movements around Santa catalina Island, California. They found that positions of sharks during the day were in deeper waters away from the island, while most of the night positions were near shore. Gruber (1982) used telemetry to examine the movements of lemon sharks, Negaprion brevirostris, around Bimini Island, Bahamas. A similar movement away from the island during the day and toward the island at night was observed. It is possible that during the day, when dissolved oxygen is very low and temperature is high, bonnethead sharks move out of the shallow flats into deeper cooler waters. The d issolved oxygen concentration of the Sea World system showed less drastic fluctuations, remaining at or very near saturation values over the entire year. This constancy can no doubt be attributed to the secondary aeration of the incoming Florida Bay water. Sharks maintained at Sea World were fed to satiation on a daily basis. The diet consisted of squid, shrimp and fish. Examinations of gut content of free ranging sharks can provided information on nutritional requirements and an indication of the degree to which the diet of captive animals approximates the natural diet. In Table 2 is shown the results of gut content examinations from 53 bonnethead sharks collected in Florida Bay. Cephalopods occurred in the diet 34% of the time and constituted the most common prey item. A diet in captivity of squid, shrimp, and fish is probably a reasonable approximation of the natural diet. However, the data suggest that crabs or lobster might also be included. Curiously it

PAGE 28

13 Table 2. Prey items found in gut content examination of bonnethead sharks from Florida Bay and Tampa Bay. Florida Bay Prey Item Cephalopoda Panulirus sp. Decapod a Osteichthyes Penaeus sp. Stomatopoda Seagrasses Unidentifiable Empty n = 53. Tampa Prey Item Callinectes sapidus Penaeus sp. Osteichthyes Stornatopoda Seagrasses Unidentifiable Empty n = 99. Bay Occurrence(%) 34 28 26 23 8 2 47 2 17 Occurrence 91 4 85 8 7 (%)

PAGE 29

14 was discovered that 47% of the animals had ingested seagrasses. These were primarily Thalassia but also included Halodule and Syringodium. In most cases, there were only a few blades of grass but in some the gut was completely filled. It is possible, that the sharks intentionally ingested the gr;-asses However, since the bonnethead feeds predominately on benthic organisms, it is more likely that the grasses were simply ingested incidental to the sharks normal feeding behavior. Reproductive Biology Maturity in Males The examination of secondary sex characteristics is often used to determine maturity in animals. In male sharks, the development of the claspers has been found useful for determining maturity in a number of species (Clark and von Schmidt, 1965; Teshima et al., 1978; Teshima, 1981; Parsons, 1 983) The claspers of an immature and mature male bonnethead shark are shown in Figure 4. In Figure 5 is shown the relationship between total length (TL) of male bonnethead sharks and clasper index (the length of the claspers expressed as a percent of the shark TL). The relationship for both Florida Bay and Tampa Bay is plotted. From about 30 to 50 em TL, the claspers represent only about 2 to 3% of TL. At about 55 to 60 em, rapid clasper development indicates that maturation has begun. In Florida Bay, development reaches its' zenith at about 68 to 70 em TL when the claspers are about 8% of the shark TL. The smallest mature males (ca 70 em) were found to have the greatest clasper index. This is explained by continued body growth after the clasper has completed

PAGE 30

' : .: .. i A .. . B .. : . .: .. : 2an Figure 4.-The claspers of (A) an immature (49.1 em TL) and (B) a mature (87.4 em TL) bonnethead shark. 15

PAGE 31

)( w c z a: UJ Q. (/) < ..J u 8 7 6 5 4 3 30 0 0 0 .,. . .. 0 0 o 40 50 60 0 0 o 0 o 00 o e e ... OCD 0 e 0-oo 0 o 0 o0 0 0 0 0 e FLORIDA BAY 0 TAMPA BAY 0 70 80 90 TOTAL LENGTH (eM) 16 Figure 5.-Development of the claspers of the bonnethead shark. This figure indicates maturation at about 68 to 70 em TL in Florida Bay and about 80 em TL in Tampa Bay.

PAGE 32

17 development, which causes a decrease in the clasper index Parsons (1983) observed the same phenomenon in the Atlantic sharpnose shark, Rhizoprionodon terraenovae. The largest male shark from Florida Bay was about 85 em TL with a 6.7% clasper index. In Tampa Bay, maturation in male bonnethead sharks proceeds somewhat differently (Figure 5). Sharks from both areas experienced rapid clasper development at about 50 to 55 em TL. However, in Tampa Bay, maturation is complete at a body size 10 em larger than the males from Florida Bay. The data suggest that Tampa Bay bonnets mature at ca 80 em TL when the clasper index is 8 to 8.3%. A parallel decrease in clasper index beginning at or near maturation is evident in the Tampa Bay shark population. The largest mature male taken from Tampa Bay was about 93 em TL with a clasper index of ca. 6 .7%. Maturity in Females-The development of the ovarian eggs is a useful indicator of maturation in female sharks (Figure 6) Parsons (1983) and others, have used this technique with success. In Florida Bay, bonnethead sharks below about 65 em TL are immature The ovary is undeveloped and the follicles are only 1 to 2 mm in diameter At 65 to 70 em TL the ovary begins a period of development. Yolk material is rapidly deposited and the first generation of ovarian eggs quickly increase in diameter First ovulation (maturation) takes place when the ovarian eggs have grown to about 2 em in diameter and when the shark is about 80 em TL. The smallest mature female taken in Florida Bay was 83 em TL. This shark was also gravid ( the embryos were approximately 20 em TL) and therefore would have been a few centimeters smaller at time of maturation. Myrberg and Gruber (1974) reported collecting a 74 em TL gravid female.

PAGE 33

2.0 CN\RIAN 1.5 EGG DIAMETER 1.0 (eM) QS Ovulation FLORIDA BAY ... ... 30405060708090100 TOTAL LENGTH (eM) 18 Figure 6.Ovarian e g g development in the female bonnethead shark. The plot indicates maturation occurs at about 80 em TL.

PAGE 34

19 When near ovulation, the ovary of the bonnethead shark is very large, vascularized, and filled with large (ca. 2 em) yellow eggs. An average of about 1 0 ovarian eggs is produced each season in Florida Bay whereas in Tampa Bay, sharks produce about 11 per season, however this difference was not significant. After ovulation, the eggs move through the oviducal gland where fertilization takes place and where the egg case is formed. From there, they move into the uterus where embryonic development begins. The suggestion has been made that ovulation in sharks may be copulation induced. Bonnethead sharks maintained in captivity provided information concerning this question. A 79.2 em TL female was collected and placed in the Sea World semi -natura l environment on January 1 5 1983. The shark was one of a colony containing all females and therefore was never given the opportunity to mate. On May 11, 1984, when the shark was 91 0 em TL, the animal was examined and 5 uterine eggs were found in each uterus. In the absence of males, ovulation proceeded which proves that ovulation does not have to be copulation induced in the bonnethead shark. It does not rule out the possibility that copulation induces ovulation in the natural environment. In Florida Bay, fertilization is efficient since only 3 infertile eggs were noted in examination of 70 embryos (4.3% infertile). A striking difference was seen in Tampa Bay. In examinations of 30 litters involving 267 embryos, 73 infertile eggs were found. An unprecedented 27% of all eggs produced were infertile and 77% of all litters contained at least one infertile egg (range = 1 to 8 infertile eggs/litter). In one case, only 2 of the 8 eggs produced were fertile.

PAGE 35

20 Parsons ( 1983) noted only two infertile eggs in examination of 315 embryos of the Atlantic sharpnose shark, Rhizoprionodon terraenovae. The cause of this infertility in the Tampa Bay population is not known. Precocity or senescence cannot be used to explain these observations since a correlation between size of the adult and the incidence of infertility could not be demonstrated. Since the production of ovarian eggs is energetically expensive, natural population regulation by infertility would be very inefficient and does not seem likely. In Figures 7 and 8 are shown the size distributions of adult females in collections Florida Bay and Tampa Bay. The modal size for Florida Bay adult females fell in the 90 to 95 em size class (Figure 7). Almost 75% of the adults collected were in the 85 to 95 em size range. The smallest adult female was 83 em TL and the largest was 103.7 em. In Tampa Bay, no mature females below 85 em T L were collected (Figure 8). The smallest mature female was 87.6 em TL and the largest was 111 .0 em TL. Modal size for Tampa Bay adults fell within the 95 to 100 em size class and abou t 83% of all adult females were between 95 and 105 em TL. This striking dissimilar! ty in size between populations of animals from widely separated latitudes has been noted before (Allee, 1963) but has not been reported in sharks. The larger body size may be an adaptati on to lower environmental temperature. This phenomenon is discussed in more detail in Chapter 6. Mating, Embryonic Development, and Parturition. The presence of full size ovarian eggs and uterine eggs can be used to determine mating season in females. In Florida Bay, ovarian eggs approaching full size were noted in females as early as March 3, 1983. In late March,

PAGE 36

20 18 1 5 14 F R 1 2 E Q u 10 E N 8 c y s 4 2 fWill1 I 80 85-80 FLORIDA. BAY I 90-95 TOTAl LENGTH (em) D non gravi d fY:d gravid 100 21 Figure 7 -The d istribution of adult females coll e cted in Florida Bay. Most individuals were within the 85 to 95 em size c lass.

PAGE 37

20 18 16 1 4 F R 1 2 E Q u 10 E N c 8 y e 4 2 n 8 5 90-95 IOODS TOTAL LENGTH (em) 22 TAMPA BAY I 10510 D non gravi d gravi d I 1101 5 Figure 8 The distribution of adult females collected in Tampa Bay. Most individuals were within the 95 to 105 em size class.

PAGE 38

23 uterine eggs were present but no embryonic development was macroscopically apparent. The smallest embryos examined were about 7 em TL and were observed in In Figure 9 is shown the embryonic development of the bonnethead shark for Florida Bay and Tampa Bay during the year 1983. Extrapolating the almst linear Florida Bay growth curve back to zero provides an estimate of the latest possible time for fertilization which was placed at (Figure 9). This is a rough approximation which assumes the growth rate at fertilization is the same as that throughout most of gestation. This, combined with the first appearance of uterine eggs suggests that the peak in mating and fertilization in Florida Bay takes place between late March and early April. In Florida Bay, embryonic growth is rapid (Figure 9). Embryos examined on May 20, 1983 averaged about 7.5 em TL and by early August were approximately 24 em TL. This represents an extremely fast, almost linear increase of 6 em/month. From sharks he l d in captivity, the size at birth and the time of parturition was determined. Two gravid females collected in July, 1983 were placed in the Sea World facility. On August 25, both sharks gave birth to a total of 21 pups The were measured immediately and ranged from 25. 2 to 29.4 em TL (mean= 27.5 em 0.6 C.I.). This observations were made again in 1984 with remarkably similar results. Five gravid females gave birth to approximately 53 pups between August 3 and August 17, 1984 (several females "pupped" simultaneously on August 3). Fourteen of the pups, randomly sampled, were immediately measured and were found to range from 24.4 to 29.6 em TL (mean = 27.2 em 0 8 C. I.). Pooling the animals born in 1983 and 1984 produced an average size at birth of

PAGE 39

35 30 25 20 15 10 5 , LATEST POSSIBLE , I I /21 101 2 57/ I I FLORIDA I I 38' I .'--TAMPA BAY I Ia go/ I. 8 20 I t I I I I I I mean 3 J I / I I I I 95% confidence interval I I 0 parturition I M A M J J A 24 s Figure 9.-Embryonic development of the bonnethead shark from fertilization to parturition. Embryonic growth through gestation is estimated at about 8.0 em/month in Tampa Bay, while in Florida Bay the growth rate is 5 5 em/month

PAGE 40

25 27. 4 ern ( 0 5 C I.) and 82. 3 g ( 8 1 C I.). Myrberg and Gruber (1971) reported bonnethead sharks born in captivity at a mean of 24 em TL during the third week of August. From the above, several important life history parameters were gleaned The bonnethead shark requires a remarkably short gestation period i n Flori da Bay of 4 to 5 months. The data strongly suggest that a massive parturition occurs during a few weeks in August It is n oteworthy that on two separate occasions (August 1983 and 1984 ) several sharks gave birth simultaneously This suggests an environmental cue which determines the time of birth. The possibility also exists t hat one individual giving birth could induce others nearby to do the same. Bonnethead sharks in Florida Bay are born at about 27 em ( 0 5 C I ) TL. The mean length of litters from 1983 and 1984 were statistically identical. The growth rate from conception to parturition takes plac e at about 5 5 em/month. In comparing embryonic deve lopment of Tampa Bay sharks with Florida Bay (Figure 9) an interesting contrast is seen Mating within the Tampa Bay population takes place approximately one month later. Back extrapolation (Figure 9) places the latest possible time of fertilization i n late May. Full size ovarian eg g s were observed in late April and by mid-May uterine eggs were observed, suggesting that mating probably peaks around the end of April up to about mid-May. Although conception is delayed in the Tampa Bay population, the difference is made up by a more rapid growth rate. The growth rate from when embryos averaged about 1 0 em TL to when embryos were an average 25 em TL, was an exceedingly rap i d 9 em/month. The pupping season in Tampa Bay was determined by sampling female

PAGE 41

26 bonnethead sharks in late August and early September. Full term embryos averaging 34.7 em( 1.2 C.I.) and 171.8 g ( 20.6 C.I.) were collected in early September (Figure 9). In some individuals collected in mid September, the highly vascularized areas of placental attachment could be discerned on the uterine wall indicating recent parturition. Unfortunately, no new:..born pups were collected. Early September parturition would mean a short gestation period of between 4. 0 and 4. 5 months in Tampa Bay. The growth rate from conception to parturition, assuming the size at birth in Tampa Bay is the same as Florida Bay, would therefore be about 7. 8 to 8.8 em/month. Size at birth in the Tampa Bay population is significantly larger than in Florida Bay. Figure 9 shows that as development progresses, the two growth curves tend to converge on the same point although in Tampa Bay the embryos grow to a larger size. Florida Bay development proceeds more slowly over a longer period of time, while in Tampa Bay development is faster over a shorter period. Figure 10 graphically demonstrates the difference in growth rate between the two populations. The embryos depicted, one from Florida Bay the other from Tampa Bay, are each approximately 2 months into gestation. A 4 or 5 month gestation period is the shortest gestation period thus far reported for any shark species. What mechanism(s) allow embryonic development to be compressed into a slightly shorter season in Tampa Bay? Olsen (1984) reported a short 6 month gestation period for the school shark, Galeorhinus australis in southern Australia. In the same publication he suggests that the short gestation is a consequence of elevated environmental temperatures. This explanation is reasonable considering

PAGE 42

A B 1cm ..,.._.... 27 Figure 10.-Bonnethead shark embryos demonstrating the difference in growth rate between the two populations. Embryo A from Florida Bay and Embryo B from Tampa Bay are both appro x i mately 2 months old.

PAGE 43

28 temperatures' accelerating effect on growth in fishes. The geographic variability within a particular trait is generally thought to be directly related to temperature. However, the results of this paper show that embryonic growth is more rapid and gestation may be slightly shorter in the warm bonnethead population of Tampa Bay (Figure 2). There is, rather, an inverse relationship between environmental temperature and embryonic development. Perhaps the adult females of Tampa Bay, due to their larger body size (Figures 7 and 8) are better cornpeti tors, are more energy efficient and are able to invest more of their energy resources into embryonic development (see Chapter 6 for a bioenergetics discussion of this). This "reproductive plasticity" has never before been demonstrated in sharks. The litter size of the bonnethead shark was examined in each area. In Tampa Bay, litter size averaged 8 .9 pups/season. The mode for the Tampa Bay population was 8 with a range of 2 to 16 pups. In Florida Bay, an average of 9.3 pups/litter/season, with a mode of 9, and a range of 7 to 14 was found. Although the sample size from Florida Bay is small, the data suggest that litter size in each population is statistically the same. Since the gestation period is only 4 to 5 months, the potential for 2 litters per year exists. However, no evidence was found to indicate this occurs. Structure and Function of the Umbilical Appendi culae. Examination of the umbilica l appendiculae of the bonnethead s hark reveals several interesting features. The light micrograph (Figure 11) shows the "bushy" appearance of a section of umbilical cord The umbilical section is approximately 17 ern in length. The appendiculae are arranged in a radial fashion around the umbilical cord and occur along

PAGE 44

Figure 11.-A section of the umbilical cord of the bonnethead shark, S tiburo. 29

PAGE 45

30 the entire length of the umbilical. The appendiculae vary from 0.1 to 1 8 em in length and are approximately 1 8 mm in diameter at their widest point. Micro-dissection reveals these appendages to be filled with an unidentified matrix. In cross section several vessels lie just under the surface of the appendage Vascularization is absent from the interior of the structure. In Figures 12 and 13 seen scanning electron micrographs of the umbilical appendiculae. The low power micrograph (Figure 12) shows the structure of a single appendiculae Numerous pores are distributed across the surface of the extension. Although not evident in the micrograph, a greater concentration of pores occurs near the tip of the extension. Figure 13 shows an enlargement of the surface of the appendiculae. At approximately 600X magnification, the cuboidal cells which form the covering of the appendage are clearly seen. The cells are from 12 to 23 urn in diameter The numerous pores which are seen on the surface of the structure are 7 to 12 urn in diameter The opening of some of these pores appears to be filled with a coagulated substance. Umbilical appendiculae have been noted in only five s hark species; the bonnethead shark (Sphyrna tiburo), the scalloped hammerhead (S. lewini), the Atlantic sharpnose (Rhizoprionodo n terraenovae), the Telok Anson shark (Scoliodon laticaudus) and Paragaleus pectoralis. The function of these extensions is unknown, hypotheses have been forwarded. Alcock (1890) but at least two proposed that the structures are nutritive in nature. Budker's (1971) findings supported this hypothesis and also suggested that nutritive substances prod u ced

PAGE 46

Figure 12.-Scanning appendiculae of magnification). electron micrograph of a single umbilical the bonnethead shark, S tiburo (54x 31

PAGE 47

32 13. Scanning electron of the of an umbilical appendiculae of the bonnethead S. (600x magnification).

PAGE 48

33 by the uterine epithelium are absorbed by the appendages Gilbert (1981) reported that the appendicu lae are believed to be respiratory in function. In only one species, S. tiburo, has the microstructure of the appendiculae been examined Schlernitzauer and Gilbert ( 1 966) histologically examined cross sections of the appendiculae of S. tiburo. They reported that the position of the capillaries within the extension and the heavy vascularization supported Alcock's ( 1 890) hypothesis, that the structures function in the absorption of uterine secretions. This paper reports the first SEM examination of the umbilical appendiculae of ._. tiburo. The complex surface structure of the extensions suggests exchange between the appendicu lae interior and exterior. The highly pomus surface of the expansion has not been previously reported. The coagulated substance present in several of the pores (Figure 1 3 ) may indicate that active transport of materials was occurring at the moment of sample preservation. It is possible that this coagulated material is the nutritive s ubstance which is reported to be present in-utero in this species. These limited observations support Alcock's (1890) and Budker's (1971) hypothesis. Conclusions From the results of this study, several conclusions may be drawn. The maintenance facility at the Sea World Marine Science and Conservation Center is very similar to the environment of free-ranging Florida Bay sharks. In situations where research programs require

PAGE 49

34 approximation of the natural environment, this type of system may be used successfully. In examining the reproductive biology of two geographically separated populations, a number of differences were noted In Florida Bay males and females mature at a smaller size than in Tampa Bay. Fecundity in the. two populations was not statistically different (about 9 pups born in each area) but the incidence of infertile eggs was higher in Tampa Bay. Embryonic growth and development is faster and size at birth is greater in the Tampa Bay population SEM examination of the umbilical appendiculae suggest the structures may be nutritive in nature.

PAGE 50

35 CHAPTER THREE: PATTERNS OF ACTIVITY AND SWIMMING VELOCITY Introduction Rhythms of activity, e.g. circadian, circalunarian, and seasonal cycles, have been demonstrated among a wide range of animal groups. Within fishes, the rhythmic patterns may manifest themselves as changes in swimming speed, catchability, heart rate, feeding activity and a host of behavioral and physiological parameters. Hobson (1972, 1973, 1974) has contributed greatly to our understanding of the behavior and adaptations of nocturnal and diurnal reef fishes. Olla and Studholme (1978) compared activity rhythms of several fish species by monitoring swimming velocities over a period of several days. They were able to show how patterns of activity were modified by environmental changes. Priede (1978), utilizing ultrasonic telemetry, monitored heart rate of brown trout, Salmo gairdneri, in the field over a period of 24 hours. His findings suggest a circadian rhythm of cardiac activity which corresponded closely with locomotor activity. Due to the inherent problems in studying large active fish species, the bio-rhythms literature is understandably biased toward smaller fishes. This problem is particularly acute when sharks are under consideration. However, a few research works have been carried to fruition. Nelson and Johnson ( 1970) examined activity rhythms in the horn shark, Heterodont us francisci and the swell shark,

PAGE 51

36 Cephaloscyllium ventriosum. Longval et al. (1982) found cyclic patterns of food intake in the lemon shark, Negaprion brevirostris. The lemon shark has also been found to have an apparent circadian rhythm of activity and metabolism (Gruber 1984). The bonnethead shark, Sphyrna tiburo is a small species, easily maintained in captivity making it an excellent subject for the examination of activity rhythms and swimming velocity. In this study, the activity pattern of the bonnethead shark is examined by monitoring changes in swimming velocity over a 24 hour period and by examining catch rates of sharks over the same time period. In addition, the manner in which size affects swimming velocity is also examined. This information is important if the energy budget of the bonnethead shark is to be described. Materials and Methods Bonnethead sharks maintained in captivity at the Sea World Marine Science and Conservation Center from July 1982 to December 1986 were used for this study. The sharks were maintained in an enclosure roughly annular in shape and measuring approximately 30 meters long, 15 meters wide and 1 meter deep. Water was supplied to this enclosure by aerating and filtering water pumped directly from Florida Bay in the vicinity of Long Key in the Florida Keys. See Chapter 2 for a complete description of the maintenan ce of these animals in captivity. Swimming velocity of these sharks was estimated by two methods. During August 1982, August 1984 and May 1986 volitional swimming speeds were determined for sharks held in the above described captive environment.

PAGE 52

37 Two lines, a measured distance apart were placed across the enclosure near the waters An observer, properly positioned could then determine the time required for the nose of the shark to pass perpendicular to the lines. Sharks that did not pass perpendicular to the lines in straight ahead swimming were not recorded. During nighttime observations a low level fluorescent light source was used. During a session, from 12 to 40 individual swimming events were recorded. Sessions lasted from 15 minutes to 1 hour and were held during most hours of the day and night. Temperature during these sessions ranged from 26 to 30C. Velocities of animals were recorded by size classes to minimize the effect of size on swimming velocity. This method provided an average swimming velocity for similarly sized individuals. From 28 to 30 November 1986 and 5 to 7 December 1986 swimming velocities were determined by a second method. Sharks ranging from 34 to 95 em total length were removed from their enclosures, quickly placed in a large respirometer (see Chapter 6 for a description of this apparatus) and swimming velocities measured as the sharks swam between the acrylic windows of the device. The time required for the shark to cover the 60 em distance between windows was used to compute swimming velocities. Velocities were recorded under constant light conditions. From 20 to 40 swimming events were recorded in sessions lasting from 5 minutes to 4 hours during most hours of the day and night. Temperature during these sessions averaged 25C ( 2C). To allow an adjustment period, sessions were never held during the first hour of the animals stay in the respirometer. Likewise, velocities of animals in obvious stress were not recorded. This method provided

PAGE 53

38 swimming velocities of individual sharks over an extended time period. As a second indicator of a ctivity, catch rates of bonnethead sharks were calculated from gill net collections made in Florida Bay in the vicinity of Long Key, Florida and in Tampa Bay, in the vicinity of Coquina Key. In Florida Bay a 91 meter, 7. 6 em stretch mesh gill net was placed approximately 500 meters from shore directly behind the sea World Marine Science and Conservation Center. The net was checked at 2 to 8 hour intervals o ver periods of from 20 to 43 hours during November 1 9 82, and January, March, May and July 1983. In Tampa Bay a 175 meter 7. 6 e m stretch mesh gill net was employed. The net was c h e c ked at 1 to 2 hour intervals during daylight hours over periods.of from 1 to 7 hours. These collections were made monthly from May 1983 to December 1984 (excluding December 1983 and January 1984) The number of animals captured and the time period over which the net was fish ed was used to calculate catc h rates. All catch data were converted to catc h per unit effort, i.e. catch/100 meter gill net/hour. Results A total of 1171 swimming events were recorded during this study. In Figures 14 and 15 the average swimming velocity of bonnethead shark s is plotted against time of day for sharks of from 65.9 to 77.2 e m total length to minimize the effec t of size on velocity. Velo cities were d etermined for animals swimming freely in the semi-natural environment at sea World and those contained in the respirometer. Using a polynomial smoothing technique a line was fit to these data.

PAGE 54

u -E u > !:: 0 0 ..J w > 80 70 85 72 20 0 1 4-5 12 22 3.4 8-9 12-13 TIME (hours) 39 mean I range 0 95%confi dence interval N=680 52 .47 27 16-17 20-21 24-0 Figure 14. Changes in the swimming velocity (em/sec) of the bonnethead shark Sphyrna tiburo over a 24 hour period.

PAGE 55

85 72 -63 u a. (I) 29 :c 32 > t: 0 0 6 0 ..J w > 0.3 0-1 4-5 12 mean I range D95%confidence interval N=680 22 52 8-9 34 12-13 TIME (hours) 27 16-17 4 I 61 20-21 40 47 24-0 Figure 15. Changes in the swimming velocity (bl/sec) of the bonnethead shark Sphyrna tiburo over a 24 hour period.

PAGE 56

41 Swimming velocities of these animals varied from about 25 em/sec (0.37 body lengths/sec) to about 78 em/sec (1.0 bl/sec). Peak velocities were recorded during the period 0800 to 1000 hours with perhaps a secondary peak during the period 1900 to 2000 hours (Figures 14 and 15). Velocities were averaged over each hour and then crepuscular hours tested against the rest of the day and daylight hours tested against nighttime hours. Using analysis of variance no significant difference (p= 0.05; F= 0.736; df=15) was found between average swimming velocities recorded during crepusclar hours (ca. 0500 to 0800 and 1800 to 2100 hours) and those recorded during the rest of the day. However, average velocities recorded during the 0500 to 1000 hour period were significantly (p= 0.05; F:c 11.02; df= 15) higher than during the rest of the day. Over the entire 24 hour period an average swimming velocity of about 40.4 em/sec was observed. This corresponds to about 0 5 to 0 6 b l/sec for sharks ranging in size from 65.9 to 77 2 em total length. Average catch rates of Florida Bay bonnethead sharks are plotted in Figure 16. Catch varied from about 0.01 to 0.55 sharks per hour. The figure suggests a bimodal relationship. Catch was highest during the 0600 to 0700 hour period and the 2000 to 2200 hour period, but remained high during all hours of the night. Catch rates during nightime hours (2000 to 0800) were significantly higher (p= 0 .05; F = 56.9; df= 1 ,22) than those recorded during the day (0800 to 2000). Lowest catches were recorded during midday hours ( ca 11 00 to 1 4 00 hours). The catch rate of Tampa Bay sharks was also examined. Catch rates ranged from 0. 30 to 4. 7 sharks/hour. Although sampling was limited to daylight hours, the data suggest a similar bimodal

PAGE 57

42 0 55 .... 7 6 7 7 0 .44 -... z; ., 0 33 ... ca .c '-r 1\ 1\ .I. rv v 8 en ::r: 0 2 2 0 0 0 .11 6 ./, 6 4 r--,z_ / \. iS r--/ 4 8 12 16 20 24 TIME (hours> Figure 16. Avera g e catch rates of bonnethead sharks taken in Florida Bay over a 24 hour period.

PAGE 58

43 relationship (Figure 17). Catch was highest during the early morning and late afternoon (3.5 sharks/hour and 4.7 sharks/hour, respectively) even though these hours were sampled the least. The mid-day hours were sampled most heavily, yet produced the fewest sharks. Velocities of individual animals recorded during respirometry determinations were examined using analysis of variance and a statistically significant (p<0.05; F= 11258.7; df= 640) relationship between fish length and volitional swimming speed was found (Figure 18). The data indicate that over the 34 to 95 ern size range, bonnethead sharks swim at about 19 to 67 em/sec. The nonlinear regression equation; U = 4 9L0.496 0 ( 1 ) where U0= volitional swimming speed in em/sec and L= total length in centimeters, was fit to the data (r= 0.972). This equation now permits the estimation of swimming velocity from total length. Discussion The activity patterns and swimming velocities of fishes may be profoundly affected by a number of environmental and physiological variables. Few studies have quantitatively examined swimming velocities and rhythmic activity in sharks. The results of this paper allow inferences to be made as to how the bonnethead shark activity pattern varies with time of d ay and also the manner in which body size affects swimming velocity. The swimming velocity analysis for the bonnethead shark (Figures 14 and 15) suggest a peak of activity during early to mid-morning and perhaps during late evening. This type of

PAGE 59

44 5 .71 -' 2 4 57 ........., --... .1:! ., 12 3 .43 ... fca .1:! U) :c 2 29 0 6 ..--0 1.14 21 ---., 1. 20 I Tf I I 4 8 1 2 16 20 2 4 TIME (hours) Figure 17. Average catch rates of bonnethead shark s taken in Tampa Bay during the per iod 0800 t o 1700

PAGE 60

u ., -E u -0 ::l. 70 30 45 58 56 62 130 42 Uo = 4 9 L 0.45 20 -----------40-----this study 65 19 48 40 Uo= 5 .7L0.45 Weihs, 1977 50 60 ----N=641 70 80 TOTAL LENGTH (em) 90 100 Figure 18. Tile relationship between voluntary swimming speed (U0 ) and total length for the bonnethead shark.

PAGE 61

46 activity pattern is suggestive of a crepuscular species although the pattern of activity was not statistically significant. Hobson (1968) reported that Carcharhinus sp. are characterized by crepuscular feeding activities. Reynolds and Casterlin (1979) found highest activity levels at dawn and dusk, and lowest at midday for the yellow perch, Perea flavescens. Additional support for a diel pattern of rhythmicity in the bonnethead was provided by examining catch rates of sharks over 24 hours. Figure 16 suggests a similar bimodal relationship with catch rates highest at dawn and dusk and lowest during midday. Catch rates, significantly higher at night, suggest a nocturnal activity pattern. The high catch rates during early morning and early evening when light levels were high suggest that net avoidance was not a significant factor in the above observed catch rates. Tampa Bay catch rates also demonstrated a decrease during midday. Catch rates in Tampa Bay were alrrost 10 times the rate observed in Florida Bay which reflects a greater availability of sharks in Tampa Bay. In their examination of the behavior of the bonnethead shark, Myrberg and Gruber (1974) were able to obtain swimming velocity data during daylight hours. An apparent increase in velocity with a peak late in the day suggested to them a diurnal rhythmicity. The results of examining catch rates of the bonnethead shark suggest that this shark is predominately nocturnal. Not surprisingly, the swimming velocity analysis failed to demonstrate a clear cyclicity. This may be explained by the fact that in some cases velocities for a particular hour of the day were determined during different times of the year and then pooled. Seasonal differences in daylength and temperature may

PAGE 62

47 have partially obscured the activity pattern. Although the range of temperatures over which the velocities were determined was not great, temperatures' effect on volitional swimming speeds of sharks has not been examined. Dizon et al. (1977) reports that the volitional swimming speeds of Katsuwonus pelamis and Euthynnus affinis are temperature independent within mst of their zone of thermal tolerance, most likely because of these fishes ability to retain heat and thus thermoregulate. However, Schaefer (1986) found a significant positive relationship between temperature and swimming speed within the range of thermal tolerance for Scomber japonicus. Considering the narrow temperature range of this study, it is probable that there was little temperature effect. The above findings are not unique in that a number of shark species have been suggested to be nocturnal. Springer (1963) observed that tiger sharks, Galeocerdo cuvieri, feed more often at night than during the day. Limbaugh (1963) found that nurse sharks, Ginglymostoma cirratum, secreted in coral caves during daylight abandoned this habit at nightfall. More quantitative evidence for a nocturnal activity pattern was provided by Nelson and Johnson (1970). They found that the horn shark, Heterodontus franscisci, and Cephaloscyllium ventriosum, both nocturnal, the swell shark, differed in that the former exhibited an exogenous and the latter an endogenous rhythm. In Gru bers' ( 1984) comprehensive study of the lemon shark, Negaprion brevirostris, there was reported an increase in activity and oxygen consumption at night. Particularly relevant to this study are Hobsons' (1968) observations on predatory behavior of shore fishes. He found that

PAGE 63

48 nocturnal predatory teleosts feed primarily on "small motile invertebrates, particularly crustaceans" which are also nocturnal. The bonnethead shark feeds almost exclusively on blue crabs, Callinectes sapidus, in Tampa Bay and cephalopods, lobster, and crabs in Florida Bay (see Chapter 6). These invertebrates are nocturnal which suggests that the bonnethead sharks activity pattern is feeding motivated. Over the entire 24 hour period, bonnethead sharks ranging in size from 65.9 to 77.2 em total length voluntarily swam at about 40 em/sec or 0.5 to 0.6 bl/sec. Webb and Keyes (1982) examined the swimming kinematics of several shark species and reported that bonnethead sharks of 84 to 102 em total length swam at a mean velocity of 0.84 0 .11 bl/sec. A significant relationship between swimming speed and body size was found in this study. Figure 18 demonstrates that voluntarily swimming speed (U0 ) increases with body Weihs (1977, 1981) developed a predictive equation for the relationship between voluntary swimming speed and size based upon an "energetic cost optimization." His equation was developed upon the hypothesis that voluntarily swimming will be at a speed which "requires minimum energy per unit distance traversed which occurs when the total metabolic rate is approximately double the standard rate." This predictive approach has been corroborated in both fresh water and salt water species. Brett and Blackburn (1978) reported that the routine metabolic rate of the dogfish, Squalus acanthias is twice the standard metabolic rate. Weihs (1981) found that the voluntary swimming velocities of the sandbar shark, Carcharhinus plumbeus and the bull shark, C. leucas could also be predicted from this energetic approach. In examining

PAGE 64

49 Figure 18, it is evident that Weihs (1977) equation and that generated from bonnethead shark swimming velocities are in close agreement. This suggests that the bonnethead shark swims at an optimal velocity and perhaps does not vary its activity pattern over a 24 hour period. With this equation it is now possible to estimate voluntary swimming speed for any size bonnethead shark. Additionally, an estimate of standard metabolic rate may be made from routine metabolism using Weihs' (1981) approach (see Chapter 6). This information is of critical importance in modeling the energy budget of the bonnethead shark. It is possible that the bonnethead shark does not alter its swimming velocity over a 24 hour period. This would explain the failure to demonstrate a clear cyclic pattern in swimming velocity and would also explain the close agreement between the theoretically determined optimum velocity (Weihs 1977) and the observed velocity. If this is the case, the differences in catch rate between night and day would indicate a cyclic pattern of habitat utilization rather than an increase in activity. Conclusions From this study it may be conc luded that the bonnethead shark is nocturnal but may have a crepuscular component of activity superimposed onto this pattern. Catches of bonnethead sharks are high at night, peak during dawn and dusk and are lowest during midday. The bonnethead shark swims at a velocity that can be predicted based on energetic cost optimization theory (Weihs 1977, 1981). I conclude that the bonnethead shark swims at optimu m velocity during the majority of

PAGE 65

50 the bonnethead shark swims at optimum velocity during the majority of the day and night and only occasionally increases above this value. This would explain the failure in this study to detect a clear cut cyclic pattern of swimming activity. The cyclic pattern in catch rate would therefore not indicate an increase i n swimming velocity but a cyclic pattern of habitat utilization. The bonnethead shark cruises (optimally) onto the feeding grounds (the shallow grass flats) at dusk, where it spends the night-time hours searching for prey and then leaves at dawn. Only during brief encounters with predators, during mating and during actual attacks upon prey items would there result in an increase above this optimal velocity. This means that routine metabolism (see Chapter 6) may be a reasonable estimate of the general level of metabolism in nature. Predators are often classified into two types; the si t-'and-wai t predator and the energy speculator. Continuously active predacious fishes are considered energy speculators. They are betting that a large energy investment in activity will reap a larger amount of energy in return. By remaining highly active, they increase their encounter rate with prey items that are likewise highly active and often patchy in distribution. Once these animals locate prey items an additional energy investment is required for chase and capture. The bonnethead shark may represent a third predatory type. These sharks are feeding on benthic prey items that are often very abundant, certainly more limited in their mobility and are perhaps more uniformly distributed. Rather than engage in high levels of activity, a constant optimal cruising velocity wherein distance covered is maximized would result in the greatest energy rewards These sharks

PAGE 66

51 and perhaps many other species might rrore appropriately be termed "predatory grazers".

PAGE 67

CHAPTER FOUR: AN EXAMINATION OF THE GROWTH INHIBITORY EFFECT OF TAGGING AND TETRACYCLINE TREATING BONNETHEAD SHARKS, Sphyrna tiburo Introduction 52 Accurate age determinations are a prerequisite for growth studies, mortality determinations and for assessing general population dynamics. Beamish and McFarlane (1983) pointed out how non-validation of the aging method in two species has resulted in considerable management errors. Tetracycline has become the method of choice for validating aging techniques in fishes. This drug incorporates in tissues where active calcification is occurring (otoliths, spines, ve rtebrae, etc.) producing a reference mark reflecting the time of drug administration. When the tissue is viewed under UV light, the number of cyclic structures (daily rings, annuli) which are formed beyond that reference mark can be compared to the known elapsed time to validate the time increment. Weber and Ridgway ( 1962) first used the method for marking the bones and scales of fish. With the advent of the tetracycline method and the demonstration of daily ring formation in the otoliths of larval fishes (Pannella, 1971), a number of studies have appeared in which this method was used to validate daily ring formation. Campana and Neilson (1982) examined daily growth increments in otoliths of the starry flounder,

PAGE 68

53 Platichthys stellatus, using tetracycline marking Ralston and Miyamoto (1983) did the same for juvenile Hawaiian snapper, Pristipomoides filamentosus. Peters and McMichael (In. Prep.) used tetracycline to validate daily growth increments in black drum, Pogonias cromis, and Parsons and Peters (In Review) did the same for the sheepshead, Archosargus probatocephalus. The tetracycline method has also been used for validating otolith annuli. Wild and Foreman (1980) examined the relationship between otolith increments and time in yellowfin and skipjack tuna using tetracycline. Beamish et al. (1983) used tetracycline to validate age determination for sablefish, Anoplopoma fimbria. To a limited extent, tetracycline has also been applied to the question of age validation in elasmobranchs. Holden and Vince (1973) first applied the method to an elasmobranch when they validated the vertebral ring method in the thornback ray, Raja clavata. Gruber and Stout (1983) used tetracycline to examine age and growth in the lemon shark, Negaprion brevirostris. Smith ( 19 84) validated the vertebral band method of age determination in the leopard shark, Triakis semifasciata, using tetracycline injections. Beamish and McFarlane ( 1985) examined annulus development in the second dorsal spine of spiny dogfish, Squalus acanthias, via the tetracycline method. Finally, tetracycline was used to validate the vertebral ring method in the bonnethead shark, Sphyrna tiburo (see Chapter 5). Many researchers have been eager to apply the tetracycline technique to age validation. Unfortunately, the effect that heavy doses of tetracycline have on the growth of fish, has not been examined. In this study, tetracycline was administered to bonnethead

PAGE 69

54 sharks, Sphyrna tiburo in doses typically used for validation studies. The efficacy for marking the vertebrae was ascertained. More importantly, the effect of tetracycline and tagging on growth was monitored. Materials and Methods Tetracycline experiment I. Litters of bonnethead pups born in August 1984 at the Sea World Marine Science and Conservation Center in the Florida Keys were utilized for this experiment. The animals were maintained in a shallow enclosure approximately 15 meters long, 5 meters wide and 1 meter deep. Water was supplied to the enclosure by aerating and filtering water pumped directly from Florida Bay. Water quality parameters were monitored on a daily basis throughout the study. The parameters examined were temperature, dissolved oxygen, salinity, pH and ammonia. The animals were fed to satiation a diet of capelin (Mallotus villosus), squid (Loligo sp.) and shrimp (Penaeus sp.) on a daily basis. On 14 December 1984, nine pups randomly selected were weighed to the nearest 5 grams, measured to the nearest millimeter (total length, TL) and then tagged using Dennison tags inserted in the dorsal fin. The animals were divided into three groups of three sharks each. One group received 1 o mg/kg body weight oxytetracycline hydrochloride (Medamycin MED-TECH INC.), one received 50 mg/kg and one received 75 mg/kg. The drug was administered intra-peritoneally using cc tuberculin syringes with 25 gage needles. The experimental animals were returned to the captive environment and maintained as a group

PAGE 70

55 along with approximately 25 other pups, all born at the same time, that were not tagged or injected. On 23 March, 1985, 99 days later all pups were again weighed and measured, and representatives from each group sacrificed for vertebral examination. The vertebrae were frozen for storage, sectioned using a Beuhler Isanet low speed saw, and viewed at 10 to 100x magnification using fluorescence microscopy. Tetracycline experiment II. The animals used for this experiment came from litters of sharks born at the Sea World Center in August 1985. The animals were maintained in captivity as previously described. On 1 3 September, 1985, 30 pups of similar size were weighed to the nearest 0. 1 gram using an Ohaus top-loader balance, measured (total length) to the nearest 0.1 mm, and tagged using Dennison tags. The animals were divided into three groups of ten sharks each such that males and females were equally represented within each group. The first group received an injection of 0 mg/ k g tetracycline, the second an injection of 12.5 mg/kg and the third, 2 5 mg/ kg. The drug solution was made by dissolving powdered oxytetracycline hydrochloride in a normal saline solution to produce a 25 mg/ml concentration. This solution was then administered to both experimental groups by intra-muscular injection (in the region of the dorsal musculature) using 1 cc tuberculin syringes with a 25 gage needle. The control group received injections of normal saline alone. Before injection, the volume in each syringe was adjusted such that each animal received the same volume (0 .2 cc). The animals were returned to the captive environment and maintained as a group along with approximately 20 other pups, all of the same age that received no tag or tetracycline. On 19 December, 97 days later, all pups were again weighed, measured,

PAGE 71

56 and representatives from each group sacrificed for samples of vertebrae. The vertebrae were processed as in experiment I. Results Marking effectiveness. One animal from each of the 25, 50, and 75 mg/kg groups was sacrificed and the vertebrae were found to possess very distinct marks when viewed under UV light. One animal that received 10 mg/kg was sacrificed and was found to possess a very faint tetracycline mark. Effect on growth Table 3 shows the results of tetracycline experiment I The treatment groups originally started out with 3 animals in each group but several tags were shed after 99 days Analysis of variance demonstrated that there was no statistical (p>0.05) difference between the mean total lengths or mean weights of the treatment groups at the beginning or end of the experiment (Table 3, columns 2 through 4) The most interesting result shown in Table 3 is the apparent decrease in growth with increasing tetracycline dose. Examining the last 4 columns in the table it is seen that among the animals receiving tetracycline, the sharks receiving the lowest dose (10 mg/kg) grew the fastest, and growth rate decreased with increasing tetracycline. The sharks receiving 75 mg/kg showed the slowest growth both in terms of length and weight However, analysis of variance demonstrated there was no statistical difference (p>0.05) between the 10, 50 and 75 mg/kg tetracycline groups The data from these groups were therefore pooled and a mean for all the drug treated animals obtained. The last row in Table 3 shows growth characteristics for the animals that were neither

PAGE 72

Table 3. Tetracycline I This examined the effect of 1 0, 50 and 75 mg/kg injections of tetracycline on of the bonnethead Unless indicated, in deviations. Tetracycline Dose (mg/kg) 10 (n-2) 50 (n-3) 75 (n) No tetra-cycline tag. Size at Injection Mean Total Mean Weight -Length (em) (gm) 39.5 230 (sa7,3) (s170) 41.3 277 (s-3.5) (ss73 7) 44.0 290 (s= 1.7) (s=14.1) 38. 8 219 (s-3.89) (s=72.7) (n=35) ( !1_=<15_}_ Size 99 Days Mean Total Mean Weight Mean --Length (em) (gm) Rate (em/d) 43. 7 340 0 043 (s-7.8) (sa 163) (s0.005) 45.1 343 0.038 (s=3 5) ( s=85.8) (s-0.004) 46.4 320 0.024 (ss0,64) (s-21 .2) (s-0.011) 45.6 339 0.069 (s=5.13) (ss124 7) (n=25) (n=25) -Mean Rate (gm/d) 1. 11 (s-0.07) 0.66 (s-0. 19) 0.30 (s-0.07) 1.21 Mean %6L 10. 9 (s=0.8) -9.1 (s=1.4 ) 5.4 (5=2. 6) 17. 5 Mean %t.W 68. 0 (s-53) 23.8 (s.0) 10.3 (s-2.0) 54. 8 \.11

PAGE 73

58 tagged nor treated with the drug. Since no tag was utilized, individual growth characteristics could not be determined and a standard deviation could not be calculated for these values. These values were therefore treated as single observations and were tested against the mean obtained by pooling the drug treated groups using a modified t-test. This t-test compares a single observation with the mean of a sample (Sokal and Rolf, 1981). Growth in length and the percentage change in length were found to be statistically higher (p<0.05) among the sharks receiving no tag and no injection of tetracycline, than those tagged and injected. However, perhaps due to the high variability associated with weight measurements, there was no statistical difference among the groups when growth in weight or percentage change in weight was examined. The results of tetracycline experiment II are shown in Table 4. A gain, analysis of variance revealed there was no significant difference in length or weight between the groups at the beginning of the experiment. After 97 days, the 0, 12.5 and 25 mg/kg groups were tested for differences in mean total length, mean weight, and growth characteristics (Table 4, columns 4 to 9) using analysis of variance. No statistically significant differences (p > 0.05) were observed. Again, this allowed the treatment groups to be pooled into one data set. This pooled data set could then be compared against the growth data for animals that received no tetracycline and were not tagged. A Students' was used to test for differences between the means of the pooled treatment groups and the group that received no tag or treatment. The sharks that were not tagged or treated were significantly (p < 0.05) larger in weight but not in length. When growth

PAGE 74

Table 4. Tetracyc line experiment II. This experiment examined the effect of 0 12. 5 and 25 mg/kg injections of tetracycline on growth of the bonnethead shark Unless otherwise indicated, numbers in parentheses are standar d dev iations. Tetracycline Dose (mg/kg) 0 (n 6) 1 2 5 (n 8) 25 (n=8) No tetra-cycline or tag. Size at Injection Mean Total Mean Weight Length (e m ) (gm) 30.6 1 03. 4 (s-1 .51) (s=17 4) 30.2 107. 2 (s=l .5) (s=19 9) 31.5 120.6 (s= l 6) (s= 18.2) 30. 5 108.0 (sal .66) (s=19 6) (n=30) (n 30) S i ze After 97 Days Mean Tot a l Mean Weight M ea n G rowth Length (em ) (gm) Rate (em/d) 37.2 204. 3 0 068 (s=4.4) (s=63 3) (s-0. 034) 38.3 216. 2 0 .083 (s=3 2) (s=57. 9) (Sz0 029) 39.9 240.9 0 088 (s=3 3) (s=51 8) (s=0 025) 40.5 260.9 o. 10 ( s=2. 76) ( s=58 1) (n=19) (n=19) G rowth Characteristic s Mean Growth Rate (gm/d) 1.04 (s-0. 06) 1.1 2 (s .. 0.45) 1.24 Mean %L.lL 21.6 (s0.1) 26. 8 (s-0. 08) 26. 7 Mean %6W 97. 7 101 5 (S=0. 4) 99.8 (s-0.48) (s=0 ; 08) (s-0. 34) 1.58 32. 8 157. 6 V1 \0

PAGE 75

60 t"ate and pet"centage change in length (Table 4, colurms 6 to 9) were tested as previously described, the untagged, untreated group had a significantly higher' growth rate in terms of length, and the pet"centage change in length and weight was also higher. Mean gr'owth rate in terms of weight for' the untreated, untagged animals was not significantly different (p>0.05) than the tagged, drug-treated animals. Discussion The increasing popularity of the dr'Ug tetracycline as a means for' validating age deter'mination in fishes has necessitated an examination of its effect on growth. To date, only two studies have pr'ovided information concet"ning this affect. Weber' and Ridgway ( 1 967) reported that Pacific salrocm, Oncor'hynchus sp. fed a diet containing oxytetracycline (2 g of tetracycline per' kilogram body weight) had sur'vival t"ates and growth t"ates similar' to untreated fish. However', fish that were tagged (adipose fin clip) and given oxytetr'acycline wet"e smaller' than those that received only clips or only tetracycline. Since their' study repot"ted no statistics associated with their survival and gr'owth t"ates, it is not clear' how they ar't"ived at their conclusion. Beamish et al. ( 1 983) found that an oxytetracycline dose of 1 00 mg/kg caused inct"eased mot"tali ty of tagged and released sablefish, Anoplopoma fimbt"ia. They observed that tagged fish given 100 mg/kg had a significantly lower' r'ecapture r'ate than fish given no injection. They also found that this level of drug significantly t"educed growth in the wild.

PAGE 76

61 The results of this paper allow us to make several conclusions concerning the use of oxytetracycline for age validation in the bonnethead shark. Doses of 25 mg/kg or greater appear to be best for providing a detectable mark. Although only a single animal in each treatment group was the animal receiving 10 mg/kg produced a very faint mark. The permanence of this mark over a long period of time would be in doubt. Beamish et al. (1983) found that 50 mg/kg gave the highest percentage of marks on otoliths of sablefish. Gruber and Stout (1983) reported that dosage in the range of 5 to 100 mg/kg was sufficient for providing a distinguishable mark in the vertebrae of lemon sharks, Negaprion brevirostris. The data in Table 3 suggest that bonnethead shark growth is inhibited by the 50 and 75 mg/kg doses of tetracycline. However, perhaps due to the very small sample sizes in this experiment, there was no statistically significant difference between the treatment groups. Similarly, there was no significant difference in growth between the treatment groups in Table 4. The most significant factor affecting growth of bonnethead sharks in these experiments was the presence or absence of a tag. In Tables 3 and 4, animals that received no tag and no tetracycline had higher growth rates and greater percentage change in length and weight than tagged-treated animals. The statistical significance of these differences was tested by treating the growth data as individual observations. This has inherent difficulties since these values are actually calculated from the average lengths of untagged animals. Since these values have no associated standard deviations they could not be tested by the usual methods. Although this approach may not be statistically appropriate,

PAGE 77

62 it at least provides the reader with a method for comparing these data. The average weight of untagged, untreated animals in Table 4 was significantly higher than the tagged treated group. This difference was tested using a t-test that compares two means since the size data does have associated standard deviations. However, the mean total lengths in Table 4 and both mean lengths and weights in Table 3 were not significantly different. The failure to demonstrate differences in mean lengths and weights in Table 3 is probably due to the small sample sizes. Total length is perhaps a less sensitive indicator than weight which may explain why mean lengths in Table 4 were not significantly different. Conclusions The above experiments suggest that tetracycline doses as high as 25 mg/kg do not impair growth of the bonnethead shark. However, there may be an inhibition of growth at doses of 50 mg/kg and higher. Interestingly, Beamish et al. (1983) tested the e ffect of a dosage of 100 mg/kg and 25 mg/kg on mortality and growth of sablefish in the laboratory and found no significant difference from controls. He concluded that the laboratory experiments were not comparable to the natural situation since fish injected with the same drug dosage, tagged, and released, suffered higher mortality and slower growth than those not drug treated. The tag used to identify individuals in the present study significantly inhibited growth in weight. The tag may result in increased drag, a decrease in the swimming efficiency and a resulting greater energy investment in swimming. More energy invested

PAGE 78

63 in swimming would mean less available for growth. The trauma o f being captured, handled and injected could also impair growth although this would likely be of relatively short duration. In future experiments of this type, it is suggested that flourescent dyes applied to the outside of the fish be used to.tag these animals. This method has been used suc cessfully in short term experiments to mark bony fishes and would likely serve well in this application. The presumed problem of excessive drag would be eliminated with these dyes.

PAGE 79

64 CHAPTER FIVE: AGE DETERMINATION AND GROWTH Introduction For most Gulf of Mexico shark species, growth rates, longevity and general population characteristics have never been adequately examined. The accumulation of this information has been severely impeded by the absence of a reliable age indicator in these animals. There is at present no easy technique for aging elasmobranchs. This problem has been circumvented in a number of ways. Accurate age and growth data have been obtained by mark and recapture methods (Kato and Carvallo, 1967; Thorson and Lacy, 1982). Parsons (1985) estimated the parturition to maturation growth curve for the Atlantic sharpnose shark, Rhizoprionodon terraenovae using collection data to separate discrete age groups and size classes. Gruber (1981) measured growth rates of captive lemon sharks and used these data to construct a simulated growth curve. Holden (1974) modified the von Bertalanffy growth equation and developed a method maturation for elasmobranchs. His method for determining age at assumes pre-natal and post-natal growth characteristics are equal and allows the estimation of age at maturity. Francis (1981) applied the method to Mustelus species and calculated growth rates and ages at maturity for seven species. He concluded that Holden's method gives von Bertalanffy growth parameter estimates that are good enough for preliminary

PAGE 80

65 fishery management purposes. However, Parsons ( 1985) reported that Holden's method considerably overestimates growth and may not be valid for estimating age at maturity for larger shark species. Analysis of growth rings appearing on the vertebral centra holds promise as a technique for aging sharks. However, validation of the technique is required in each species examined. In a timely report, Beamish and McFarlane (1983) found that in 75 studies published between 1965 and 1980, in which fish age determinations were used, only 40% attempted age validation and none validated all age groups included in the study. They also outlined case histories of how non-validation in two species resulted in considerable management errors. Although much work has yet to be done, there is mounting evidence supporting the annual nature of vertebral rings in sharks. Holden and Vince (1973) used tetracycline injections, to show that North Sea populations of the thornback ray, Raja clavata, exhibit annual rings. Tetracycline, when injected into an organism, is adsorbed in areas where calcification is occurring. Therefore, a shark injected, tagged, and released, will carry a marker reflecting the time of injection. When these animals are recaptured, the excised vertebrae can be viewed under ultraviolet light and the position of the tetracycline marker noted. If sufficient time has passed, the frequency of ring formation may be determined and direct validation obtained. Stevens (1975) compared vertebral ring age estimates of blue shark, Prionace glauca with Aasen's (1966) Petersen distribution age estimates and found almost identical values for age one through four. Thorson and Lacy (1982) assumed the rings to be annual and estimated

PAGE 81

66 age and growth parameters in the bull shark, Carcharhinus leucas. Parsons (1983) compared age estimates from vertebral rings with estimates from length-frequency data and found good agreement for Age Groups 0 and 1 in the sharpnose shark, Rhizoprionodon terraenovae. Pratt and Casey (1983) hypothesized that two rings may be formed on the vertebral centrum each year in the shortfin rnako, Isurus oxyrinchus. However, vertebrae from four tagged/recaptured makos gave inconclusive results. As yet, the only reports to validate the annual nature of the vertebral rings is Smith's (1984) work on the leopard shark, Triakis semifasciata, and Gruber's (1984) work on the lemon shark, Negaprion brevirostris. McFarlane and Beamish (1987) validated the dorsal spine method of age determination for the spiny dogfish, Squalus acanthias. In this report, the suitability of the vertebral ring method of age determination in the bonnethead Sphyrna tiburo, is ascertained. Growth determined from the vertebral ring method is compared with growth of captive animals, of tagged/recaptured animals and with growth determined from back-calculated length at age. Geographic variability in growth is also examined. Materials and Methods From July 1982 until December 1986 bonnethead sharks, Sphyrna tiburo were collected from Tampa Bay at 27'N, 82'W and from Florida Bay at 24'N, 81'W (Figure 1 ). The majority of sharks were collected over grass flats using gill nets fished continuously for as long as 48 hours. The nets ranged from 175 to 350 meters in

PAGE 82

67 length and were 7 ern stretch mesh. Nets were checked at 1 hour intervals for live specimens and up to 8 hour intervals when specimens were needed for dissection. Validation of aging technique. The frequency of vertebral ring formation was examined using the antibiotic tetracycline. From 3 July 1983 until 19 May 1985, 70 bonnethead sharks were tagged, injected, and released into Tampa Bay. Sharks captured in good condition were tagged using Dalton rototags inserted in the dorsal fin, injected with oxytetracycline HCl (intra peritoneally), sexed, and measured (total, fork, and standard lengths). The message on each rototag instructed the f isherrnan to save the shark and contact the university. Tetracycline dosage ranged from 20 to 50 rng/kg and was estimated using a length-weight realtionship. When sharks were recaptured, they were weighed and their total length measured. Precaudal vertebrae numbers 25 to 40 were consistently removed. These sections of vertebral column were cleaned of all tissue and frozen. One vertebra was removed from each section of vertebral column, generally numbers 38, 39 or 40. These vertebrae were sectioned at approximately 500 urn using a Buehler Isornet low speed saw and the sections mounted on microscope slides using histomount. The sections were viewed using fluorescence microscopy at 25 to 100x magnification. By alternately viewing the sections under UV and visible light the position of the tetracycline mark relative to the vertebral rings could be determined (Figure 19). The radius of each ring, the radius of the tetracycline mark, and the vertebral radius were each measured to the nearest 0.1 rnm using an ocular micrometer. Maintenance of animals in captivity at the Sea World Marine

PAGE 83

68 Figure 19. A photomicrograph of a sectioned vertebral centrum from a bonnethead shark previously injected with tetracycline. The fluorescent label is clearly visible.

PAGE 84

69 Science and Conservation Center also provided the opportunity for age validation (see Chapter 2 for a complete description of the maintenance of these animals in captivity). Bonnethead sharks were measured, weighed, tagged, injected with tetracycline, and held in captivity from 9 August 1982 until 4 May 1986. Rototags were used for sharks greater than about 60 em total length and Dennison tags for smaller animals Animals born in captivity were also used in this study and provided the unique opportunity to examine the vertebral ring method in animals of known age Tetracycline dosage ranged from 1 2. 5 mg/kg to 1 00 mg/kg and was ad.minis tered both intraperi toneally and intramuscularly. Sections of vertebral column were collected at various time periods and processed in the manner previously described. Growth rate and age determination. The growth rate of the bonnethead shark was examined by three methods : (1) the examination of vertebral rings of wild-caught animals (2) maintenance of animals in captivity at the Sea World Marine Science and Conservation Center and (3) tagged released and recaptured animals Vertebrae harvested from wild caught animals were used to examine growth Vertebrae were cleaned and preserved in 1 O% formalin, 95% ethanol or f r ozen Formalin preserved specimens were later changed to 50% isopropyl alcohol. A single vertebra was removed from the section, allowed to dry for about 10 minutes, and placed in 5% sodium hypochlorite for 15 to 30 minutes depending on size. This procedure thoroughly cleaned the connective tissue from the face of the centrum. The vertebra was then washed in tap water for 30 minutes and stored in 50% isopropyl or 95% ethyl alcohol. Using Parson's (1983) microtopographic technique the vertebral rings were accentuated

PAGE 85

70 (Figure 20) The rings were counted using a dissecting microscope their radii measured in some cases, and the centrum radius measured. Ring counts were made independently by two different readers. When counts were not in agreement, those vertebrae were recounted in an effort to resolve the differences. If the differences could not be resolved, those vertebrae were excluded from the data set. Measurements of ring radii allowed the back-calculation of length at ring formation. Maintenance of animals in captivity at the Sea World Marine Science and Conservation Center from 20 July 1982 until 7 December 1986 provided a second method for the examination of growth rate . On each of 24 trips to the center, bonnethead sharks were weighed and measured (total, fork and standard lengths). Smal l individuals ranging from about 24 to 50 em were weighed to the nearest 0.1 g and larger individuals were weighed to the nearest 10 g All length measurements were made to the nearest 0. 1 em. Tagged individuals allowed the examination of individual growth rates. Litters of sharks, born in captivity, were segregated from other animals and average growth rates were obtained for the entire g roup The final method used for examining growth rate was tag and release of wild-caught animals. The tagging method has already been described in the age validation section. The length and weight of recaptured animals provided growth of animals in the wild. Results Validation of aging technique. Seventy bonnethead sharks were tagged,

PAGE 86

71 Figure 20. A picture of a bonnethead shark vertebrae in which the vertebral rings have been accentuated using the penciling technique.

PAGE 87

72 tetracycline injected, and released into Tampa Bay. Five animals were recaptured. A tetracycline reference mark could not be located on the vertebrae of four of these animals. One animal showed a distinct tetracycline mark. The animal was tagged and released on 17 November 1984 and recaptured on 29 November 1985. Upon sectioning and examining a vertebra from this animal using ultraviolet microscopy, the tetracycline reference mark was located and its position in relation to the vertebral rings was determined. The reference mark formed in close association with ring #4, and ring #5 was just forming when the animal was recaptured. Maintenance of animals in captivity provided a great deal more information concerning the validity of the vertebral ring method of aging. Litters of pups were born in August of 1983, 1984 and 1985 (Table 5). Individuals from these litters were sacrificed at various time periods and their vertebrae examined for evidence of ring formation. Table 5 demonstrates that animals held in captivity for one year had two vertebral rings and those held for two years had three rings. In one case (tag #7179) a single ring formed over approximately one year. One animal (tag /17848) sacrific ed shortly after birth but before its first winter possessed a single ring. One and two year old animals sacrificed in early December of 1986 when water temperatures were still high (ca. 25C) were scored as 2+ and 3+ suggesting that formation of rings 3 and 4 was imminent. Mean marginal increment (Table 5) was calculated for 5, 1 year old animals sacrificed in June and 5, 1+ year old animals sacrificed in December. The mean marginal increment was 1.34 mm in December and 0.76 in June. Tetracycline injection of wild-caught animals admitted into

PAGE 88

Table 5. The results of examining the vertebral rings of sharks held in captivity for periods of from 0.3 to 2.3 years. Tag U Date of Size at Date of Size at Known Vertebral Ring Radii (mm) Vertebral Mean Marginal Birth Birth Sacrfice Sacrifice Age Ring U 1 2 3 Radius (mm) Increment -721l7 Aug. 85 27.5 12/06/86 63.1l 1.3 2+ 2.1 21._ 5.5 --7877 " It 59.2 1.3 2+ 2.0 _1&._ ll.7 -1.31l 7880 It It It 65.1 1.3 2+ 2.1 .2!2_ 5.1l n --" It It 62.1l 1.3 2+ 1.9 3.6 5.0 7882 It It It 57.1 1.3 2+ 2.0 ll.2 --261 Aug. 81l 27. 2 09/13/85 66. 9 1.08 2 1.7 3-3 5.6 1.7 n 262 " It 51l.5 1.08 2 2.1 3.3 ll.3 71lll3 08/25/83 27.5 06/29/81l 53.9 0.85 2 1.7 3-3 ll.2 7179 It 27.5 It 48.0 0.85 1 1.7 3-3 7178 27.5 53.0 0.85 2 1.9 3.3 lj. 1 0.76 n-5 7441 27.5 53.2 0.85 2 2.5 3.8 ll.2 71l40 25.8 60.Il 0.85 2 2.5 3.6 4.6 71l38 27.5 60.0 0.85 2 2.5 3.3 ll.O 7848 Aug. 85 30.7 11/23/85 34.2 0.27 1 2.3 -2.5 -7233 27.2 03/24/85 49.2 0.60 2 1.6 2.7 --271 27.2 12/06/86 72.2 2.31 3+ 2.5 3.9 6.0 6.7 272 It 27.2 89.1 2.31 3+ 2.5 4.1 6.1l 8.1 1 1 n 273 27.2 75.9 2.31 3+ 1.8 3.0 6.0 6.9 -..1 w

PAGE 89

74 captivity and newborn animals also provided the means for examining vertebral ring formation. By examining Table 6 the position of the tetracycline reference mark in relation to the vertebral rings may be determined. In most cases the number of rings formed beyond the reference mark was an indicator of time elapsed. When the time elapsed extended through the winter, even if the period was less than one year, a ring formed (tag /17883 and 7221 ) When tetracycline was administered during winter months, the reference mark consistently formed in close association with a vertebral ring. Likewise, when the drug was administered soon after birth (tag #7886, 7883 and 7223) the mark was closely associated with the first vertebral ring. In two animals were the results contrary to the above. Animal #105 formed two rings beyond the reference mark when only one should have formed and #7223 apparently failed to form ring #1. Growth rate and age determination. The validation of the vertebral ring method of aging in the bonnethead shark allowed the construction of g rowth curves for wild caught animals. Vertebrae were harvested from 143 animals collected in Tampa Bay and 96 animals collected in Florida Bay (supplemented with animals held in captivity). The method and a nonlinear regression technique (see Vaughan and Kanciruk, 1981) were used to estimate the parameters of the von Bertalanffy ( 1938) growth equation. The generalized form of the von Bertalanffy equation and the resulting estimates are shown in Table 7. The estimates obtained using the nonlinear regression technique were used to construct growth curves. All growth curves were significant at p<0.05. Appendix 1 may be referred to for the standard deviations around the von Bertalanffy estimates. The von

PAGE 90

Table 6. The results of the study to validate the vertebral ring aging technique in the bonnethead shark. These animals were injected with tetracycline, held in captivity for 0.08 to 1.45 years and then examined for ring formation (Tetra.-tetracycline). Tag H Date of Date of Vertebral Vertebral Tetra. Sacrifice Ring # Radius Injection (mm) 7/20/!;2-" 102 1/15/83 10/6/83 4 7.3 1.7 101 7 /22/83 4 7.1 2.3 108 2 /17/843 /10/84 6 7.5 1.9 1 /15/!:S3-107 2/17/84 3 /10/84 6 7.7 2.1 3119/!;3-106 2/17/84 3/7/84 6 8.8 2.3 103 1 /15/83 7/22/83 6 8.2 2.1 105 1/15/83 6/29/84 6 7.7 2.3 7886 9/13/85 10/85 1 2.2 2.0 7883 9 / 13/85 5/4/86 2 3.6 1.7 7176 5 / 11/84 6/29/84 2 4.0 1.5 7221 12/14/84 3/24/85 1 2 9 1.7 7223 12/14/84 3/24/85 1 3.0 Ring Radii (mm) 2 3 4 5 2.9 4.8 6.3 3.3 4 6 5.9 2.9 4 2 6.1 6.7 3.3 4.8 6.2 6.7 3.4 5.6 7.5 8.2 3.4 4.8 6.1 6.5 3.6 5.0 6.3 7.1 3.0 3.1 2.8 2.6 6 7 3 7. 1 8.6 7 5 7.2 Radius of Tetra. Mark(s) (mm) 5.4/6.2 5 .2/5.9 7.3 6.7/7.4 8.2/8.6 7.5 6.3 1.9 1.7 3 5 2.7 2.5 Time E l apsed Between Tetra. Injection & Sacrifice 1.24 yrs. 1.01 tl 0 .06 1.15 1.03 tl 0.51 tl 1.45 0.08 tl 0.64 tl 0 .13 tl 0.27 0 .27 tl Rings F ormed Between Tetra. Injection & Sacrifice 1 1 0 1 1 0 2 1 I 1 I 0 1 1 -----.:1 \Jl

PAGE 91

76 Bertalanffy curves plotted in Figures 21 and 22 show growth of Tampa Bay males and females, respectively. Comparing the figures it is evident that growth of males and females is similar through age 1+. At approximately age 2+ the curves diverge males grow little after age 3+ to 4+ (ca. 85 em TL) whereas females continue to increase in length until reaching a theoretical maximum size of about 1 1 5 em. Males in Tampa Bay apparently have a maximum age of about 6+ years whereas females were aged at 7+ years. Over the ages reported here, males grew at a rate of about 10 em/year and females at about 11 em/year. Similar comparisons may be made between males and females from Florida Bay (Figures 23 and 24) As before, a divergence of growth is seen at about age 2+. Males attain a theoretical maximum size of about 82 em TL growing very little after about age 4+ (Figure 23). Females continue to grow up to about age 5+ to 6+ and have a maximum size of 103 em TL. Maximum age in males was 5+ years and females 7+. Over the ages reported here males grew at a rate of about 11 em/year and females at about 10.5 em/year. Using the sizes at maturity reported in Chapter 2 it was possible to determine age at mat urity. In Tampa Bay, females mature at about 2.2 years and males at about 2 0 years. In Florida Bay, females mature at about 2 3 years and males at about 2 0 years. The Ford-Walford and nonlinear regression estimates shown in Table 7 were generally in good agreement The largest discrepancies were noted between the Ford-Walford and nonlinear estimates for males. Little difference was noted between the 2 estimates for females. The estimates generated using the nonlinear technique have associated statistics which make possible comparisons between estimates. Table 7

PAGE 92

-E () ::J: t-CJ z w ..J .... ;! 0 t77 92 72 52 cf TAMPA BAY 3 4 I 20 I 7 7 I 6 I I L= 88.8(1-e0.58[R-(-0.77)]) -Fitted Actual N=48 r =0.84 B 1 o+ VERTEBRAL RING NUMBER AGE
PAGE 93

78 132 S? TAMPA BAY 13 112 3 17 8 E I I u :c 15 I 92 1-" z UJ ...1 ...1 72 :5 l = 115. 0 ( 1-e0 34 [R-(-1.1 )] ) 0 1- -Fitted 52 Actual N=96 r =0.94 32 B 1 2 3 4 5 6 7 o+ 1+ 2+ 3+ 4+ s+ 6+ VERTEBRAL RING NUMBER AGE
PAGE 94

-E () X ._ CJ z UJ ...I ...I ._ 0 ._ if FLORIDA BAY 83 5 63 11 I 43 79 6 10 9 I L = 81. 5 ( 1-e0.53 [R-(-0 .64)]) -Fitted Wild caught Held in captivity N=44 r = 0 .90 1 2 3 4 5 o+ 1+ 2+ 3+ 4+ B VERTEBRAL RING NUMBER AGE
PAGE 95

80 120 BAY 9 12 3 100 6 E 11 I (,) 80 :z: C!J z w 60 L = 103.3 ( 1 e0.37 [R-(0.60)]) ..J ..J -Fitte d 0 Wild c aught 4 0 Hel d in capt i vity N =61 20 B 1 2 3 5 o + 1 + 2 + 3 + 4 + 5 + 5 + VERTEBRAL RING NUMBER AGE Figure 24 Von Bertalanffy growth curve of female bonnethead sharks collected from Florida Bay and sharks held at least part of their lives in captivity (B = birth, R = vertebral ring number, L = length).

PAGE 96

81 Table 7. Ford-Walford and nonlinear regression estimates of the von Bertalanffy growth equation. The coefficients of the equation are presented for sharks from Tampa Bay, Florida Bay and captive anima.ls. A. Generalized form of the von'Bertalananffy growth equation. L t -where Lt. equals length (em) at age t, equals asymptotic length (em), K is a growth coefficient per year, t0 is theoretical age (year) at zero length and R is vertebral ring number (R equals age in years for captives). B. Estimates Tampa Bay Florida Bay Sex K t r Sex K t r M 86.2 0.73 0.87 M 76.7 0.74 0.97 F 115.2 0.35 -1 .99 0.99 F 103.9 0.33 -2.07 0.99 c. Nonlinear regression estimates. Tampa Bay Florida Bay Sex K t r Sex K t r M 88.8 0.58 --0.77 0.84 M 81.5 0.53 0.90 ( (.3) (:55) ( (.2) (:55) F 115.0 0.34 .1 0.94 F 103.3 0.37 -0.60 0.93 ( ;9) (.08) (;4) (;3) (.1) (:4) D. Nonlinear regression estimates. In Captivity Sex K t r F 98.2 0.58 '-'0.56 0.98 (;5) ( 06) (;06)

PAGE 97

82 shows that females have a larger asymptotic size (Leo) and a lower growth coefficient (K) than males Comparing geographic localities, Tampa Bay sharks are larger than Florida Bay sharks. The 95% confidence intervals around the estimates show that the Leo of females is significantl y greater than males but K values do not differ significantly. The larger size (Leo) of Tampa Bay animals when compared with Florida Bay animals was not statistically significant. However, the length-frequency distribution of adult females from Tampa Bay was statistically different than that for Florida Bay (Figures 7 and 8), which indicates that adult females from Tampa Bay are larger in length. While the von Bertalanffy equation provided a good fit for growth in length, the best fit for growth in weight was provided by a logistic model (Figures 25 through 28 and Table 8). The general form of the logisti c equation is found in Table BA. Appendix 2 may be referred to for the standard deviations around the logistic weights -at-age estimates. Figure 25 shows the growth of Tampa Bay female sharks. The logistic growth equation was fit to this data set using a nonlinear regression technique A significant fit was produced (r =0.93, p<0.05). The a g e-weight relationship for Tampa Bay males is found in Figure 26. The logistic equation again provided a significant fit (r =0.80, p<0.05). In similar fashion, a logistic model of growth in weight provided the best fit for sharks collected in Florida Bay. Figure 27 shows the growth curve for females (r 0.85, p<0.05) and Figure 28 the same for males (r =0. 93, p <0.05). In Tampa Bay, females grow at a rate of about 1100 g /year and males at a rate of about 500 g /year. In Florida Bay, females grow at

PAGE 98

83 x1000 8 BAY I I 6 0) 1- :1: Cl w 4 I I ...1 :! 0 17526. 3 W= 2 1 + e -[0. 65 R + (-2 .54)] -Fitted Actual N=93 r=0.92 0 2 3 4 5 6 7 B 1 o+ 1+ 2+ 3+ 4+ 5+ 6+ VERTEBRAL RING NUMBER AGE Figure 25. Growth in weight of female bonnethead sharks collected from Tampa Bay (B =birth, R =vertebral ring number, W =weight).

PAGE 99

84 (x1000) 4 cJ TAMPA BAY W= 2500 1 +e-[1.34 R +(-2.45)] I 3 I C) I 1-I CJ I w 2 3: I ..J 0 1- 1 -Fitted Actual N=47 r =0.80 0 B 1 2 3 4 5 6 o+ 1+ 2+ 3+ 4+ s+ VERTEBRAL RING NUMBER AGE (years) Figure 26. Growth in weight of male bonnethead sharks collected from Tampa Bay (B = birth, R = vertebral ring number, W =weight).

PAGE 100

85 (x1000) 6 FLORIDA BAY I 5 3610.9 W= 1 + e -[1.46 R + (-3.83)] en -4 1-:I: C) I UJ 3: 3 I I ..J 0 I 12 -Fitted I Actual 1 Held in captivity N=60 r = 0.85 0 B 1 2 3 4 5 6 7 o+ 1+ 2+ 3+ 4+ 5+ e+ VERTEBRAL RING NUMBER AGE
PAGE 101

86 (x100) 20 0) ... 15 ::1: (!) -w 3: ...J 10 0 ... 5 cJ FLORIDA BAY W= 1964.9 1 +e -[1.22 R +(-2. 75)] I I -Fitted Actual Held in captivity N=30 r = .93 1+ 3+ VERTEBRAL RING NUMBER AGE
PAGE 102

87 Table 8. Nonlinear regression estimates of the logistic growth equation. The coefficients of the equation are presented for sharks from Tampa Bay, Florida Bay and captive animals. A. Generalized form of the logistic growth equation. W = a/[1 + e-(bR+c)] where W equals weight (grams), R equals vertebral ring number (R equals age in years for captives), and a, b, and care constants. B. Nonlinear regression estimates. Tampa Bay Sex a M 2500 ( ) F 7526 () b c 1.34 -2.45 0 .80 ( 72) ( :36) 0.65 -2.54 0.92 (.18) ( .4) C. Nonlinear regression estimates. In Captivity Sex a b c F 3459 1.91 0.97 () (.66) (.28) Florida Bay Sex a b c M 1965 1.22 2.75 0. 93 () (.48) (.94) F 3611 1.46 -3.83 0 .85 () ( 7 6) ( 1 86)

PAGE 103

88 about 660 g/year and males at about 450 g/year. In Tampa Bay, the asymptotic size for females is about 7.5 kg and for males about 2.5 kg. In Florida Bay, the growth curve for females is asymptotic near 3.6 kg and for males near 2.0 kg. The data in Table 8B demonstrate that there are statistically .significant differences in asymptotic weight between the sexes and between geographic regions. Females are significantly heavier than males and Tampa Bay females are significantly heavier than Florida Bay females. Asymptotic weight of males was not significantly different between the two areas. Maintenance of animals in captivity provided valuable information concerning the growth rate of individual sharks as well as average growth rates for litters of animals born in captivity. The growth records of individual sharks ar-e plotted in Figur-e 29. In Figur-e 29A the growth of individual animals born in captivity is plotted. The growth rate of these animals over the appr-oximate 10 to 11 months in captivity aver-aged about 2.7 e m/month. Fr-om birth in August at 25 to 28 em TL the sharks increased to 53 to 60 em by late June of the following year-. In Figur-e 29B the growth of indi victuals captur-ed in Flor-ida Bay and intt"'duced into captivity is shown. Two animals, 70 and 75 em TL, captured in July 1982 grew to about 89 em after a year in captivity. This represents a gr-owth r-ate of 1.2 to 1.6 em/month. A 79 em animal, captured in January, 1983 increased to 89. 5 em afterone year (0.9 em/month) One animal 84 em TL captured in ear-ly August gr-ew to 96 em in one year( 1 em/month growth rate) and another animal 92 em captur-ed at the same time grew to 100 em over the same time per-iod (0. 7 em/month). An 89.5 em animal captured November1982 gr-ew at 0 4 em/month over one yearand a 90.5 em animal gr-ew at 0.2 em/month over

PAGE 104

90 80 E (,) ::1: t-CJ 60 z w ...I ...I 0 40 .... ... -"' "' "' "' "' "' lr ... A J A S 0 N 1983 89 ....... 106F .....,..108F o-o 1KF &--/:> 102 w - 5KF .. .. 101 w ....... 07429 07440 07443 <>--<> 07430 ........ 07438 &--/:> 07441 F M A M J J A S 0 N o: J F M 1984 I 1985 Figure 29. Individual growth records of sharks held in captivity.

PAGE 105

90 the same time period. Finally, a 102 em animal captured January 1983 had grown to about 104 em by January 1984 (an increase of 0.2 em/month). Seasonally, growth rates were lowest during the winter months and highest during the late summer and fall. Litters of pups born in captivity during August of 1983, 1984 and 1985 were used for the determination of average growth rates. The growth of animals born in August 1983 is depicted in Figures 30 and 31. Over ca. 10 months these animals grew at an average of 2. 7 em/ month and 62.8 g/month. From birth in August at 27.5 0.6 em and 69.3 8.5 g, the sharks increased to an average of 54.5 3.1 em and 697 136.4 g in June 1984. Rapid growth immediately after birth in fall was followed by a slowing of growth in winter. During their first spring or summer, the male and female growth curves begin to diverge, although the difference at that time was not significant. In Figures 32 and 33 are shown the growth of sharks born in August 1985. These animals increased from an average of 30.5 0.6 em, (108 7.3 g) in September 1985, to 64.9 1 9 em ( 1023 96.3 g) in December 1986. This represents a growth rate of 2.3 em and 66.1 g per month. A slight divergence in growth in weight was noted during the animals first summer, but it was not statistically significant. Growth data provided by the August 1984 litters (Figures 34 and 35) were the most informative. These animals were born at a mean of 27.2 0.77 em TL (101.7 6.7 g) and grew at a rate of 2.8 em/month (121 .1 g/month) over 13 months. This growth rate is slightly higher than the growth of the two litters previously discussed. Figures 34 and 35 demonstrate that at some point between March and September 1985 the male and female growth curves began to diverge. By December,

PAGE 106

-E u :I: 1z w _. _. 0 191 5 60 16 4 fL 20/ I 13/ / I AMale 9 Female 1 Comb i ned 30 21;1 I 50 40 ASONDJ FMAM..J..JA 1983 1984 Figure 30 G r owth in length of bonnethead sharks born in captivity in August 1983. The mean and 95% co nfidence interval are shown. .

PAGE 107

92 5 800 C) .... l: 600 w w 19 400 I c( a: 13/ w 1of 6 Male 200 9) Female Combined 21,.f I AS 0 N D J F MAMJ J A 1983 1984 Figure 31. Growth in w e i ght of bonnet h ead sharks born in captivit y in August 1 983. The mean and 95% confidence interval are shown.

PAGE 108

93 12 I 60 -E () :z:: 50 1-CJ z w 19 _J I _J 40 t! 30/ 0 .... 30 I ASONDJ FMAMJJASONDJ 1985 1986 Figure 32. Growth in length of bonnethead sharks born in captivity in August 1 985. The mean and 95% confidence interval are shown.

PAGE 109

C) --w 3: w a: w 94 5 7 Female Combined ASONDJ FMAMJJASON.DJ 1985 1986 Figure 33. Growth in weight of bonnethead sharks born in captivity in August 1985. The mean and 95% confidence interval are shown.

PAGE 110

0 0 35 w IX: :::;) !;( IX: w a. w ..... -E u J: t0 z w ..J ..J 0 t 6 Male Female Combined 95 A S 0 N D !.J F M A M .J .J A S 0 N D !.J F M A M J J A S 0 N D :J F 1984 I 1985 I 1986 I Figure 34. Growth in length of bonnethead sharks born in captivity in August 1984. Also shown is a plot of water temperature data for the environment in which the sharks were held captive. The arrows indicate the time of annulus (=ring) formation in these animals. Means and 95% confidence intervals are shown.

PAGE 111

p 35 w a: :::l a: w D. ::e w ... c:n ... 25 2500 8 1500 iii 3: 500 25 35---14---- t:. Male Female Combined 96 3 I A S 0 N 0 !J F M A M J J A S 0 N Di J F M A M J J A S 0 N Dl J 1984 I 1985 I 1986 I Figure 35. Growth in weight of bonnethead sharks born in captivity in August 1984. Also shown is a plot of water temperature data for the environment in which the sharks were held captive. The arrows indicate the time of annulus (=ring) formation in these animals. Means and 95% confidence intervals are shown.

PAGE 112

97 growth in weight between the sexes was statistically different (Figure 35). Growth in length had diverged significantly by May 1986 (Figure 34). By December 1986 females had increased to an average of 87.4 6.68 em and 2840 166.5 g weight. Males had grown to 75 6 2.67 em TL and 1650 247 g weight. Over the 2.3 years in captivity, females averaged 2.2 em/month and 99.0 g/month and males averaged 1 .8 em/month and 56.3 g / month. Figure 34 demonstrates that growth in length slowed during the winter months when water temperatures were lowest and was most accelerated during the warmer months. Growth in weight (Figure 35) slowed during the winter of 1 984 but the winter of 1 985 saw a 1 2% decrease in weight for females and a 30% decrease for males. Figure 35 demonstrates that this decrease was preceded by a period of very rapid growth and coincided with low environmental temperature. No decrease in weight was observed in December 1986 most likely due to the fact that water temperatures were still high (27C). The animals of Figure 29A are of known age and using the data from Table 6 we can assign ages to the animals in Figure 29B. These data, and those for growth of captive animals shown in Figures 30 through 35, were used to construct a c omposite growth curve for all females held in captivity (Figure 36). The von Bertalanffy equation was fit to this data using a nonlinear regression tec hnique. Appendix 1 shows the standard deviations around the length-at-age estimates. Captive animals had an of 98.2 em, a growth constant (K) of 0.58 /yr, and a t of -0.56 yr. Growth in weight of captives was examined by the same 0 method and is shown in Figure 37. Appendix 2 shows the standard deviations around the estimates. This composite growth

PAGE 113

-E (.) l: 1-Cl z UJ ..J ..J 0 198 100 80 60 40 20 0 B L= 98 2 ( 1 _8-0.SBlA-(-0.56)]) -Fitted Actual N=168 r = 0 98 s+ AGE
PAGE 114

99 (x 1000) C) t:c Cl -w 4 3 2 1 W= 3459. 1 1 +e-f1.91 A +(-3.09)] -Fitted Actual N=128 r = 0 .97 AGE (years> Figure 37. A composite growth curve of female bonnethead sharks held in captivity. This plot was constructed as in Figure 19. (B = birth, A = age in years, W =weight).

PAGE 115

100 curve was constructed using a logistic growth equation. Asymptotic size was calculated to be 3459.1 g. Tables 7D and 8C show the coefficients of the von Bertalanffy and logistic growth equations, respectively for the above discussed composite growth curves. The associated statistics allow values for captive animals to be compared with the values for wild-caught females from Florida Bay (since all animals used to construct the composite growth curve were female). The growth constant K from the von Bertalanffy equation is significantly higher for captive animals than for wild-caught animals whereas Lm values were not different (Table 7C and 7D). The coefficients of the logistic equation (Table 8C) for captive animals are not significantly different than those for wild-caught animals (Table 8B). These comparisons suggest that captive animals may grow slightly faster in terms of length than wild-caught animals but do not attain larger sizes. Growth in weight of captive females was not statistically different from growth in weight of wild caught females. Back-calculations of size at age. In order to back-calculate size at age using vertebral rings it was necessary to examine the relationship between vertebral radius and total length. This relationship is shown in Figures 38 and 39. A linear regression of total length on vertebral radius was highly significant in both cases. In examining the figures it is noted that little difference exists between males and females. Likewise, the equations generated from Tampa Bay and Florida Bay are almost identical (Figures 38 and 39). Measurements of vertebral ring radii allowed the back-calculation of by direct proportion. Lee's phenomenon (a condition in

PAGE 116

-E (,) J: t- 101 Figure 38. The relationship between vertebral radius and total length for Tampa Bay sharks (L =length, V =vertebral radius).

PAGE 117

-E u ::c 1z w ...J ...J t! 0 192 72 52 L= 27.7 + 7 .5V r = 0 92 N=91 male female 2 4 VERTEBRAL RADIUS (mm) 102 Figure 39. The relationship between vertebral radius and total length for Florida Bay sharks (L =length, V = vertebral radius).

PAGE 118

103 which back-calculated lengths from older fish are smaller than lengths calculated from younger fish for the same presumed ring) was not detected in this data set. Table 9 shows the results of the calculations for each area and for sharks held in captivity. lengths at age agreed well with those predicted by the von Bertalanffy equation and with observed lengths at age. In Table 9A back-calculated sizes for males and females from Tampa Bay may be compared with Figures 21 and 22. Back-calculated size at birth for both sexes was about 35 em Which compares favorably with the observed/estimated size at birth (see Chapter 2). The results found in Table 98 suggest that Florida Bay sharks are born at about 32 em TL Which is slightly larger than the observed size at birth in captivity (ca. 27 em TL). Back'"'calculated size at birth for captives (25.9 em TL, Table 9C) was in good agreement with the observed size at birth in captivity (27 em TL, Chapter 2). In each of the above, the aforementioned divergence in growth at about age 2+ between the sexes is noted. Back'"'calculations in Table 9B and 9C agree well with observed lengths at age and von Bertalanffy estimates (Figures 23 and 24) excepting the apparent overestimation in size at birth shown in Table 9B. Using known-age animals, born in captivity, it was possible to the time of ring formation. Figures 34 and 35 show that the first ring formed in early February and the second ring in late November. In Figure 35 the formation of the second ring was closely associated with a dramatic decrease in body weight. The single animal tagged, released, and recaptured in Tampa Bay provided information on the growth rate in the field. This animal was

PAGE 119

104 Table 9. Back-calculated total lengths (em) determined from vertebral rings for sharks from Tampa Bay, Florida Bay and those held in captivity (B = birth). Ring (age in years) B 0+ 1 + 2+ 3+ 4+ A Tampa Bay Males Mean 35.2 61 .2 75.4 80.2 S .D. 1 7 8 8.6 7.6 n 21 21 15 5 Females Mean 35. 4 58.8 76. 6 89. 3 96.7 104.0 S .D. 7.2 6 7 3.8 n 41 41 37 32 24 2 B. Florida Bay Males Mean 32. 4 46.2 55.8 66.8 76.6 S D 8.8 5.7 2 9 n 13 13 12 12 2" Females Mean 31.6 48.7 62.5 78.1 84.8 92.2 S .D. 7 7 2.7 5.1 n 15 15 14 14 11 6 c. Captives Females Mean 25. 9 42.6 59.7 74. 7 81.7 88.1 S.D. 4;9 6 0 8 6 n 28 24 11 7 5 5

PAGE 120

105 105.5 em total length at release and had increased to 107.7 em after 377 days at large. This represents a growth rate of about 2.2 em/year (0. 18 em/month). Discussion Validation of aging technique. At present the vertebral ring method of age determination has been proven valid in only three species of sharks. With the results of this study, a fourth species may be added to this list. A single animal tagged, released and recaptured in Tampa Bay produced one ring during its 377 days at large. However, the most revealing results were provided by the maintenance of animals in captivity. Animals born in captivity and sacrificed after one year possessed two vertebral rings and those animals two years old possessed three rings (Table 5). Wild caught animals injected upon capture and held in captivity for a year produced one vertebral ring beyond the tetracycline reference mark (Table 6). Prior to this report the vertebral ring method of aging had been proven valid in only two species (Gruber, 1983; Smith, 1984). However, in neither study were all age groups validated. Gruber (1983) validated the method for the lemon shark, Negaprion brevirostris, up to age 30 months and Smith (1985) did the same for leopard sharks, Triakis semifasciata, 5+ to 16 years old. This study provided validation of all age groups (age 0 to 6+) since no sharks were observed in excess of 6+ years of age. Deviations in the expected ring number were observed in three instances. One animal ( 117179, Table 5) was much smaller than other animals of the same age and apparently did not form ring /12. Its

PAGE 121

106 failure to form this ring could be related to its slow growth. Animal #105 (Table 6) produced an additional ring. Due to the difficulty in interpreting rings at the centrum margin it is possible that the centrum edge was erroneously counted as a ring. The apparent failure of shark #7223 to form ring #1 cannot be explained. animals injected with tetracycline were observed to produce a fluorescent marker located in close proximity to the first vertebral ring (Table 6). This "birth mark" phenomenon has been observed in a number of elasmobranch species. Berry (1965) found that clearnose skates, Raja eglanteria, "n" years of age had "n+1" vertebral rings. Other species reported to possess this birth mark include the lemon shark, Negaprion brevirostr-is, (Gruber and Stout, 1983), the blue shark, Pr-ionace glauca, thresher shark, Alopias vulpinus, and shortfin mako shark, Isurus oxyrinchus, (Cailliet et al., 1983), the sharpnose shark, Rhizopr-ionodon terraenovae, (Parsons, 1983) and the sandbar shark, Car-charhinus pl urn be us, (Casey et al. 1985). The first ver-tebr-al r-ing may prove to be a bir-th mark in most species of elasmobr-anchs. The process(es) responsible forthe formation of this r-ing have not been examined. The sudden change from the benign environment in-utero to the mor-e competitive fr-ee-living existence may contr-ibute to its formation. The tetr-acycline method was found useful for examining seasonal variation in vertebr-al ring deposition. Tetr-acycline incor-por-ated concurrently or in close association with ver-tebr-al r-ings when injections wer-e administered dur-ing wintermonths (Table 6). In contrast, injections given during summermonths (#102 and 101, Table 6) created marks located between r-ings. The obvious interpretation of

PAGE 122

107 these results is that the ring structure fonns during winter months. Gruber ( 1983) reported that in the lemon shark a fall and a late winter band (= ring) appear to be fonned annually. Similarly, Smith (1985) found that hyaline edged centra (= ring) predominated in fall and winter. Caillet and Raqtke (1986) used electron microprobe analysis for age determination and verification in the grey reef shark, Carcharhinus arnblyrhynchos, and the thresher shark, Alopias vulpinus, and reported that calcium and phosphorus concentrations were lowest on the centrum edge during winter months. These regions of reduced calcium and phosphorous represent the translucent band referred to as a ring in this report. Mean marginal increment analysis (Table 5) shows that wide marginal increments were observed in early December and narrow ones in June. This lends further support to the hypothesis of winter ring formation. The tetracycline method will prove useful for age validation in most shark species but the shortcomings of the technique should be noted. In some cases no incorporation occurred. Four of the five animals recaptured in Tampa Bay that had received injections showed no fluorescent mark. The estimation of tetracycline dosage using a length-weight relationship could have resulted in an underestimation such that no incorporation occurred (the animals were estimated to have received 30 to 50 mg/kg. However, the dynamic nature of calcium in cartilaginous tissue could also account for this. If calcium in cartilage is mobilized at certain times, then the tetracycline mark could be obliterated. Failure to incorporate the tetracycline may reflect the decrease in calcification that is seen during the winter months although several captive animals injected during winter

PAGE 123

108 produced strongly fluorescent markers. While tetracycline validation is gaining increasing popularity among fisheries biologists and has been used to estimate age and growth rate for many species, the topic that perhaps most urgently needs to be addressed, is the effect of tetracycline on growth in fishe. s (see Chapter 3). Growth rate and age determination. Only recently have the determination of growth rates and age in elasmobranchs been the subject of research Most studies of this nature have examined growth in length by the von Bertalanffy growth equation fit to vertebral ring age data (Stevens, 1975; Thorson and Lacy, 1982; Pratt and Casey, 1 983; Gruber and Stout, 1983; Cailliet et al. 1983; Waring, 1984; Casey et al., 1985; Parsons, 1985; Branstetter and McEachran, 1986). A number of techniques are available for estimating the parameters of the von Bertalanffy equation but, perhaps due to its ease of application, the Ford-Walford method, a linear technique, has gained the most popularity. Vaughan and Kanciruk (1982) compared two linear methods of applying the equation with an iterative, regression method. They found that the non-linear procedure yielded the most accurate and precise growth parameter estimates and concluded that linear methods are now obsolete. In this work, the results of applying the von Bertalanffy growth equation to bonnethead shark age-length data is reported. Both Ford-Walford estimates and non-linear regression estimates are compared (Table 7). A Walford line produced correlation coefficients ranging from 0.87 to 0.99 indicating a good fit (Table 7B). Non-linear regression estimates also produced a good fit (r =0.84 to 0.94, Table 7C) and are assumed to be the most reliable based on the results of Vaughan and Kanciruk (1982).

PAGE 124

109 Von Bertalanffy growth curves are plotted in Figures 21 through 24. The parameters used are those found in Table 7C. Some interesting differences were noted between the sexes and within sexes from different geographic locales. There are several important differences in growth characteristics between males and females. Females grow to a larger size than males and are apparently longer lived. Females have been observed to be larger than males for numerous shark species and separate growth curves are normally constructed (Moss, 1972; Thorson and Lacy, 1982; Cailliet et al., 1983; Pratt and Casey, 1983; Casey et al., 1985). Male and female growth in length is similar early in life but then diverge. In Tampa Bay the greatest rate of change in the male growth curve occurs between ages 1+ to 2+ years (ca. 65 to 70 em TL) while female growth slows at about ages 3+ to 4+ (75 to 80 em). Similar growth characteristics were also observed when growth in weight was examined (Figures 25 to 28) although there was a high degree of variability in weight within a particular age class. This is especially evident among the adult female age classes, where total weight varied by as much as 3 kg for the same age class. A decrease in the rate of growth in weight occurs at around age 2+ to 3+ for males and around 3+ to 6+ for females. It is noteworthy that the greatest rate of change in growth in length corresponds quite well with the size at maturity (see Chapter 2). There is an apparent selection for rapid growth to maturity and then a slowing of growth thereafter except in Tampa Bay females where growth rate was relatively high after maturity. This might reflect a change from the production of somatic tissue to the production of reproductive tissue (see Chapter 6) and would also explain the variability observed in adult female

PAGE 125

110 growth in weight. Each adult female in Tampa Bay annually produces an average of 8.9 pups and each pup averages 171.8 g. In Florida Bay, the average litter size is about 9.3 pups and each weigh an average of about 82.5 g (see Chapter 2) Which means that in Tampa Bay, 1529 g and in Florida Bay, 767. 3 g of reproductive t issue is annually produced and then lost over an approximate 4 to 5 month period. A sudden loss of tissue would account for a part of the variability observed in growth in weight in adult females. This degree of variability was not observed in males and would not be expected since the male investment in reproductive tissue is considerably smaller. Differences in the growth constant K were noted between males and females (Table 7C). Males have a K approximately 30 to 40% higher than females. This suggests that males complete their growth curve 30 to 40% faster than females. This is primarily accomplished by possessing a smaller Lm (Table 7C) than females since growth rates are similar. Few studies have examined geographic variability among populations of sharks. Springer (1960) disc ussed distribution, migratory behavior and abundance of Eulamia milberti, (=Carcharhinus plumbeus) in the Western North Atlantic. Thorson et al. (1966) examined the populations of bull sharks, Carcharhinus leucas, in Lake Nicaragua. Springer (1967) reported on some of the general features of shark distributions in the Gulf of Mexico and Atlantic coastal waters. Sex and depth segregation in populations of the marbled catshark, Galeus arae, have also been examined (Bullis, 1967). Until now, no study had examined how populations of sharks differ between tropical and temperate regions. The differing environments between the two study areas apparently had a significant effect on the

PAGE 126

111 growth characteristics of the bonnethead shark. Environment apparently had the greatest effect on asymptotic size. L= for Tampa Bay females was 10% higher than Florida Bay females and Tampa Bay males had an L= 9% higher than Florida Bay males. However, these asymptotic length values were not statistically significant. When length-frequencies are compared, Tampa Bay females are significantly (p < 0.01, t-statistic = 8.39, df = 105) larger in terms of length. Environmental differences also had a statistically significant effect on asymptotic weight. Tampa Bay females were 3 times heavier than Florida Bay females. Males from Tampa Bay were larger (22%) than Florida Bay males, but the differences were not significant (Table 8B). This is probably best explained by differences in reproductive effort between the sexes (Chapter 6) The tendency for members of a species to be larger at higher latitudes has been widely noted but has not been reported among sharks. Growth plasticity is a common feature among fishes and is generally attributed to food limitation or differences in environmental temperature. The observation that captive shark growth was not statistically different from wild-caught animals suggests that food is not the limiting factor (see Chapter 6 for a bioenergetics discussion of this phenomenon). By maintaining sharks in captivity, both and litters born in captivity, it was possible to examine individual and seasonal growth differences. Several conclusions may be drawn concerning individual growth rates. It is evident that, although all animals were born at remarkably similar sizes, individual growth rates vary The widening confidence intervals of Figures 30 to 35 and the growth of

PAGE 127

112 individual animals shown in Figure 29A demonstrate this variability. Figure 29B shows that animals introduced into captivity at similar sizes grew at varying rates. Despite individual variability, average growth of litters born in 1983, 1984 and 1985 were almost identical. Male and female growth was not statistically different for approximately the first year of life, although the data suggest that growth begins to diverge during the first summer after birth. The growth of male and female sharks became statistically distinct after about 12 to 20 months (Figures 34 and 35). This lends support to the divergence in growth between age 1 + and 2+ that was observed in the growth analysis by vertebral rings. Sharks grew most rapidly during late summer and fall, and slowest during the winter (Figures 34 and 35). Growth in weight of the 1984 litter (Figure 35) was dramatically refduced by the winter of 1985. There was a large decrease in weight at this time but surprisingly growth in length was not affected. These results suggest that weight may be a more sensitive barometer than length for monitoring the effect of environment on growth. This is also evident in the examination of Table 8B where asymptotic weight for females was found to be statistically different between geographic locales while asymptotic length (Table 7C) was not. The composite growth curves shown in Figure 36 and 37 were constructed using data from captive animals. They may be compared with the growth data obtained from vertebral ring analysis. Table 7B and 7C show the parameters of the von Bertalanffy growth equation estimated from vertebral ring data. Since all animals used to construct the composite growth curve were female, we can compare the values of the

PAGE 128

1 1 3 von Bertalanffy shown in Table 7D with those for Florida Bay females in Table 7C. Maximum size (Lm) and t were not statistically different 0 between captive and wild-caught females. The K value for captives was significantly higher than that for wild-caught females. The degree of variability within these estimates suggests that the animals in captivity were growing at similar rates. The 95% confidence interval around the in-captivity K value is much narrower than those values estimated from the field. Since the composite growth curve was in part constructed from measurements of the same animals over a period of time, this would reduce the variability around the estimate of K. However, the captive animals were fed to satiation daily and might be expected to grow at similar rates. The higher degree of variability for animals in the natural environment perhaps reflects the spectrum of competitive abilities that would exist in nature. Competitively superior animals might have a growth constant similar to the captive growth constant but less competitive animals must certainly have a lower K Figure 37 shows the composite curve for growth in weight of captive sharks. Table 8C gives the parameters of the logistic growth equation and the 95% confidence intervals about each. These may be compared with the values for Florida Bay females in Table 8B. There is no statistical difference in the parameters of the logistic growth equation when captive females are compared with wild caught females In addition, the growth rate of the single tagged, released and recaptured animal (0.18 em/month) is in relatively good agreement with the growth rate observed for animals of similar size held in captivity (0.2 em/month).

PAGE 129

1114 The growth rate of captive sharks has frequently been reported to be accelerated. Wass (1971) reports that captive grey reef sharks, Carcharhinus amblyrhynchos grew 1 0 times faster than those in the field, while Gruber and Stout (1983) found that lemon sharks, Negaprion brevirostris grew 9 times faster in captivity. In comparing the growth characteristics of captive bonnethead sharks with wild-caught sharks the growth constant K alone was found to be statistically different. Growth in weight, asymptotic weight, and asymptotic length were not different. It is possible that larger species of sharks, such as lemon and grey reef, are more food limited in the wild than smaller species. This woold explain the observed growth acceleration of larger species when introduced into the relatively food rich environment of captivity. Gruber (1983) reported that 80% of hook and line captured lemon sharks had empty stomachs. Gut content examinations of bonnethead sharks in Florida Bay and Tampa Bay reveal that only 17% and 7% respectively, of guts were empty (see Chapter 6). The bonnethead shark is small, occupies a lower trophic level, is apparently less food limited and, perhaps for these reasons, does not experience greatly accelerated growth in captivity. The close approximation of the captive environment used in this study to the natural environment (see Chapter 2) probably contributes to the similarity in growth of captive and sharks. With the validation of the vertebral ring method of aging and the demonstration of a linear relationship between shark length and vertebral radius, the stage was set for back-calculations of size at age (Table 9). Back-calculated lengths-at-age were in good agreement

PAGE 130

115 with the growth curves of Figures 21 through 24 and Figure 36. Pratt and Casey ( 1983) and casey et al. ( 1985) backcalculated lengths at age from vertebral ring data for the shortfin mako, Isurus oxyrinchus and the brown shark, Carcharhinus plumbeus, respectively. In both cases back-calculated lengths at age compared favorably with lengths-at-age estimated using the von Bertalanffy in the former and observed length-at-age in the latter. Back calculated size at birth suggests that bonnethead sharks in Tampa Bay are born at about 35 em TL. This is in good agreement with the observed size at birth. In Florida Bay, size at birth was back-calculated to be about 32 em for animals in the wild and about 26 em for captives. Back-calculated size at birth for captives is very close to the observed size at birth in captivity (ca. 27 em TL). Conclusions Several important conclusions concerning the age and growth of the bonnethead shark may be drawn from this study. (1) Evidence presented here confirm that the vertebral ring method of aging is valid for the bonnethead shark. Most of the validation information was provided by the maintenance of animals in captivity. This suggests that sharks in captivity produce vertebral rings useful in aging. Casey et al. (1985) . also observed growth bands in the vertebrae of two sandbar sharks held under controlled conditions in the New England Aquarium. Additionally, since validation was accomplished at lower latitudes where seasonality may be dampened, there is good reason to expect the method to be valid for animals of more temperate waters. (2) The vertebral rings form

PAGE 131

116 during winter when water temperatures are lowest. Relative position of the tetracycline reference mark, mean marginal increment analysis and back...;calculated time of ring fonnation each support this conclusion. (3) The von Bertalanffy equation provides a good fit to age-length data but growth in weight is described by the logistic equation. (4) Growth characteristics differ between males and females. Males are smaller as adults and have a higher K value. (5) Bonnethead sharks of Tampa Bay are larger than those of Florida Bay. (6) Bonnethead sharks experience an acceleration of growth in captivity but do not grow to larger sizes even though fed to satiation daily. This suggests that food limitation is not controlling the maximum size of the animal. The results presented in this report provide valuable information concerning the age and growth of the bonnethead shark. This data is necessary for determining the energetics of somatic tissue production. This information can now be incorporated into the bonnethead shark energy budget.

PAGE 132

CHAPTER SIX: THE BIOENERGETICS OF THE BONNETHEAD SHARK, Sphyrna tiburo: DESIGN AND APPLICATION OF A RESPIROMETER FOR DETERMINATION OF METABOLIC RATE Introduction 117 The study of fish energetics is considered to have begun with the classic work of Winberg (1956). In the short time since that report numerous attempts have been made to examine the energy budgets of fishes. Most studies have utilized the general energy budget equation: I P + R + E where I equals the energy content of food ingested, P is the energy invested in production of somatic and reproductive tissues, R is the energy used in respiration whic h is further decomposed into standard, routine, active and feeding metabolism, and E is the energy lost in excretory products. A variety of approaches have been used to estimate these parameters. Estimates of food consumption are generally made from laboratory feeding studies (Solomon and Brafield, 1972; Elliot, 1976; Gruber, 1982) or by the examination of gut content of wild-caught animals (Clarke, 1978; Baird and Hopkins, 1981; Hopkins and Baird, 1985). In some cases gut evacuation rates have also been

PAGE 133

118 used to estimate consumption (Davis and Warren, 1970). Production (P) is the most straightforward quantity to determine. With an estimate of the caloric content of sana tic and reproductive tissues the rate at which energy is sequestered into these tissues can be easily described (Jobling, 1985; Wooton, 1985) The most problematical quantity in the energy budget to determine is the estimation of metabolism (M). Metabolism can be estimated from measurement of heart rate (Priede and Tytler, 1977), change in body composition during sustained performance (Brett, 1973) as well as direct and indirect calorimetry. In fishes, indirect calorimetry has been used most often. Indirect calorimetry exploits the relationship between oxygen consumption and the energy liberated from oxidation of foodstuffs. Respiration rates have been determined for many species of fish, using various types of respirometers (Brett, 1965, 1973; Farmer and Beamish, 1969; Wohlschlag and Wakeman, 1978; Torres et al., 1979; Beamish, 1981). Few studies have examined the metabolic rate of elasmobranchs (Winberg, 1956; Lenfant and Johansen, 1966; Piiper and Schumann, 1967; Hughes and Umezawa, 1968; Chan and Wong, 1977; Brett and Blackburn, 1978; Gruber, 1984). These studies primarily concerned temperate species and animals of a benthic nature. At present, no study has examined respiration of a species of shark that demonstrates obligate ram ventilation and no study has examined metabolism over the full spectrum from juvenile to adult. The major stumbling block has been the design of a respirometer that could accommodate the continuous swimming activity of these large animals. In the present study a respirometer was designed large enough for respiration measurements in both juvenile and adult bonnethead sharks,

PAGE 134

119 Sphyrna tiburo. The bonnethead shark is a small species of tropical to temperate distribution. It is easily handled and was found to perform well in respirometry studies. The measure of oxygen consumption in the bonnethead shark was used to determine metabolism (M) which was included in the general budget equation. An estimate of ingestion rate (I) and production (P) provided enough of the energy budget such that the remaining quantity, excretion (E) could be determined by difference. Energy budgets were constructed for summer months for age classes 0 through 3 for male and female sharks from both Tampa and Florida Bays. The energy budget of the bonnethead shark was constructed such that the bioenergetics of two geographically separated populations could be compared The life history characteristics of fishes may be profoundly influenced by proximal factors (Stearns, 1977). The same species of fishes in contrasting environments may exhibit among others, differing age compositions, growth characteristics, and reproductive characteristics. However, an interpretation of these differences can be difficult. An energy budget provides a common frame-work upon Which various life history "strategies" might be examined. Life history parameters expressed in terms of energy transformations can provide information concerning the effect of environment on the physiology of organisms and more importantly, may provide insight into the process of natural selection. The aim of this modelling exercise was not to construct the most precise energy budget possible for a rigorously defined set of conditions, but to make reasonable estimates of each parameter in the model such that general energetic trends might be examined.

PAGE 135

120 Materials and Methods Determination of metabolic rate (M). From 28 November 1986 to 7 December 1987, bonnethead sharks held in captivity at the Sea World Marine Science and Conservation Center were utilized for metabolic rate determinations. The animals were of both sexes and ranged in size from 95 to 4650 g (29.1 to 95.4 em total length). A respirometer was designed to ac c omodate the complete size range of bonnethead sharks. The device was constructed of 1 .9 em (3/411) thick plywood, sealed with silicon caulk and coated with epoxy paint (Figure 40). Plexiglas windows set into the respirometer lid allowed examination of the interior. The lid was fitted with a thermometer and a fitting for insertion of a polarographic oxygen electrode. Access to the interior of the respirometer was through a plexiglas box sealed to the top of the device. The bottom of the box was cut to leave an 8 e m collar which surrounded a 25 b y 25 em hole cut in the lid. A separate piece of plexiglas cut to fit this collar was used to seal the entrance to the device. By applying a heavy coat of silicon grease to the collar the sheet of plexiglas could be firmly sealed. Weights were placed along the seal to insure its' integrity. The box surrounding the entrance allowed seawater to rise several centimeters above the level of the entrance. This reduced the chance of introducing air bubbles into the respirometer when an excited animal was placed in the device and, in addition, made sealing the entrance much easier. The respirometer was filled with 35 /oo salinity sea water by

PAGE 136

plexiglass box chart recorder oxygen elect rode plexiglass thermometer windows 121 Figure 40. A diagram of the respirometer used for determining metabolic rate in the bonnethead shark.

PAGE 137

122 using a diatom filter equipped with an ultraviolet sterilizer. Water pumped directly from Florida Bay was passed over a biological filter before being diatom filtered, UV treated and pumped into the respirometer. The respirometer was filled on a slight incline such that any air pockets would be forced toward the entrance where they could be squeegeed out. A polarographic electrode, amplifier and strip chart recorder were utilized to monitor oxygen levels. Animals were dip-netted from the captive environment and placed immediately in the respirometer. Animals were not fed for at least 24 hours prior to their placement in the respirometer. Time of day, chart speed and temperature were recorded. During each experiment, an estimate of swimming velocity was made at one to two hour intervals by determining the time required for the animal to swim between the plexiglas windows. At the end of the respirometry determination the length, weight, and sex of the animals were recorded. Between runs, an oxygen cylinder was used to return the dissolved oxygen level in the respirometer to near saturation. During this time a small submersible pump was used to circulate water in the respirometer to insure the dissolved oxygen concentration was uniform. After about 24 hours of use, a complete change of the water in the device was made even though ammonia levels were undetectable. As a control, the device was prepared as described above, minus the experimental animal, and monitored for four hours with no change in oxygen content detected. Oxygen consumption was calculated using the general equation; C bsv/w

PAGE 138

123 where C equals oxygen consumption in mg02/kglhr, b is the rate of change of dissolved oxygen observed during the respirometry run (the slope taken from the chart recorder printout), s is the solubility of oxygen calculated at the experimental temperature and pressure, v equals the chamber volume (851.4 liter), and w equals the wet weight of the animal. Determination of food intake (I). Determination of food intake was facilitated by the maintenance of animals in captivity and by the examination of gut content of animals. Utilizing animals of similar size born in captivity, the ingestion rate was determined. One group of three animals averaging 55 em in length and 700 grams weight, and another group averaging 74 em and 1773 grams were fed to satiation a diet of capelin, Mallotus villosus, each day over a three week period during June and July 1986. The total weight of food ingested by each group was recorded each day such that an average daily rate could be calculated. To convert wet weight consumption to dry weight the dry weight values reported for osteic hthyes (22.3%) by Kushlan et al. (1986) was utilized. Dry weight consumption rates could then be converted to caloric consumption rates using the caloric content of rainbow smelt, Osmerus mordax, (5.8 kcal/g dry weight) reported by Foltz and Norden (1977). Caloric values for osteichthyes typically range from about 4. 7 to 5. 6 kcal/g dry weight (Kushlan et al. 1986). Food intake in the field was determined by examining gut contents of animals captured in Florida Bay and Tampa Bay. Gut contents were identified to species when possible and wet weight was determined by

PAGE 139

124 the appropriate length-weight relationship. By drying animals in a 60 C oven, the appropriate values for conversion of wet weight to dry weight was obtained. Dry weight of gut content was converted to caloric values using the caloric values reported by Kushlan et al. ( 1986). Detennination of production (P). Estimates of production were made possible by directly determining growth curves (see Chapter 5) and by the examination of average litter size (see Chapter 2). To detennine dry weight, three animals, 69.9, 77.9 and 139.7 g were placed in a 60 C oven and dried to constant weight. Using the caloric value estimated by Gruber ( 1984) for lemon shark, Negaprion brevirostris, tissue (5.07 kcal/g dry weight) and the dry weight of bonnethead shark tissue, caloric growth could be calculated. The caloric investment in producing offspring could also be estimated. Assuming the dry weight and caloric content of embryonic tissue is similar to those reported above, the average weight at birth and average litter size reported in Chapter 2 could be used to calculate the energetic investment in reproduction. Results Detennination of metabolic rate (M). Oxygen consumption rates of 20 sharks, weighing 95 to 4650 g, was measured. Animals weighing 1000 to 1600 g performed best in the respirometer such that most respiration data were obtained from this size group Respiration rates were measured over periods of 0.5 to 11 hours, during all times of the day, at temperatures averaging 26 ( ) C. Animals were removed from the

PAGE 140

125 respirometer either upon their expiration or when possible, just prior to it. The smallest animals (0.2 kg) reduced the partial pressure of oxygen in the respirometer to levels as low as 49.3 rnrn Hg, about 30% of saturation, before signs of stress developed Sharks 1 to 2.5 kg reduced the oxygen level in the respirometer to as low as 71.3 mrn Hg, about 47% of saturation. Larger sharks, 2.8 to 4.7 kg, produced only 1 to 2 hours of respiration data and were able to reduce the oxygen concentration to only about 143 rnm Hg, about 94% of saturation. During respiration measurements, animals swam at an average of 35.9 (.9 C.I.) em/sec or 0.53 (.055 C .I.) body length/sec. The average respiration rate was 240 (.8 mg Several variables were examined for their effect on oxygen consumption in the bonnethead shark. Oxygen consumption was regressed on time of day, partial pressure of oxygen, swimming velocity and wet weight. In this multiple regression, only wet weight was found to significantly explain the variability in oxygen consumption. In Figure 41 is shown the relationship between wet weight and absolute oxygen consumption (mg02 /kg) The linear regression was highly significant indicating an increase in oxygen consumption with increasing size. When standardized to weight a different relationship is seen. Figure 42 shows an exponential regression of relative oxygen consumption on body weight The regression equation; c (5 .61-0.11W) e . where C equals oxygen consumption in mg02/kg/hr and W equals weight in kilograms, was significant at the p = 0.001 level (df = 67; r =

PAGE 141

....... BOO '-.t::: ........ ON t:J) E 600 ....... <'! 0 ..... ...... a.. 400 Cl) <'! 0 (.) <'! 200 lLJ {!) >-)( 0 I . . .. .. .... ....... I . . .' .. .. :. .. 0 C = 68.9+177.8W ....... .. / 1 n = 6 9 .. ( r =0.87 0 1 2 3 4 WEIGHT (kg) 126 5 41. A of absolute o x ygen consumption (C) on body weight (W). The dotted lines the 95% confidence limits.

PAGE 142

127 370 0) 0) E -z 0 -t. .. .. .. .. . ... . -. --... I ... ... I :J tn z 0 0 220 : -. .... z w >< 0 :c 170 I ... ......... --....... .. C = e ( 5 .61-o.nw) n=68 r =-0.39 ... --........................ ....... . .. ..... .... .... .... WEIGHT (kg) Figure 42. An exponential regression equation of oxygen consumption (C) on body weight (W). The dotted lines are the 95% confidence intervals.

PAGE 143

128 ..;0.39). Using the oxycalorific coefficient for fish (3.25 cal/mg0 2 ) reported by Brafield and SOlomon (1972), the above equation may be used to estimate the metabolic rate of the bonnethead shark. This allowed the construction of the equation; where M equals metabolic rate in calories/kg/hr and the remaining variable is as before. The above equation provided the best fit to the data set but for comparison with other data the equation c = awb was also fit to the data. This equation was significant at the p=0.05 level (df =67, r The constants were estimated at a= 243 and b -0.08. Using the above equation metabolic rate was calculated for age class 0 to 3 years, for both sexes and for each geographic area. Captive female bonnethead shark s of age class 0 to 3 years had metabolic rates of 1 2. 9 to 50.3 kcal/day (Table 1 0) Male sharks of t h e same age classes from Florida Bay had rates of 11.3 to 32.2 kcal/day (Table 11). Female sharks of age class 0 to 3 from Tampa Bay had rates of 24.1 to 58.3 kcal/day and males had rates of 18.8 to 39.6 kcal/day (Tables 12 and 13). Table 14 shows the metabolic rates and the confidence intervals around those rates for each age class. When animals of similar weight were considered, swimming velocity was found to have a significant effect on oxygen consumption and metabolism. This allowed the construction of a power performance curve for the bonnethead shark. For this analysis only sharks between 1 and 1.6 kg in weight were utilized to minimize the effect of body weight

PAGE 144

129 Table 10. Summer (May through August) energy budget for captive female bonnethead sharks of age class 0 to 3 years. All values are in kilocalories per day and numbers in parentheses are percentages of the total budget that each component represents. (I = ingestion, P and P = production of somatic and reproductive tissue, M z routine metabolism and E = energy excreted as waste). r (A). Ingestion (I) estimated from observed data and excretion (E) determined by difference. Age Class 0 1 I 49. 4 ( 1 00) 3 1 ( 6 3) 66. 5 (100) 4.9 (7.4) E 12.9 (26.1) 33.4 (67.6) 35.3 (26.1) 26.2 (39.5) 2 83.2 (100) 1.7 (2.0) 7.2 (8.9) 47.7 (57.3) 26.6 (31.9) 3 88.0 (100) 0.3 (0.3) 7 2 (8. 2) 50.3 (57 2) 30. 2 (34.3) (B). Assuming excretion (E) equals 27%. Age Class 0 I 2 1 9 ( 1 00) 3 1 ( 1 4 2) 55.1 (100) 4.9 (8.8) E 12.9 (58.9) 5.9 (27.0) 35.3 (63.7) 14.9 (27.0) 2 77. 5 (100) 1.7 (2.2) 7.2 (9.3) 47.7 (61.5) 20. 9 (27.0) 3 79. 2 (100) 0.3 (0.5 ) 7.2 (9.1) 50.3 (63.5) 21.4 (27.0)

PAGE 145

130 Table 11. Summer (May through August) energy budget of male bonnethead sharks from Florida Bay of age class 0 to 3 years. All values are in kilocalories per day and numbers in parentheses are percentages of the total budget that each component represents. (I = ingestion, P s and P r = product i on of somatic and reproductive Mr = routine metabolism and E energy (A). Ingestion (I) estimated from observed data and excretion (E) determined by difference. Age Class 0 2 3 I 48.4 ( 1 00) 1.3 (2 7) 54.5 (100) 1.6 (2.9) 60 0 3 ( 1 00) 1. 0 ( 1 .6 ) 63.4 (100) 0.4 (0 6) ? ? (B). Assuming excretion (E) equa l s 27%. Age C lass 0 2 3 I 17.3 ( 1 00) 1.3 (7 9) 31. 2 (100) 1 6 (5.1) 40.8 ( 1 00) 1.0 (2.4 ) 44. 7 (100) 0 4 (0 9) See text. (2 0 7?) (4. 1?) E 1 1 3 (23. 3) 35. 8 (74 0) 21. 2 (38. 9) 31.7 (58 2) 28. 8 (47. 8) 30. 5 (50 6) 32. 2 (50 8) 30.9 (48.7) E 1 1 3 (65 3) 4 7 (27. 0) 21. 2 (67 9) 8.4 (27.0) 28.8 (70.6) 32. 2 (72.0) 11.0 (27.0) 12 0 1 (27 .0)

PAGE 146

131 Table 12. Sunnner (May through August) energy budget for female bonnethead sharks frOI!l Tampa Bay of age class 0 to 3 years. All values are in kilocalories per day and numbers in parentheses are percentages of the total budget that each component represents. (I = ingestion, P s and P r = production of somatic and reproductive tissue, respecr,i vely, Mr = routine metabolism and E = energy excreted as waste). (A). Ingestion (I) estimated from observed data and excretion (E) determined by difference. Age Class 0 I 56.6 (100) 2.2 (3.9) 67.7 (100) 3.2 (4.7) E 24.1 (42.6) 30.3 (53.5) 36.2 (53.5) 28.4 (41 .9) 2 85.5 (100) 3.8 (4.4) 14.5 (17.0) 48.6 (56.8) 18.6 (21.8) 3 110.2(100) 3 .7 (3.4) 14.5 (13.2) 58.3 (52.9) 33.7 (30.6) (B). Assuming excretion (E) equals 27%. Age Class 0 1 I 36.0 (100) 2.2 (6.2) 54.0 (100) 3.2 (5.9) E 24.1 (66.9) 9.7 (27 .0) 36.2 (67.0) 14.6 (27.0) 2 91.6 (100) 3.8 (4.1) 14.5 (15.8) 48.6 (53.1) 24.7 (27.0) 3 104.8 (100) 3.7 (3.5) 14.5 (13.8) 58.3 (55.6) 28.3 (27.0)

PAGE 147

132 Table 13. Summer (May through August) energy budget for male bonnethead sharks from Tampa Bay of age class 0 to 3 years. All values are in kilocalories per day and numbers in parentheses are percentages of the total budget that each component represents. (I = ingestion, P and P = production of sana tic and reproductive rMr = routine metabolism and E = energy (A). Ingestion (I) estimated from observed data and excretion (E) determined by difference. Age Class 0 1 2 3 I 52.9 (100) 2.5 (4.7) 62.4 (100) 2 2 (3.5) 69. 1 ( 100) 0. 9 ( 1. 3) 71.7 (100) 0.3 (0.4) ? ? (B). Assuming excretion (E) equals 27%. Age Class 0 1 2 3 I 29.2 (100) 2 .5 (8 6) 45.6 (100) 2 2 (4.8) 52 6 ( 1 00 ) 0 9 ( 1 7 ) 54.7 (100) 0.3 (0.5) See text. (3. 1?) (4.2?) E 18.8 (35.5) 31.6 (59.7) 31.1 (49. 8) 29.1 (46.6) 37.5 (54.3) 30.7 (44.4) 39.6 (55.2) 31. 8 (44.4) E 18.8 (64.4) 7.9 (27.0) 31.1 (68 2) 12. 3 (27 0) 37.5 (7 1. 3) 39. 6 (72.4) 14.2 (27 .0) 14.8 (27.0)

PAGE 148

133 Table 14. Metabolic rates (in kcal/day) for sharks from Florida and Tampa Bays. The values in parentheses represent the 95% confidence intervals. Age Class Florida Bay Tampa Bay Female Male Female Male 0 12.9 11.3 24.1 18.8 ( 1 2 0-13 2 ) (10.5-12.0) (22.8-25.4) (17.7-19.7) 35.3 21.2 36.2 31.1 (33.1-38.6) (20.2-22.5) (33.4-38.9) (29.1-32.4) 2 47. 7 28.8 48.6 37. 5 (42.3-53.8) (27.3-30.5) (43 1-56.1) (35 2-40 6) 3 50.3 32.2 58.3 39.6 (43.8-57.1) (30.3-34.2) (48.9-75.8) (36.6-43.0)

PAGE 149

134 on respiration and also, because this size shark performed best in the respirometer. This power perfonnance curve is plotted in Figure 43. The exponential regression equation; where C equals oxygen consumption and bl equals swimming velocity in body lengths/sec provided an ANOVA significant fit (p = 0.039; f = 6.04; df = 8; r = 0 .66). This equation predicts that oxygen consumption at 0 body lengths/sec is mg02/kg/hr. This corresponds to a standard metabolic rate of 3.17 kcal/day. Determination of food intake (I). The rat e of food intake was estimated by maintenance of animals in captivity. Sharks 11 months old, averaging 55 em TL and 700 g weight consumed 32.6 g/day wet weight of cape lin, Mal lotus villosus. This is a consumption rate of 4.66% body weight per day Sharks 23 months old, averaging 74 em TL and 1773 g weight consumed 62.1 g/day wet weight which is about 3.5% body weight per day Using the dry weight value for osteichthyes (22 .3%) and an average caloric value for rainbow smelt Osmerus mordax, (5.8 kcal/g dry weight) the above ingested materials were converted to caloric values. The smaller size group consumed 7.27 g dry weight/day or 42.2 kcal/day, and the larger group 13.8 g dry weight/day or 80.0 kcal/day. The collection of sharks from Florida Bay and Tampa Bay provided information on stomach content and allowed an estimate of ingestion rate. The frequency of occurrence of prey items in the diet of Florida Bay sharks was shown in Table 2 Chapter 2. In examination of 53 guts

PAGE 150

135 .. .r::. 340 ., E -z 300 0 .... a. :E :;) CIJ 260 z 0 0 z w 220 )( 0 C= e ( 3.70+3.49 bl) n = 10 r = 0.66 0 .46 0.49 0.52 0.55 0 .58 SWIMMING VELOCITY (bl/sec) Figure 43. A power performance curve for bonnethead sharks of 1 to 1 .6 kg in weight.

PAGE 151

136 from Florida Bay sharks, cephalopods, lobster, crabs and fish in that order, were found to predominate. Seventeen percent of guts examined were empty. The results of gut content examination of Tampa Bay sharks is also shown in Table 2. These sharks feed on blue crabs, Gallinectes sapidus, almost exclusively. Only 7% of guts examined were empty in Tampa Bay. Figure 44 shows a linear regression of gut content weight (expressed as percent total body weight) on total body weight. The regression was significant (p=0.025; f=5.46; df=35; r=-0.37) indicating a general decrease in gut content as percent body weight with increasing size. Figure 45 shows an exponential regression of total gut content weight on total body weight. The regression equation; where I equals total weight of ingested food in grams and W equals body weight in kilograms. This regression was significant at p = 0.05 (f = 13.6; df = 35; r = 0.53). Figures 44 and 45 also include the average quantity of ingested material for captive animals. These values fall within the range of gut content values observed for Tampa Bay sharks. The almost monophagous diet of the bonnethead shark in Tampa Bay made an estimation of the caloric value of gut content a simple matter. Blue crabs were found to be 33.9% (.8 S.E.) dry weight. The average caloric value for crustacea (3.99 kcal/g dry weight) reported by Kushlan et al. (1986) was used to convert blue crab dry weight to caloric content. The equation;

PAGE 152

137 8 .... X n=36 (!) -r = 0 37 UJ 3: Tampa Bay ...1 Keys Capt i ves 6 0 .... .... z UJ 0 4 a: UJ a. en c( .... z UJ 2 .... z 0 0 .... :::;:) (!) 00 2 4 6 8 TOTAL WEIGHT (kg) Figure 44. A linear regression of gut content weight (expressed as percent of total body weight) on total body weight.

PAGE 153

138 250 t7) 1-::E: 200 Cl iii 3: 1-150 z w 1z 8 100 1-::l Cl 1 =e(3.46+0. 21W) n =37 r = 0.53 Tampa Bay Keys Captives 2 4 6 TOTAL WEIGHT (kg) Figure 45. An exponential regression of gut content wei ght on total weight. (I equals total weight of ingested food and W equals body weight). 8

PAGE 154

139 I where I equals ingested material in kilocalories and W equals weight in kilograms, provides an estimate of the caloric value of gut content for various sized sharks. The high diversity of items in gut content of Florida Bay sharks prevented the estimation of caloric value of gut content in the above manner. However, by assuming that the frequency of feeding in wild-caught sharks in Tampa Bay and Florida Bay is similar to captive animals, the above equation can be used to estimate daily ration. Since the caloric value of crab and capelin is similar ( 1 .3 versus 1.35 calories/g wet weight, respectively) the equation may be used without modification for both captive and wild-caught animals. The close agreement between the quantity of ingested material in captive animals and in wild-caught animals (Figures 44 and 45) also supports the application of this equation for estimating an average meal size in both captive and wild animals. Determination of production (P). The growth equations reported in Chapter 5 describe growth in weight of sharks from both Tampa and Florida Bays, of both sexes, and of captive females (Table 8, Chapter 5). Dry weight of the bonnethead shark was found to be 22.4% ( 90 S .E.). This value and the caloric value reported by Gruber (1984) (5.07 kcal/g dry weight) can be used to describe caloric growth of the bonnethead shark. The logistic growth equation; W a/[1+e-(bR+c)J

PAGE 155

140 where W equals weight in grams, R equals vertebral ring number (age A, is substituted for R in the case of captive females) and a, b, and c are constants, accurately describes growth in the bormethead shark. Appendix 2 may be referred to for an estimate of the variability around the calculations of sanatic production. The values found in Table 8, Chapter 5 and the above dry weight and caloric values may be used to calculate the caloric investment in somatic tissue for a particular age class. The method assumes that caloric value and percentage dry weight of tissues do not change appreciably over the life of the animal (see Discussion for an evaluation of this assumption) The energetic investment in producing reproductive tissues was also estimated. Florida Bay adult females produce an average of 9.3 pups every year. Sharks born in captivity in 1983 and 1984 were born at an average of 82.3 g (.1 C.I.). Using the caloric and dry weight values already reported, the estimation of net caloric value of an average litter is possible. In Florida Bay, an average litter represents an investment of 869 kilocalories. In Tampa Bay, the average reproductive output is 8.9 pups each year and the pups average 171.8 g (.6 C.I.). An average litter in Tampa Bay is an energetic investment of 1737 kilocalories. This method again assumes that the caloric value and dry weight of embryonic tissue is not different from larger animals. General energy budget. With estimates of metabolism (M), food intake (I), and production (P), the energy budget of the bormethead shark may be constructed. As metabolic rate determinations were made at 26 ()

PAGE 156

141 C, and food intake determinations for captive animals were made during June and July, energy budgets were only calculated for the months May to August when water temperatures in both Tampa and Florida Bays ranged from ca. 25 to 30C and averaged about 28C. Separate budgets were constructed for sharks of age class 0, 1, 2 and 3 years, from both geographic areas (Tables 10 through 13). In one case, average daily food intake (I), production of sana and reproductive tissue (Ps and Pr), and metabolism (M) were estimated, and excretion (E) was determined by difference (Tables 10A through 13A). In a second case, excretion was assumed to be 27% (Brett and Groves, 1979) and the corresponding ingestion rate (I) determined by subtraction (Tables 10B through 1 3B) In comparing Tables 10A through 13A several general trends may be recognized. Gross conversion efficiency is highest in the smallest age classes and decreases in the older age groups, except among Tampa Bay females where production was relatively constant through all age classes examined. At its highest level, growth of somatic tissue accounted for 7.4% of the energy budget. In females, a general shift from production of somatic tissue to production of reproductive tissue was seen at age group 2 with reproductive tissue accounting for 8.9 to 17% of the budget. In males, no estimation of reproductive investment was possible. Routine metabolism generally accounted for the greatest percentage of the budget. Metabolism among age class 0 animals during summer consumed 23.3 to 42.6% of the energy budget. Older age classes were estimated to consume 38.9 to 57.3% of their budget during routine metabolism. The energy lost as waste (E) was estimated by subtraction and was found to be very high in age class 0 (ca. 54 to 74%) and lower

PAGE 157

142 in the older ages (ca 22 to 58%). Ingested energy (I) was estimated to be 48.4 to 110.2 kilocalories per day which when converted back to . wet weight equals about 5 6 to 1 .8% of body weight depending on size. When energy lost as waste (E) was assumed to equal 27% (Tables 108 through 13B), several components of the budget were altered. The amount of ingested energy (I) decreased to range from 17.3 to 104.8 kilocalories per day When converted back to wet weight, this represents about 2. 3 to 1 7% of body weight per day Production of somatic tissue (Ps) was again highest in the age class 0 animals (ca 9-14%) and decreased with increasing age. In age class 3 production of soma accounted for 0 5 to 0.9% of the energy income except for Tampa Bay females which were still growing fairly quickly at age class 3 (3 .5% of energy budget). Production of reproductive tissue (Pr) equalled about 9% for Florida Bay females and about 14 to 15% for Tampa Bay females. Male reproductive effort was estimated at 3 t o 4% (see Discussion for an explanation of these estimates). When waste was assumed to be 27%, the energy investment in routine metabolism (Mr) increased. In addition, over each age class the percentage of the energy income consumed in routine metabolism was more consistent. Routine metabolism varied from about 51% to about 72%. The age _class 2 and 3 males from both Tampa and Florida Bays apparently invested the greatest percentage of their energy budget in routine metabolism (70 to 72%). Discussion Evaluation of assumptions. In modelling the energy budget of the

PAGE 158

143 bonnethead shark a number of assumptions were made. Perhaps the most important assumption made was that the routine respiration rates observed in captive sharks at 26 C would not be drastically different from the summer (May to August) respiration rates of animals from Tampa and Florida Bays. Water temperatures in Tampa Bay during summer averaged 28.2 C ( 0.66) and in Florida Bay averaged 28.8 C ( 0.28) (Figure 2). Even though metabolic rate may be slightly underestimated, this should not affect the general trends in the energy budget components such that comparisons between the various components of the budget would be meaningful It may be argued that the bonnethead never or seldom experiences the highest temperatures recorded in Tampa and Florida Bays (ca. 31 C) and that 26 C may lie closer to its temperature preference than the average for the environment. Crawshaw and Hammel ( 1973) report that horn sharks, Heterodontus francisci utilize behavioral temperature regulation to maintain a body temperature of 24 c. Murray (1971) reported that fish can detect temperature changes of less than 0.5 c and are able to select a particular temperat ure or range of temperatures. Thermoregulation is frequently cited as an important factor in fish movements (see Pickering, 1981). The absence of bonnethead sharks the very warm midday waters of summer was pointed out in Chapter 3. During summer, the bonnethead shark utilizes the shallow, grassflats alm:>st exclusively at night, presumably moving to deeper water during the day. Although most likely feeding motivated (see Chapter 3) this behavior would also have important thermoregulatory advantages. These animals would therefore experience temperatures during summer no higher than the warmest

PAGE 159

144 values recorded at night, which would bring the summer averages closer to the 26 C at which respiration rates were measured. Assuming no significant temperature effect, the possibility of an inherent difference in metabolism between the two populations has to be considered. In this study it was necessary to assume that weight would be the most significant factor affecting metabolism. Tampa Bay sharks by virtue of their larger size at age would have higher total metabolic rates than sharks from Florida Bay. Since metabolic rate determinations of sharks from Tampa Bay were not made, the possibility exists that higher or lower metabolic rates could have been selected for in the Tampa Bay population, but this possibility could not be addressed. Salinity differences between Florida and Tampa Bays could affect metabolic rate. However, the bonnethead shark is relatively stenohaline and avoids areas of low salinity. The Florida Bay population rarely if ever experiences salinities much below about 34 0 loo. In Tampa Bay salinity fluctuations are greater but bonnethead sharks were collected in areas where the salinity never dropped below 28 /oo and average d about 33 /oo. Salinity changes of this magnitude would not be expected to cause drastic changes in metabolic rate. A second assumption necessary for this energy budget analysis is that the level of activity observed during respirometry determinations is similar to the routine activity level and that the routine activity level represents the general level of activity in nature. During respirometry measurements sharks of about 66 to 76 em total length (1 to 1.6 kg weight) were observed to swim at 0.46 to 0.59 bl/sec (Figure --43). Measurements of activity levels in nature were not possible but

PAGE 160

145 in Chapter 3 the swimming velocity of captive bonnethead sharks was monitored. Assuming the routine activity levels observed in captivity are similar to the activity levels in nature, the bonnethead shark averages about 0.5 to 0.6 bl/sec for sharks ranging in size from 65. 9 to 77.2 em total length. The velocity measurements made during respirometry determinations are in close agreement with those observed in captivity. Additionally, the close agreement between velocities observed during respirometry measurements and those predicted by Weihs' (1981) theoretical approach (Chapter 3) suggest that velocities are optimal (routine) over all sizes examined. A fishes activity level and metabolism would rise above the routine level during migration, feeding, escap e and mating. The bonnethead shark apparently makes an inshore/offshore migration during the cooler months in Tampa Bay but during summer months was observed to only make daily movements in and out of shallow water. The activity level of pelagic fishes such as tuna, billfishes and lamnid shark s may be greatly increased during the pursuit of prey items The bonnethead however, feeds on benthic prey and likely does not engage in a great deal of pursuit. The scenario suggested by examination of gut content is that, when foraging, this shark detects the presence of a cryptic prey item and simply attacks the bottom often ingesting seagrasses and sediment. The increase in activity during escape from predators and mating cannot be addressed. The bonnethead shark and likely all small sharks are preyed upon by larger s harks. Larger bonnethead sharks would be less sucepti ble to predation and would perhaps spend less time in escape. In estimating food intake (I) it was necessary to assume that the

PAGE 161

146 average meal size of captive animals is similar to wild-caught animals from Tampa and Florida Bays. This assumption was supported by the close agreement between the daily ration observed in captive animals and the meal size predicted by the regressions shown in Figures 44 and 45. The regression shown in Figure 45 showed a poor fit for gut content weight data for animals above 4 kg. For this reason the regression was used only for calculating gut content weight for animals below 4 kg. Additionally, it was assumed that the daily feeding rate observed in captivity would be similar to the natural rate. The low incidence of empty guts observed in both Tampa and Florida Bays (7 and 17%, respectively) suggest that the bonnethead successfully obtains prey on a regular basis such that the above assumption concerning feeding frequency may be valid. However, in examining the values of food ingested (I) and excreted material (E) in Tables 10A through 13A it is suggested that this method overestimated food ingestion rate. Since production (P) and metabolism (M) were measured directly, an overestimation of food ingested (I) would not affect these values. To balance the budget an excess of food ingested (I) would necessarily produce an overestimation of excretion (E). This was apparently the case, particularly in the younger age classes where E equalled as much as 74% of the budget. In the general energy budget for carnivorous fish reported by Brett and Groves (1979) fecal and excretion equalled 27% of the budget. Budgets for carnivorous fishes in which excretion was estimated to exceed 40% are rare (Brett and Groves, 1979). Excluding the unusually high levels of excretion for the 0 age class, the average excretion rate is about 40%. This value is relatively high,

PAGE 162

147 again suggesting that the ingestion rate was overestimated. As an alternative to the above, the assumption that excretion equals 27% was made and the amount of energy necessary to balance the budget calculated (Tables 10B through 13B). This resulted in a decrease in the estimation of food ingested (I) with daily consumption equalling 2.3 and 1.7% of body weight for the youngest and oldest age classes, respectively. Gruber (1982) reported that lemon sharks averaging 1754 g consumed 2. 7% of its' body weight each day when fed to satiation. Medved (1985) found that sandbar sharks, Carcharhinus plumbeus ranging in size from 1.2 to 2.1 kg had gut content averaging 0.96% of body weight. The average rates at which fish in nature evacuate food from their stomachs must be equal to their average rates of food consumption (Davis and Warren, 1970). Graeber (1974) reported that following lows in feeding there is a 2 to 3 day increase in food intake in the lemon shark. Longval et al. (1982) reported that laboratory lemon sharks apparently feed on a 4 day cycle. Medved (1985) found that the tire required for the gut of sandbar sharks, Carcharhinus plumbeus to be 98% evacuated was 70.7 hours for sharks fed blue crabs. The above findings prompted an evaluation of the feeding frequency in the bonnethead shark If the feeding cycle for the bonnethead is assumed to be two, three, or four days and the values of energy ingested (I) estimated in Tables 1 OA through 13A are used, there results an underestimation of energy ingested. These findings suggest that the bonnethead shark does not engage in a feast/famine feeding regime as perhaps is the case in larger sharks This shark, as already suggested, is able to acquire prey on a more regular basis than larger

PAGE 163

148 sharks. The estimates of the production (P) component of the energy budget required few assumptions and are therefore considered the most accurate. The assumption that vertebral ring data may be used to determine age and thus growth, has already been examined in Chapter 5. The caloric values of fish tissues for a number of species were found to range from 4.7 to 5.6 kcal/g dry weight by Kushlan et al. (1986) and Gruber (1982) reported that lemon shark tissue had a caloric content of 5.07 kcal/g dry weight. Assuming that bonnethead tissue caloric value is similar to the caloric value reported by Gruber is reasonable. Seasonal changes in energy content of tissues and changes in energy content of tissues with age would introduce a degree of error into the above approximations but there is no reason to expect significant deviations from the above value. For example, Foltz and Norden (1977) found seasonal changes in caloric content of about 5.5 to 6.1 k cal/ g dry weight for the smelt, Osmerus mordax. The energy invested in embryonic tissue was determined by calculating the caloric value of full term embryos. This estimate does not include the energy consumed in mating activity or the energetic investment in producing ripe oocytes, and assumes the metabolic rate of developing embryonic tissue would not differ from the metabolic rate of an equal amount of developing somatic tissue. The values thus obtained are not meant to represent the total energetic investment in reproduction, but provide an estimate of the energy sequestered in embryonic tissue. Power performance curve. The extrapolation method developed by Brett (1964) was used to determine standard metabolic rate in bonnethead

PAGE 164

149 sharks of 1 to 1.6 kg weight. The power performance curve thus constructed predicts a standard metabolic rate of 3.17 kcal/kg/day. Altman and Dittmer (1974) list 57 values of standard metabolism for fish. These values have a mean of 6.96 (.64 S.D.) kcal/kg/day and most are for fish found in climatic zones. Brett and Groves (1979) estimate that tropical fish would have a standard metabolism of about 12 kcal/kg/day at about 26 C. Gruber (1984) reports a standard metabolic rate of 11.4 kcal/kg/day for a 1 kg lemon shark, Negaprion brevirostris. The value reported here for standard metabolism in the bonnethead shark is probably an underestimation. The failure of sharks to swim over a wide range of velocities made the estimation of standard metabolism questionable (Figure 43). In Chapter 3 it was determined that routine swimming of bonnethead sharks could be predicted by the theoretical approach of Weihs (1981). Briefly stated, this approach assumes volitional swimming will be at an optimum velocity such that energy utilized over distance traveled will be minimized and that this occurs when routine metabolism was twice standard metabolism. Brett and Blackburn (1978) found that routine metabolism is twice standard metabolism in the dogfish, Squalus acanthias. Wohlschlag and Wakeman (1978) found the same relationship in the seatrout, Cynoscion nebulosus. Assuming this relationship is valid means that a 1 kg bonnethead shark would have a standard metabolic rate of about 9.5 kcal/kg/day at 26C. Comparing the energy budgets. When the energy budgets between different sexes and within sexes for different geographic areas are compared, several interesting trends were noted. For this comparison, only Tables 1 OB through 13B were considered since these budgets are

PAGE 165

150 believed to most closely approximate the natural situation. In comparing ingestion rates (I) it is observed that at a given age females require a larger daily ration than males and Tampa Bay sharks a larger ration than Florida Bay. This would be expected considering the aforementioned differences in size. The most interesting components of the energy budget were the estimates of metabolism (M) and production (P). The greatest percentage of the energy budget was consumed in routine metabolism. Metabolism among females accounted for 53 to 67% of the budget and remained relatively constant with age except for Tampa females of age class 2 and 3 (Table 128). In these situations metabolism dropped from about 68% to about 53% at maturity. This decrease may be due to a slight overestimation of sanatic production at maturation. This may have resulted because total weight of adult females, including the weight of embryonic tissue, was used in some cases to construct the curves for estimating growth of sana. This would cause a slight overestimation of somatic growth and the percentage of the budget it represents. Likewise, a slight underestimation of the percentage of the budget utilized for production of reproductive tissues and an understimation of the percentage of the budget used for routine metabolism would result. Metabolism among males from both geographic areas were in good agreement and would be expected since they are similar in weight. The percentage of the budget utilized for metabolism was almost identical for each age class. Males of age class 2 and 3 had the highest metabolic rates. This also may be a slight overestimation since the male investment in reproduction could not be measured. If significant

PAGE 166

151 production of male gonad is not accounted for in the budget, it would result in the percentage of the budget consumed in routine metabolism (Mr) and somatic production (P ) being artificially increased. s Although the variability around the estimates of weight at age was very high in some cases (Appendix 2) comparisons are nevertheless possible. The percentage of the energy budget utilized for somatic production decreased with increasing age in every case. A similarity in the somatic production estimates as percentage of the total budget is noted among males from Tampa and Florida Bays. The high gross conversion efficiency observed for age class 0 sharks from Florida Bay is in part due to the fact that these animals were captive. Although the overall growth curves for captive female and wild-caught females were not statistically different, the captive animals had a higher growth rate within the younger age classes (see Chapter 5). The relatively high somatic production observed among females reflects their continued growth through these age classes. Somatic growth in age class 3 females from Florida Bay dropped to a low 0.5% whereas in Tampa Bay somatic growth at the same age class accounted for 3.5% of the budget. Production of reproductive tissues must be factored into the energy budget at maturation. By age class 2 both males and females have matured (see Chapter 5). Tables 10B through 13B show the energetic investment in procreation for both males and females and for both areas. Females from Tampa Bay utilized ca. 14 to 16% of their total budget for production of reproductive tissues whereas Florida Bay females utilized about 9%. The larger size of females in Tampa Bay must contribute to this increased investment in reproduction. Little

PAGE 167

152 data are available for reproductive effort among other shark species. Parsons (1982) reported a positive relationship between adult total length and litter size for the Atlantic sharpnose shark, Rhizoprionodon terraenovae. More information is available for species of bony fish. Wootton (197J) found that in the stickleback, Gasterosteus aculeatus, fecundity per spawning is primarily determined by the weight of the female and Hirshfield (1980) found the same for the medaka, Oryzias latipes. Reproductive effort in the stickleback was found to range from to 39.9% of the energy content of food consumed over a breeding season. Hirshfield (1980) reported 10 to 18% investment in reproduction for the medaka. In a study of the northern anchovy, Engraulis mordax, Hunter and Leong (1981) reported that fecundity per spawning was a function of body size and mean annual reproductive effort ranged from 8 to 11%. The results of this study show that annual reproductive effort in the bonnethead shark is within the range of efforts reported for other fishes. Although total annual effort in this shark is similar to the effort for serially spawning oviparous fishes, the reproductive effort per individual is much higher in the bonnethead. A crude approximation of the reproductive effort in males was made (Tables 118 and 138). By assuming that metabolism in age class 2 and 3 would be the same as age class 1 the aforementioned overestimation in metabolism could be accounted for and attributed to male gonad production. This resulted in a 3 to estimate of reproductive effort. The cost of producing testicular tissue, when compared to female reproductive effort is likely very small for most species but the estimates are rarely reported.

PAGE 168

153 The effect of the differing environments of Tampa and Florida Bay on reproductive effort is seen in examining Tables 10B and 12B. The female reproductive effort in Tampa Bay is statistically higher than that reported for Florida Bay. In a comparison of the reproductive effort of two populations of the Gila topminnow, POeciliopsis occidentalis, Constantz (1979) found a similar situation. In an environment of nearly abiotic constancy, the topminnow reproductive effort varied between 3. 1 and 6. 5%. In an environment subject to flash-flooding, reproductive effort peaked at 9.8%. At a given length, the average female from the more unstable environment had a higher reproductive weight than fish from the constant environment. Constantz (1979) interpreted these results as a response to relative food availability concluding that the fish of the constant environment were more food limited. In the case of the bonnethead shark, food 1 imitation does not apppear to be an important factor. When female sharks were held in captivity under environmental conditions closely approximating the natural situation (see Chapter 2) and fed to satiation daily, their growth did not differ from that observed in the wild (see Chapter 5). Likewise, gravid females held in captivity still produced pups statistically smaller than those produced by females in Tampa Bay. This suggests that something other than food limitation is resulting in these differences. Examination of the energetic investment in somatic production might also provide information relating to the above observations. Female sharks in Tampa Bay are able to continue somatic growth for a number of years after maturation whereas in Florida Bay growth of soma drops to very low values at maturation. In

PAGE 169

154 short, sharks from the more stable tropical environment are not able to invest as much energy in reproduction and must decrease somatic growth after maturation. In the rrore environmentally dynamic warm""temperate situation, investment in reproduction is higher and somatic growth is possible even during the period of peak reproductive effort. This may be explained by (1) natural selection for larger body size such that larger offspring may be produced in Tampa Bay or (2) simple phenotypic variation in body size in response to conditions more favorable for growth in Tampa Bay. Interestingly, growth of males from Tampa and Florida Bays was not statistically different (see Chapter 5). Whatever factor(s) is selecting for larger females in Tampa Bay is not affecting growth of males. This is exactly what would be expected if selection is ultimately operating for production of larger offspring. An increase in the reproductive effort of males would not necessarily result in an increase in fitness of offspring. The metabolic rate and energy budgets estimated for the bonnethead shark may be compared with the same for other species. Sharp and Dizon (1978) reported that the weight exponent (b) in the equation describing the relationship between routine metabolic rate and weight (M = aWb) in yellowfin tuna, Thunnus albacares was found to be -0.2. Kitchell et al. (1978) calculated an exponent of -0.016 for the skipjack tuna, Katsuwonus pelamis. In the bonnethead shark this exponent was suggesting a decrease in weight-specific metabolism with increasing weight that is intermediate between the above species. Brett and Blackburn (1977) list metabolic rates for a number of elasmobranch species. Converting their oxygen consumption rates to

PAGE 170

155 caloric consumption produces a range of 3. 8 to 15.9 kcal/kg/day for these primarily benthic species of elasmobranchs. Gruber ( 1984) reported routine metabolism of 14.0 kcal/kg/day for the lemon shark, Negaprion brevirostris. Routine metabolic rates of the bonnethead shark range from 9.4 to 29.6 kcal/kg/day. The bonnetheads higher metabolism may be explained by this shark's higher level of activity. The bonnethead shark requires ram ventilation for ventilating the gills and must swim continuously. The other sharks discussed above are more benthic in nature and have the ability to stop swimming movement altogether. Excluding the lemon shark, all species in which metabolic rates have been determined were temperate species and were acclimated to temperatures well below the temperatures reported in this study. Brett and Groves ( 1979) constructed a general energy budget for young carnivorous fish feeding well, in Which metabolism accounted for 44% of the budget, growth for 29% and excretion for 27%. Gruber (1984) reported that 22.4% of the budget of the lemon shark is consumed in production, 49.4% in metabolism and 28.2% in excretion. In this study, . production was found to be generally lower and metabolism generally higher than those reported above. An energy budget is a balance sheet of energy income set against energy expenditure (Brafield, 1985). Energy partitioned into one component of the budget is no longer available for use in other components. The lower conversion efficiency of the bonnethead shark may be a result of its' high metabolic rate. If metabolism is consuming a large portion of the budget during the summer months, as is the case, then less energy would be available for growth. Kitchell et al. (1977) applied a bioenergetics model to simulate growth of fish

PAGE 171

156 with different temperature optima. They found that fish with a lower temperature optimum grew faster than those with a higher optimum. In this study, growth of fish with presumably the same temperature optimum in enviroments of different temperature regimes (Tampa and Florida Bays, see Figure 2) are being compared Perhaps during the spring and fall when water temperatures are lower, conversion efficiency in the bonnethead shark is higher and metabolism lower. At some seasons and temperatures, the maintenance costs of fish may be low and the food they consume may be used very effectively for growth (Davis and Warren, 1970). This may very well be the case in the bonnethead since growth remains high during the spring and fall when metabolism would be expected to be lower. Perhaps the environment of Tampa Bay is on the average closer to the temperature optimum for the bonnethead shark thus resulting in the observed differences. It is noteworthy that sharks collected in Florida Bay and held at Sea World of Orlando in a temperature regime of about 25 C year round, grew to sizes c omparable to those collected in Tampa Bay. This observation supports temperatures importance in shaping the life histories of these populations. Conclusions Several important conclusions may be drawn from this bioenergetics analysis. (1) The bonnethead shark was found to have a relatively high metabolic rate. This is due to the high temperatures during summer months and the ram ventilation requirement in this shark. Unlike benthic shark species which often cease swimming movement during the

PAGE 172

157 warmest part of the day, the bonnethead is obligated to swim continuously. (2) The requirement for continuous swimming activity may make routine metabolism an accurate estimate of the general level of metabolism in nature. Optimization theory suggests that continuously active fishes will swim at an optimum velocity that maximizes efficiency. The optimum velocity would not be expected to vary over a 24 hour period and indeed, this is what was found in the examination of swimming velocity in the bonnethead shark. However, optimum swimming velocity may vary as environmental conditions vary, such that seasonal differences in swimming velocities may result. (3) The elevated metabolic rates observed during summer months results in a decrease in growth efficiency. Growth efficiency declined with age in each case. However, Tampa Bay females were able to continue somatic growth over a longer period of time thus resulting in higher efficiencies at later ages. (4) Female reproductive effort was highest in the Tampa Bay population. The larger body size of female sharks in Tampa Bay may result in larger offspring. (5) Temperatures accelerating effect on metabolism may be the most important factor influencing the life history characteristics of these populations.

PAGE 173

158 LITERATURE CITED Chapter Two Alcock, A. 1890. Observations on the gestation of sane sharks and rays. J. Asiat. Soc. Bengal, Allee, M.H., W.J. Emerson, A.F. Park, M.W. Park and M.W. Schmidt. 1963. "Principles of Animal Ecology", 4th edition, W.B. Philadelphia. pp 119. Budker, P. 1971. The life of sharks. Columbia Univ. Press, N.Y. 304 pp. Clark, E. 1963. The maintenance of sharks in captivity, with a report on their instrumental conditioning. In: P. W. Gilbert (editor), Sharks and survival, pp. D. C. Heath and Co. Boston. Clark, E. and K. von Schmidt. 1965. Sharks of the central gulf coast of Florida Bull. Mar. Sci. 15:13-83. Corwin, J. T. 1978. The relation of inner ear structure to the feeding behavior in sharks and rays, Scanning Electron Microscopy, Vol. II pp 11 05-1112 Gilbert, P. W. 1981. Patterns of shark reproduction. Oceanus, 24(4):30-'39. Gilmore, R.G., J.W. Dodrill and P.A. Linley. 1982. Reproduction and embryonic development of the sand tiger shark, Odontaspis taurus. Fish. Bull. Gruber, S.H. 1982. Role of the lemon shark, Negaprion brevirostris (Poey) as a predator in the tropical marine environment: A multidisciplinary study. Florida Scientist 45:46-75. Keyes, R .S. 1982. Sharks: An unusual example of cleaning symbiosis. Copeia 1 :225-227. Myrberg, A.A. Jr., and S.H. Gruber. 1974. The behavior of the bonnethead shark, Sphyrna tiburo. Copeia, 2:358-374. Olsen, A M. 1984. Synopsis of biological data on the school shark, Galeorhinus australis. FAO Fisheries Synopsis No. 139, 42 pp.

PAGE 174

159 Parsons, G.R. 1983. The reproductive biology of the Atlantic sharpnose shark, Rhi zoprionodon terraenovae (Richardson). Fish. Bull. Schlernitzauer, D.A., and P.W. Gilbert. 1966. Placentation and associated aspects of gestation in the bonnethead shark, Sphyrna tiburo. J. Morph. 120:219-232. Sciarrotta, T. C. and D. R. Nelson. 1977. Diel behavior of the blue shark, Prionace glauca near Santa Catalina Island, california Fish. Bull. 75:519-528. Teshirna, K., M. Ahmad and K. Mizue. 1978. Studies on sharks-XIV. Reproduction in the Telok Anson shark collected from Perak Malaysia. Japanese Journal of Ichthyology, 25:181-189. Teshirna, K. 1981 Studies on the reproduction of Japanese smooth dogfishes, Mustelus manazo and M. griseus. The Journal of Shimonoseki University of Fisheries:29:113-199. Weihs, D., R.S. Keyes and D.M. Stalls. 1981. Voluntary swimming speeds of two species of large carcharhinid sharks. Copeia 1:219-222. Chapter Three Brett, J.R. and J.M. Blackburn. 1978. Metabolic rate and energy expenditure of the spiny dogfish, Squalus acanthias. J. Fish. Res. Bd. Can., 35:816-821. Dizon, A.E., W.H. Neill and J.J. Magnuson. 1977. Rapid temperature compensation of volitional swimming speeds and lethal temperatures in tropical tunas (Scombridae). Env. Biol. Fish., 2:83-92. Gruber, S.H. 1984. Bioenergetics of the captive and free=ranging lemon shark; Negaprion brevirostris. A.A.Z.P.A. Annual Proceedings, pp. 341-373. Hobson, E.S. 1968. Predatory behavior of some shore fishes in the Gulf of california. U.S. Fish. Wild. Ser. Res. Rep. 73, 92 pp 1972. Activity of Hawaiian reef fishes during the evening transitions between daylight and darkness. Fish. Bull. u.s., 70:715-740. 1973. Diel feeding migrations in tropical reef fishes. der wtss. Meeresunters. 24:361-370. . 1974. Feeding relationships of teleostean fishes on coral reefs in Kona, Hawaii. Fish. Bull. U.S., 72:915-1031.

PAGE 175

160 Limbaugh, C. 1963. Field notes on sharks. In: Sharks and Survival. P.W. Gilbert, ed., pp. D.C. Heath and Co., Boston. Longval, M.J., R.M. Warner and S.H. Gruber. 1982. Cyclical patterns of food intake in the lemon shark, Negaprion brevirostris under controlled conditions. Fla. Sci., Myrberg, A.A., Jr. and S.H Gruber. 1974. The behavior of the bonnethead shark, Sphyrna tiburo. Copeia, Nelson D.R. and R.H. Johnson. 1970. Diel activity rhythms in the nocturnal, bottom-dwelling sharks, Heterodontus francisci and Cephaloscyllium ventriosum. Copeia, 4:732-739. Olla, B.L. and A.L. Studhol11E. 1978. Comparative aspects of the activity rhythms of tautog, Tautoga oni tis, bluefish, Pomatomus saltatrix, and Atlantic mackerel, Scomber scombrus, as related to their life habits. In: Rhythmic Activity of Fishes, J.E. Thorpe ed., Academic Press,IN.Y., pp131-152. Priede, I. G. 1978. Behavioural and physiological rhythms of fish in their natural environment, as indicated by ultrasonic telemetry of heart rate. In: Rhythmic Activity of Fishes, J .E. Thorpe ed., Academic N.Y., pp. 153-168. Reynolds, W.W. and M.E. Casterlin. 1979. Behavioral thermoregulation and locomotor activity of Perea flavescens. Can. J. Zool. 57:2239-2242. Schaefer, K.M. 1986. Lethal temperatures and the effect of temperature change on volitional swimming speeds of chub mackerel, Scomber japonicus. Copeia, 1:39-44. Springer, s. 1963. Field observations on large sharks of the Florida-Caribbean region. In: Sharks and Survival. P.W. Gilbert, ed., pp. D.C. Co., Boston. Weihs, D. 1977. Effects of size on sustained swimming speeds of aquatic organisms. In: Scale effects in animal locomotion. T .J. Pedley (ed.), pp Academic Press. Weihs, D. 1981. Voluntary swimming speeds of two species of large carcharhinid sharks. Copeia, 1:219-222. Webb, P.W. and R.S. Keyes. 1982. Swimming kinematics of sharks. Fish. Bull. U.S. Chapter Four

PAGE 176

161 Beamish, R J and G .A. McFarlane. 1983. The forgotten requirement for age validation in fisheries biology Trans. Amer. Fish. Soc. 112:735-'743. Beamish, R J., and G.A. McFarlane. 1985. Annulus development on the second dorsal spine of the spiny dogfish (Squalus acanthias) and its' validity for age determination 42:1799-1805. Beamish, R J G.A. McFarlane and D.E. Chilton. 1983. Use of oxytetracycline and other methods to validate a method of age determination for sable fish (Anoplopoma fimbria). Pages 95-116. In : Proceedings of the International 3ablefish Symposium. Alaska Sea Grant Report 83-3. Campana, S E and J D Neilson. 1982. Daily growth increments in otoliths of starry flounder (Platichthys stellatus) and the influence of sane environmental variables in their production Can. J. Fish. Aquat Sci 39:973-942 Gruber S H., and R G Stout. 1983. Biological materials for the study of age and growth in a tropical marine elasmobranch the lemon shark, Negaprion brevirostris. In; E D Prince and L.M. Pulos (editors), Proceedings of the International Workshop Age Determi nation of Oceanic Pelagic Fishes : Tunas, Billfishes, and Sharks, p NOAA Tech. Rep. NMFS 8 Holden, M J., and M .R. Vince 1973. Age validation studies on the centra of Raja clavata using tetracycline. J. Cons. Int. Explor. Mer. 35:13..:17 Pannella, G. 1971. Fish otoliths: daily growth layers and periodical patterns. Science 173:1124-'1127. Parsons, G.R. and K.M. Peters (in review). Age determination of larval and juvenile sheepshead, probatocephalus using otolith daily growth increments. and growth Archosargus Peters, K M., and R .H. McMichael. (in preparation). Early life history of the black drum, Pogonias cromis (Pi sces: Sciaenidae) in Tampa Bay, Florida. Ralston S and G.T. Miyamoto. 1983. Analyzing the width of daily otolith increments to age Hawaiian snapper Pristipomoides filamentosus. Fish. Bull U S 81:523 -535. Smith, S E 1984. Timing of tetracycline-injected leopard 113: 308313. vertebral band deposition in sharks. Trans Am. Fish. Soc. Sokal, R.R., and F.J. Rohlf 1981. Biometry, 2nd edition. W .H. Freeman and Company publisher. 859 pp. . Weber, D D and G. J Ridgway. 1962. The deposition of tetracycline drugs in bones and scales of fish and its' possible use for

PAGE 177

162 marking. Prog. Fish. Cult., October:150-155. Weber, D.D., and G.J. Ridgway. 1967 Marking Pacific salrron with tetracycline antibiotics. J. Fish. Res. Brd. Can. Wild, A., and T.J. Foreman. 1980. The relationship between otolith increments and time for yellowfin and skipjack tuna marked with tetracycline. Inter.-Amer. Trop. Tuna. Comm. Bull., Chapter Five Aasen, 0., 1966. Blahaien, Prionace glauca (Linnaeus), 1958. Fisken og havet, 1. Beamish, R.J., and G.A. McFarlane. 1983. The forgotten requirement for age validation "in fisheries biology. Trans. Am. Fish. Soc., 112:735-743. Berry, J., Jr. 1965. The use of vertebral rings to determine age and growth of Raja eglanteria, the clearnose skate in Delaware Bay. Unpubl. M.S. Thesis, University of Delaware. Branstetter, S. and J.D. McEachran. 1986. Age and growth of four carcharhinid sharks common to the Gulf of Mexico: A summary paper. Proceedings of the second international conference on Indo-Pacific fishes. T. Uyeno, R. Aria, T. Taniuchi and K Matsuura. p. Ichthyological Society of Japan Tokyo. Bullis, H.R. 1967. Depth segregations and distribution of sex-maturity groups in the marbled catshark, Galeus arae. In: P.W. Gilbert et al., eds., "Sharks, skates and raysrr:-Johns Hopkins Press, Baltimore; Md. Casey, J.G., H.L. Pratt, and C.E. Stillwell. 1985. Age and growth of the sandbar shark (carcharhinus plumbeus) from the Western North Atlantic. Can. Jour. Fish. Aqu. Sci. Cailliet, G.M., L.K. Martin, J.T. Harvey, D. Kusher, and B.A. Welden. 1983. Preliminary studies on the age and growth of blue, Prionace glauca, common thresher, Alopias vulpinus, and shortf in mako, Isurus oxyrinchus, sharks from California waters. In: R.C. Summerfelt and G.E. Hall eds., Proceedings of the international workshop on age and growth of fish, Iowa State University Press, p. 359-370. and R.L. Radtke. 1987. A progress report on the __ analysis technique for age determination and verification in elasmobranchs. In: R.C. Summerfelt and G.E. Hall eds., Proceedings of the international workshop on age and growth, Iowa State University Press, p.

PAGE 178

163 Francis M.P. 1981. Von Bertalanffy growth rates in species of Mustelus (Elasmobranchii; Triakidae). Copeia 1:189-192. Gruber, S.H. 1981. Lemon sharks; supply-side economists of the sea. Oceanus 24:56-'64. 1984. Bioener getics of the captive and free-ranging lemon shark, Negaprion brevirostris. AAZPA Annual Proceedings, p. 339-373. and R.G. Stout. 1983. Biological materials for the study -----,.------=of age and growth in a tropical marine elasmobranch, the lemon shark, Negaprion brevirostris (Poey). In: E.D. Prince and L.M. Pulos eds. Proceedings of the international workshop on age determination of oceanic pelagic fishes: Tunas, billfishes and sharks. p. 193-'205. Holden, M.J. 1974. Problems in the rational exploitation of elasmobranch populations and some suggested solutions, p. In: Sea fisheries research. F.R. Harden Jones, eds. Scientific Books, London. and M.R. Vince. 1973. Age validation studies on the ------;------;::;-;:: centra of Raja clavata using tetracycline. J. Cons. Int. Explor. Mer. 35:13-'27. Ka to, S. and A. H. Carvallo. 1967. Shark tagging in the Eastern Pacific Ocean, 1962-65. In: "Sharks, Skates and Rays". P.W. Gilbert eds., Johns Hopkins Press, Baltimore p. 91-101. McFarlane, G.A. and R.J. Beamish. 1987. Validation of the dorsal spine method of age determination for spiny dogfish. In: R.C. Summerfelt and G.E. Hall eds., "Age and Growth of Fish"; Iowa State University Press, p. 287-300. Moss, S.A. 1967. Tooth replacement and body growth rates in the smooth dogfish, Mustelus canis (Mitchill). Copeia 4:808-811. Parsons, G. R. 1983. An examination of the vertebral rings of the Atlantic sharpnose shark, Rhizoprionodon terraenovae. Northeast Gulf Science 6:63-66. 1985. Growth and age estimation of the Atlantic __ Rhizoprionodon terraenovae: A comparison of techniques. Copeia 1 Pratt, H.L., Jr. and J .G. Casey. 1983. Age and growth of the shortfin mako; Isurus oxyrinchus, using four methods. Can. Jour. Fish. Aquatic Sci., Smith, S.E. 1984. Timing of tetracycline""injected leopard 113:308..:313. vertebral-band deposition in sharks. Trans. Am. Fish. Soc.,

PAGE 179

164 Springer, S. 1967. Social organization of shark populations. In: P.W. Gilbert et al. eds., "Sharks, Skates and Rays", Johns Hopkins Press, Baltimore Maryland, p. Stevens, J.D. 1975. Vertebral rings as a means of age determination ion the blue shark, Prionace glauca. J. Mar. Biol. Ass. U.K., 55:657-"665. Thorson, T.B., D.E. Watson, arid C.M. Cowan. 1966. The status of the freshwater shark of Lake Nicaragua. Copeia 3:385-402. and E.J. Lacy. 1982. Age, growth rate and longevity of carcharh1nus leucas estimated from tagging and vertebral rings. Copeia 1:110-116 Vaughan, D.S., and P. Kanciruk. 1982. An empirical comparison of estimation procedures for the von Bertalanffy growth equation. J. Cons. Int. Explor. Mer., 40:211-219. Waring, G.T. 1984. Age, growth, and mortality of the little skate off the northeastcoast of the United States. Trans. Am. Fish. Soc., 1134: 314'-\321 Wass, R.C. 1971. A comparative study of the life history, distribution and e cology of the sandbar shark and the grey reef shark in Hawaii. Ph.D. dissertation, University of Hawaii, Honolulu. Chapter Six Altman, P.L., and D.S. Dittmer eds., 1974. "Biological Data Book", Vol. 3, Part V, "Fishes", 2nd Ed., pp. 1624-1630. Fed. Am. Soc. Biol., Biol. Handbooks, Bethesda, Maryland. Baird, R.C., and T .L. Hopkins, 1981. Trophodynamics of the fish Valenciennellus tripunctulatus. II. Selectivity, grazing rates and resource utilization. Mar. Ecol. Prog. Ser., Beamish, F.W., 1981. Swimming performance and metabolic rate of three tropical fishes in relation to temperature. Hydrobiologia, 83:245'"'254. Brafield, A.E., and D.J. Solomon, 1972. Oxycalorific coefficients for animals respiring nitrogenous substrates. Comp. Biochem. Physiol Brafield, A.E., 1985. Laboratory studies of energy budgets. In: "Fish Energetics", P. Tytler and P. Calow eds. Johns Hopkins Press, Baltimore, Maryland. p.

PAGE 180

165 Brett, J.R., 1965. The relation of size to rate of oxygen consumption and sustained swimming speed of sockeye salrron, (Oncorhynchus nerka). J. Fish. Res. Bd. Can., ----:----'-:' 1973. Energy expenditure of sockeye salrron, Oncorhynchus nerka, during sustained performance. J. Fish. Res. Bd. Can. 30:1799-1809. and J .M. Blackburn, 1978. Metabolic rate and energy ------:"":'..,....--..:. expenditure of the spiny dogfish, Squalus acanthias. J. Fish. Res. Bd. Can., and T.D. Groves, 1979. Physiological energetics. In: "Fish Physiology", Vol. VIII, W.S. Hoar, D.J. Randall and J.R. Brett eds. Academic Press, N.Y., p. 279-352. Chan, D.K., and T.M. Wong, 1977. Physiological adjustments to dilution of the external medium in the lip-shark, Hemiscyllium plagiosum. III. Oxygen consumption and metabolic rates. J. Exp. Zool., 200:97-102. Clarke, T.A., 1978. Diel feeding patterns of 16 species of mesopelagic fishes from Hawaiian waters. Fish. Bull. u.s. 76:495-513. Constantz, G.D., 1979. Life history patterns of a livebearing fish in contrasting environments. Oecologia, 40:189. Crawshaw, L.I., and H.T. Hammel, 1973. Behavioral temperature regulation in the California horn shark, Heterodontus francisci. Brain Behav. Evol. Davis, G.E., and C.E. Warren, 1970. Estimation of food consumption rates. In: W. E'. Ricker, ed. "Methods of Assessment of Fish Production in Fresh Waters". Blackwell Scientific, 313 p. Elliot, J.M., 1976. The energetics of feeding, metabolism and growth of brown trout, Salmo trutta in relation to body weight, water temperature and ration size. J. Anim. Ecol. 45:923. . ., Farmer, G. J. and F. W. Beamish, 1969. Oxygen consumption of Tilapia nilotica in relation to swimming speed and salinity. J. Fish. Res. Bd. Can. 26:2807-2821. Foltz, J.W., and C.R. Norden, 1977. Seasonal changes in food consumption and energy content of smelt, Osmerus mor-dax in Lake Michigan, Trans. Am. Fish. Soc., 106:230-234. . Graeber, C.R., 1974. Food intake patterns in captive juvenile lemon sharks, Negaprion brevirostris. Copeia, 2:554-556. Gruber, S.H., 1982. Role of the lemon shar-k, Negaprion brevirostris as a predator in the tropical marine envir-onment: A multidisciplinar-y study. Fla. Sci., 45:46-75.

PAGE 181

166 1984. Bioenergetics of the captive and free-ranging lemon shark, Negaprion brevirostris. AAZPA Annual Proceedings. Hirshfield, M.F., 1980. An experimental analysis of reproductive effort and cost in the Japanese medaka, Oryzias latipes. Ecology, 61:282. Hopkins, T .L., and R.C. Baird, 1981. Trophodynamics of the fish Valenciennellus tripunctulatus. I. Vertical distribution, diet and feeding chronology. Mar Ecol. Prog. Ser., 5:1-10. Hughes, G.M., and S.I. Umezawa, 1968. Oxygen consumption and gill water flow in the dogfish, Scyliorhinus cani cula. J. Exp. Biol. 49:557-564. Hunter, J. R. and R. Leong, 1981 The spawning energetics of female northern anchovy, Engraulis mordax. Fish. Bull. 79-215. Jobling, M. 1985. Growth. In: P. Tytler and P. Calow eds. "Fish Energetics". John Hopkins University Press, Baltimore, Maryland. p. 213-230 Kitchell, J.F., D.J. Stewart, and D. Weininger. 1977. Application of a bioenergeticsmodel to yellow perch(Perca flavescens) and walleye (Stizostedion vitreum vitreum). J. Fish. Res. Board Can. 34:1922-:1935. W.H. Neill, A.E. Dizon, J.J. Magnuson. 1978. Bioenergetic spectra of skipjack and yellowfin tunas. In: "The Physiological Ecology of Tunas", Sharp and Dizon editors, Academic Press, N.Y., pp. 357-368. Kushlan, J. A. S. A. Voorhees, W. F. Loftus and P. C. Frohring, 1986. Length, mass and calorific relationships of everglades animals. 49:65'-79. Lenfant, C., and K. Johansen, 1966. Respiratory function in the elasmobranch Squalus suckleyi, G. Respir. Physiol. 1:13-29. Longval, M.J., R.M. Warner and S.H. Gruber, 1982. Cyclical patterns of food intake in the lemon shark, Negaprion brevirostris under controlled conditions. Fla. Sci., Medved, R.J., 1985. Gastric evacuation in the sandbar shark, Carcharhinus plumbeus. J. Fish. Biol., 26:239-253. ' Murray, R.W., 1971. Temperature receptors. In: W.S. Hoar and D.J. Randall eds., "Fish Physiology", Vol. V, Academic Press, N.Y., p. 121-133. Parsons, G.R., 1982. The reproductive biology of the Atlantic sharpnose shark, Rhizoprionodon terraenovae. Fish. Bull. U.S., 81:61-73.

PAGE 182

16' 7 Pickering, A.D., 1981. "Stress and Fish". Academic Press, N.Y., 367 p. Piiper, J., and D. Schumann, 1967. Efficiency of 0 exchange in the gills of the dogfish, Scyliorhinus stellaris. tRespir. Physiol 2:135-148 Priede, I. G., and P. Tytler, 1977. Heart rate as a measure of metabolic rate in teleost fishes; 8almo gairdneri, 8alrro trutta and Gadus morhua. J Fish Biol., 10:23 1 -242 Solomon, D J and A.E. Braf ield, 1972. The energetics of feeding metabolism and growth of perch, Perea fluviatilis. J Anim. Ecol., 41 :699 Stearns, S C 1977. The evolution of life history traits: A critique of the theory and a review of the data. Ann. Rev. Ecol. Syst., Torres, J.J. B.W. Belman and J.J. Childress, 1 979. Oyxgen consumption rates of midwater fishes as a function of depth of occurrence Deep Sea Research 26A:185-197. Winberg, G G., 1956. "Rate of metabolism and food requirements of fishes". Beloruss State Univ. Minsk., Fish. Res. Bd. Can. Transl. ser. No. 194 ( 1 960). Wohlschlag, D E. and J .M. Wakeman, 1978. Salinity stresses, metabolic responses and distribution of the coastal spotted seatrout, Cynoscion nebulosus. Contributions in Mari ne Science 21:171-185. Wooton, R .J., 1977. Effect of food limitation during the breeding season on the size, body components and egg production of female sticklebacks, Gasterosteus aculeatus. J Anim. Ecol. 46: 823. 1985. Energetics of reprodu ction. In: P Tytler and P. ----,c=-a-=1:-o-w_e_ d....,..s..:... "Fish Energetics". John Hopkins University Press, Baltimore Maryland p

PAGE 183

168 APPENDIXES

PAGE 184

169 APPENDIX 1 STANDARD DEVIATIONS OF MEAN LENGTHS CALCULATED USING THE VON BERTALANFFY GROWTH EQUATION The standard deviations appearing in this Appendix were calculated using the standard deviations for the parameters of the von Bertalanffy equation appearing in Table 7. By using a Monte Carlo simulation, one thousand combinations of the von Bertalanffy parameters were generated for each age group based on the observed standard deviations. These data were then used to back calculate one thousand total lengths and the standard deviations of those lengths calculated and tabulated below. Tampa Bay Males Age 1yr 2" 3" 4" 5" Mean 63.5cm 73.3" 75;8" 76.9" Standard Deviation 27.6 32:6 36:0 58:2 Florida Bay Males Age 1yr 2" 3" 4" 5" Mean 58":Tcm 65:2" 70.0" 75.5" Captive Females Age 1yr 2" 3" 4" 5" 6" Mean '58"":lfcm 75:8" 90.7" 94.111 95:711 Standard Deviation 14.9 14;6 1 3:4 14; 1 Standard Deviation 4.0 3;9 3;0 Tampa Bay Females Age 1yr 2" 3" 4" 511 6" Mean 73.7cm 84:3 II 92.0 98;9 II 10 2.7" 105:811 Standard Deviation 12.0 1L7 10:7 10:3 9.5 8:7 Florida Bay Females Age 1yr 211 311 411 5 611 Mean b2.1fcm 73;5 II 81.4 II 91:9" 94:9 Standard Deviation 12.6 12;6 11.8 11:5 10.7 1 0 1

PAGE 185

APPENDIX 2 STANDARD DEVIATIONS OF MEAN WEIGHTS CALCULATED USING THE LOGISTIC GROWTH EQUATION 170 The standard deviations appearing in this Appendix were calculated as explained in Appendix 2 using the standard deviations of the parameters of the logistic equation shown in Table 8 Tampa Bay Males Age Mean Standard Deviation 1yr 1326g 819 2" 1701" 845 3" 1937" 782 4" 2131" 751 5" 2196" 708 Florida Bay Males Age Mean 1yr 803g 2" 1250" 3" 151 0" 4" 1704" 5" 1777" Captive Females Age 1yr 2 3" 4" 5" 6" Mean 087g 2178" 2877" 3193" 3324" 3361" Standard Deviation 516 579 528 494 448 Standard Deviation 427 829 741 524 439 399 Tampa Bay Females Age Mean Standard Deviation 1yr 1780g 798 2" 2739" 1226 3" 3768" 1498 4" 4902" 1746 5" 5634" 1731 6" 616611 1676 Florida Bay Females Age Mean Standard Deviation 1yr 1353g 1199 2" 2007" 1347 3 2461" 1284 4" 2842" 1208 5" 3020" 1113 6" 3139" 1009


printinsert_linkshareget_appmore_horiz

Download Options

close
No images are available for this item.
Cite this item close

APA

Cras ut cursus ante, a fringilla nunc. Mauris lorem nunc, cursus sit amet enim ac, vehicula vestibulum mi. Mauris viverra nisl vel enim faucibus porta. Praesent sit amet ornare diam, non finibus nulla.

MLA

Cras efficitur magna et sapien varius, luctus ullamcorper dolor convallis. Orci varius natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Fusce sit amet justo ut erat laoreet congue sed a ante.

CHICAGO

Phasellus ornare in augue eu imperdiet. Donec malesuada sapien ante, at vehicula orci tempor molestie. Proin vitae urna elit. Pellentesque vitae nisi et diam euismod malesuada aliquet non erat.

WIKIPEDIA

Nunc fringilla dolor ut dictum placerat. Proin ac neque rutrum, consectetur ligula id, laoreet ligula. Nulla lorem massa, consectetur vitae consequat in, lobortis at dolor. Nunc sed leo odio.