Respiratory physiology of juvenile tarpon, Megalops atlanticus

Respiratory physiology of juvenile tarpon, Megalops atlanticus

Material Information

Respiratory physiology of juvenile tarpon, Megalops atlanticus
Geiger, Stephen P. (Stephen Patrick)
Place of Publication:
Tampa, Florida
University of South Florida
Publication Date:
Physical Description:
vii, 96 leaves : ill. ; 29 cm.


Subjects / Keywords:
Tarpon -- Respiration ( lcsh )
Dissertation, Academic -- Marine science -- Masters -- USF ( FTS )


General Note:
Thesis (M.S.)--University of South Florida, 1993. Includes bibliographical references (leaves 82-93).

Record Information

Source Institution:
University of South Florida
Holding Location:
Universtity of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
029493857 ( ALEPH )
29333533 ( OCLC )
F51-00101 ( USFLDC DOI )
f51.101 ( USFLDC Handle )

Postcard Information



This item is only available as the following downloads:

Full Text


Respiratory physiology of juvenile tarpon, Meqalops atlanticus. by Stephen P. Geiger A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Marine Science University of South Florida April, 1993 Major Professor: Joseph J. Torres, Ph.D.


Graduate Counctl University o f South Florida Tampa, Florida CERTIFICATE OF APPROVAL Master's THESIS This is to certify that the Master's Thesis of Stephen P Geiger with a major in Marine has been approved by the Examining Committee on 20 November, 1992 as satisfactory for the Thesis requirement for the Master of Science degree. Thesis Committee: Maj -6 r T orres Member: yo' f. tioab&e:Et'..-


Stephen P. Geiger c 1992 All Rights Reserved


ACKNOWLEDGEMENTS I wish to thank the Don Hawley Foundation which provided the initial funding for this study, especially George C. Hixon, Joseph M. Hixon III, and Mike Collins, as well as Game Conservation International. This study was funded in part by interagency agreement C7491 and EC044 between the Florida Marine Fisheries Commissiion and the Florida Marine Research Institute. Special thanks to Dr. c. Mangum for providing data on the respiratory properties of tarpon blood, and to J. Donnelly for unlimited help around the lab. I would like to thank my committee members R. Crabtree and P. Motta and Major Professor J. Torres for guidance and direction during this study. Most importantly, I would like to thank my wife, Leslie, for her loving encouragement throughout all we have shared.


TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES ABSTRACT INTRODUCTION Historical perspective Air-breathing Oxygen consumption Oxygen transport Hydrogen sulfide swim bladder OBJECTIVES MATERIALS AND METHODS Collection and maintenance of specimens Obligate air-breathing Air-breathing Oxygen consumption Hydrogen sulfide oxygen transport Swim bladder i iii iv v 1 3 3 10 13 16 19 21 22 22 25 26 29 34 36 37


RESULTS Obligate air-breathing Air-breathing Opercular ventilation Oxygen consumption Hydrogen sulfide oxygen transport Swim bladder DISCUSSION Obligate air-breathing Oxygen consumption Air-breathing Opercular ventilation Hydrogen sulfide Oxygen transport Summary: Tarpon as air-breathers LITURATURE CITED Appendix I ii 39 39 40 47 50 67 62 62 65 66 67 70 73 76 78 81 82 94


LIST OF TABLES Table 1. Total number of breaths recorded during each observation period. Ten fish observed in each period. Observation periods measured in hours. 40 Table 2. Air-breath Frequency of tarpon within given ranges of oxygen concentration and temperature. 41 Table 3. TWO-WAY ANOVA for opercular ventilation frequency based on oxygen concentration in parts per million and temperature C. 47 Table 4. Mean opercular ventilation frequencies in ventilations per minute as grouped by the appropriate temperature and dissolved oxygen concentration. 50 Table 5. Hematocrit values from field caught tarpon and tarpon held in the lab for periods exceeding twelve months. 63 Table 6. Appendix 1 Blood characteristics of several fishes. 95 iii


LIST of FIGURES Figure 1. Schematic of annular respirometer. outer radius = 60 em, inner radius = 30 em, depth= 20 em; total volume= 41.7 liters. 24 Figure 2. Linear regression models for air-breathing frequencies versus dissolved oxygen. 42 Figure 3. Linear regression model for air-breathing frequency versus temperature. 45 Figure 4. Non-linear regression model for opercular ventilation frequency versus dissolved oxygen concentration. 48 Figure 5. Linear regression models for opercular ventilation frequency versus temperature. 51 Figure 6. Linear regression model for standard oxygen consumption versus weight. 54 Figure 7. Linear regression model for metabolic cost of swimming versus swimming speed. 55 Figure 8. The effects of oxygen concentration upon routine oxygen consumption 57 Figure 9. The effects of oxygen concentration upon swimming speed. 58 Figure 10. Regression models for opercular ventilation frequency versus sulfide concentrations 60 iv


RESPIRATORY PHYSIOLOGY OF JUVENILE TARPON, MEGALOPS ATLANTICUS. Stephen P. Geiger An Abstract Of a thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Marine Science University of South Florida April, 1993 Major Professor: Joseph J. Torres, Ph.D. v


Two modes of respiration have been described for juvenile tarpon: aquatic, where tarpon obtain the oxygen used in metabolic processes only by using the gills, and aerial, where tarpon obtain oxygen using an accessory respiratory structure, the swim bladder. This study describes the influence of dissolved oxygen, temperature, pH, and H2S on air-breathing frequency in tarpon. Air-breathing frequency, an indicator of oxygen uptake through the swim bladder, increased as dissolved oxygen concentration decreased at temperatures below 29C. In normoxic water at l9C, tarpon air-breathe about once every two hours increasing to more than four times per hour in severely hypoxic water at 29C. At 33C air-breathing frequency was consistently above 4 breaths per hour regardless of P02 Air-breathing frequency also increased with increasing concentrations of sulfide and decreasing pH. The opercular ventilation frequency increased rapidly over the range of 0 -1 parts per million of dissolved oxygen concentration, then showed little change in increasingly well oxygenated water. Opercular ventilation frequency decreased with decreasing temperature and pH, and increasing concentrations of sulfide. Tarpon were shown to survive acute exposure to H2S concentrations exceeding vi


100 uM. Ten tarpon survived for eight days without access to atmospheric oxygen when held in normoxic water; seven of those ten tarpon survived fourteen days. The fact that all required oxygen was obtained solely through the gills for a sustained period in ten individuals led to the conclusion that the tarpon is a facultative rather than an obligate air-breather. Tarpon routinely take up an estimated 27% of their required oxygen from the air, using the swim bladder. The standard metabolic rate in terms of ml 02 hr-, Y, was related to weight, W, by the following equation: y = 1 57 wtJ.833 (Significance= 0.001, r2 = 62.4 %) The cost of swimming, Y, in ml 02 kg-hr-was related to swimming speed, X, in body lengths per minute by the following equation: Y = 3.09X + 73.5 (Significance= 0.001, r2 = 44.4 %) Abstract Approved:,...Major Professor: Joseph J. Torres, Ph.D. Professor, Department of Marine Science Date Approved: vii


Respiratory physiology of juvenile tarpon, Megalops atlanticus (Valenciennes) Introduction Fishes of the genus Megalops include the tarpon, M atlanticus, and the ox-eye tarpon, M cyprinoides (Wade,1962). Tarpon are found throughout the tropical Atlantic, and expatriates occur as far north as Ireland and Nova Scotia {Twoney and Byrne, 1985). ox-eye tarpon are found in the western tropical Pacific and tropical regions of the Indian ocean {Coates, 1987). These two species belong to the Elopiformes, a order which also includes the ladyfishes, and bonefishes. The infradivision, the elopomorpha, which includes eels, the Anguilliformes, and spiny eels, the Notocanthiformes is united by a unique larval stage known as the leptocephalus (Smith, 1984). Tarpon spawn near the continental shelf break, with males starting at an age of seven to ten years and females first spawning at an age of ten to thirteen years {Cyr, 1991). Spawning males are usually more than 22 kg, and females greater than 29kg (Cyr, 1991). Tarpon produce one million or more eggs (Babcock, 1936; Cyr,1991) which form leptocephali measuring less than 3 mm when they hatch {Crabtree, et al., in press). Stage II leptocephali and metamorphosing larvae have occasionally been captured in nearshore waters and estuaries. Small juveniles are found 1


in mangroves, salt marshes and stagnant pools of water in coastal regions (Breder,1939; Harrington,1958). Juveniles are occasionally found far up-river in fresh water (Harrington and Harrington, 1960). Juvenile tarpon spend about one year after metamorphosis in mangroves and salt marshes (Cyr,1991). Individuals living in mangroves must be able to withstand a broad range of temperatures (minimum of 12 to a maximum of 36C), salinities (minimum 5 to maximum 40 ppt), pH's (minimum 5.7 to maximum 8.8), and oxygen concentrations (minimum 0.0 to maximum 10.6 mg/1) (data from J.David, st.Lucie County mosquito control, personal communication). Moffett and Randall (1957) showed tarpon could survive in waters of 15 to 40C; sudden temperature changes were observed to cause stress. 2 Juvenile tarpon are capable of withstanding wide variations in dissolved oxygen including exposure to hypoxic conditions (2.8 and 1.02 ppm 02 ) (Schlaifer and Breder ,1940; Schlaifer, 1941; and Breder, 1942) by breathing air. Schlaifer (1941) concluded that tarpon were obligate airbreathers. In his experiments, tarpon survived a maximum of 128 hours when denied access to the surface. A common method of coping with severe hypoxia in fishes is the utilization of atmospheric oxygen (Das, 1940; Carter and Beadle, 1929;Johansen, 1966). Tarpon obtain oxygen by breathing air into their swim bladder. The swim bladder is


3 moderately well vascularized and filled with lung-like tissue which is assumed to be the site of 02 uptake (Babcock,1936). Liem (1989) suggested that the degree of vascularization in the tarpon swimbladder, when compared to the lungfish swimbladder, indicates that tarpon use the swim bladder equally for air-breathing and buoyancy. Tarpon may rely on the swimbladder increasingly for buoyancy after the juvenile stage, especially when adults undergo long migrations in highly oxygenated offshore waters. The ox-eye shows little modification in its swim bladder to facilitate air breathing (Liem, 1989). Historical Perspective Air-breathing A variety of mechanisms are used by fish to deal with hypoxia. Numerous species breathe air directly using modified gills, vascularized buccal and branchial regions, stomach and intestines, and swimbladders (Johansen, 1966). Kramer (1983) discussed a common alternative to airbreathing called aquatic surface respiration in which fishes ventilate their gills with oxygenated water from the thin surface layer. Numerous authors have shown that fish increase ventilation of the gills at the onset of hypoxia in an attempt to maintain oxygen uptake (Graham, 1978;


4 Hughes and Singh, 1970; Itazawa and Takeda,1978). Alternatively, fishes may show reduced activity including a reduced rate of gill ventilation, thereby reducing oxygen demand during hypoxia (Wu and Woo, 1978; Jordan,1976). Air-breathing in fishes has evolved with two major modes. One mode of respiration, seen most commonly in intertidal fishes, is often accomplished through highly vascularized skin and unspecialized gills (Al-Kadhomiy and Hughes, 1988; Martin and Lighton, 1989). The more common mode of air-breathing is observed in freshwater fishes which utilize specializations in existing organs including the swim bladder, intestinal tract, and bucchal cavity (Johansen, 1966). Liem {1989) has shown that air-breathing evolved in three of the four major groups of Teleostei. Air breathing is wide-spread in the Osteoglossomorpha and the Euteleostei, but absent from the Clupeomorpha. In the Elopomorpha only the two species of Megalops breathe air. Megalops atlanticus being better adapted for taking up atmospheric oxygen through its swim bladder than M cyprinoides. In most teleosts air-breathing is accomplished by using the buccal pump to force air into the swim bladder or digestive tract but some fish use a less efficient cough mechanism at times (Liem,1989). Channa sp. uses the cough mechanism when it travels out of water (Liem, 1984). In water, air-breathing teleosts exhale while the head is above


5 the swim bladder as the fish approaches the surface. Hydrostatic pressure and the reduced pressure created by the expansion of the buccal cavity both work to expel the air. Air is drawn into the mouth as the fish surfaces and the buccal cavity continues to expand. The head is then lowered and both buccal pump and hydrostatic pressure help force the bubble into the swim bladder or other airbreathing organ (Liem, 1989). Excess air is expelled from behind the opercula so that aquatic respiration can continue. Many species of fish have been classified as either an obligate air-breathing species, requiring access to the surface at all times, or as facultative air-breathers, utilizing the capacity to obtain atmospheric oxygen to expand the available habitat either spatially or temporally. Most experimenters employ an experimental design which prevents. access to the surface while maintaining normoxic conditions to determine if a species requires access to the surface. Authors have classified fishes as obligate after periods as short as 10 minutes (Willmer, 1934), 1 week (Graham et al, 1977; Dubale, 1951) and as long as 2 weeks (Magid, 1966; Jordan, 1976). carter and Beadle (1929), and Willmer (1934) first studied air-breathing in the fishes of South American swamps. Carter and Beadle stated that fishes which could survive 12 hours without access to the surface were facultative air-breathers. Gee and Graham (1978)


6 considered both Hoplosternurn thorocaturn and Brochis splendens, both South American catfish, to be continuous facultative air-breathers, species which carne to the surface constantly but were capable of meeting their oxygen requirements in norrnoxic by aquatic respiration alone. Beta splendens survived at least 48 hours without access to the surface and fl. thorocaturn survived up to seven days. Some species are considered obligate air-breathers only at high temperatures (Rahn et al, 1971; Stevens and Holeton, 1978). Fishes which are obligate air-breathers only under certain environmental conditions such as Lepisosteus osseus(Rahn et al, 1971) are more appropriately termed facultative airbreathers. Schlaifer (1941) presented data suggesting that tarpon are obligate air-breathers. Schlaifer placed fish measuring 10 -13 ern into small 20 ern by 184 ern aquaria with a wire mesh 6 ern from the bottom. The tarpon were restricted to below the wire mesh. Fish used in his study survived an average of 59.2 hours without access to the surface (range 7 -128 hours; standard deviation = 41.3 hrs, n=13), although all thirteen died. In nine of the trials aeration was provided and dissolved oxygen concentration was near air saturation. In four of the trials no aeration was provided and (02 ) at death was 1.9 ccjliter, roughly 35% of air saturation.


Schlaifer and Breder (1940) and Schlaifer (1941) found that at cool temperatures the frequency of air-breathing in tarpon was directly related to dissolved oxygen concentration (oxygen partial pressures of 25%, 50% and 100% of air saturation) but did not study the relation at temperatures greater than 22C. The authors studied air-breathing at temperatures ranging from 18 to 40C. The rate of airbreathing increased up to 35C and then decreased. 7 The importance of atmospheric oxygen in respiration can be studied by observing the air-breathing frequency as temperature and oxygen concentration vary. The increased air-breathing frequency (fAa) corresponds to an increased oxygen uptake at accessory respiratory organs as oxygen consumption rises with increasing temperature or when available oxygen in the water decreases. Many authors have found that air-breathing frequencies increase as temperature rises. Some air-breathing fish that exhibit this behavior are the bowfin, Amia calva (Horn and Riggs, 1973; Johansen, 1970) the South American fishes Ancistrus chagresi and Hypostomus plecostomus (Graham and Baird, 1982) and the gar, Lepisosteus osseus (Rahn et. al., 1971). Bimodally respiring fish, those species which can take up oxygen from the atmosphere and the water, have been shown to increase air-breathing frequency when exposed to hypoxic water including the South American catfish Coryadoras aenus


8 (Kramer and McClure, 1980), the bowfin (Mckenzie, 1991; Johansen, 1970), as well as ff. plecostomus and A chagresi (Graham and Baird, 1982). The African lungfish, Protopterus aethiopicus, may be so dependent on airbreathing, due to its poorly developed gill structure, that no change in fAs can be observed as P02 declines (Lenfant and Johansen, 1968; Johansen and Lenfant, 1968) although Jesse (1967) observed a slight increase in air-breathing frequencies in moderate hypoxia. Few authors have examined how the air-breathing frequency of fish may change in relation to both oxygen tension and temperature. Graham and Baird (1982) showed that the threshold partial pressure below which A chagresi, a facultative air-breather, would air-breathe decreased from a P02 of 57 mmHg at 15C to 33 mmHg at 25 to 30C. Rahn (1971) showed that L osseus could take up sufficient oxygen at the gills at temperatures below about 22C, even down to a P02 of 40 mmHG, but that above 22C the swim bladder might serve to supply the body with 75% or more of the total oxygen needed for routine metabolism. The presence of predators can act to inhibit airbreathing. The bowfin, Amia calva, exhibited maximum fAs at night, presumably to minimize predation from visuallyoriented wading or diving birds (Horn and Riggs, 1973). Kramer and Graham (1976) showed that schools of airbreathing fishes decreased their fAs when the surface above


them was agitated, also a potential mechanism for avoiding predators. Kramer (1986) and Kramer (1983) have shown that the presence of herons near man-made ponds containing air-breathing fish caused a reduction in airbreathing behavior. 9 Aquatic oxygen uptake can be estimated using the opercular ventilation frequency (fov) Ventilation frequency can be used as an indicator of opercular ventilation volume, the volume of water flowing over the gills from which oxygen might be removed. The relation between f Aa and fov is not always direct. Both types of oxygen uptake increase with the onset of hypoxia (Jesse, 1967; Hughes and Singh, 1970; Gee and Gee, 1991; McKenzie, 1991) but many fishes decrease opercular ventilation below a threshold P02 (Johansen, 1970, Hughes and Singh, 1970; Bicudo and Johansen, 1979). Decreased water flow across the gills may help to prevent deoxygenation of blood previously oxygenated at the airbreathing organ (Johansen, 1966). McCormack {1967) showed that in the Florida spotted gar, Lepisosteus osseus, f0v doubled from 20 to 40 beats per minute in the interval between air-breaths. The gar was presumably increasing the oxygen uptake at the gills as the oxygen in the swim bladder was exhausted. Hughes and Singh {1970) published similar observations for the climbing perch, Anabas testudinus, as did Johansen and Lenfant (1967) for the lungfish,


10 aethiopicus. Oxygen consumption oxygen consumption rates vary widely among fishes. In non-air-breathing fish routine oxygen consumption (V02 ) ranges from about 45 ml 02 kg1 hr1 in such sluggish fish as carp (Lumholt and Johansen, 1979), and the pumpkinseed, Lepomis gibbosus (Brett and Sutherland, 1965), to values such as 240 ml02 kg1 hr1 in the mosquitofish, Gambusia affinis (Cech et al., 1985), and 560 ml02 kg1 hr1 in juvenile sockeye salmon, Oncorhyncus nerka (Brett, 1964). Active metabolic rate may be as much as ten times the routine rate in some fishes (Brett and Sutherland, 1965). Air-breathing fish have oxygen consumption rates that fall in the range of purely gill breathing fish. The airbreathing erythrinid fish, Hoplias microlepis, consumes 54 ml02 kg1 hr1 (Dickson and Graham, 1986) and another South american fish, Heteropneusteus fossilis, routinely consumes 40-60 ml02 kg1 hr1 (Arunachalam and Pandian, 1976). In the most active air-breathing fishes, on a weight specific basis, the amphibious blenny Alticus kirki (Martin and Lighton, 1989) and Channa striatus, one of the snakeheaded fishes (Singh, 1986) routine oxygen consumption exceeds 2 0 0 ml02 kg 1 hr1


11 Changes in temperature and oxygen concentration may both affect oxygen consumption rates. Temperature increases cause increased oxygen consumption in salmon (Brett and Glass, 1973) and declining oxygen concentration has been shown to cause increased respiration rates in the gourami, Trichogaster trichoptera (Burgrenn, 1979) and in both the red grouper, Epinephalus akaara, and the black sea bream, Mylio macrocephalus (Wu and Woo, 1984). Lagodon rhomboides, the pinfish, exhibited non-linear increases in routine swimming activity as temperature changed from 10 to 30 C but that increases in metabolic rate were linear over the 20 degree change (Wohlschlag et al., 1968). Todd and Ebeling (1967) showed that the mudsucker (Gillichthys mirabilis) became quiescent at the onset of hypoxia but increased activity when 02 concentrations dropped below a critical level, about 3.0 mgjl (1.9 ml/1). Graham et al. (1990), Lai et al. (1990), Parsons (1990), Piiper et al. (1977) and others have shown a positive correlation between swimming speed and oxygen consumption in fishes. Few authors have studied the effects of oxygen concentration on swimming activity. Todd and Ebeling (1967) observed that the activity of G.mirabilis increased during hypoxia. The increased activity would presumably lead this amphibious fish to more favorable environmental conditions in other bodies of water or possible help to aerate small


bodies of water through agitation. Gee and Graham (1978) stated that the rate of air-breathing and overall activity increased as P02 declined in Hoplosternum thoracatum. Kramer {1983) states that some fishes may emigrate from areas of low oxygen content but that this behavior is most likely beneficial either for fishes with no other means of survival or in those fish having a good chance of finding more highly oxygenated water. 12 Schlaifer (1941) found that the swimming activity of tarpon decreased from 32.0 em/min. in water with 2.8 ppm 02 to 20.3 em/min. in water with 1.02 ppm. The observed decrease in activity may have been an attempt to conserve oxygen by reducing overall metabolic needs. Animals which change oxygen consumption based on available environmental oxygen are known as oxyconformers. Measuring oxygen consumption rates of tarpon at varying oxygen concentrations would help to determine if tarpon are oxyconformers, or under what conditions tarpon use their air-breathing ability to live as oxyregulators, maintaining a fairly constant oxygen consumption rate regardless of available aquatic oxygen. Determinations of'oxygen consumption, in particular a comparison of aquatic and bimodal respiration, may be of use in determining a species need for access to atmospheric oxygen. Several authors have shown that obligate airbreathing fish exhibit a decrease in total oxygen


13 consumption when forcibly submerged. Liem et al. (1984) showed that in Gymnotus carapo, the banded knifefish, oxygen consumption decreased by an average of 31% when forcibly submerged. Ghosh et al. (1986) showed that total oxygen consumption decreased from 131.1 to 56.8 ml 02 when another obligate air-breathing fish Notopterus chitala, the featherback, was forcibly submerged. The large drop in oxygen consumption indicates the inability of the featherback's gills to supply the body with sufficient oxygen. Graham et al. (1977) on the other hand showed that a facultative air-breathing fish, Piambucina festae, a characin, exhibited no difference in overall respiration rates whether respiring aquatically or bimodally. The observation that festae can maintain oxygen consumption when submerged indicates that the gills can supply the body with sufficient quantities of oxygen to maintain routine rates of oxygen consumption. oxygen transport The oxygen binding properties of blood are basic to describing the respiratory physiology of a species. oxygen affinity, oxygen carrying capacity, and the effects of C02 on oxygen binding characteristics of the blood are some parameters that help describe a fish's ability to transport oxygen from the external media to the tissues. Blood oxygen


14 affinities for several fish are shown in Appendix 1. The Pso (oxygen concentration at which the blood is 50% saturated) ranges from 6.1 to 21 in low PC02's and from 14 to 100 at higher PC02's. The oxygen capacity (in volume percent) ranges from 4.9 to 22.5. Note that in many species oxygen affinity drops when blood is exposed to increased concentrations of C02 Jacquez (1979) states that teleost blood 02 capacity is a direct function of erythrocyte content of the blood. Appendix 1 shows that much of the variation in 02 carrying capacity of fish bloods can be described by the variation in hematocrit by the following regression: y = 0.116 + 0.323 X where y = oxygen carrying capacity and x = hematocrit in volume percent ( r2 = 0.612; significance= 0.01). Fish may increase oxygen capacity of the blood by increasing the number of red blood cells. Each fish's blood will react differently to changes in gas concentrations of their environment (Phelps et al., 1979). Buffering capacity, hemoglobin types, organic phosphate concentrations (and the fish's ability to undergo short and long term changes in organic concentrations in their blood) and even the environment that the fish are acclimated to as important in controlling how gases are transported in air-breathing fishes.


Tarpon spend part of their lives in swamps and other fresh waters, as well as in estuaries and pelagic ocean environments, and face a wide range of physical environments. Tarpon may experience wide ranges of temperature, oxygen concentration, and pH, all of which 15 affect oxygen binding by the blood at respiratory surfaces. Mangum {1991, personal communication) has examined the f blood of tarpon for a Bohr shift, and found that the P50 changed little over the normal range of physiological pH {6.48.0), showing only a very slight Bohr shift. She also found that the P50 was high for a teleost, ranging from P02 of 12 to 3.5 torr in the pH range studied. These results indicate that the blood of tarpon has a high affinity for oxygen and that in whole blood the affinity was not strongly affected by pH changes. High oxygen affinity and the ability for blood to withstand changes in pH are beneficial in waters where oxygen is likely to be limiting (Powers, 1979?). The natural environment a fish lives in may be the most important factor in determining how the blood is affected by C02 Powers et al {1979) showed that the concentration of C02 in the blood of air-breathing fishes from Amazonian swamps was generally higher than that of non-air-breathing fishes. Air-breathing fishes were more sensitive to C02 and had higher concentrations of C02 in their blood than their non-air-breathing relatives (Farmer, 1979). Species


16 which cannot access atmospheric oxygen must rely on specializations in blood-oxygen binding characteristics in order to survive hypoxic conditions whereas air-breathing fishes can reduce the threat created by hypoxia and thereby reduce the need for specializations in blood oxygen binding characteristics. Among air-breathers, the effect of carbon dioxide is largest in those fishes with well developed swim bladders and choroid retes (Farmer, 1979). These authors suggest, as did Wittenberg and Wittenberg (1974), that the Bohr effect was first developed to supply oxygen to the eye via the choroid rete and that fish later developed the rete mirabilia in the swim bladder. Phelps et al. (1979) caution against generalities when describing fish blood because each fish's blood will react differently to changes in gas concentrations in the environment. Effects of Hydrogen sulfide The effects of hydrogen sulfide upon most species of Florida's fishes, including tarpon, have not been previously studied. Hydrogen sulfide is a potential toxin for fish in any habitat where even trace amounts are present. Smith, Oseid and Olson (1976) showed that Pimaphales promelus, the fathead minnow, exhibited a decrease in growth, survival, and fecundity when subjected to hydrogen sulfide


17 c .oncentrations of 0.1 uM. Bonn and Follis {1967) showed that concentrations of 20 40 uM H2S were lethal to Ictalurus punctatus, the channel catfish. Torrans and Clemens (1982) found that a key enzyme in the aerobic metabolic pathway, cytochrome c oxidase ( the terminal cytochrome in the electron transport system), was inhibited 40 to 74% by H2S concentrations of 2.5 uM in punctatus which resulted in respiratory arrest in as little as four minutes. Smith {1976) observed sublethal effects of H2S at concentrations of 0.025 uM and mortality at 6 uM when studying Lepomis macrochirus, the bluegill. Bagarinao and Vetter {1989) studied the effects of H2S on 13 species of estuarine fish with a range of habits and habitats. The most sensitive fish studied was Engraulis mordax, the northern anchovy, which suffered 100% mortality at H2S concentrations of 1 -2.7 uM. The least sensitive fish studied was Gillichthys mirabilis, the longjaw mudsucker, which suffered no mortality at H2S concentrations of up to 53 uM. The striped mullet, Mugil cephalus, which is often found in foul pools with tarpon, suffered little mortality until H2S concentrations reached 40 uM, beyond which 100% mortality occurred. In a study by Abel et al. {1987), hydrogen sulfide concentrations of 0.6 to 20.6 uM were observed in a Florida mangrove swamp. Two fish species, Gambusia affinis, the mosquitofish, and Rivulus marmoratus, the rainwater


18 killifish, were collected in sites where H2S exceeded 2.9 uM and seven species were collected from sites where the concentration of H2S was less than 1.5 uM. No fish were caught when hydrogen sulfide exceeded 11.8 uM. Juvenile tarpon are frequently captured near the study sites used by Abel et al. In fact, tarpon are also known to prey on most of the species caught by Abel ( Harrington and Harrington, 1962). Carlson et al. (1988) have observed that hydrogen sulfide may reach 100 uM in isolated portions of the Indian River lagoon, in regions where juvenile tarpon are frequently captured. Luther et al. (1976) and Vetter et al. (1989) showed that hydrogen sulfide could reach concentrations of 3500 uM in the sediments of salt marshes. Mixing by wind and tides may release H2S from sediments in quantities sufficient to reach toxic levels in the water column especially in hypoxic or anoxic bodies of water (Roden and Tuttle, 1992). Natural creeks in Florida's Indian River lagoon have concentrations of sulfides ranging from less than 1 uM while adjacent manmade creeks had much higher sulfide concentrations, up to 1640 uM (Rey et al., 1992). Millero (1986) showed that the half time for oxidation of H2S (15 160 uM) to sulfates varied from 15 to 40 hours, more concentrated solutions of sulfides taking the longest to be completely oxidized. Millero's study indicates that sulfides may persist for periods at least as long as a tidal cycle once introduced to


19 the water column. Chen and Morris (1972) showed that the oxidation of H2S was especially slow when pH drops below 7. Somero et al. (1989) showed that H2S was more toxic than or other combined forms of sulfide. Once introduced to an acidic swamp, H2S is likely to limit the distribution of those fishes which are sensitive to the toxic effects. Tarpon may be more resistant to hydrogen sulfide than non-air-breathing predators; if so, then tarpon could exploit a habitat which is most likely unavailable to many predators. Diversity in a Florida marsh decreased after impoundment for mosquito control due to reduced water quality (Harrington and Harrington, 1982). Hydrogen sulfide may approach one mM near the organic rich sediment-water interface in this habitat (Carlson et al., 1989). The pH in these mangrove swamps is often below 7, a point at which H2S and Hs-are roughly in equilibrium. The unionized state is believed to be the more toxic phase (Somero, 1989). Swim bladder volume Most air-breathing fish exchange about 50% of their swim bladder air with each breath. These include the bowfin, Amia calva, 55% (Johansen,1970); the African lungfish, Protopterus aethiopicus, 50% (Lenfant and Johansen, 1968); and gar, Lepisosteus osseus, 40% (Rahn


20 e .t. al. 1971), all fishes which breathe air in a manner similar to tarpon. The volume of air exchanged in each breath varies among fish from a low of about 14% of the swim bladder volume in Piambucina festae, a lebiasinid (Graham 1977), to a high near 130% of the swim bladder volume in jeju, Hoplerythrinus uniteaniatus, an erythrinid (Kramer, 1978). The jeju inhales before exhaling, unlike most air-breathing fishes. A survey of the literature on air-breathing in fishes shows that show that roughly 60% of the oxygen is removed from each breath (Bicudo and Johansen, 1979; Burggren, 1979; Johansen and Lenfant, 1967). Given that oxygen represents about 21% of the volume of gas inspired we can see that the volume of oxygen gained from a breath would be approximately 6.3% of the volume of the swim bladder in most air-breathing teleosts.


OBJECTIVES This study examines the role of air breathing in juvenile tarpon by testing the following hypotheses: 1) Tarpon are obligate air breathers. 21 2} Rates of air-breathing, opercular ventilation, and oxygen consumption vary as a function of oxygen concentration and temperature. 3} The circulatory system of tarpon show specializations to hypoxia including morphological adaptations, increased hematocrit values, and decreased sensitivity to C02 4) Tarpon reduce water flow across their gills or increase their reliance on air-breathing as hydrogen sulfide concentrations increase.


22 MATERIALS and METHODS Collection and maintenance of specimens Juvenile tarpon were collected from 3 locations in Southern Florida: Ozona in Pinellas County, just south of Collier Seminole State Park in Collier County, and Jack Island mosquito impoundment in St. Lucie County. Thirtythree fish were collected during the fall of 1989, 35 during September of 1990 and three in December of 1991 from the above locations. Five additional tarpon were collected during May of 1991 from Jack Island for a study of their blood. Juveniles were studied because this life history stage is most common in the highly variable environmental conditions of the mangrove swamps. The leptocephalus larva of tarpon are prohibitively difficult to collect and maintain for detailed and controlled physiological study, and the adult stages prohibitively large. The tarpon used in this study were maintained in a recirculating seawater system consisting of three 500 liter wood and fiberglass aquaria and ten 40 liter glass aquaria in a running seawater table at the University of South Florida. Study fish were fed to satiation once weekly on a diet of chopped herring, squid and shrimp. A minimum of six


23 weeks acclimation to the laboratory conditions preceeded any experiments. Tarpon used in respiration trials were held for ten to eighteen months prior to being used in experiments. Studies of air-breathing or opercular ventilation frequency were conducted in the 40 liter glass aquaria. Forced submergence and oxygen consumption rate studies were conducted in a 41.7 liter annular respirometer (Fig. 1). Tarpon were allowed to acclimate to their experimental chambers for at least 12 hours before observations were begun, but not longer than 24 hours. Acclimation to a given experimental temperature for a minimum of four weeks preceeded studies of air breathing frequency (fAs) or opercular ventilation frequency (fov) Oxygen consumption rates were determined at 25C (range= 24 to 26.5). Oxygen concentrations were recorded using an array of Clark polarographic electrodes (Clark, 1956) connected to a Hewlett-Packard data acquisition system (Torres and Somero, 1988). During studies of air-breathing and opercular ventilation oxygen concentrations were recorded once per minute. Oxygen concentrations were recorded once every five minutes during studies of oxygen consumption. Electrodes were calibrated in air saturated water (high voltage) and in nitrogen saturated water (low voltage) and a linear increase in voltage with increasing oxygen concentration was assumed. Nitrogen saturated water is virtually oxygen free. The


ANNULAR OXYGEN / ELECTRODE/ OXYGEN ELECTRODE RESPIROMETER INFRARED LED's > TOP VIEW 60cm OUTFLOWING W ATER I N FLOWING WATER PHOTOTRANSI STOR Figure 1. Schematic of annular respirometer. Outer radius = 57cm, inne r radius = 26cm, height = 19cm. Total volume = 41.71. Inflow and outflow are ports for peristaltic pump line which circulates wate r across oxygen electrodes. Two lap counters represented by LED's and phototransistors. 1-.:>


current obtained in sampled waters is proportional to the 'difference between high and low values obtained during calibration of the probe (Torres and Somera, 1988). 25 Oxygen concentrations were repeatedly monitored for one minute periods during studies of hydrogen sulfide. Hydrogen sulfide may cause polarographic type oxygen electrodes to malfunction, and therefore exposure of the probe to sulfides was limited to periods of one minute at a time. Accurate measurements of dissolved oxygen concentrations in hydrogen sulfide concentrations of up to 40 uM were possible but at higher concentrations of sulfide, innaccurrate readings seemed to occur, as readings higher than those normally assocciated with oxygen were observed in the voltage between anode and cathode. During initial experiments oxygen concentration did not drop with the addition of low concentrations of sulfide. Subsequently, the level of aeration provided by an air stone was deemed sufficient to maintain normoxic conditions ( > 3 ppm) although oxygen concentrations were not measured. Obligatory Air Breathing Behavior Ten forced submergence experiments were conducted, seven within a temperature range of 22 to 26C, and three within a temperature range of 29 to 33C. Tarpon were fed twice during each week of each experiment. Uneaten food was


26 removed within twelve hours. Forced submergence studies were conducted using a 41.7 liter annular respiration chamber (Figure 1). The chamber was completely enclosed, effectively preventing access to the surface. The chamber was connected to a flow-through water source during forced submergence studies, which allowed dissolved oxygen concentration to remain near air saturation. Each fish was acclimated to the chamber for 24 hours before the two week forced submergance trial was begun. During the acclimation period a small air space allowed the tarpon to air-breathe. The fish were then held for an additional two weeks without access to the surface. The fish were checked daily for signs of stress such as erratic swimming behavior or obvious physical injuries. Environmental Effects on Air-Breathing The air-breathing behavior of tarpon was observed over a temperature range of 18.5 to 33C and at oxygen concentrations between 0.05 and 5.5 parts per million (0.5% and 100% of air saturation). Ten tarpon were placed in 39 liter aquaria, one per aquaria, during each air-breathing frequency trial. Seven observation periods of one, one and one half, two(2), two and one half, three, and six hours were used to give a total of 18 hours of observations at each


27 temperature. Time trials of different duration were compared by two-way ANOVA and were not significantly different. All seven experimental durations were grouped and subdivided into temperature and oxygen concentration categories for further analysis. Air-breathing frequencies were studied by linear regression analysis between oxygen concentrations and temperatures and also studied by multiple ANOVA. One minute counts of the fish's opercular ventilation frequency were conducted after each air-breathing frequency experiment. Opercular ventilation was analyzed by nonlinear regression analysis between oxygen concentrations and linear regression analysis between temperatures. Opercular ventilation frequencies were further analyzed by multiple ANOVA. Regression models show non-linear relations which were estimated by FISHPARM (Prager 1989) and then iterated to least squares fit by STATGRAPHICS (STSC,Inc, 1989) Flowing water for all ten aquaria was from a common water source. Oxygen concentration in the aquaria was manipulated by bubbling nitrogen into the supply water and by randomly altering the height of the inflow pipe to each aquaria from approximately 15 ern above the surface to 15 ern below the surface. Randomization allowed for a double-blind study, and reduced potential bias due either to the observer knowing the oxygen concentration that each fish was in or to


repeated experimental conditions in a given aquaria. Double-blind observations produce unbiased results (Martin and Bateson, 1986). The observer could watch all ten fish and not know the concentration of oxygen in the aquaria. Observations were later compared to oxygen concentrations which were recorded beyond sight of the observer. 28 Continuous observation periods were used for recording air-breathing frequency in juvenile tarpon. Martin and Bateson (1986} discussed observation techniques for estimating behavior frequency (ie.: taking a b reath) and duration (ie.: time between breaths). The authors suggest that the best methods for estimating behavior duration are either: 1} Randomly timed instantaneous observations. 2} Continuous observations. Instantaneous observations are used for long duration events since rapid behaviors may be missed. Liem (1980} indicated that air-breaths in teleosts could occur in less than 1 second, a rapid event best observed by continuous periods of observation. A mean value for air-breathing frequency from within each oxygen concentration range was calculated for each experiment. Data were arbitrarily grouped as follows according to oxygen concentration for air-breathing frequency experiments; 0-0.99ppm 1-1.99ppm; 2-2.99ppm; 3-3.99ppm; 4-4.99ppm; > 5ppm. Repeated measurement of the


29 air breathing frequency would have constituted pseudoreplication (Martin and Bateson, 1986) since there was no difference in experimental conditions between some observations. For example, measuring the length of the same fish twice does not yield two data points but a more accurate measure of the same data point. During each trial, a maximum of six data points per fish were generated, one from each oxygen concentration range. oxygen Consumption All oxygen consumption determinations were conducted at 25C ( 1.5C ). A total of thirty four aquatic respiration runs and ten bimodal respiration runs were conducted. Swimming activity of tarpon was monitored during twenty two of the aquatic respiration trials and all ten of the bimodal respiration trials. Mean length of the tarpon used in respiration trials was 28.5 em. Mean weight was 159.0 grams. A 41.7 liter annular respirometer was used for oxygen consumption determinations. The respirometer was constructed with clear acrylic and measured 60 em in diameter and 19 em high. Diameter of the inner wall was 30 em. Two doors allowed access through the top of the chamber, which also housed fittings for oxygen electrodes and circulation hoses. Clear acrylic construction allowed


30 observation of swimming activity simultaneous with determination of routine oxygen consumption rate (Figure one) Two infrared lap counters (modified from light gates consisting of three infrared light emitting diodes, three focusing lenses, and three infrared detecting photodiodes) originally described by Cripe (1975) monitored activity while oxygen consumption was simultaneously monitored. Computer software was developed for recording data from the light gates in tandem with the oxygen consumption measurements. This progam allowed a Hewlett-Packard computer system (models HP 86B and 87XM) in tandem with a digital multimeter (Kiethley 195A) and scanner (Kiethley 705) to monitor the voltage in the light gate circuit roughly 100 times per minute on each of two light gates. The program detected when the beam between an LED/transistor pair was broken, indicating the fish was between them, by the change in voltage. The program could also count the number of times in a given period (either three or five minutes) that the signal from each gate was broken. The computer recorded the total number of breaks (indicating a lap swum by the fish) in each three or five minute period. After each period of light gate readings the program broke to a subroutine that scanned the oxygen probes at a rate of roughly one probe per five seconds. Light gates were scanned at rates of approximately 100


31 times per minute during the gate scanning subroutine. The frequency at which each light gate was scanned makes it unlikely that any laps by a fish would have been missed, especially when fish swim at less than one bodylength per second (which represents a majority of the time from any given run). Some swimming activity may have been missed while the oxygen probes were scanned but this period was minor, approximately seven percent of the total observation time per cycle. Swimming speeds and respiration rates calculated from visual observation in 16 of the oxygen consumption trials were compared to 28 trials with infrared lap counters by ANOVA to estimate the difference between the two techniques. The chamber was sealed during oxygen consumption trials. A slow circulation was created across the oxygen electrodes by either a peristaltic pump during studies of solely aquatic respiration or by a magnetic stir bar during studies of bimodal respiration. The respiration chamber was completely filled with water during aquatic-only respiration trials. A 1.7 or 2.7 liter volume of air was left during studies of bimodal oxygen consumption, depending on the size of the tarpon and planned duration of the respiration trial, to allow the tarpon to breathe air. I assumed for data analysis that the small air space remained in equilibrium with the water. This assumption was based on the small volume of air to surface area ratio (2120 cm2 : 1700 or 2700


32 cm3 or 1.25 cm2 surfacejcm3 volume air in most experiments), the stirring motion of the magnetic stir bar, the vigorous mixing of the tarpon surfacing for air, and the observation that the reading on the oxygen electrodes changed little when the chamber was shaken at the end of each run. The readings of oxygen concentration and estimates of oxygen consumption took one to four hours at the beginning of each bimodal run to stabilize as opposed to one to two hours in aquatic respiration trials. However, after a stable reading was achieved there was less variability in the oxygen readings using magnetic stir bars than with those trials where stirring was created only by the action of the peristaltic pump. A reliable rate of oxygen consumption was achieved within five to twelve hours. Oxygen consumption runs were terminated when the tarpon began to show signs of stress, such as frantic attempts to reach the surface. Blank runs were conducted to estimate microbial contribution to respiration in the chamber. Oxygen consumption was studied at 25()C and was corrected to 25 degrees using an assumed Q10 value of two. The Q10 value of 2 assumes that metabolic rate (as a sum of all metabolic processes) will double with each ten degree increase in temperature. Routine metabolic rates (ml02kg-1hr-1 ) were calculated


33 for each run in one hour blocks by observing the change in oxygen partial pressure over time. The large volume of the respiration chamber relative to the small rate of change of oxygen concentration and slow rate of mixing prevented analysis in shorter time blocks. In some runs the first hour was not analyzed because the tarpon would swim in erratic non-linear patterns which could not be accurately monitored with the lap counter. A comparison of swimming speeds and oxygen consumption over a range of oxygen concentrations was conducted using data from each one hour block as a separate data point, with two to twelve one hour block data points from each run. Analysis of oxygen consumption in small blocks allowed cost of swimming to be calculated in some trials by correlating the number of laps swam to the volume of oxygen consumed. In those runs where sufficient data was collected, standard metabolic rate could be estimated. Standard metabolic rates for most runs were calculated from a compiled curve generated by calculating a single mean cost of swimming from each of the runs. The cost of swimming in the 34 aquatic respiration trials and the 10 bimodal respiration trials was calculated separately.


34 Hydrogen sulfide A total of fifty observations of opercular ventilation frequency in five trials of sulfide tolerance were conducted. Sulfide concentrations ranged from zero to 150.9 uM H2S (232.1 uM total sulfides}. In each trial pH was determined and ranged from 6.83 to 7.8 in sulfide trials. The pH determines the ratio between sulfide forms (Millero, 1986}. Seventy-three observations of the experimentally manipulated pH were determined in seven trials with pH ranging from 4.1 to 8.9. The response of tarpon to hydrogen sulfide was tested in a manner similar to that of Bagarinao and Vetter (1989}. Hydrogen sulfide stock (approximately 10 roM} was introduced in 15 20 ml aliquots roughly once every 20 minutes (until levels of 100 -250 uM were reached} into an aerated seawater aquarium containing one tarpon. A control aquaria which also held one tarpon was used. No sulfide was added to the control aquaria. The control aquaria helps to estimate changes in opercular ventilation frequency which occurred and were not related to sulfide exposure, such as handling stresses. Opercular ventilation frequency of each tarpon was observed for one minute immediately prior to taking a sample for determination of sulfide concentration and air-breathing frequency noted for 5 to ten minutes over the 20 minute period between sulfide additions. Thirty minute observations of fAa on three fish exposed to sulfide


35 were conducted. Hydrogen sulfide samples were taken after each count of fov The sulfide experimental tank and control tank were both stirred after the addition of sulfide to the experimental aquaria to help remove the effects of stirring only the experimental aquaria. No blank addition to the control aquaria was used. Sulfide concentrations were determined using the methods of Cline (1969). Diamine reagent (N,N-dimethyl-p-phenylene-diamine oxalate) and ferric chloride were dissolved in 50 % HCl A 50 ul sample was mixed with 2 ml of diamine reagent. Absorbance was determined spectrophotometrically at 670 nm after 20 minutes. Absorbance was compared to a standard curve. Standards were prepared by mixing preweighed amounts of washed sodium sulfide into de-oxygenated, distilled water. The actual concentration of hydrogen sulfide as H2S and Hswas calculated by using the solubility constants reported by Chen and Morris (1972}, and Millero (1986) as well as the observed pH. Sulfide determination by the Cline method (1969) was very reliable. Absorbance of the sample was used to determine total sulfide concentration based on a standard curve generated from ten standards, replicated twice each. Absorbance gave sulfide concentration by the following equation (r2 = 0.997}: (Absorba nce 0.1495) (5) ---------------------= 0.00217


36 During the addition of sodium sulfide pH increased in the experimental aquaria from 7.4 to approximately 8. In order to study the effects of only H2s, pH was adjusted with the addition of small amounts of either hydrochloric acid or sodium hydroxide. The experiment thus needed controls for acid and base additions. To completely understand the reaction of tarpon the following conditions were studied; 1) Addition of sodium sulfide. 2) Addition of hydrochloric acid. 3) Addition of sodium hydroxide. 4) No addition of any kind as a control. In each of the four aquaria, sulfide was measured as well as pH. All four aquaria were stirred manually after the addition of sulfide, acid, or base to ensure that no gradients in sulfide concentrations or pH had arisen. Opercular ventilation frequency was monitored for one minute periods and air-breathing behavior noted for twenty minutes after each addition of sulfide, acid or base. oxygen transport Hematocrit values were determined for five tarpon from Jack Island mosquito impoundment. Jack Island is a stagnant body of water where the smell of sulfur gas is obvious and the water is clouded with organic material. Dissolved


37 oxygen concentration is commonly 2 ppm or less (J. David, St.Lucie County Mosquito control, personal communication). Fish were held in water from their collection site for less than one day before blood was drawn. Blood samples were drawn from severed vessels in the gill arches. Five fish were sampled which had been held in the lab for eighteen months. The laboratory water was clean and filtered and DO typically was at 80% saturated or more ( >4ppm). Two of the laboratory held fish were bled from the gill arch and the remaining three were bled from the dorsal aorta in the caudal peduncle. The hematocrit of tarpon blood was determined by filling a heparinized capillary tube 3/4 full, sealing, and spinning for 5 minutes on a hematocrit centrifuge (Ottis et al, 1980). The hematocrit is then determined as the percentage of blood which is composed of blood cells and the percentage of blood which is plasma. The length of each fraction in the capillary tube was measured to the nearest millimeter. Swim Bladder Volume Swim bladders were measured in length and width to the nearest mm. Width was measured at one centimeter intervals along the length of the swim bladder by gently stretching the membrane to its full extent. I then assumed that this


38 measure corresponded to one half the circumference of the cylindrical swim bladder. Volume of each truncated cone was calculated and a total volume estimate for the swim bladder was reached by adding the volume of the separate cones. The volume of the swim bladder tissue was measured to the nearest 0 1 ml by displacment in a graduated cylinder. Volume of the tissue was then subtracted from the total estimated volume, to arrive at an estimate of the volume of air that the swim bladder held. Most air-breathing fish have swim bladders with thick spongy walls that are easily reinflated or studied by more exacting histological techniques. Air can be injected into the swim bladder until it appears full. In tarpon, the wall of the swim bladder is much thinner than that of most teleosts, with only a thin membrane present, precluding the use of gas injection to determine volume. The volume of the small anterior portion of the swim bladder was not measured.


39 RESULTS Obligate air-breathing Seven of ten tarpon survived forced submergence until experiments were terminated at two weeks and all tarpon survived at least eight days before dying. The mean submergence time for the three fish that died was nine days. All three fish that died were from trials at 22 to 26 C. At the termination of all ten trials the tarpon had a deflated swim bladder, indicated by a loss of buoyancy. In those three fish which did not survive the entire two weeks, the body cavity was dissected, and swim bladder examined. No air was detectable in any of the three swim bladders. All ten tarpon suffered from abrasions on the anterior-most portion of the mandible, and numerous abrasions on the ventral surface of the body, including torn fins. The most severe wounds were on the ventral lobe of the caudal fin. Five of the ten fish used in forced submergence trials ate at least once during their two week trial. All three of the fish that died were among those five that did not eat.


40 Environmental effects on air-breathing The duration of the observation period did not significantly alter the resulting observed frequency of air-breathing (Kruskall-Wallis test; t s = 0.632, Ucrit = 1.96; df = 31,46; 0.1) (Table 1). All 7 experiments from each temperature were pooled for further analysis. The air-breathing frequency was negatively correlated with dissolved oxygen concentration at the lower temperatures but this correlation decreased with increasing temperature until at high temperatures no correlation was present. Air-breathing is more frequent at low dissolved oxygen concentrations, at temperatures below 29 C (Figure 2a) (Table 2). At 33 C the relation between f A n and P02 is not significant (Figure 2b). At higher temperatures oxygen Table 1. Total number of breaths recorded during each observation period. Ten fish observed in each period. Observation periods measured in hours. Observation period (hours) 1 1.5 2 2 2.5 3 6 Temp C # a fb Number of Breaths 18.5 167 (0.11) 20 32 28 36 3 29 19 22.0 157 (0.11) 8 8 14 8 7 56 56 26. 0 227 (0.08) 39 10 60 31 33 23 31 29. 0 520 (0.03) 3 8 25 55 95 26 148 133 33. 0 640 (0.03) 15 50 29 66 94 162 224 Total 1711 (0.05) 120 125 186 236 163 418 463 f 0.05 0 .06 0.05 0.04 0.07 0.04 0.06 a : # = number of air-breaths observed at each temperature b : f (frequency) = Total number of observed breaths Total hours of observation


41 Table 2 Air-breath Frequency of tarpon within given ranges of oxygen concentration and temperature. Mean value for the total number of air-breaths observed at a particular oxygen concentration and temperature devided by the number of hours fish were observed under that given condition. Data from all seven observation periods at each temperature are included. Mean value for each temperature in column marked all; mean value for each dissolved oxygen concentration in row marked all. Overall mean air-breath frequency = 4.05 breaths per hour. Dissolved oxygen concentration (ppm) All 0-1 1-2 2-3 3-4 > 4 Temperature All 4.05 5.52 4.05 5.71 2.96 2.45 18.5 -19.5 2.23 5.24 2.87 1. 06 0.78 0.99 22 1. 96 4.97 2.57 2.12 1. 33 1. 23 25.0 -26.5 2.58 7.90 3.58 1. 74 1. 01 2.28 28.0 -29.5 3.41 4.01 3.82 4.61 2.31 3.15 32.9 8.90 7.81 5.84 11.8 8.00 11.4 uptake at the gills did not appear to meet minimum requirements, driving tarpon to air-breathe about every 8 to 12 minutes regardless of oxygen concentration (Table 2). Air-breathing frequency can be predicted by regression curves for temperatures ranging from 19 to 29C, and the model explains 24 % of the variation in the observations (Figure 2a). At these temperatures fAo remained low until the falling oxygen concentration drops below two parts per million and climbs very rapidly at concentrations below one part per million. There was no significant difference between the four lower temperatures studied (19, 22, 26, 29) allowing this data to be pooled for regression analysis. The regression model for 33 C was not significant, and the


Figure 2. Linear regression models (solid line) for airbreathing frequencies in breaths per hour, Y, versus dissolved oxygen concentration in parts per million, X. Dashed lines represent 95% confidence intervals. a -Temperatures = 18. 5 -2 9 C. Y = 0. 9 Gx-0 59 ; significance= 0.001, r2 = 24.35% b -Temperature = 33C. Y = 1.62X + 4.89; significance = 0.391, r2 = 0.3%


""' '-:I 0 "" Jl u c: Ql :j lT ill '-u.. 01 c: ... til Ql '-.a I '--<1: ""' '-::1 0 v Jl u c: Ql :I lT Ql '-u.. 01 c: ... Ill ill '-CD '-<1: 16 12 8 4 a 16 14 12 10 8 6 4 2 a 0 a I I I .: \. \ ... 1 .. .:: . .. Figura Two 2& 18.5 -29 q: ... .... 2 3 4 5 6 Oxygen Concentration (ppm> : ,.. .. . . ----------1 . -:. : 2 3 4 Oxygen Concentration 5 a 43


44 best models based on dissolved oxygen concentration explained less than 5% of the variation in observations. Air-breathing frequency appeared random at high temperatures and ranges between once every two hours to more than twelve times per hour at all oxygen concentrations (Figure 2b). Air-breathing frequency increased with temperature only in waters where dissolved oxygen concentration exceeded 1 ppm (Table 2) (F=4.07, df=4,623; F crit=3.48; cx:=0.01). Regression models can successfully predict the variation in f A B over the range of temperatures 18.5 to 33 C for concentrations of 3-5.5ppm, 2-3ppm, and 1-2ppm. The percentage of variation explained by the model decreases with decreasing oxygen concentration until at concentrations below 1ppm, the regression model does not explain a significant portion of the variation in fAs The 95% confidence limits become narrower with increasing power in regression models, and are very near the regression model of air-breathing in normoxic waters, but less so for the model of air-breathing in hypoxic water (Figure 3, a-d). The slope of the regression model decreases from normoxic to hypoxic conditions and approaches zero below 1 ppm (Figure 3, a-d). The observed slope of near zero in Figure 3a indicates that at low dissolved oxygen concentrations temperature is less important than aquatic oxygen concentration in determining how often tarpon air-breathe.


Figure 3. Linear regression model (solid line) for airbreathing frequency in breaths per hour, Y, versus temperature, T. Dashed lines represent 95% confidence intervals. a -Dissolved oxygen concentration = 0 -1 parts per million. Y = 0.08T + 3 .65; significance = 0.290, r2 = 1. 46. b. -Dissolved oxygen concentration = 1 -2 parts per million. Y = -2.73T us; significance= 0.010, r2 = 5. 9. c. -Dissolved oxygen concentration = 2 -3 parts per million. Y = -6.74T2 28 ; significance= 0.001, r2 = 17.42. d: -.dissolved oxyqen = 3 -5 5 parts per m1.ll1.on. Y = e -z.095'"+o.097T; s1.gn1.f1.cance = 0.001, r2 = 24.9.


"" L :::l 0 .r: 16 '-" 12 ::n u c QJ :::l r:r 8 l.i.. C'l c .r: .... m 4 L .0 I L

47 Environmental effects on opercular ventilation frequency Multiple Analysis of Variance (MANOVA) showed that significant changes in opercular ventilation frequency occured over the ranges of temperatures and oxygen concentrations studied and the interaction between temperature and oxygen was significant. Opercular ventilation frequency increased slowly as dissolved oxygen concentration decreased from five and one half to two parts per million. As dissolved oxygen concentration drops below 1 parts per million fov drops rapidly. The relationship is significant all temperatures. Mean fov is highest at a dissolved oxygen concentration between one and two ppm (Figure 4; Table 5) at all temperatures studied. The significance of the regression models decreased as temperature increased so that at 28-33C the model predicted less of the variation in fov than the regressions for 19 and 20-27C (Table 6). The 95% confidence intervals of the mean cannot be calculated for a non-linear curve. Table 3. ANOVA for opercular ventilation frequency based on oxygen concentration in ppm and temperature C. Source ss d. f. MS F Sig. Main 17227 8 2153 18.7.0001 oxygen 6657 4 1664 14.8.0001 Concentration Temperature 7487 4 1871 16.3.0001 Interaction 3229 16 201 1.8.0350 Residual 45958 400 114 Total 66415 424


Figure 4. opercular Y, versus million. Non-linear regression model (solid line) for ventilation frequency in ventilations per minute, dissolved oxygen concentration, X, in parts per y = Significance= 0.001, r2 = 12.31% (77. 9 X) {1 + (( X/0.71 )A1.26)} b 20 27C. Significance= 0.001, r2 = 14.85% (79.8 X) y = {1 + (( X/0.85 )A1.38)} c. -28 33C. Significance = 0.001, r2 = 19.61% (103.9 X y = {1 + (( X/0.72 )A1.26)}


,.... c e 80 '-" 70 >u 60 z w 50 ::::> 0 w 40 a::: LL z 30 0 20 H .... ,.... c e 80 70 '-" >60 u z w 50 ::::> 0 40 w a::: LL 30 z 0 20 H .... ,.... c e 80 '-" 70 >u 60 z w ::::> 50 0 w 40 a::: LL z 30 0 20 H .... 4a botC .. 0 1 2 3 4 5 6 OXYGEN CONCENTRATION (ppm) f. .. I : 4b 20 27 c . ... . . ... ... .. . . -. I . -\ ... : .. . : . 0 1 2 3 4 5 O XYGEN CONCENTRATION (ppm) 4c 28 33 c , : . ... . . . . .. .. . : .... .. 1 2 3 4 5 0 OXYGEN CONCENTRATION (ppm) 6 49


50 Table 4. Mean opercular ventilation frequencies in ventilations per minute as grouped by the appropriate temperature and dissolved oxygen concentration. Mean value for each temperature in column marked all; mean value for each dissolved oxygen concentration in row marked as all. overall mean ventilation frequency = 34.5 ventilations per minute. Dissolved oxygen concentration (ppm) Temperature All 0-1 1-2 2-3 3-4 >4 All 34.5 25.1 38.5 37.6 36.7 33.3 18.5 -19.5 30.6 25.7 31.1 31.5 36.0 30.9 22 33.2 27.8 34.3 35.5 33.2 32.4 25.0 -26.5 32.5 20.9 39.2 36.5 33.3 41.2 28.0 -29.5 35.4 25.2 36.4 46.3 35.5 49.0 32.9 42.9 30.7 47.4 41.6 44.2 42.1 Opercular ventilation frequency increased linearly with increasing temperature over the 14 degree range studied in normoxic waters, greater than 1ppm (Table 5) (Figure 5). Ventilation frequency is stable in hypoxic waters, less than 1 ppm, over the observed temperature_range, with a mean of 25.2 beats per minute. oxygen consumption Routine oxygen consumption (V02 ) decreased slightly from 105.9 ( + 39.6, x SD ) in aquatically respiring tarpon to 94. 6 ( 22.5 ) ml02 kg-1 hr-1 in bimodally respiring tarpon but the two were not significantly different. Activity levels within the annular respirometer were uniformly low (mean swimming speed= 0.18 body lengths per second) suggesting that the fish were not in any way


Figure 5. Linear regression models (solid line) for opercular ventilation frequency in ventilations per minute, Y, versus temperature C, T. Dashed lines represent 95% confidence intervals. a. -Observations conducted at dissolved oxygen concentrations of 0 -1 ppm. Y = 0.17X + 21.17. Significance = 0.581; r2 = 0.43% b. -Observations conducted at dissolved oxygen concentrations of 1 -6 ppm. Y = 0.856X + 15.32. Significance= 0 .001; r2 = 14.7%


:J'I u 70 60 ::::l cr Ql ..t 50 c 0 ..... c Ql ::> (.. ., ::::l u 40 30 20 10 a. 0 :J'I u c Ql ::::l cr Ql (.. LL.. c 0 ..... ., ..... c Ql ::> (.. ., ::::l u (.. Ql a. 0 0 70 60 50 40 30 20 10 0 Sa F igure 5 0 -1 ppm o2 52 ---... -----_ ___ ___ .......... -........ -.. --.. .. --.. ------. .. r ..... ................... .. ......... : ..... ........... -.......... ; ... -.... -....... ; .. 19 22 25 28 31 34 Temperature 5b 1 -5.5 ppm o2 .. . . : ... t ..... ... r -... ... .. ... -. .. : ... -.. -!: ..... .. --. 19 22 25 28 31 34 Temperature


stressed, which would be indicated by high or erratic activity patterns. 53 The standard metabolic rate for juvenile tarpon can be described by the following allometric equation: ( 2 ) y = 1. 57 xo .s33 where Y = oxygen consumption in ml 02 hr (V02 ) and X = weight in grams (Figure Six). This regression should only be used within the range of the fish studied, 50 to 336 grams. Separate points are indicated for bimodal and aquatic data but the regression line given was calculated from all thirty nine points. Dashed lines indicate 95% confidence intervals for the true mean at any given point on the line. Mean swimming speed of fish in aquatic trials was 10.9 bodylengths per min. and only 4.8 bodylengths per min. during bimodal trials. ANOVA comparison of three methods of studying oxygen consumption (aquatic only with infrared light gates, ALG; aquatic only without light gates, ANG; bimodal without light gates, BNG) showed no significant difference between observed swimming speeds in ALG and ANG trials or between ANG and BNG trials. Significant differences were observed between ALG and BNG trials (F=7.18, Sig. = 0.0028). The cost of swimming in tarpon was calculated using both ALG and ANG for aquatic respiration trials and BNG for bimodal respiration trials. The cost of swimming at


54 STANDARD OXYGEN CONSUMPTION 30 ,... t. .J::. 25 .. E .._, w 1-20 . + = BIMODAL >< + 0 = AQUATIC -0 0 100 200 300 400 WEIGHT (grams) Figure 6. Regression model (solid line) for standard oxygen consumption Y, in milliliters of oxygen per hour, versus weight, w, in grams. Dashed lines represent 95% confidence intervals. Y = 1.57 W 0 833 significance = 0.001; r2 = 62.4% Bimodal respiration (where tarpon could extract oxygen from air and water simultaneously) represented by 11 "; aquatic respiration (where tarpon could only extract oxygen from the water) represented by +


z 0 H z 0 u z w X 0 180 150 120 90 60 30 + 4 COST OF SWIMMING in TARPON 8 12 16 + = BIMODAL = AQUATIC 20 SWIMMING SPEED (BODY LENGTHS PER MINUTE) 55 24 Figure 7. Linear regression model (solid line) for metabolic cost of swimming in terms of milliliters of oxygen consumed, Y, versus swimming speed X, in body lengths per minute. Y = 3.09 X+ 73.5. Significance= 0.001; r2 = 44.43. Dashed lines represent 95% confidence intervals. Bimodal respiration (where tarpon could extract oxygen from air and water simultaneously) represented by "; aquatic respiration (where tarpon could only extract oxygen from the water) represented by +


low speeds can be predicted by: (3) Y = 3.20X + 77.8 (aquatic) (r2 = .672) (4) Y = 6.78X + 61.6 (bimodal) (r2 = .839) where Y is dissolved oxygen consumption (ml oxygen per kg per hour) and X equals the swimming speed in bodylengths per minute (r2 = 0.420, n=26) (Figure Seven). 56 Neither routine oxygen consumption rate nor standard oxygen consumption rate varied significantly as a function of oxygen concentration. The large variation due to changes in activity overshadows any subtle changes which might have occurred as a result of decreasing oxygen content in the water. Routine oxygen consumption and swimming speed both drop slightly at oxygen concentrations below 1 ppm, but these changes were not significant (Figures 8 and 9). The observation that swimming speed and oxygen consumption do not change significantly showed that tarpon remain active until death is imminent and are capable of respiring aquatically at a normal rate even in very hypoxic water. Hydrogen sulfide The frequency of opercular ventilation was related to sulfide concentration as: (6) fov = e .oux (total sulfides) (Fig lOa) (7) f ov = e 3 7220 026 X (hydrogen sulfide) (Fig lOb) where X = sulfide concentration in micromoles. The


57 EFFECTS of OXYGEN CONCENTRATION on OXYGEN CONSUMPTION 150 m I I E 120 I I I I z 0 H 90 00 z 0 u 60 z w > X 0 30 w z H 0 0 0 5 1 1.5 2 2 5 3 3 5 4 OXYGEN CONCENTRATION (ppm) Figur e 8 The effects of oxygen concentration, parts per million, upon routine oxygen consumption, in milliliters of oxygen per hour. Error bars represent 95% confidence intervals. Oxygen consumption values grouped in one half ppm increments (0.25-0.75 ppm 02 0.75-1.250 ppm 02 etc.).


'"' 16 :;, c: E 'OJ Q. 12 .c .... 0) c: OJ 8 0 .0 '"" 0 w w a.. 4 (/) (!) z H :c :c 0 (/) 0 5 58 EFFECTS of OXYGEN CONCENTRATION on SWIMMING SPEED 1 1.5 2 2 5 3 3.5 4 OXYGEN CONCENTRATION (ppm) F igure 9. The effects of oxygen concentration, parts per million, upon swimming speed, in body lengths per minute. Error b ars r epresent 95% confide n c e intervals. Oxygen consumption values grouped in one half ppm increments (0.25-0.75 ppm 02 0.75-1.250 ppm 02 etc.).


59 regression model for total sulfides predicts fov better than hydrogen sulfide, but both are significant. The regression models both predict fov would drop to zero at high sulfide concentrations. My observations indicate that some level of ventilation is maintained at extremely high sulfide concentrations. The normal fashion of gill ventilation is replaced by infrequent coughs, or flow reversals across the gills. Opercular frequency showed little change between o and 10 uM H2S but dropped rapidly thereafter and reached a low of three beats per minute when H2S reached a concentration near 70 uM (Figures Ten and Eleven). No lethal effects of sulfide (H2S < 175 uM) have been observed while studying tarpon. Air breathing frequency rose to a high of once every three minutes, higher than that observed for tarpon in anoxic water with no sulfide present. However, both environmental factors co-occur and would be difficult to resolve. Opercular ventilation frequency is affected by changes in pH, although this change is minor compared to that of sulfide or oxygen. Opercular ventilation frequency decreases by about three movements per minute with each drop


Figure 10. Exponential regression models (solid line) for opercular ventilation frequency in ventilations per minute, Y, versus sulfide concentrations S Dashed lines represent 95% confidence intervals. a. -Opercular ventilation frequency versus H2S concentration in uM. f ov = e; r2 = 67.73%; significance= 0.001. b. -Opercular ventilation frequency versus Hsconcentration in uM. fov = e 3.722-0.026X; r2 = 56.00% ; significance = 0.001.


EFFECTS of SULFIDES ,.... 10 a c e 80 '-' + >+ + u .t z LIJ + :::l 60 + 0 LIJ 0::: ..... + LL + + + + z + + ... ++ .. ++ + 0 40 '-+ '+ + + + H +... + ..... + + + '+ + 0::: u z LIJ :::l 0 LIJ 0::: LL z 0 H ..... 0:::

62 of one pH unit between a range of pH nine and pH 4 (Figure Twelve). This change can be expressed as: (8) f ov = 2. 21 pH 0 66 7 ( r2 = 2 3 7 6 % n=9 5; sig. = 0.05} where fov is the number of opercular ventilations per minute. Tarpon will air-breathe about once every ten minutes in normoxic water of pH 4, six to twelve times more often than tarpon in normoxic seawater at pH of 8. oxygen transport Laboratory held tarpon have lower hematocrit values than recently captured tarpon (Table Seven) Field caught tarpon (# 1-5) had a mean hematocrit of 45.5 volume percent and laboratory held fish had a mean hematocrit of 29.7 volume percent (# 6-10}. swim bladder volume The mean swim bladder volume obtained from six fish was 56 ml per kilogram of fish 11 ml per kg ( standard deviation). The range varied from 68 ml per kg in a 0.136 kilogram tarpon to 33 ml per kg in a 0.155 kilogram tarpon. no weight specific changes were seen over the range of weights studied, 0.117 -0.322 kilograms.


63 Table 5. Hematocrit values from field caught tarpon, numbers 1 -5, and tarpon held in the lab for periods exceeding twelve months, 6 10. Mean hematocrit from field caught fish was 39.6 volume %; mean hematocrit from lab held fish was 29.7 volume% In samples 5 and 7 only one measurement was taken, as indicated by ", no mean was calculated for these two sample. Number Hematocrit Mean ( volume % 1 33.3 33.3 33.3 2 41.1 36.4 31.6 3 41.0 43.2 45. 5 4 46.8 46.8 46.8 5 65.5 6 29. 7 30.0 30.3 7 26. 9 8 28.5 29. 4 30.4 9 32.2 31.4 30.7 10 26.0 25.5 25.0


64 EFFECTS of pH on OPERCULAR VENTILATION FREQUENCY 60 >u z w + ++ :::l + 0 0:: 50 ++ LL + z + + 0 H + + 1+ 0::

65 DISCUSSION In aquatic environments with restricted circulation biotic processes may reduce dissolved oxygen concentration to levels where it is inaccessable to some organisms. Each species has an ability to regulate oxygen consumption until the partial pressure of oxygen drops to some critical level. This critical level varies from near saturationto nearly anoxic conditions depending on the species studied. Many aquatic organisms have developed means of accessing the atmosphere, where the P02 is essentially constant. Some fishes utilize the thin surface layer, which remains oxygenated by diffusion processes. Many fishes also trap air bubbles, either in mouth, bucchal cavity, digestive tract or swim bladder, and take atmospheric oxygen below the surface with them. These two adaptations are common in fresh water habitats. In salt water habitats. the nost common means of gaining atmospheric oxygen is through emergence, which may be forced upon some species by receeding tides. In intertidal species air-breathing is usually a combination of cutaneous respiration and oxygen uptake at the gills, which are kept moist. All fishes which utilze atmospheric oxygen maintain some degree of aquatic


66 respiration and are termed bimodal. Some species are required to breathe air at all times and are termed obligate air-breathers. Bimodal respiration in open ocean fishes is apparently rare, tarpon being the only known species to do so. Obligate air-breathing Tarpon were found to be facultative air-breathers at the temperatures and dissolved oxygen concentrations studied. Ten tarpon survived for periods exceeding one week without access to atmospheric oxygen. Oxygen needs for routine metabolism were supplied by the gills alone, in normoxic water. Tarpon must rely on air-breathing in hypoxic waters. I suggest that the tarpon-are continuous facultative air-breathers, in accordance with Gee and Graham's description of the term (1976), rather than obligate air-breathers as suggested by Schlaifer (1941). When denied access to the surface, tarpon became negatively buoyant after one to two days, indicating that the swim bladder was deflated. Tarpon expelled air from the swim bladder as they approached the surface in a manner similar to that of other air-breathing fishes. When a tarpon expels the air from its swim bladder and then finds the surface inaccessible, it cannot reinflate the swim bladder. Surfacing in tarpon at temperatures of 26 C and


67 below may be related to the need to maintain buoyancy, as in the South American catfish Corydoras aenus (Kramer and McClure, 1980). A negatively buoyant tarpon would suffer greatly in competing for resources with other tarpon, and other species. Decreased buoyancy would also require that a higher than normal percentage of the daily energy budget be dedicated to swimming activities (foraging, migrating, escape behaviors, etc.) (Gee, 1976) and leave less energy for growth. oxygen consumption Measurement of oxygen consumption provides insight to the understanding of both air-breathing and opercular ventilation frequencies by defining the rates at which oxygen is extracted from either the air in the swim bladder or from the water crossing over the gills. I attempted to provide. a low stress environment for all experimental procedures. The annular respirometer allowed fish to travel for extended distances without encountering barriers, much longer than would be the case in small square respiration chambers. The annular respirometer did not force the fish to swim, as in a swim tunnel respirometer, but allowed estimation of cost of swimming at low speeds. The small range of oxygen consumption rates observed and the high correlation coefficient obtained for the


68 relation between swimming speed, albeit slow, and oxygen consumption suggests that the calculated values for standard metabolic rates are reliable. Activity levels within the annular respirometer were uniformly low (mean swimming speed = 0.18 body lengths per second) suggesting that the fish were not stressed, which would be indicated by high or erratic activity patterns. Slow swimming speeds are commonly observed when volitional swimming speeds are studied (Brett, 1964). The small range of oxygen consumption rates observed also indicates that a good estimate of standard metabolic rate has been obtained. The circular path of swimming was assumed to be at the midpoint of the chamber, having a circumference of 141 em the actual path having a possible circumference of 94 to 188 em Observations indicate that the fish generally avoid contact with the walls of the respirometer, but error in the estimate may have occurred due to underestimation of swimming distance in the horizontal and vertical planes. Weihs (1981) showed that swimming in a circular motion overestimates the cost of swimming but the percent error would be small at the speeds observed in our experiment. Rapid turning about in the chamber, which was observed but not common, may cause an overestimate of distance swum due to repeated passage through the same lap counter.


69 The small observed difference in the predicted cost of bimodal and aquatic respiration may be due the smaller mean size of the fish tested during aquatic respiration trials. Biologists have recognized that within species weight specific respiration rate declines with an increase in weight (Kleiber, 1932; Roy and Munshi, 1986). The drop between aquatic and bimodal respiration represents a decline of only fifteen percent, which was not significant, but is opposite from that expected in an air-breathing fish (Hughes and Singh, 1971; Arunachalam and Pandian,1976; Liem, 1984), which generally use more oxygen when respiring bimodally. The increased oxygen consumption reported in the literature for bimodally respiring fish is associated most commonly with small benthic fish (Kramer and McClure, 1981) where considerable energy is expended during frequent trips to the surface. The relative cost of transport to the surface for a large surface oriented fish such as tarpon would be considerably less. Increased oxygen consumption has been observed in at least one other continuous facultative air-breathing fish, Polypterus senegalensis, during forced submergence (Magid, 1966). Increased oxygen consumption in submerged facultative air-breathers may be associated with increased swimming behavior in searching for the surface. The observation that oxygen consumption increases indicates that the gills can maintain oxygen uptake above levels required


for routine metabolism giving support to my belief that tarpon are facultative air-breathers. Air-breathing frequency 70 A high rate for air-breathing in tarpon is once every three minutes at 25C in either hypoxia or the presence of a toxic compound such as hydrogen sulfide. Tarpon may rise for air once every minute for short periods of time (personal observation), regardless of dissolved oxygen concentration, temperature, or other environmental parameters. Most air-breathing fish exchange approximately 50% of their swim bladder (or other accessory air-breathing organ) per breath (Johansen et al., 1970; Rahn et al.,1971). These authors and others have shown that the expired gas has 60% of the oxygen removed (Bicudo and Johansen, 1979; Burggren, 1979; Johansen and Lenfant, 1967). Given that oxygen makes up 20.94% of the atmosphere, a typical air-breathing fish takes up 8.8% of the volume of the swim bladder as oxygen in each breath. This relation is demonstrated in the following equations: swim Bladder Volume 0.5 = 0.5 SBV = Volume present after expiration


0.5 SBV 0.4 0.2094 SBV 0.5 0.2094 = o. 0419 sav1 = Volume oxygen present after expiration = 0.1047 SBV2 = Volume inspired as oxygen 71 0.0419 SBV1 + 0.1047 SBV2 = 0.1466 SBV3 0.1466 SBV3 0.6 = Volume present as oxygen after inspiration = 0.088 SBV4 = Volume of oxygen removed from each breath (ml 02*kg-1 ) The volume of the swim bladder in tarpon was shown to be 56 ml*kg-1 The estimate of oxygen gained from each breath using the above calculations would be 4. 93 ml 02 kg-1 per breath. 0. 088 SBV4 56 = 4. 93 ml 02 kg-1 per breath Thus an undisturbed tarpon rising to breathe once every three minutes might be expected to gain 98.6 ml 02 kg -1 hr -1 and an excited tarpon breathing air once every minute up to 296 or more milliliters of oxygen per hour from the swim bladder alone.


72 Since tarpon were observed to decrease opercular ventilation in hypoxic waters at all temperatures, and anaerobic respiration is not a viable alternative for long durations, which tarpon must endure, I propose that tarpon routinely extract more oxygen per breath than is common for most air-breathing fishes. At the highest temperatures I observed, 33C, undisturbed tarpon rise to breathe at the surface every 10 to 20 minutes, regardless of dissolved oxygen concentration. Using the above estimates from other species's ability to extract oxygen from each breath, tarpon are taking up an estimated 39.4 milliliters of oxygen per hour through the swim bladder when air-breathing eight times per hour, a normal fAB at 25 C. The oxygen consumption rate of tarpon was 70 mlo2 kg1 hr at 25 C If tarpon are gaining only 39.4 ml02hr1 by air-breathing, one might conclude that in hypoxic waters a resting tarpon would either be extracting as much as 30 milliliters of oxygen per hour at the gills against a very strong gradient, or respiring anaerobically. The high affinity of this species's blood for oxygen (C.Mangum, personal communication) suggests that tarpon could indeed extract oxygen from all but the most hypoxic water. Tarpon exposed to high concentrations of sulfide airbreathe as often as once every three minutes. The estimated value of 4. 93 ml 02 kg1 per breath predicts that tarpon would take up 98.6 ml 02 kg hr in sulfide rich water.


This rate of oxygen uptake is sufficient to meet standard metabolic needs and may represent an increased metabolic rate during the stress of sulfide exposure. opercular Ventilation Frequency 73 In most fishes, falling dissolved oxygen concentrations will cause an increase in ventilation volume, the volume of water passing over the gills (Kerstens et al., 1979; Farrell and Daxboeck, 1981). Ventilation volume may be increased by increasing the amplitude of each opercular beat or by increasing the opercular ventilation frequency. The total amount of oxygen extracted for respiration will increase as ventilation volume increases (Hughes, 1973) but the percentage of oxygen removed from the water passing over the gills declines (Prosser, 1991). The fish must pump more water for each milliliter of oxygen extracted in order to maintain a constant rate of aquatic oxygen uptake, termed oxy-regulation. Increasing ventilation volume is a costly metabolic function. Approximately 10 to 20% of the total metabolic cost when at rest is related to gill ventilation (Hughes, 1973). In non-air-breathing fish the oxygen demand of the respiratory system will become greater than the gills can supply when dissolved oxygen reaches a critical point, called the critical oxygen concentration (Pc). Most non-


74 air-breathing fish begin to reduce oxygen consumption and become oxy-conformers below the Pc (Basu,1959). Many oxygen conformers enter into a torpor and may recover when returned to normoxic water, as when the returning tide refills a tidal pool, a reaction which allows some individuals to survive brief encounters with hypoxia. However, in most cases hypoxia is progressive. If hypoxia becomes severe fish will often lose equilibrium. Death will ensue shortly after this point for most fish. In air-breathing fish, accessory respiratory organs become the major oxygen source when dissolved oxygen falls below the P c The decrease in opercular ventilation frequency and reduced metabolic costs related to the opercular pump may confer an energetic advantage to tarpon in hypoxic waters in comparison to those species which rely on greatly increased ventilation volumes for increased oxygen extraction during hypoxia. Tarpon change their ventilation frequency in response to low P02 in a manner similar to both facultative air-breathers and non air-breathing fishes. Ventilation frequency increases to a critical point, levels off, and then drops quickly as hypoxia becomes severe. This is unlike obligate air-breathing fish which exhibit little change in air-breathing and ventilation frequencies with changing oxygen partial pressure (Johansen et al., 1978}. The relationship between fov and P02 is significant at all temperatures. for tarpon, largely because the steep drop in


75 frequency always occurs at oxygen concentrations below about 1 part per million. The slope of the curve at higher concentrations changes slightly, but not significantly. The reduced opercular ventilation observed in tarpon in hypoxic waters confers an energetic savings which is at least as important if not more so than the potential loss of oxygen to the water, which should be minimal. The two effects of reduced opercular ventilation should in any case be mutually beneficial. Johansen (1966) proposed that ventilation of the gills in air-breathing fish was reduced when facing hypoxia to reduce the loss of oxygen from the blood (oxygenated at the air-breathing organ) to the water through the gills. However, oxygenation of the water passing over the gills has never been shown experimentally; in lungfishes, for which the proposal was based, or any other species. In normoxic water the Bohr effect will facilitate oxygen uptake at the gills and would decrease loss of oxygen in hypoxic water. The high oxygen affinity of tarpon blood would serve to prevent loss of oxygen to all but nearly anoxic water. Randall et al. (1981} state that there is often an optimal ratio between ventilation and perfusion of respiratory organ, allowing for maximization of gas transfer per unit energy expended. In the case of air-breathing fish, there is little chance of oxygen transfer to the already oxygenated blood in hypoxic water; therefore, both


76 ventilation and perfusion of the gills should be reduced to that which is necessary for transfer of carbon dioxide to the water. During excretion of carbon dioxide (or bicarbonate) at the gills the pH of blood will increase slightly and the oxygen carrying capacity of the blood will be increased (Brittain, 1987), reducing the potential for loss of oxygen to the surrounding water. HYDROGEN SULFIDE This concentration is much higher than the lethal limits observed by Bagarinao and Vetter (1989) for seven species of fish.Temperature and oxygen are not the only factors that drive tarpon to air-breathe. Reducing conditions present near organic substrates underlying hypoxic water result in dangerous concentrations of hydrogen sulfide. Sulfide concentrations found in Florida's estuarine swamps (Carlson et al., 1988; Abel et al., 1987) may exceed the lethal limits of many fishes (Bagarinao and Vetter, 1989; Torrans, and Clemens, 1982; Bonn and Follis, 1967) The pH of estuarine swamps where tarpon are common ranges near 7 (Zale and Merrifield,1987). Sulfide at pH 7 is roughly in equilibrium as an ionized, non-toxic form, Hs-1 and a very toxic form, H2S. Decreasing pH was shown to stress tarpon by causing an increased fAs and a decreased


77 f0v especially below a pH of 5. Such low pH would be uncommon in normoxic water and the effects of an acidic media would most likely be overshadowed by the extremely toxic sulfide, present as almost 100% H2S below pH 6. The ability of a species to withstand any of these environmental extremes would be influenced by the occurrence or absence of compounding environmental factors. Tarpon clearly can withstand exposure to H2S for short periods and may be able to survive in waters where other species cannot. Snook have been shown to tolerate moderate hypoxia (Peterson and Gilmore, 1991), but no data is available on sulfide tolerance. Snook migrate out of hypoxic mangrove habitats at a small size, approximately 100-150mm. Bagarinao and Vetter (1989) found that two basses (barred sandbass, Paralabrax nebulifer, and kelpbass, clathratus) found in Californian estuaries suffered mortality at 14 uM H2S. Sulfide exposure in other estuarine carnivores remains poorly studied. Freshwater gars, which can tolerate hypoxia and low salinities (common where juvenile tarpon are captured) as well as eN-(cyanide ) (Smatresk, 1986), which disrupts aerobic metabolism similarly to sulfide, may be a competitor, and predator, of tarpon in some estuaries. The combination of sulfide tolerance in tarpon and abundant prey suggests that tarpon utilize the mangrove swamp during their critical juvenile stage, until such time


78 as they would be large enough to avoid most predators and small species could no longer supply sufficient food sources for larger fish. While as many as 45 species of carnivorous fishes pass through Florida's salt marsh habitat (Durako et al., 1983) Lewis et al. (1983) collected only 14 species of carnivore from the upper mangrove swamp, including tarpon. The poor water quality of mangrove swamps may serve to exclude many species of fishes, but not tarpon. Habitats where juvenile tarpon are found appear to possess abundant prey (Lewis et al., 1985) such as poeciliids, cyprinodintids, and juvenile mullet. These species may be tolerant of hypoxia and sulfide exposure (Bagarinao and Vetter, 1989), or else efficient at utilizing the thin surface layer which is more oxygenated (Kramer, 1983) and would thus possess less sulfide than deeper in the water column (Roden and Tuttle, 1992). oxygen transport Hypoxia acclimated tarpon had a significantly higher hematocrit than normoxia acclimated tarpon. An increase in hematocrit would give the blood the ability to carry more oxygen through two possible means: reducing phosphate concentration within red blood (which effectily increases the hemoglobins affinity for oxygen) cells when red blood


79 cells swell andjor increasing the number of red blood cells in the blood, thereby increasing hemoglobin available for oxygen carrying {Satchell, 1992}. In areas where environmental hypoxia prevents full saturation of a fish's blood, an increase in hematocrit is commonly observed (Weber et al., 1979}. The increase in hematocrit allows the circulatory system a greater capability of meeting the tissues demand for oxygen. Hemoglobin of normoxia acclimated tarpon has a high affinity for oxygen with a P5 0 around 8 mmHg at physiological pH's {Mangum, personal communication). This high affinity for oxygen indicates that tarpon blood would be at least partially saturated even in nearly anoxic waters. Additionally, the blood was shown to be insensitive to changes in pH, having only a small Bohr shift. The blood would thus be functional even in acidic waters, where hypoxia is most likely to occur. The insensitivity to carbon dioxide and pH, which are directly related in water, is more common in non-air-breathing fish from hypoxic habitats than from air-breathers (Powers et al. 1979). The observation that tarpon blood is adapted to hypoxic habitats in ways more similar to non-air-breathing fishes than air-breathing fishes suggests that tarpon exist primarily as aquatic respirers and only as air-breathers in extreme hypoxia in a facultative manner as needed. Non-airbreathing fish must maintain a flow of water over the gills


80 at all times and must have adaptations in the blood to cope with acidification; such as increased carbonate buffering. Air-breathing fish can partially isolate the gills when water quality dictates and may require less ability to buffer against pH changes. It may be that tarpon efficiently transfer oxygen from the blood to the tissue when the blood is nearly saturated (as also indicated by the very slight Bohr shift; Mangum, personal communication). The observed behavior of tarpon becoming primarily an airbreather at P02's higher than that required to result in oxygen binding by their hemoglobin supports this hypothesis. Summary: tarpon as bimodal respirers The results reported here indicate that tarpon are facultative air-breathers under most conditions, but when facing high temperatures andfor low dissolved oxygen concentrations may be forced to depend on oxygen uptake through the swim bladder. Tarpon have several specializations to hypoxic environments including a vascularized swim bladder capable of providing oxygen for metabolic processes, blood with high affinity for oxygen and resitant to fluctuations in carbon dioxide, and an ability to partition the rate of oxygen uptake at its two respiratory organs, the gills and the swim bladder. The ability to air-breathe may have also allowed this species to


81 evolve tolerance to other limiting factors in swamp-like areas, such as low pH and the presence of hydrogen sulfide, a naturally occurring toxin. Juvenile tarpon may spend one year or more in stagnant tropical swamps. The fresh, brackish, hypersaline ponds they inhabit are often hypoxic. The harsh and highly variable environment may exclude some species, but apparently do not limit the distribution of juvenile tarpon. The vulnerable first year juveniles may use swamps as a refugia from predators which is rich with potential prey.


82 LITERATURE CITED Abel,D.C., C C .Koenig, and W P .Davis. 1987. Emersion in the mangrove forest fish Rivulus marmoratus: a unique response to hydrogen sulfide. Env.Biol.Fishes. 18(1): 67-72. Al-Kadhomiy,N.K. and G .M.Hughes. 1988. Histological study of different regions of the skin and gills in the mudskipper, Boleopthalrnus boddati, with respect to their respiratory function. J.mar.biol.Ass. U .K. 68, 413-422. Arunachalam,E.V. and T.J.Pandian. 1976. Food intake, conversion and swimming activity in the air-breathing catfish Heteropneusteus fossilis. Hydrobiologica 51(3): 213-217. Babcock,L.L. 1936. The Tarpon, 4th edition. Bagarinao,T. and R.D.Vetter. 1989. Sulfide tolerance and detoxification in shallow-water marine fishes. Mar.Biol. 103 : 291 302. Basu,S.P. 1959. Active respiration of f ish in relation to ambient concentrations of oxygen and carbon dioxide. J.Fish.Res. Bd .Can. 16(2): 175-212. Bicudo,J.E.P.W. and K .Johansen .1979."Respiratory gas exchange in the airbreathing fish, Synbranchus marmoratus.", Env.Biol.Fish, vol.4, no.1, 55-64. Bonn,E.W. and B J .Follis. 1967. Effects of hydrogen sulfide on channel catfish, Ictalurus punctatus. Trans.Am.Fish.Soc. 96: 31-37. Breder,C.M.Jr. 1939. On the trail of tarpon. Bull.N.Y.Zool.Soc. 42(4): 99-110. Breder,C.M.Jr. 1942. Social and Respiratory behavior of lare tarpon. Zoologica XXVII. 1-4. Breder C.M.Jr. 1944. Materials for the study of the life I history of Tarpon atlanticus. Zoolog1ca 42(4); 217-252. Brett,J.R. 1964. Respiratory metabolism and swimming performance of young sockeye salmon. J.Fish.Res.Bd.Can.


83 21(5): 1183-1226. Brett,J.R. and N.R.Glass. 1973."Metabolic rates and critical swimming speeds of sockeye salmon (Oncorhyncus nerka) in relation to size and temperature.", J.Fisheries Research Board of canada, vol.30, no.3, 379-387. Brett,J.R. and D.B.Sutherland. 1965. Respiratory metabolism of pumpkinseed (Lepomis gibbosus) in relation to swimming speed. J.Fish.Res.Bd.Can. 22{2): 405-409. Burggren,w.w. 1979. "Bimodal gas exchange during variation in environmental oxygen and carbon dioxide in the air breathing fish Trichogaster Trichopterus.", J.Exp.Biol. 82, 197-213. Carlson,P. B.Sargent, H.Arnold, L.Yarbro, and J.David. 1988. The effects of water management practices on impoundment water quality. In, Workshop on Salt Marsh Management and Research, organized by Subcommitee on managed marshes and the Florida Coordinating Council on Mosquito Control. Univ.Fla. Vera Beach. Carter,G.S. and L.C.Beadle. 1931. "The fauna of the swamps of the Paraguayan Chaco in relation to its environmental. -II. Respiratory adaptations in the fishes.", J.Linn.Soc. London. Zoology, Vol.XXXVII, 321-367. Cech,J.J.Jr., M.J.Massingill, B.Vondracek, and A.L.Linden. 1985. Respiratory metabolism of mosquitofish, Gambusia affinis: effects of temperature, dissolved oxygenn, and sex difference. Env .Biol.Fish. 13(4): 297-307. Chen,K.Y. and J.C.Morris. 1972. Kinetics of oxidation of aqueous sulfide by 02 Env.Sci.Tech. 4(6): 529-537. Clark,L.C., R.Wolf, D.Granger, and A.Taylor. 1953. Continuous recording of blood oxygen tensions by polarography. J.Appl.Physiol. 6: 189-193. Cline,J.D. 1969. Spectrophotometric determination of hydrogen sulfide in natural waters. Limnol.Oceeanogr. 4(3): 454-458. Coates,D. 1987. Observations on the biology of tarpon, Megalops cyprinoides (Brousssonet) (Pisces:Megalopidae), in the Sepik River, Northern Papau New Guinea. Aust.J.Mar.Fresmw.Res. 38, 529-535.


84 Crabtree,R.E., E.c.cyr, R.E.Bishop, L.M.Faulkenstein, amd J .M.Dean, in press. Age and growth of tarpon, Meqalops atlanticus, larvae in the Gulf of Mexico, with notes on relative abundance abd probable spawning areas. Environmental Biology of Fishes. Cripe,C.R., J.H.Cripe, and R.J.Livingstone. 1975. Apparatus for the quantitative determination of locomotor activity patterns of aquatic organisms using infrared LED's. J.Fish.Res.Bd.Can. 32{10): 1884-1885. Cyr.E.C. 1991. Aspects of the life history of the tarpon, Meqalops atlanticus, from south Florida. Unpub. Doctoral Dissertation, 138 pp. Das,B.K. 1940. "Presedential address -Nature and causes of evolution and adaptation of the air-breathing fishes {a resume).",Proceedings of the Indian Science congress, vol.27, 215-260. Day,R.A. and A.L.Underwood. 1980. Quantitative analysis, 4th ed. Prentice Hall, inc. Englewood. 660pp. Dickson,K.A. and J.B.Graham. 1986. Adaptations to hypoxic environments i the erythrinid fish Hoplias microlepis. Env .Biol.Fish. 15{4):310-308. Dubale,M.S. 1951. "A comparative study of the extent of gill-surface in some representative Indian fishes, and its bearing on the origin of the air-breathing habit.", Univ. of Bombay,Journal, vol.19, 90-101. Durako,M.J., J A .Browder, W.L.Kruczynski, C.B.Subrahmanyan, and R.E.Turner. 1983. Salt marsh habitat and fishery resources of Florida. Pages 189 280 in Florida Aguatic Habitat and Fishery Resources, W.Seaman,ed. Fla. Am.Fish.Soc., Eustis. 543 pp . Farmer,M., H.J.Fyhn, U.H.Fyhn and R.W.Noble. 1978. Occurrence of Root effect hemoglobins in Amazonian fishes. Comp.Biochem.Physiol. Vol.62A. 115-124. Farrell,A.P. and C.Daxboeck, 1981. Oxygen uptake in the lingcod, Ophiodon elongatus, during progressive hypoxia. Can.J.Zool. 59, 1272-5. Gee,J.H. 1976."Bouyancy and aerial respiration: factors influencing the evolution of reduced swim bladder volume of some Central American catfishes {Trichomycteridae, Callichtyidae, Loricariidae, Astroblepidae).", Can.J.Zool., vol.54, 1030-1037.


Gee,J.H. and P.A.Gee. 1991. Reactions of gobiod fishes to hypoxia: bouyancy control and aquatic surface respiration. Gee,J.H. and J.B.Graham. 1978. Respiratory and hydrostatic functions of the intestine of the catrfishes Hoplosternum thoracatum and Brochis splendens. J.exp.Biol. 74: 1-16. Ghosh,T.K., A.Moitra, G.K.Kunwar, J.S.D.Munshi. 1986. 85 Bimodal oxygen uptake in a freshwater air-breathing fish, Notopterus chitala. Jap.J.Ictyol. 33(3): 280-285 Ghosh,T.K. and J.S.D.Munshi. 1987. Bimodal oxygen uptake in relation to body weight and seasonal temperature of an air-breathing climbing perch, Anabas testudinus (Bloch). Zool.Beitr.N.F. 31(3): 357-364. Graham,J.B. and T.A.Baird. 1982. The transition to airbreathing in fishes. I. Environmental effects o the facultative air-breathing of Ancistrus chagresi and Hypostomus plecostomus. J.exp.Biol. 96: 53-67. Graham,J.B., D.L.Kramer and E.Pineda. 1978. "Comparitive respiration of an air breathing and a non air-breathing characoid fish and the evolution of aerial respiration in characins.", Physiol.Zool, vol.51, 279-288. Graham,J.B., D.L.Kramer and E.Pineda. 1977."Respiration of the air breathing fish Piambucina festae.", J.Comp.Physiol. 122, 295-310. Harrington,R.W.Jr. 1966. Changes through one tear of growth rates of tarpon, Megalops atlanticus (Valenciennes), reared from mid-metamorphosis. Harrignton,R.W. and E.S.Harrington. 1982. Effects on fishes and their forage organisms of impounding a Florida salt marsh to prevent breeding by salt marsh mosquitoes. Bull.Mar.Sci. 32(2): 523-531. Hillman,s.s. and P.C.Withers. 1987. Oxygen consumption during aerial activity in aquatic and amphibious fish. Copiea 1: 232-234. Hochachka,P.W. and G.N.Somero. 1984. Biochemical Adaptation. Princeton University Press, Princeton. 538 pp. Horn,M.H. and C.D.Riggs. 1973. Effects of temperature and


86 light on the rate of air-breathing of the bowfin, Amia calva. Copiea 4: 653-637. Hughes,G.M. 1973. Respiratory responses to hypoxia in fish. Amer.Zool. 13: 475-489. Hughes,G.M. and B.N.Singh. 1971. Gas exchange with air and water in an air-breathing catfish, (Heteropneusteus) Sacchobranchus fossilis. J.Exp.Biol. 55: 667-682. Hughes,G.M. and B.N.Singh. 1970."Respiration in an airbreathing fish, the climbing perch Anabas testudinus Bloch I. Oxygen uptake and carbon dioxide release into air and water.'',J.Exp.Biol. 53, 265-280. Hughes,G.M. and B.N.Singh. 1970.''Respiration in an airbreathing fish, the climbing fish, Anabas testudinus II. Respiratory patterns and the control of breathing. J.Exp.Biol., 53, 281-298. Isaacks,R.E. and D.R.Harkness. 1980. Erythrocyte organic phosphates and hemoglobin function in birds, reptiles, and fishes. Amer.Zool. 20:115-129. Itazawa,Y and T.Takeda. 1978. "Gas exchange in the carp gills in normoxic and hypoxic conditions. Resp.Physiol. 35, 263-269. Jesse,M.J., C.Shub and A.P.Fishman. 1967. Lung and gill ventilation in the African lungfish. Resp.Physiol. 3:267-287. Johansen,K., D.Hanson and C.Lenfent. 1970. ''Respiration in a primitive air breather, Amia calva.", Resp.Physiol., 9, 162-174. Johansen,K. 1966. Air-breathing fishes. Sci.Am. 219, 102111. Johansen,K. and C.Lenfant. 1967. Respiratory function in the South American lungfish ,Lepidosiren paradoxa (Fitz). J.Exp.Biol. 46: 205-218. Johansen,K. and C.Lenfant. 1968. Respiration in the African lungfish Protopterus aethiopicus. II. Control of breathing. J.exp.Biol. 49:453-468. Jordan,J. 1976.''The influence of body weight on gas exchange in the air breathing fish, Clarias batrachus.", Comp. Biochem. Physiol., vol.53A. 305-310.


Kerstens,A., J.P.Lumholt, and K.Johansen. 1979. The ventilation, extraction, and uptake of oxygen in undisturbed flounders, Platichthys flesus: responses to hypoxia acclimation. J.exp.Biol. 83, 169-179. 87 Kleiber,M. 1932. Body size and metabolism. Hilqardia 6: 315-353. Kramer,D.L. 1987. Dissolved oxygen and fish behavior. Env. Biol.Fishes. 18(2): 81-92. Kramer,D.L. 1983. Aquatic surface respiration in the fishes of Panama: distribution in relation to risk of hypoxia. Env.Biol.Fish., vol. 8, no.1, 49-54. Kramer,D.L. 1978. Ventilation of the respiratory gas bladder in Hoplerythrinus unitaeniatus. Can.J.Zool. 56:931-938. Kramer,D.L. and J.B.Graham. 1976. synchronous air-breathing, a social component of respiration in fishes. Copiea 4:689-697. Kramer,D.L. and M.McClure. 1981. The transit cost of aerial respiration in the catfish Corydoras aenus. Physiol. Zool. 54(2): 189-194. Kramer,D.L. and M.McClure. 1980. Aerial respiration in the catfish Corydoras aeneus (Callichthyidae). Can.J.Zool. 58:1984-1991. Lai,N.C., J.B.Graham, and L.Burnett. 1990. Blood respiratory properties and the effect of swimming on blood gas transport in the leopard shark Triaka semifasciata. J.exp.Biol. 151:161-173. Lenfant,c. and K.Johansen. 1968. Respiration in the African lungfish, Protopterus aethiopicus. J.exp.Biol. 49:437-452. Lenfant,c., K.Johansen and G.C.Grigg. 1966/67."Respiratory properties of blood and pattern of gas exchange in the lungfish Neoceratodus forsteri (Krefft) .'', Resp. Physiol. 2, 1-21. Lenfant,C. and K.Johansen. 1968."Respiration in the african lung-fish Protopterus aethiopicus I. Respiratory properties of blood and normal patterns of breathing and gas exchange". J. Exp.Biol., 49, 437-452. Lewis,R.R.,III, R.G.Gilmore, Jr., D.W.Crewz, and W.E.Odum.


88 1983. Mangrove habitat and fishery resources in Florida. Pages 281-336, in Florida Aquatic Habitat and Fishery Resources. W.Seaman, Jr. Ed. Fla.Arn.Fish.Soc., Eustis. 543 pp. Liem,K.L., B.Echlander and W.L.Fink. 1984. Aerial respiration in the banded knifefish Gyrnnotus carapo {Teleostei: Gyrnnotoidei). Physiol.Zool. 57{1): 185-195. Liem, K.F. 1980. Air ventilation in advanced teleosts: Biomechanical and evolutionary aspects. In Environmental physiology of fishes. ed. M.A.Ali. Plenum Pub. Corp. New York. Liem,K.F. 1984. The muscular basis of aquatic and aerial ventilation in the air-breathing teleost fish Channa. J.exp.Biol. 113: 1-18. Liem,K.F. 1989. Respiratory gas bladders in teleosts: Functional conservation and morphological diversity. Amer.Zool. 29: 333-352. Lumholt,J.P. and K.Johansen. 1979. Hypoxia acclimation in carp -how it affects 02 uptake, ventilation, and 02 extraction from water. Physiol.Zool. :38-49. Luther,G.W.III, T.M.Church, J .R.Scudlark, and M.Cosman. 1986. Inorganic and organic sulfur cycling in salt-marsh pore waters. Science 232: 746-749. Magid,A.M.A. 1966."Breathing and function of the spiracles in Polypterus senegalus.", Anim.Behav., 14, 530-533. Martin,P. and P.Bateson. 1986.Meauring behavior: An Introductory Behavior. Cambridge Univ. Press, Cambridge, 199 pp. Martin,K.L.M. and J.R.B.Lighton. 1989. Aerial C02 and 02 exchange during terestrial activity in an amphibious fish, Alticus kirki {Blennidae). Copiea 3:723-727. McCormack,B. 1967. Aerial respiration i the Florida spotted gar. Quart.J.Fla.Ac.Sci. 30: 68-72. McKenzie,D.J.,S.Aota and D.J.Randall. 1991. Ventilatory and cardiovascular responses to blood pH, plasma C02 blood 02 content, and catecholamines in an air-breathing fish, the bowfin {Amia calva). Physiol.Zool. 64{2): 432-450. Millero,F.J. 1986. The thermodynamics and k inetics of the

PAGE 100

hydrogen sulfide system in natural waters. Mar.Chem. 18: 121-147. 89 Moffett, A. and J. E. Randall. 1957. "The Roger Firestone Tarpon Investigation Progress Report". Ottis,K., J.F.Pritchett and L.C.Wit, 1980. Experimental Animal Physiology, 3rd ed. Kendall/Hunt Publ. Comp. Dubuque, 226pp. Parsons,G.R. 1990.11Metabolism and swimming efficiency of the bonnethead shark Sphyrna tiburo.",Mar.Biol., 104, 363-367. Perez-Pinzon,M.A. and P.L.Lutz. 1991. Activity related costs of osmoregulation in the juvenile snook (Centropomus undecimalis). Bull.Mar.Sci. 48(1): 58-66. Peterson,M.S. and R.G.Gilmore. 1991. Ecophysiology of juvenile snook Centropomis undecimalis (Bloch): Lifehistory implications. Bull.Mar.Sci. 48(1): 46-57. Phelps,c., M.Farmer, H.J.Hyhn, U.E.H.Fyhn, R.L.Garlick, R.W.Noble and D.A.Powers. 1979. Equilibria and kinetics of oxygen and carbon monoxide binding to the haemoglobin of the South American lungfish, L.Paradoxa. Comp.Biochem.Physiol. 62A, 139-143. Piiper,P., M.Meyer, H.Worth, and H.Willmer. 1977. Respiration and ciculation during swimming activity in the dogfish Scyliorhincus stellaris. Resp.Physiol. 30: 221-239. Powers,D.A. 1980. Molecular ecology of teleost fish hemoglobins: strategies for adapting to changing environments. Amer.Zool. 20: 139-162. Powers,D.A., H.J.Fyhn, U.E.H.Fyhn, J.P.Martin, R.L.Garlick and S.C.Wood, 1979. A comparative study of oxygen equilibria of blood from 40 genera of Amazonian fishes. Comp. Biochem.Physiol. 62A, 67-85. Prager,M.H., S.B.Saila, and C.W.Reckseik. 1989. FISHPARM: a microcomputer program for estimation of non-linear models in fishery science, second edition. Old Dominion University Oceanography Technicl Report 87-10. Prosser,C.L. 1991. Environmental and Metabolic Animal Physiology. Comparitive Animal Physiology. Fourth Edition. Wiley-Liss, inc. New York. 578 pp. Rahn,H., K.B.Rahn, B.J.Howell, C.Gans and S.M.Tenny. 1971.

PAGE 101

Air breathing of the garfish (Lepisosteus osseus). Resp.Physiol., 11, 285-307. Randall,D.J., W.W.Burggren, A.P.Farrell, and M.S.Haswell. 1981. The Evolution of Air Breathing in Vertebrates. Cambridge University Press, Cambridge. 133p. 90 Rey,J.G., J.Shaffer, T.Kain, R.Stahl, and R.Crossman. 1992. Sulfide variation in the pore and surface waters of artificial salt-marsh ditches and a natural tidal creek. Estuaries 15(3), 257-269. Rickards,W.L. 1968.11 Ecology and growth of juvenile tarpon in a Georgia salt marsh";Bull.of Mar.Sci., vol. 18, 220-239. Robins,C.R. 1977. "The tarponUnusual biology and man's impact determine its future"; Proc. 2nd Ann.Mar.Rec.Fish Symp., San Fransisco, CA; 105-112. Roden,E.E. and J.H.Tuttle (1992) Sulfide release from estuarine sediments underlying anoxic bottom water. Limnol. oceanog. 37(4), 725-738. Root,R.W. 1931. "The respiratory function of the blood of marine fishes.",Biol.Bull., 61, 427-456. Roy,P.K. and J.S.D.Munshi. 1986. Morphometries of the respiratory organs of a freshwater major carp, Cirrhinus mrigala in relation to body weight. Jap.J.Ichthyol. 33(3): 269-279. Satchell,G.H. 1991. Physiology and Form in Fish Circulation. cambridge University Press. Cambridge. 235 pp. Saxena,D.B. 1963. A review on ecological studies and their importance in the physiology of air breathing fishes. Ichthyologica 2, 116-128. Schlaifer,A. 1941. Additional social and physiological aspects of respiratory behavior of small tarpon. Zoologica 26(11); 55-60. Schlaifer,A. and C.M.Breder. 1940. Social and respiratory behavior of small tarpon. Zoologica 25; 493-512. singh,B. R M.S.Prasad, and A.P.Mishra. 1986. Oxygen uptake through water during early life in Channa striatus (Bloch). Pol.Arch.Hydriobiol. 33(1): 97-104. Smatresk,N.J. 1986. Ventilatory and cardiac reflex responses to hypoxia and NaCN in Lepisosteus osseus, an air-

PAGE 102

91 breathing fish. Physiol.Zool. 59(4): 385-397. Smith, D.G. 1983. Elopiformes, Notocanthiformes, and Anguilliformes: Relationships. in ontogeny and Systematics of Fishes. H.G.Moser, editor in chief. Am. Soc.Ich.Herp. Allen Press, Lawrence. 760 pp. Smith,L., H.Kruszyna, and R.P.Smith. 1977. The effect of methemoglobin on the inhibition of cytochrome c oxidase by cyanide, sulfide or azide. Biochem.Pharm. 26, 22472250. smith,L.L. Jr., D.M.Oseid, G.L.Kimball, and S.M.El-Kandelgy. 1976. Toxicity of hydrogen sulfide to various life history stages of bluegill (Lepomis macrochirus). Trans.Am.Fish.Soc. 105: 442-449. Smith,L.L. Jr., D.M.Oseid, and L.E.Olson. 1976. Acute and chronic toxicity of hydrogen sulfide to the fathead minnow, Pimephales promelas. Env.Sci.Tech. 10(6): 565-568. Soivo,A., M.Nikinmaa, and K.Westman. 1980. The blood oxygen binding properties of hypoxic Salmo gairdnaeri. J.Comp.Physiol. 136: 83-87. Somero,G.N., J.J.Childress, and A.E.Anderson. 1989. Transport, metabolism, and detoxification of hydrogen sulfide in animals from sulfide rich marine environments. CRC Crit.Rev.Ag.Sci. 1(4): 591-614. Stevens,E.D. and G.Holeton. 1978.''The partitioning of oxygen uptake from air and from water by the large obligate air breathing teleost pirarucu (Arapaima gigas).", Can. J. Zool. 56, 974-977. Todd,E.S. and A.W.Ebeling. 1966. Aerial respiration in the longjaw mudsucker Gillichthys mirabilis (Teleostei: Gobiidae). Biol.Bull., 130, 265-288. Torrans, E.L. and H.P.Clemens. 1982. Physiological and biochemical effects of acute exposure of fish to hydrogen sulfide. Comp.Biochem.Physiol. 71C(2): 183190. Torres,J.J. and G.N.Somero, 1988. Metabolism, enzymic activities and cold adaptation in Antarctic mesopelagic fishes. Mar.Biol. 98: 169-180. Twoney,E. and P.Byrne. 1985.11A new record for tarpon, Tarpon atlanticus (Osteichtyes-Elopiformes-Elopidae) in the

PAGE 103

North Atlantic"; J.Fish Biol. (26); 359-362. Vetter,R.D., P.A.Matrai, B.Javor, and J.O'Brien. 1989. Reduced sulfur compounds in the marine environment: analysis by HPLC. 92 Wade,R.D. 1962. The biology of the tarpon, Megalops atlanticus, and the ox-eye, Megalops cyprinoides, with emphasis on larval development. Bull.Mar.Sci. of the Gulf and carr. 12(4). Weber,R.E., G.Lykeboe, and K.Johansen. 1976. Physiological properties of eel hemoglobin: hypoxic acclimation, phosphate effects, and multiplicity. J/exp.Biol. 64:75-88. Weihs,D. 1981. Effects of swimming path curvature on the energetics of fish motion. Fish.Bull. 79(1): 171-174. Willmer,E.N. 1934. Some observations on the respiration of certain tropical fresh water fishes. J.exp.Biol. 11: 283-306. Wittenberg,J.B. and R.L.Headrich. 1974. The choroid rete mirabile of the fish eye. II. Distribution and relation to the pseudobranch and to the swimbladder rete mirabile. Biol.Bull. 146: 137-156. Wittenberg,J.B. and B.A.Wittenberg. 1962. Active secretion of oxygen into the eye of fish. Nature. 194: 106-107. Wittenberg,J.B. and B.A.Wittenberg. 1974. The choroid rete mirabile of the fish eye. I. oxygen secretion and structure: comparison with the swim bladder rete mirabile. Biol.Bull. 146:116-136. Wohlshlag,D.E. 1968."Seasonal changes in the respiratory metabolism of the pinfish (Lagodon rhomboides)" contr.Mar.sci., vol. 13, 89-104. wu,R.S.S. and N.Y.S.Woo. 1984. "Respiratory responses and tolerance to hypoxia in two marine teleosts, Epinephelus akaara (Temminck & Schlegal) and Mylio macrocephalus (Basilewsky) .'' Hydrobiologia 119, 209217. Zale,A.V. and S.G.Merrifield. 1989. Species profiles: Life histories and environmental requirements of coastal

PAGE 104

93 fishes and invertebrates (South Florida); Ladyfish and Tarpon. U.S. F ish and Wildlfie, Biological Report 82(11.104). 17pp.

PAGE 105


PAGE 106

Table 6. Blood characteristics of several fresh (fw) and saltwater (sw) fishes. Ability to breathe air indicated by a yes (y) or no (n) as is observed Root shift. Oxygen carrying capacity (02 cap.) determined without C02 present. All observations by author(s) cited. Appendix 1. (continued) Species Env. RS AB P s o b P s o Myleus setiger fw y 15 55 Hydroleus scombroides fw y 10 22 HoQlias malabaracius fw y 10 35 HoQlosternum littorale fw n 12 20 ElectroQhorous electricus fw n y 14 19 HyQostomous SQ. fw y 10 35+ PterygoQlichthyes SQ. fw y Neoceratodus fosteri fw n y 11 2 5 ProtoQterus aethioQicus fw y y 11 25 LeQidosiren Qaradoxa fw n y 12 14 Osteoglossum bichirrosum fw n y 6 AraQaima gigas fw y 21 Symbranchus marmoratus fw y OQsanus tao sw y n 14 80 --0.2 Hem. cap. 10.8 10.2 6.5 18.1 19.8 9-11 8.6-11 21-22 31 7.7 25 6.8 14-19 5-7 28 11.0 30.8 10.4 6.2 19.5 Author Willmer, 1934 Willmer Willmer Willmer Willmer Weber, et al. 1979 Weber, et al. Lenfant et al. 66/67 Len. & Joh., 68 Joh. & Len., 67 Joh. et al., 78 Joh. et al., 78 Weber, et al. 1979 Root, 1931 \D U1

PAGE 107

Appendix 1. (continued) Prionotus carolinus sw y n 17 1oo 7.7 24 Root Scombei: scomber sw y n 17 65 15.8 37.1 Root LoRhius Riscatorius sw n 5.1 15.5 Root stenotomus chrysoRS sw n 7.3 32.6 Root SRheroides maculatus sw n 6.8 17.5 Root -The partial pressure of oxygen at which blood is 50 % saturated with oxygen. -The P50 with no C02 present. P50 b -The P5 0 with 25 mmHg C02 present. Env. -freshwater, fw; or saltwater, sw. RS -Root shift detected, y; no root shift detected, n. o2 cap. -oxygen capacity of the blood in volume % Hem. -Hematocrit in volume % PC02 = 3 0 mmHg. + PC02 = 10 mmHg. AB -Ability to breath air, Y i no ability to breathe air known, n. \0 0\


Download Options

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


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.


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.


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.


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.