<%BANNER%>

Respiratory Allocation and the Resting Rate of Metabolism in the African Lungfish Protopterus aethiopicus


PAGE 1

RESPIRATORY ALLOCATION AND THE RESTING RATE OF METABOLISM IN THE AFRICAN LUNGFISH Protopterus aethiopicus By ASHLEY W. SEIFERT A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Ashley W. Seifert

PAGE 3

I dedicate this thesis to my parents and my family. More than anyone else, their undying encouragement has helped keep me sane. Their love and belief in my potential pushed me to believe in myself. I also dedicate this work to my grandparents, Evelyn and Nathan Epstein and Louise and Randolph Seifert. While Lou has had the opportunity to share in my triumphs along the way, I know that the others are smiling from somewhere in the cycle of life. I only hope that one day I can fulfill the destiny that all of you have seen for me. It is often I recall your memories and they keep me smiling one day at a time.

PAGE 4

ACKNOWLEDGMENTS I am sure that this thesis would not have been possible without the support and commitment of my advisor, Dr. Lauren Chapman. Her knowledge and guidance as both a scientist and mentor have been truly inspiring. Her passion towards science served as a daily reminder of what it means to truly love ones work. It is a lasting impression that will stay with me forever. I would also like to thank the other members of my committee, Dr. Brian McNab, Dr. Steven Walsh, and Dr. James Albert. I appreciate their sacrifice of both time and energy to serve on my committee. The combined wisdom of these men has helped me shape both my science and myself. My conversations with each of them challenged me to rigorously question my work, and have deeply affected my application of the scientific method. Lastly, I would like to thank Dr. Frank Nordlie. From the very beginning of my tenure here at the University of Florida he has served as both scientific advisor and sage. Our trips to Cedar Key and our time together in Africa are burned in the annuls of my memory. Perhaps more than anyone, his carefree being reminded me that I might be in the right place after all. In addition, I would also like to thank all those at Lake Nabugabo, Uganda, for their assistance and friendship. Kiberu Jackson and Mutebi Jackson, in particular, treated me as a brother. Together we bridged cultural boundaries that ingrained in me the belief that we are all brothers and sisters on this tiny blue sphere. I am indebted to Dr. Colin Chapman for his patience in entertaining my seemingly endless questions about anything and everything. How we seemed to have free spots of time that always overlapped is still iv

PAGE 5

a mystery. Of course, it goes without saying that his passion for science should stand as a pillar for us all. I would like to thank Mac Kobza, Tom Gillespie, Chrissy Apodaca and Nat Seavy for making my time in Kibale National Park both enjoyable and a wee bit nerve-racking. Last, I would like to acknowledge the support of Sigma Xi and the Raney Fund for contributing funds to help support my research. This work was also supported by the National Science Foundation (IBN 0094393), the Wildlife Conservation Society, and Ocean Optics support to Dr. Lauren J. Chapman v

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES ...........................................................................................................viii LIST OF FIGURES ...........................................................................................................ix ABSTRACT .........................................................................................................................x CHAPTER 1 GENERAL INTRODUCTION....................................................................................1 2 RESPIRATORY ALLOCATION AND THE RESTING RATE OF METABOLISM IN Protopterus aethiopicus...........................................................................................7 Introduction...................................................................................................................7 Materials and Methods...............................................................................................10 Study Site.............................................................................................................10 Collection and Maintenance of P. aethiopicus....................................................12 Respirometry.......................................................................................................13 Results.........................................................................................................................18 Field Laboratory Data..........................................................................................18 UF Laboratory Data.............................................................................................18 Chamber Effect....................................................................................................23 Resting rate of metabolism..................................................................................24 Discussion...................................................................................................................26 Metabolic Profile.................................................................................................28 Methodological Issues.........................................................................................31 3 THE RELATIONSHIP BETWEEN TEMPERATURE AND THE RESTING RATE OF METABOLISM IN Protopterus aethiopicus.......................................................35 Introduction.................................................................................................................35 Materials and Methods...............................................................................................37 Statistics......................................................................................................................38 Results.........................................................................................................................39 Discussion...................................................................................................................41 vi

PAGE 7

4 SUMMARY AND CONCLUSIONS.........................................................................47 LIST OF REFERENCES...................................................................................................49 BIOGRAPHICAL SKETCH.............................................................................................54 vii

PAGE 8

LIST OF TABLES Table page 2-1 Respiratory allocation of Protopterus aethiopicus from the Lwamunda Swamp, Uganda, between water and air at high levels..........................................................19 2-2 Respiratory allocation of Protopterus aethiopicus from the Lwamunda Swamp, Uganda, from air at the aquatic respiration plateau (representing the level of DO 2 at which individuals respired almost completely aerially)...........................................23 3-1 Q 10 values calculated for various species of teleost fishes.......................................44 viii

PAGE 9

LIST OF FIGURES Figure page 2-1 Lwamunda Swamp, Lake Nabugabo, Uganda. Red stars indicate two collection sites where specimens of Protopterus aethiopicus were captured...........................11 2-2 Respirometry system used to measure the resting rate of metabolism. The entire system is submerged in a water bath to maintain temperature. The system is connected to a labtop that records the data...............................................................14 2-3 Bilogarithmic relationship between resting rate of metabolism (RRM) and body mass of Protopterus aethiopicus from the Lwamunda Swamp, Uganda, measured at high levels of DO 2 in the field..................................................................................20 2-4 The decline in oxygen partial pressure over time for Protopterus aethiopicus from the Lwamunda Swamp, Uganda, in a two-phase, closed respirometer, where fish had access to both air and water simultaneously......................................................21 2-5 The relationship between the degree of curvilinearity (b1 from the equation y=b 1 e b0*t ) and body mass (g) for Protopterus aethiopicus from the Lwamunda Swamp, Uganda.......................................................................................................22 2-6 The bilogarithmic relationship between RRM (mg O 2 hr -1 ) and body mass (g) for the African lungfish, Protopterus aethiopicus, collected from the Lwamunda Swamp, Uganda. Data represent RRM at low (ecologically relevant) DO 2 levels..25 2-7 The bilogarithmic relationship between resting rate of metabolism (RRM) (mg O 2 hr -1 ) and body mass (g) of Protopterus aethiopicus from the Lwamunda Swamp, Uganda, measured at levels of DO 2 prior to the plateau in aquatic respiration.......27 3-1 Bilogarithmic relationship between resting metabolic rate (RRM) and body mass for six individuals of Protopterus aethiopicus from the Lwamunda Swamp, Uganda, measured at three different temperatures: 30 C, 25 C, and 20 C..........40 3-2 Mean resting rate of metabolism adjusted to 110.5 g fish (mean for six individuals) for three temperatures: 30 C, 25 C, and 20 C......................................................40 ix

PAGE 10

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science RESPIRATORY ALLOCATION AND THE RESTING RATE OF METABOLISM IN THE AFRICAN LUNGFISH Protopterus aethiopicus By Ashley W. Seifert May 2004 Chair: Lauren J. Chapman Major Department: Zoology Air breathing in fishes has a very rich evolutionary history, having evolved independently many times in several different phylogenetic lineages. The ancient air-breathing lungfishes are of particular interest because of their proximate relationship to the ancestor of tetrapods. The use of lungs as the primary site of oxygen exchange and the seemingly increased physiological dependence upon aerial respiration across ontogeny provide an excellent opportunity to study the physiology of these fishes along an evolutionary vector similar to that which the tetrapods might have experienced as they began to invade terrestrial environments. However, there remain some key gaps in our understanding of the respiratory profile of lungfishes, in particular respiratory allocation, and metabolic demand of juveniles in the context of ecologically relevant aquatic oxygen levels. This objectives of this study were to (1) investigate respiratory partitioning and quantify the resting rate of metabolism among juvenile Protopterus aethiopicus that vary x

PAGE 11

in body mass across a dissolved oxygen (DO 2 ) range typical of that found within hypoxic wetlands and (2) quantify the effects of temperature on resting rates of metabolism of juvenile P. aethiopicus through calculation of Q 10 values across a range of temperatures representative of those found within their natural habitat. Chapter 1 quantifies the relationship between respiratory allocation (air versus water) and the resting rate of metabolism (RRM) in the primitive air-breathing lungfish, P. aethiopicus. Simultaneous measurements of oxygen consumed from both air and water were made to determine the RRM at ecologically relevant levels of dissolved oxygen for juveniles 2 to 221 g. Aerial oxygen consumption averaged 98% (range = 94% to 100%) of total respiratory allocation. This finding contradicts the long-held belief that small juvenile lungfish primarily depend on aquatic respiration to meet their aerobic metabolic demand. Measurements of oxygen consumption made across a gradient of dissolved oxygen from normoxia to anoxia showed that P. aethiopicus maintains a stable RRM despite a change in respiratory allocation between water and air. Chapter 2 examines the effect of temperature on the resting rate of metabolism in P. aethiopicus. We present a Q 10 value of 3.14 for six individuals ranging in body mass from 42 to 222 g across an ecologically relevant 10 C temperature range. This is the first study to our knowledge to determine a Q 10 value for a non-teleost fish and the first to describe this relationship for an air-breathing fish. A comparison of literature-derived Q 10 values for both tropical and temperate fishes indicated a higher Q 10 in tropical species (n=3) than in temperate species (n=10) across an ecologically relevant thermal range. xi

PAGE 12

CHAPTER 1 GENERAL INTRODUCTION The transition from water to land stands as a pivotal stage in vertebrate evolution. Any study of this transitional period undoubtedly alludes to the importance of the atmosphere as the new prime source of metabolic oxygen and fishes were the first vertebrates to rise to the surface and begin to break free of their bonds to aquatic existence. Air breathing in fishes has a rich evolutionary history, having evolved independently numerous times in several different phylogenetic lineages and with various structures for gas exchange (Randall et al. 1981; Graham 1997). Air-breathing fishes are thought to have had their origins in the Late Silurian, based on fossil traces of lung-like structures in placoderms and ostracoderms (Gardiner 1980; Little 1990, Coates and Clack 1991; Gordon and Olson 1994; Long 1995; Graham 1997). The early evolution of lungs and their proposed widespread occurrence among ancient fishes are also strengthened by the observation that extant representatives of mostly extinct groups are bimodal breathers, who use either paired lungs (ancestral condition) or a single lung as the primary air-breathing organ (Randall et al. 1981; Burggren et al. 1986; Ultsch 1996). The phyletically ancient air breathers are of particular interest because of their proximate relationship to the ancestor of tetrapods. Their use of lungs as the primary site of oxygen exchange and their seemingly increased physiological dependence upon aerial respiration across ontogeny provide an excellent opportunity to study the physiology of these ancient fishes 1

PAGE 13

2 along an evolutionary vector similar to that which the tetrapods might have experienced as they began to invade the land. Surely the most infamous of air-breathing fishes are the lungfishes. Aside from the discovery of the coelacanth, Latimeria chalumnae, in the early part of the 19 th century, the near simultaneous discovery of both Lepidosiren paradoxa and Protopterus annectens in the 1830s and the discovery of Neoceratodus forsteri in 1870 stand as three of the most important zoological discoveries of the last two centuries (Burggren and Johansen 1987). Their exquisite fossil record provides a good picture of the interspecific relationships of both extinct and extant forms. However, ever since their discovery, their taxonomic position in relation to the tetrapods has remained a source of rich scientific debate. Lungfishes belong to the Ceratodontimorpha and, along with Latimeria (the coelacanth), remain the only extant representatives of the Sarcopterygii, a group from which tetrapods later emerged. Within the Ceratodontimorpha, the Lepidosireniformes, comprised of the African family Protoperidae (four species) and the South American family Lepidosirenidae (one species), appear largely unchanged from the ancestral Dipterus of the Carboniferous period (Long 1995; Graham 1997). Neoceratodus forsteri is the lone species representative of the Ceratodontidae and is considered the most primitive of the three genera based on several phylogenetic characters (Marshall 1987). Of the three genera of lungfishes, two are obligate air-breathers, while the Australian species, Neoceratodus forsteri, uses aerial respiration as a supplementary means of gas exchange (Grigg 1965; Graham 1997). Fossil evidence and their relative morphological stability through evolutionary time allow for the conjecture that modern day

PAGE 14

3 Lepidosireniformes may be physiologically similar to a tetrapod ancestor (Burggren et al. 1986). Are these interesting creatures the closest extant relative of the tetrapods? Regardless of when or whether scientists will come to answer this question, the fact remains that lungfish represent a very special organism for evolutionary biologists working from an ecological and physiological perspective. The respiratory and cardiovascular similarities that both Lepidosiren paradoxa and Protopterus species share with terrestrial amphibians make them interesting transitional species to investigate these systems in the context of vertebrate evolution. Beginning with Homer Smith in the 1930s, the majority of early research on Protopterus aethiopicus, and lungfishes in general, centered on cardio-respiratory adaptations and the physiological effects of aestivation. Given that P. aethiopicus shifts from having strongly developed external gills as small juveniles to poorly developed internal gills and well-developed paired lungs as adults, researchers hypothesized a strong shift in respiratory allocation between air and water. Indeed, this is precisely what early research uncovered. A comparison of early studies that measured oxygen consumption in lungfishes yields a profile whereby juveniles use primarily aquatic respiration to meet metabolic demands, and adults rely almost exclusively on aerial respiration (Smith 1930; Sawaya 1946; Johansen and Lenfant 1967; Lenfant and Johansen 1968; Lenfant et al. 1970; McMahon 1970; Johansen et al. 1976; Babiker 1979). Most of these investigations centered exclusively on either juveniles or adults. However, a study done by Johansen et al. (1976) examined changes in respiratory allocation between water and air across a wide range in body size from 4 g to 500 g. Their experiments established a two-phase

PAGE 15

4 respiratory profile for P. amphibius based on these experiments where the smallest juveniles relied on dissolved oxygen to meet metabolic demands, while larger animals acquired their oxygen aerially. Based on their findings they concluded that the altered dependence on water and air-breathing during the life-cycle of P. amphibius will most likely set territorial limits to the distribution of larvae and young fish within the distribution range of the species (Johansen et al. 1976. p. 399). This statement is striking in that it predicted the opposite of what is known of juvenile and adult African lungfish based on dissolved oxygen levels within their natural habitat. A careful consideration of the ecology of juvenile lungfish highlights this interesting paradox. Juvenile species of African lungfish are born within nests that are found within swamp regions surrounding larger bodies of water or in portions of rivers where current is slow. Measurements of dissolved oxygen concentrations made by Greenwood (1958) at the surface of swamps surrounding Lake Victoria ranged from 1.01 to 1.71 mg l -1 and within the breeding nests from 0.98 to 2.66 mg l -1 Research from the Lwamunda Swamp surrounding Lake Nabugabo (Uganda) found that juveniles remain within fringing swamps surrounding the main lake until they reach a body size of approximately 40 cm TL (Goudswaard et al. 2001). Individuals smaller than 40 cm TL are rarely found in the open waters of Lake Nabugabo where larger lungfish supported a thriving fishery before the introduction of Nile Perch in the 1960s. Dissolved oxygen levels in the Lwamunda Swamp are extremely low, and it is the younger juveniles (nestlings to juveniles of 40 cm TL) that are found under these hypoxic conditions (Goudswaard et al. 2001). With these ecological parameters in mind, it is curious that the youngest juveniles, who seemingly experience the greatest respiratory stress, would rely

PAGE 16

5 mainly upon aquatic respiration, while adults who move into open water habitats would suddenly transform into animals dependant on aerial respiration. These previous findings set up an interesting paradox created by the seemingly opposing forces of dissolved oxygen levels and respiratory physiology of these animals. If the transition from water to land in early tetrapods was mediated by hypoxic stress within ancient freshwater habitats, it was of great interest to me to determine if small, juvenile lungfish were primarily air-breathers and if their respiratory physiology was determined by strong environmental gradients. Given the ecological characteristics of Lake Nabugabo with a fringing hypoxic swamp surrounding a main lake of much higher DO 2 and the findings of Johansen et al. (1976), I sought to resolve the above stated paradox and set out to examine the respiratory allocation of Protopterus aethiopicus. My objectives were to (1) investigate respiratory partitioning and quantify the resting rate of metabolism among juvenile Protopterus aethiopicus that vary in body mass across a dissolved oxygen (DO 2 ) range typical of that found within hypoxic wetlands and (2) quantify the effects of temperature on resting rates of metabolism of juvenile P. aethiopicus through calculation of Q 10 values across a range of temperatures representative of those found within their natural habitat. Chapter 2 explores respiratory allocation and the resting rate of metabolism in juveniles of the African lungfish (Protopterus aethiopicus). The objective of this chapter is to quantify respiratory allocation between water and air and the rate of oxygen consumption over the size range typically found in marginal hypoxic wetlands in the Lake Victoria basin (nestlings to 40 cm TL, Goudswaard et al. 2001). To meet this objective, the rate of oxygen consumption and relative allocation to air and water

PAGE 17

6 breathing was measured across a broad range of dissolved oxygen concentrations including those reflective of the natural habitat. I also explored the effect of chamber acclimation time and chamber size on respiratory allocation and rate of oxygen consumption. In Chapter 3, I quantify the relationship between temperature and the resting rate of metabolism in Protopterus aethiopicus through a temperature acclimation study. Fishes, as the most species-rich group of chordates, lack a comprehensive synthesis of metabolic data that includes both phyletically ancient (non-teleost) and advanced groups. In particular, studies examining metabolic rates and the effects of temperature in non-teleost fishes are completely absent from the literature. This chapter determined the Q 10 value for P. aethiopicus. This Q 10 value is discussed in relation to those discerned from the literature for both temperate and tropical teleosts.

PAGE 18

CHAPTER 2 RESPIRATORY ALLOCATION AND THE RESTING RATE OF METABOLISM IN Protopterus aethiopicus Introduction Air breathing in fishes has a rich evolutionary history, having evolved independently numerous times in several different phylogenetic lineages and with various structures for gas exchange (Randall et al. 1981; Graham 1997). Air-breathing fishes are thought to have had their origins in the Late Silurian, based on fossil traces of lung-like structures in placoderms and ostracoderms (Gardiner 1980; Little 1990; Coates and Clack 1991; Gordon and Olson 1994; Long 1995; Graham 1997). The early evolution of lungs and evidence of their widespread occurrence among ancient fishes is supported by the observation that extant representatives of mostly extinct groups are bimodal breathers, using either paired lungs (ancestral condition) or a single lung as the primary air-breathing organ (Randall et al. 1981; Burggren et al. 1986; Ultsch 1996). Despite air breathing having evolved independently several times in more derived fishes, primitive air breathers are of interest because of their proximate relationship to the ancestors of tetrapods. Their use of lungs as the primary site of oxygen exchange and apparent increased physiological dependence upon aerial respiration across ontogenetic stages provides an excellent opportunity to study the physiology of these ancient fishes along an evolutionary vector similar to that which tetrapods might have initially experienced during the transition to a terrestrial existence. However, very little is known about energetics of primitive air-breathers under ecologically relevant oxygen levels and the 7

PAGE 19

8 relationship between respiratory allocation (water vs. air) and body size. In this study, I examined the energetics of juvenile African lungfish (Protopterus aethiopicus) across a range of body sizes to detect ontogenetic shifts in respiratory allocation and to measure resting rate of metabolism. Of the three families of lungfishes, the Lepidosirenidae (comprised of one species, Lepidosiren paradoxa) and the Protopteridae (comprised of four species of Protopterus) are obligate air-breathers, whereas the Ceratodontidae (comprised of the Australian species, Neoceratodus forsteri), uses facultative air breathing as a supplementary means of gas exchange (Grigg 1965; Graham 1997). Obligate air-breathing lungfishes possess highly partitioned, paired lungs that originate from the esophagus, have separate pulmonary circuits, and circulatory systems with a high degree of septation in the heart, resembling that of extant amphibians (Burggren and Johansen 1987). The majority of early research on P. aethiopicus, and lungfishes in general, focused on cardio-respiratory adaptations in larger fish (Sawaya 1946; Johansen and Lenfant 1968; Lahiri et al. 1970; Lenfant et al. 1970; McMahon 1970). The few studies that explored ontogenetic variation in respiratory strategies reported shifts in respiratory allocation with body size. For example, Johansen et al. (1976) found that small juveniles (3.7 g) of Protopterus amphibius relied on aquatic respiration to meet 70% of their oxygen requirement, whereas aquatic respiration accounted for as little as 10% of the oxygen uptake in larger specimens (300 g to 500 g). Similarly, Lenfant et al. (1970) found that aerial respiration accounted for nearly 97% of the total oxygen demand in large individuals of Lepidosiren paradoxa, but only 26% for small juveniles. The combination of these studies created a physiological portrait of the Lepidosirenidae and

PAGE 20

9 Protopteridae as animals with two distinct respiratory life stages (Jesse et al. 1967; Lenfant et al. 1970; McMahon 1970; Johansen et al. 1976; Babiker 1979). Small, juvenile lungfish were thought to be primarily dependent on aquatic respiration for survival, while larger individuals became increasingly dependent on aerial respiration as they grew, ultimately relying on aerial respiration for over 90% of their metabolic demand. This scenario, however, seems inconsistent within the ecological context of their habitats. In the field, nestling P. aethiopicus have been observed to begin air breathing at a length of 23 to 27 mm while still possessing external gills (Smith 1931; Greenwood 1987). At this size, they are associated with either floating mats of papyrus roots or amongst heavily vegetated patches within the swamp (Greenwood 1987). In lakes Victoria and Nabugabo (East Africa), Goudswaard et al. (2001) reported that P. aethiopicus smaller than 40 cm were very rare in open waters and almost exclusively associated with the fringing swamp. There, dissolved oxygen concentration can be extremely low (Chapman et al. 2002). Both Greenwood (1958) and Johnels and Svensson (1954) suggested that the shallowness of P. aethiopicus nests may be connected with the air-breathing behavior of the nestlings. In these nests, young have been observed resting on the roots near the surface (Greenwood 1987). Dissolved oxygen measurements made by Greenwood (1958) at the surface and at the bottom of the nests ranged between 1.01 and 1.71 mg l -1 and between 0.98 and 2.66 mg l -1 respectively. Although the smallest size of aestivating individuals is unknown, Protopterus dolloi, a West African lungfish species, has been found in cocoons at lengths of 40-50 mm (Johnels and Svensson 1954). The fact that individuals this small can aestivate suggests that they are capable of sustaining themselves solely on aerial respiration. In conditions such as these, and with

PAGE 21

10 the capacity to acquire oxygen aerially, it would appear that small, juvenile lungfish might use aerial respiration in their natural environment to an extent previously unrecognized. In earlier studies, measurements of oxygen consumption in small, juvenile P. aethiopicus were made under conditions of near-saturated dissolved oxygen, which do not reflect the observed natural ambient conditions at this life-history stage. The objective of this study was to quantify respiratory allocation between water and air and the rate of oxygen consumption in juvenile African lungfish (Protopterus aethiopicus) over the size range typically found in marginal hypoxic wetlands in the Lake Victoria basin (nestlings to 40 cm TL, Goudswaard et al. 2001). To meet this objective, we measured rate of oxygen consumption and relative allocation to air and water breathing across a range of dissolved oxygen concentrations reflective of the natural habitat. We also explored the effect of chamber acclimation time and chamber size on respiratory allocation and rate of oxygen consumption. Materials and Methods Study Site Specimens of the African lungfish, Protopterus aethiopicus were obtained from a natural population within the Lwamunda Swamp that surrounds Lake Nabugabo, Uganda (31 o 50 E; 31 o 56 S; to 0 o 20 E; 0 o 25 S, Fig. 2-1). Lake Nabugabo lies just south of the equator and is a small satellite lake (surface area = 24 km 2 mean depth = 4.5 m) that was isolated from Lake Victoria approximately 4000 years ago (Greenwood 1965). The lake lies within the extensive Lwamunda Swamp that was formerly a bay on the western shore

PAGE 22

11 Figure 2-1. Lwamunda Swamp, Lake Nabugabo, Uganda. Red stars indicate two collection sites where specimens of Protopterus aethiopicus were captured. of Lake Victoria (Worthington 1932; Greenwood 1965; Ogutu-Ohwayo 1993). The lake margin is dominated by hippo grass (Vossia cuspidatum), Miscanthidium violaceum, water lilies (Nymphaea lotus and N. caerulea), and small stands of papyrus (Cyperus papyrus). The dense interior of the swamp (dominated by M. violaceum) is characterized by numerous small lagoons and channels that serve as important habitats for P. aethiopicus (<40 cm total length, Goudswaard et al. 2001). Dissolved oxygen concentrations (DO 2 ) at several sites within a small swamp lagoon were measured over a 1-year period. Monthly DO 2 measurements in the upper 20 cm of water averaged 1.5 mg l -1 (26.8 mm Hg) in the morning and 1.8 mg l -1 (32.9 mm Hg) during the afternoon

PAGE 23

12 (Chapman et al. 2002). Water temperatures at the sites averaged 22.3 C in the morning and 23.8 C in the early afternoon (Chapman et al. 2002). DO 2 values in the Lwamunda swamp are particularly low in the dry season. In a survey of 12 small lagoons within the Lwamunda Swamp in the early dry season Chapman et al. (2002) reported an average DO 2 level of only 0.6 mg l -1 (11 mm Hg, upper 20 cm of water) in the early morning and 1.3 mg l -1 (24 mm Hg) in the mid-day period. In samples taken just above the sediments, DO 2 averaged 0.3 mg l -1 (5 mm Hg) in the early morning and 0.4 mg l -1 (7 mm Hg) in the mid-day (Chapman et al. 2002). Collection and Maintenance of P. aethiopicus Individuals were live-captured using baited minnow traps set at various places within the lagoons and left overnight. Traps were collected the following morning and each lungfish placed in its own separate holding container. This method was appropriate for catching individuals ranging in mass from 10 g to 250 g. Smaller individuals (<5 g) were obtained by dip netting vegetated areas within the lagoons and in all cases were caught in dense vegetation mats. Individual lungfish were held in 1.2-l buckets with lids while in Uganda and then transported to the University of Florida. Once at the University, individual lungfish were held in partitioned 208.2-L tanks with no two lungfish occupying the same space within a tank. Water temperature was maintained at 23 0.5 C, with a 12/12 photoperiod. Dissolved oxygen concentration averaged 8.16 0.05 mg l -1 Temperature, dissolved oxygen, pH, conductivity, ammonia, and nitrite levels were measured weekly to maintain appropriate water quality. Individual fish were held under these conditions at least 1 year before experimental trials. Fish were fed once a week on fresh chicken or beef liver and,

PAGE 24

13 in the case of the smallest individuals, blood worms. Our experimental protocols for P. aethiopicus were approved by the University of Florida (IACUC project #Z090). Respirometry Oxygen consumption was measured using a two-phase respirometer (Fig. 2-2). The respirometer was constructed using PVC, oxygen impermeable vinyl tubing and brass compression fittings (Swagelok ). Several chambers of different sizes were built and were interchangeable within the system. The chamber where fish were placed consisted of a long length of PVC connected to a T-junction that served as an air chamber. The T-junction was constructed from clear PVC to allow periodic observation of fish during respiratory trials, while the air chamber itself was constructed from opaque PVC. The length of PVC that constituted the bulk of the water chamber was opaque. A 1.5 amp Mag-drive water pump, flow meter, oxygen-sensing probe, and thermistor were all placed in series for the water phase of the system. A stopper, with an oxygen probe, thermistor, and small fan inserted through it, was used to seal the air portion of the system. Thus, each phase of the respirometer contained an associated oxygen sensing probe and thermistor that provided for simultaneous measurement of oxygen consumption and temperature from both aerial and aquatic phases of the respirometer. The respirometer utilized Ocean Optics Oxygen Sensing technology to measure oxygen concentration in both water and air (Dunedin, FL; http://www.oceanoptics.com). Oxygen detection using these probes is based on excitation of a ruthenium complex suspended in a sol-gel matrix at the ends of the probes. Light is emitted from a blue LED source at ~470 nm that travels via fiber optic cables to the tip of the probes.

PAGE 25

14 Figure 2-2. Respirometry system used to measure the resting rate of metabolism. The entire system is submerged in a water bath to maintain temperature. The system is connected to a labtop that records the data. (a) Water chamber; (b) air chamber; (c) 2.5 amp Mag Drive pump; (d) flowmeter; (e) three stop-cock compression fittings that allows the system to convert from a closed system to a flow-through system; (f) Ocean Optics oxygen sensing probe to measure DO 2 in water phase; (g) thermistor measuring water temperature; (h) thermistor and Ocean Optics oxygen sensing probe to measure temperature and DO 2 in air phase; (i) spectrophotometer.

PAGE 26

15 The ruthenium is excited when exposed to the light, and emits energy at ~600 nm that is returned via the fiber optic cables to a spectrophotometer. As oxygen molecules come in contact with the tip of the probe, the energy associated with the excitation via the blue LED is quenched. This quenching (a non radiative transfer) results in a decrease of intensity. An inverse relationship therefore exists between the intensity of excitation associated with the ruthenium and the presence of oxygen in a given medium. This information in intensity units is converted to percent saturation of oxygen in the medium based on probe calibration (Krihak and Shahriari 1996). The probes were calibrated to 0% DO 2 in water containing sodium sulfite. The probes were calibrated to 100% saturation for the water chamber in air-saturated water and for the air chamber in air. The probes were calibrated on a scale of percent saturation in the medium. Percent saturation was then converted to units of mg l -1 with corrections for ambient air pressure and water vapor pressure. The respirometer was designed as a closed system with the capacity to flush out and replace the contents of either phase. This allowed the removal of any metabolic waste products that accumulated during the acclimation period. Probes were calibrated prior to each run. Individual fish were placed in the system containing a mixture of aerated tap water and filtered tank water. Controls were run on the system with no fish present to determine biological activity of the water. In all cases microbial O 2 consumption in the water was determined to be negligible. Controls were also run to determine the diffusion of oxygen between the air and water phases under various levels of dissolved oxygen. These controls showed that the transfer of oxygen between the two phases was undetectably small.

PAGE 27

16 The resting rate of metabolism (RRM) was measured both at the Makerere University Biological Field Station in Uganda and in our laboratory at the University of Florida (UF). Resting rate of metabolism was measured as a function of oxygen consumed over time. We refer to our measurements of oxygen consumption as RRM, although we believe our final estimates represent minimum rates for these fish (measured in the dark). Movement of the fish during trials was almost exclusively associated with respiration and therefore considered a portion of maintenance metabolism since it is necessary for survival. The protocols differed slightly as described below. For those fish measured in the field (N=16), individual fish were acclimated for 3-h periods inside the chamber with free access to air. After the acclimation period the air chamber was sealed and the run was initiated. Oxygen consumption was recorded over a DO 2 range from 7.21 (130 mm Hg) to 5.77 mg l -1 (104 mm Hg) at 23 C. Measurements were made within 3 wks of capture, on post-absorptive fish, and we refer to these as our field estimates. For those fish measured in the laboratory (N=14), individuals were acclimated overnight to the system with free access to air. The DO 2 concentration in the water chamber was allowed to drop to near 0.0 mg l -1 the approximate level at which aquatic oxygen consumption ceased. An overnight acclimation was used because oxygen consumption measurements revealed that the RRM measured after a 3-h acclimation dropped approximately two-fold when compared to measurements made after a 16-h acclimation. The effect of a short acclimation time on the RRM only became apparent after laboratory tests were run at the UF where temperature control over an extended period could be maintained. Prior to each run in the morning the system was flushed with air-saturated water until the DO 2 concentration in the system was greater than 2.57 mg l -1

PAGE 28

17 (46.4 mm Hg). This served to remove any metabolic wastes that may have accumulated overnight and to raise aquatic DO 2 The air chamber was then closed and the run initiated. Oxygen measurements were made every 10 sec over the course of the run and recorded automatically by the computer software running the system. Each run was monitored and the percent saturation of oxygen in the air chamber was never allowed to drop below 608.0 mm Hg. When oxygen concentration within the air chamber approached 608.0 mm Hg, the stopper was removed and fresh air allowed to mix into the air chamber to return the oxygen concentration to 100% saturation. This mixing did not increase DO 2 within the water chamber. Once the DO 2 concentration within the water chamber reached a plateau representing zero uptake from water, the run was allowed to continue for 1 h and was then terminated. Oxygen consumption was recorded over a DO 2 range between 4.29 mg l -1 (77.4 mm Hg) and 0.00 mg l -1 at 23 C ( 0.5C). Temperature was maintained throughout the run by placing the entire system inside a Rubbermaid cooler that served as a temperature bath. Most of the system was submerged in the cooler and the entire apparatus was covered with a black sheet to shield the fish from any movements of the investigators. Individuals were starved for 7 d prior to each experiment to ensure that they were post absorptive. Runs lasted an average of 12 h. At the end of each run, individuals were drip dried and weighed (0.0 g). An additional experiment was run in the laboratory to investigate any changes in the RRM at near saturated levels of DO 2 compared to those at ecologically relevant DO 2 levels. We used the overnight laboratory acclimation period to provide comparability with our runs at lower DO 2 Following acclimation, the entire chamber was flushed and the DO 2 level in the chamber returned to near saturation. These runs lasted 6 h or less and

PAGE 29

18 were terminated when DO 2 concentration in the water phase was less than 6.0 mg l -1 (108.4 mm Hg). Analysis of covariance was used to compare the bilogarithmic relationships of RRM and body mass among size groups and DO treatments. A t-test was used to compare means when body size was not required as a covariate. Results Field Laboratory Data Measurements of oxygen consumption on lungfish taken within 3 wks of capture and measured at our field laboratory (N=16) at high DO 2 concentrations revealed a strong shift in respiratory allocation with body mass (Table 2-1). Individuals less than 55 g were almost exclusively water breathers (percent allocation from water 95.7 0.9%). The allocation to air breathing increased from an average of 4.3 0.9% in individuals less than 55 g to 50.7 9.0% in individuals 55 to 111 g, and to 79.7 2.7% in individuals greater than 111 g (Table 2-1). Measurements of oxygen consumption also revealed a shift in the bilogarithmic relationship of RRM and body mass (Fig. 2-3). There was no difference in the slopes of the bilogarithmic relationship of RRM (mg O 2 hr -1 ) and body mass (g) between individuals < 83 g and 83 g (F = 4.187, p = 0.063). However, the intercepts (representing adjusted mean RRM) differed (F = 20.743, p = 0.016). Individuals < 83 g had a 5.4-fold higher adjusted mean RRM of 10.26 mg O 2 hr -1 compared to individuals 83 g (1.89 mg O 2 hr -1 RRM adjusted to the mean mass of 82.0 g). UF Laboratory Data The RRM of P. aethiopicus (N=14) measured in the laboratory comprised both aerial and aquatic oxygen consumption: Figs. 2-4 A,B illustrates these separate

PAGE 30

19 components as oxygen consumed (mm Hg) over time (min) for two individuals. While aerial consumption of oxygen fit a linear function, aquatic oxygen consumption fit an exponential function (Figs. 2-4 A,B). To examine if the degree of curvilinearity within the water phase varied with body size, we fit an exponential function (y=b o e b1*t ) to individual curves and compared the exponents (b 1 ) among individuals as a function of body mass. A comparison of the exponents indicated the degree to which the relationship was curved or departed from a linear relationship. The degree of curvilinearity was positively correlated with body mass (r = 0.598, p = 0.031, Fig. 2-5). The smallest individual (2 g) was excluded from this analysis. Lungfish reduced aquatic respiration to less than 1% of total oxygen consumed at low aquatic PO 2 represented by the plateau point where aquatic respiration was nearly Table 2-1. Respiratory allocation of Protopterus aethiopicus from the Lwamunda Swamp, Uganda, between water and air at high (130 mm Hg 104 mm Hg) levels of DO 2 and measured at an average temperature of 23 C. Measurements were made within 3 wks of capture at the Makerere University Biological Field Station, Uganda. Mass (g) Percent O 2 from Water Percent O 2 from Air 17 96 4 21 97 3 27 94 6 55 68 32 77 67 33 83 19 81 89 65 35 107 61 39 108 19 81 111 46 54 120 25 75 123 31 69 149 16 84 160 14 86 169 20 80 180 16 84

PAGE 31

20 -1.00.01.02.00.01.02.03.0Log10 body mass (g)Log10 RRM (mg O2 hr )-1 Figure 2-3. Bilogarithmic relationship between resting rate of metabolism (RRM) and body mass of Protopterus aethiopicus from the Lwamunda Swamp, Uganda, measured at high levels of DO 2 in the field [(7.21 (130 mm Hg) to 5.77 mg l -1 (104 mm Hg) at 23 C]. Individuals were acclimated for 3 h prior to trials. Closed squares represent individuals 81 g, and open squares represent individuals 83 g. Individuals 81 g were measured in 2.5 cm chambers, and fish 83 g were measured in 5.1 cm chambers.

PAGE 32

21 020406080100050100150200250300350400Time (min)PO2 (mm Hg) 22 gA 600700800W AirWater -40-200204060801000100200300400500600700800Time (min)PO2 (mm Hg) B221 g 700800W AirWater Figures 2-4. The decline in oxygen partial pressure over time for Protopterus aethiopicus from the Lwamunda Swamp, Uganda, in a two-phase, closed respirometer, where fish had access to both air and water simultaneously. The curvilinear function y=b 0 e b1*t for aquatic respiration is indicated for two representative individuals A)22 g B)221 g).

PAGE 33

22 00.00050.00100.00150.00200.00250.00300.0035050100150200250Body mass (g)b1 Figure 2-5. The relationship between the degree of curvilinearity (b1 from the equation y=b 1 e b0*t ) and body mass (g) for Protopterus aethiopicus from the Lwamunda Swamp, Uganda. Curvilinearity describes the decline in oxygen partial pressure over time in the aquatic phase of a two-phase, closed respirometer where fish had access to both air and water simultaneously. zero on each graph (Table 2-2). Whereas some variation among individuals occurred, fish 87 g reached this plateau at a significantly lower DO 2 level than individuals 81 g (t-test, t = 3.182, p=0.030). Active aquatic respiration ceased at a mean DO 2 of 0.48 0.2 mg l -1 (8.7 mm Hg) for fish 87 g and 2.58 0.7 mg l -1 (46.5 mm Hg) for fish 81 g.

PAGE 34

23 Table 2-2. Respiratory allocation of Protopterus aethiopicus from the Lwamunda Swamp, Uganda, from air at the aquatic respiration plateau (representing the level of DO 2 at which individuals respired almost completely aerially). This plateau point was considered representative of ecologically relevant levels of DO 2 Water temperature was approximately 23 C. Mass Percent O 2 from air 2.0 100.0 9.0 99.0 22.0 100.0 41.5 98.2 50.0 97.7 81.0 99.6 87.0 100.0 112.0 100.0 134.0 97.7 139.5 94.0 144.0 100.0 206.5 97.0 208.0 98.2 221.8 100.0 Chamber Effect In the field, chamber size was altered among specimens to maintain a similar ratio between chamber volume and fish mass following previously published protocols for other fishes. In the laboratory, three individuals (2, 9, and 22 g) were measured within a 2.5 cm (1-inch) PVC chamber (total water volume=266 ml), while the remaining individuals were measured in 5.1 cm (2-inch) PVC chambers (total water volume=1175 ml). To explore potential effects of chamber volume, we tested the 9and 22-g fish in both 2.5 cm and 5.1 cm chambers to compare RRM. For these two individuals we replicated each chamber size twice at two different levels of DO 2 corresponding to two different ratios of respiratory allocation (pre-plateau and plateau). The RRM was calculated as two rates for each chamber size: from the portion of the curve describing the water phase before it plateaus when lungfish used both aquatic and aerial respiration (pre-plateau), and from the plateau of the curve when the lungfish consumed primarily

PAGE 35

24 atmospheric oxygen (plateau). We adjusted RRM for each DO 2 regime by removing effects of body mass based on the equation described by Ultsch (1995), [VO 2 (adjusted)=(mean weight) b-1 (observed weight) 1-b (observed VO 2 )]. This allowed us to compare the adjusted mean RRM for each chamber size for both the pre-plateau and plateau. At pre-plateau DO 2 levels, individuals run in the 2.5 cm chamber had a 2.4-fold higher RRM when compared to the same individuals run in the 5.1 cm chamber (RRM, 2.5 cm: 0.147 0.02, 5.1 cm: 0.06 0.01). At plateau DO 2 levels, RRM values were similar (RRM, 2.5 cm: 0.046 0.01, 5.1 cm: 0.045 0.01). When these smaller lungfish were respiring from both media (which generally occurs above field-relevant DO 2 levels) there was a distinct chamber effect, with higher RRM values in the smaller chamber. When respiratory allocation was primarily aerial (at ecologically relevant levels of DO 2 ), there was no detectable chamber effect. Resting rate of metabolism The main objective of this study was to collect metabolic data for Protopterus aethiopicus at ecologically relevant DO 2 levels to provide a more realistic physiological portrait of how animals survive under natural conditions in their respective environments. The RRM was calculated as total oxygen consumed (mg O 2 hr -1 ) from both the air phase and aquatic phase of the respirometer. We used the aquatic phase of the curve at its plateau where the lungfish is respiring almost completely aerially because this is a good representation of the level of DO 2 found in habitats used by small-sized lungfish in the field, and chamber effects were not detectable (Goudswaard et al. 2001; Chapman et al. 2002). With the exception of the 2-g individual, we used data for individuals run in the 5.1 cm chambers. Total metabolic rate was positively correlated with body mass

PAGE 36

25 (r 2 =0.9392, p=0.031, Fig. 2-6), with a scaling exponent of 0.78. Individuals showed a mean percent allocation to air breathing of 98.7 0.48%. -2.0-1.00.01.00.01.02.03.0Log10 body mass (g)Log10 RRM (mg O2 hr )-1 Figure 2-6. The bilogarithmic relationship between RRM (mg O 2 hr -1 ) and body mass (g) for the African lungfish, Protopterus aethiopicus, collected from the Lwamunda Swamp, Uganda. Data represent RRM at low (ecologically relevant) DO 2 levels (measurements made at plateau point where oxygen consumption in the aquatic phase was almost 0%; this was taken to represent ecological conditions because DO 2 levels would either not drop any lower or were in the range found in the field). Measured at 23 C.

PAGE 37

26 Individuals were also run at high (near saturated) levels of DO 2 to test for an effect of DO 2 on RRM. No difference in slopes of the bilogarithmic relationship between RRM (mg O 2 hr -1 ) and body mass (g) occurred between high and low levels of DO 2 (F = 3.177, p = 0.094). In addition, the intercepts (representing the adjusted mean RRM) did not differ between the two oxygen levels (F = 0.895, p = 0.357). However, respiratory allocation differed with body size and percent respiratory allocation from water decreased with an increase in body size for fish tested in both the field (r 2 =0.822, p<0.001) and the laboratory (r 2 =0.547, p=0.058). We also measured pre-plateau RRM approximating bimodal respiration before oxygen acquisition from the aquatic phase plateaued. The level of dissolved oxygen at which the pre-plateau RRM was measured differed with each individual according to the plateau level. No difference in slopes of the bilogarithmic relationship between RRM (mg O 2 hr -1 ) and body mass (g) was found between pre-plateau levels and low levels of DO 2 (F= 0.265, p= 0.612). No difference in the intercepts (representing the adjusted mean RRM) occurred between groups (F= 0.242, p=0.626, Fig. 2-7). For the above two comparisons the 2-g individual was not included, because it had to be run in a different chamber size. Discussion The results of this study clearly indicate the importance of considering ecological conditions when measuring physiological parameters. Juveniles between 2 and 221 g (including externallygilled individuals) respired almost exclusively aerially at low

PAGE 38

27 -0.500.5111.52Log10 body mass (g)Log10 RRM (mg O2 hr -1) 2.5 Pre-plateau Low Figure 2-7. The bilogarithmic relationship between resting rate of metabolism (RRM) (mg O 2 hr -1 ) and body mass (g) of Protopterus aethiopicus from the Lwamunda Swamp, Uganda, measured at levels of DO 2 prior to the plateau in aquatic respiration versus low levels (plateau point at which fish is respiring almost completely aerially; ecologically relevant). Measured at 23 C. concentrations of DO 2 which is contrary to earlier studies on small juveniles at high oxygen levels. Protopterus aethiopicus also showed a remarkable ability to maintain its RRM over the entire range of DO 2 from 100% saturation to near anoxic conditions. The relationship between aquatic gas exchange and DO 2 is described by a curvilinear relationship with the degree of curvilinearity increasing as a function of body mass. This result, along with both the findings related to acclimation time and chamber size, stress

PAGE 39

28 the importance of protocol in determining the RRM and respiratory allocation to water and air in air-breathing fishes. These results differ from previous studies where both total oxygen consumption and respiratory allocation were measured at high DO 2 levels. To better understand the importance of these results, they need be examined in light of what is known of respiration in lungfish. Metabolic Profile A comparison of previous studies that measured oxygen consumption in lungfishes, yielded a profile whereby juveniles used primarily aquatic respiration to meet metabolic demands, and adults relied almost exclusively on aerial respiration to meet metabolic oxygen demands (Smith 1930; Sawaya 1946; Johansen and Lenfant 1967; Lenfant and Johansen 1968; Lenfant et al. 1970; McMahon 1970; Johansen et al. 1976; Babiker 1979). Johansen et al. (1976) reported that immediate post larval and early juvenile Protopterus amphibius depended on aquatic respiration for 70% of their oxygen demand, whereas fish between 300 and 500 g obtained as little as 10-15% of their oxygen from water (Johansen et al. 1976). When measuring respiratory allocation between water and air in Protopterus annectens Babiker (1979) found that small specimens < 200 g utilized less than 10% aerial respiration to meet metabolic demands, whereas individuals > 400 g utilized aerial respiration for 83% of their total oxygen uptake. In the only other published study to measure juvenile lungfish and report body mass, Johansen and Lenfant (1967) found that one juvenile L. paradoxa (150 g) used aquatic respiration to meet 64% of its metabolic demand. Oduleye (1977) found that 4 to 150 g juvenile P. annectens acquired about 94% of their oxygen aerially, however variation among size classes was not reported in this study.

PAGE 40

29 These previous findings mirror RRM measures that we made on lungfish within 3 wks of capture under high levels of DO 2, when juvenile P. aethiopicus (< 83 g) used aquatic respiration to account for 84 7% of their total oxygen uptake. Similarly, measurements in our laboratory at UF after long-term acclimation to normoxia and taken at high DO 2 indicated higher allocation to aquatic respiration with juveniles < 83 g acquiring approximately 66 10% oxygen uptake from water. These results differed from laboratory measures at low aquatic oxygen levels that indicated a mean allocation towards aerial respiration for individuals between 2 and 221 g of 98.7 0.46%. Our laboratory measurements of oxygen consumption made across a gradient of DO 2 from normoxia to anoxia showed that P. aethiopicus maintains its RRM despite a change in respiratory allocation between water and air. While respiratory allocation differed depending on size, with smaller individuals acquiring more oxygen from the aquatic phase at higher levels of DO 2 P. aethiopicus displays a remarkable ability to maintain its RRM despite these drastic changes in respiratory allocation. Larger individuals still satisfy the bulk of their oxygen demand aerially, while the smallest individuals apparently rely more on aquatic respiration at higher levels of dissolved oxygen. This difference may represent the ability of the gills to efficiently extract oxygen from water in smaller individuals. Based on the curvilinear relationship between aquatic oxygen consumption and increasing body size, it would appear as if smaller individuals are more capable of satisfying their oxygen demand via the gills and or skin. The curvilinear pattern of oxygen uptake observed in P. aethiopicus in the aquatic phase of respiration may indicate a passive adjustment to falling levels of DO 2 that would support active adjustments to aerial respiration as the primary mechanism to maintain a relatively

PAGE 41

30 constant RRM. For a metabolic regulator breathing from one medium, the relationship between oxygen consumed as a function of time would be linear on a given ample oxygen to meet the metabolic demand. Measurements made on both terrestrial air breathers and aquatic water breathers yield this result (McNab 2002). The exponential nature of the relationship shown here suggests that the acquisition of oxygen from the aquatic phase may be fueled by a diffusion gradient, maintained across the gills, skin, or both, and related to total respiratory surface area. Smaller individuals have a larger surface area to body mass ratio than do larger individuals and thus a passive diffusion of oxygen across multiple gas exchange surfaces may be ample to satisfy the oxygen demand at higher levels of dissolved oxygen. A large diffusion gradient that would help to facilitate oxygen uptake in hypoxic water would cause increased rates of oxygen diffusion under higher levels of DO 2 due to the passive nature of the diffusion gradient. Although aquatic respiration in the lungfish has been assumed to occur mainly at the gills, recent studies on Lepidosiren paradoxa have indicated that the skin may play a more prominent roll in the aquatic exchange of gases (Abe and Steffensen 1996; Sanchez and Glass 2001). No direct experiment has been performed on lungfishes to determine the contributions of both skin and gills to aquatic respiration. However, Sacca and Burggren (1982) investigated this possibility in Calamoichthys calabaricus, a primitive air-breathing polypterid and determined that the skin and gills contributed 32% and 28%, respectively, to oxygen uptake. Spaces between the large, thick ganoid scales of C. calabaricus therefore permitted aquatic gas exchange suggesting the relatively smaller scales of P. aethiopicus would pose little resistance to oxygen uptake. In fact, in similar shaped, elongate fishes, such as Electrophorus (Farber and Rahn 1970), and

PAGE 42

31 Saccobronchus (Singh and Hughes 1971), which are all capable of air breathing, the skin has been shown to be an equally important gas exchange organ. The frequency of gill ventilation in lungfishes is known to remain constant with increasing hypoxia, and thus it seems reasonable to suggest that the increase in aquatic respiration can be attributed to diffusion across the skin (Johansen and Lenfant 1968; Sanchez et al. 2001). Furthermore, when burrowing in soft mud, the gills of P. aethiopicus would be severely compromised, although burrowing individuals efficiently excrete both CO 2 and metabolic wastes. Under conditions of extreme hypoxia, it is possible that the skin plays an even greater role in oxygen exchange. Our observations of individuals before and after trials indicated that P. aethiopicus might actively alter vascularization patterns to the skin. The junction of both the pelvic and pectoral fins with the body were always very red, and the color intensified in some individuals over the course of a trial. This red color also extended along portions of the appendages. In some cases, lighter-colored individuals showed an increase in vascularization over extensive portions of the body. Lenfant and Johansen (1968) documented a progressive vasodilation of the skin in response to prolonged air exposure. An increased vascularization of the skin could facilitate increased oxygen uptake and/or increased offloading of carbon dioxide. Future investigations comparing total respiratory surface area to body mass in the lungfish along with direct measurements of gas exchange across the skin will be needed to determine more clearly the role of the skin in oxygen uptake in P. aethiopicus. Methodological Issues In the process of evaluating the RRM in P. aethiopicus, two methodological issues were discovered that influence metabolic rate measurements in this species: chamber size and acclimation time. Both a change in chamber size with body mass and the short

PAGE 43

32 acclimation time used in the field led to differences between our field and laboratory measurements of RRM in P. aethiopicus. At our field lab, small juveniles exhibited elevated metabolic rates (when total metabolic rate was graphed as a function of body size) compared to larger individuals, with small juveniles having a 5.9-fold higher RRM than those individuals 83 g. However, the shift was coincident with our shift from a small to large chamber size. When we evaluated chamber size effects in the laboratory, we found that use of a smaller chamber size for fish < 22 g coincided with an elevation in the RRM at high DO 2 but not when fish were breathing primarily atmospheric air (plateau level). Thus, we attribute the shift in adjusted mean RRM that we observed between small and large individuals in our field lab to reflect, at least in part, the change in chamber size. The smaller chambers may be perceived by smaller individuals as confining and illicit a stress response in these smaller individuals. Similar results were found, in both Protopterus amphibius (Johansen et al 1976) and in Protopterus annectens (Babiker 1980) with elevated RRM in smaller individuals. Methods used in these studies indicated that alternate chamber sizes were also used. Many previous studies also report a short acclimation time in the chamber. Our acclimation comparisons in the laboratory revealed at least a two-fold decrease in RRM between a 3-h and 16-h acclimation period. This suggests that chamber acclimation time should be carefully considered in any design or meta-analysis of fish metabolic rates. Both methodological issues may reflect behavioral responses. A short acclimation time potentially increases stress levels. Within the various chambers the smallest individuals were seen to hover at the water/air interface, whereas large individuals spent the majority of their time hidden within the opaque portion of the water chamber except

PAGE 44

33 when moving to breathe air. Prodding of smaller individuals both in the chamber and in holding tanks elicited a burst of activity. This behavior resembled an escape mechanism or predator avoidance. While all juveniles measured in this experiment were from the swamp, the smallest fish, particularly the gilled individuals, were often found associated with dense vegetation in the field. When disturbed in the field, they behaved in a similar manner. Their reflex is to retreat rapidly to the bottom of the water column, and this behavior occurs when an animal is confined or threatened. The close confines of the smaller chamber could have increased the stress level of these individuals enough to account for the increase in metabolic rate. The effect disappeared when individuals switched to an almost complete dependence on aerial respiration. This suggests that the increase in RRM may be attributed to an increase in aquatic respiration brought about by a stress response to a confined chamber. This conclusion is reinforced by the finding that chamber size apparently had no effect on aerial respiration. It is clear from this study that Protopterus aethiopicus is able to maintain a relatively constant RRM over a range of DO 2 levels from normoxia to hypoxia. Measurements of the RRM at normoxic conditions, pre-plateau levels, and with individuals respiring almost completely aerially did not differ. Although respiratory allocation differs across the range of dissolved oxygen from normoxia to extreme hypoxia, adjustments in oxygen uptake serve to balance the overall uptake of oxygen to stabilize RRM. Since smaller juvenile specimens of Protopterus aethiopicus are clearly taking in higher amounts of dissolved oxygen at near saturated levels of DO 2, it will be interesting to determine if these adjustments are passive or active and contribute to surplus energy uptake, or if they are related to both carbon dioxide excretion and

PAGE 45

34 nitrogenous waste removal. Future experiments will be necessary to define the role of the skin in oxygen uptake and how this relates to changing levels of DO 2

PAGE 46

CHAPTER 3 THE RELATIONSHIP BETWEEN TEMPERATURE AND THE RESTING RATE OF METABOLISM IN Protopterus aethiopicus Introduction Few of all environmental factors known to affect metabolism are as significant as temperature. The study of this relationship, from both a molecular and whole organism perspective, yields insight into the intimate connection between animal energetics and the environment. A derivation from first principles of the association between temperature and the resting rate of metabolism has been a topic of rich conversation ever since August Krogh and his co-workers attempted to formulate such a relationship for poikilothermic organisms (Krogh 1914, 1916). His quantitative expression of this relationship for a single species, the goldfish (Cassius auratus) led to his derivation of what came to be known as Kroghs normal curve (Ege and Krogh 1914). The normal curve expressed an exponential increase in metabolic rate as a function of increasing temperature. Winberg (1956), examining resting metabolic rates in fishes, compared values drawn from the literature to Kroghs normal curve in an effort to compare metabolic rates from different species measured at different temperatures. Using the normal curve for orientation, Winberg (1956) derived a series of coefficients that could be used to adjust metabolic rates of temperate fishes by applying the Vant Hoff equation, [Q 10 =(K 2 /K 1 ) (10/ t2-t1) ] where K 1 and K 2 are the metabolic rates measured at two different temperatures t 1 and t 2, for changes in temperature across intervals of 10 C. This permitted adjustment of 35

PAGE 47

36 metabolic rates measured at a specific temperature to a second specific temperature of biological interest and a description of how an individuals resting metabolism responded to an increase or decrease in temperature. As long as an ectotherm is acclimated for an adequate length of time to each new temperature, this rate reflects the long-term compensatory effects associated with metabolism affecting a homeostatic balance within the individual despite a change in environmental conditions. The Q 10 coefficients for fishes derived by Winberg were based largely on data collected for temperate fishes. The degree to which these coefficients apply to metabolic data collected for fishes that inhabit tropical or polar waters and that experience a smaller range of natural temperature fluctuation is largely unknown. Researchers investigating species from these systems are faced with the decision to calculate their own Q 10 values or to use published values measured on other species. Based on the experiments of Johnston et al. (1991) and Clarke and Johnston (1999), utilizing Q 10 relationships for organisms with different lifestyles and/or from unrelated taxa can clearly lead to erroneous adjustments to calculated metabolic rates. Fishes, as the most species-rich group of chordates, lack a comprehensive synthesis of metabolic data that includes both ancient relic fishes (non-teleost) and teleosts, and that adequately represents tropical, temperate, and polar regions. In particular, studies examining metabolic rates and the effects of temperature in non-teleost fishes are completely absent from the literature. The objective of this study was to determine the Q 10 value for the African lungfish, Protopterus aethiopicus, an ancient air-breathing fish from a tropical freshwater habitat. Here we present the Q 10 values for six individuals ranging in body mass from 42 to 222 g evaluated across a biologically

PAGE 48

37 relevant 10 C temperature range. These Q 10 values are discussed in relation to values drawn from the literature for both temperate and tropical teleosts. Materials and Methods Six juvenile lungfish ranging in size from 44 to 222 g were used in this study. Individuals had been housed at the University of Florida (UF) since their collection in June 2001 from Lake Nabugabo, Uganda. Lake Nabugabo lies just south of the equator and is a small satellite lake (surface area = 24 km 2 mean depth = 4.5 m) that was isolated from Lake Victoria approximately 4000 years ago (Greenwood 1965). The lake lies within the extensive Lwamunda Swamp that was formerly a bay on the western shore of Lake Victoria (Worthington 1932; Greenwood 1965; Ogutu-Ohwayo 1993). In the Lake Nabugabo system, Goudswaard et al. (2001) found that 90% of the lungfish captured in minnow traps in the swamp were between 20 cm and 45 cm TL. Larger lungfish in the system are found in more open waters (L. Chapman, personal observations). Water temperatures measured throughout the year and within the swamp averaged between 21.3 C and 25.2 C in the early morning and between 21.3 C and 27.2 C in the early afternoon (Chapman et al. 2002). Prior to metabolic trials, individuals were held at 23 C .5 C (12/12 photoperiod) in four partitioned 20-gallon tanks with no two lungfish occupying the same space within a tank. Tank temperatures were controlled with aquarium heaters, and temperatures were measured daily. Dissolved oxygen, pH, conductivity, ammonia, and nitrite levels were measured weekly to maintain appropriate water quality, and tanks were maintained on a 12/12 photoperiod. Individuals were fed fresh chicken liver once a week. Our experimental protocols for P. aethiopicus were approved by the University of Florida (IACUC project #Z090).

PAGE 49

38 Six individuals were acclimated across the temperature range of 20 C to 30 C over a 5-month period beginning in January 2003. Individuals initially at 23 C were gradually increased to 30 C over a 1-wk period. Once at 30C, individuals were acclimated for at least 2 wks with a daily fluctuation in temperature of less than .5 C. Each lungfish was assigned a number, and a random number table was used to select the order in which individuals were measured for their resting rate of metabolism (RRM). For fishes the resting rate of metabolism is approximately the same as the standard rate of metabolism (SRM) and is the respiration rate of an unfed fish resting quietly in the experimental chamber (Clarke 1991, 1993). Here we use the (RRM), which we believe accurately reflects the standard rate of metabolism for Protopterus aethiopicus. Individuals were starved for 1 wk prior to their metabolic run. The RRM was measured via the measurement of oxygen consumption in a two-phase respirometer following the protocol described in Chapter 2. Oxygen consumed from both water and air was measured simultaneously. Respirometer temperature was maintained to within .5 C during each experimental run. Once all six individuals were measured for the RRM, individuals were lowered to the next temperature regime over a 3-d period and again acclimated for at least a 2-wk period. In this way, the above metabolic experiments were performed at 30 C, 25 C, and 20 C. Statistics The resting rate of metabolism (RRM) for each individual was calculated as an adjusted mean by controlling for the effects of body mass based on the equation described by Ultsch (1995) [VO 2 (adjusted)=(mean weight) b-1 (observed weight) 1-b (observed VO 2 )]. We used the scaling exponent b=0.78 from Chapter 2 that describes

PAGE 50

39 the bilogarithmic relationship between metabolic rate and body mass for this population of lungfish across a broad size range of 14 juvenile fish. Repeated measures analysis of variance was used to calculate the adjusted mean resting rate of metabolism at each water temperature. Following Potvin et al. (1990), Mauchlys criterion was used to test for the compound symmetry of the variance-covariance matrix. The Q 10 value relating RRM to water temperature over a 10C range was calculated using the equation: Q 10 =(K 2 /K 1 ) (10/ t2-t1) where K 2 and K 1 are metabolic rates at temperatures t 2 and t 1 The determination of the Q 10 value indicates an adjustment in RRM to temperature changes across a known temperature gradient (Winberg 1956) and was determined for Protopterus aethiopicus over the range of 20 C to 30 C. All errors reported are standard errors. Results Protopterus aethiopicus showed a positive bilogarithmic relationship between the rate of oxygen consumption and body mass (range=44 to 222 g) at all three water temperatures (20 C: r 2 =0.798, P=0.016; 25 C: r 2 =0.847 P=0.009; 30 C: r 2 =0.862, P=0.007, Figure 3-1). Because of the relatively small sample size for which we could repeatedly measure metabolic rates across a temperature gradient, we used the slope of the bilogarithmic relationship described in Chapter 2 (0.78) to derive adjusted mean resting metabolic rates at each of the three temperatures. Adjusted mean RRM ranged from 0.00833 mg O 2 hr -1 at 20 o C to 0.01567 mg O 2 hr -1 at 25 o C and 0.02617 mg O 2 hr -1 at 30 o C (adjusted to a mean body mass of 110.5 g, Fig 3-2). The mean RRM at each temperature interval was significantly different from the preceding interval (20 C vs 25 C: p=0.006, and 25 C vs 30 C: p=0.019, repeated contrasts, Fig. 3-2).

PAGE 51

40 -0.500.000.501.001.502.002.50 Lo g 10mass ( g ) Log RRM (mg O hr ) 10 2 -1 Figure 3-1. Bilogarithmic relationship between resting metabolic rate (RRM) and body mass for six individuals of Protopterus aethiopicus from the Lwamunda Swamp, Uganda, measured at three different temperatures: 30 C, 25 C, and 20 C. Lungfish were acclimated for at least 2 wks prior to measurement. Triangles represent 30 C, squares represent 25 C, and closed circles represent 20 C. 00.0050.0100.0150.0200.0250.030202530 Water temperature ( C)Adjusted mean RRM (mg O hr ) 2-1o Figure 3-2. Mean resting rate of metabolism adjusted to 110.5 g fish (mean for six individuals) for three temperatures: 30 C, 25 C, and 20 C. A within-subject contrast found the mean RRM at each temperature interval to differ significantly from the preceding interval: 20 C vs 25 C p=0.006, and 25 C vs 30 C p=0.019. Error bars are less than .002.

PAGE 52

41 The relationship between RRM and temperature was calculated as the Q 10 value for the temperature interval of 10C between 20C and 30C. The Q 10 value was 3.14. Discussion The primary focus of this study was to examine the response in RRM of Protopterus aethiopicus within a biologically relevant range of water temperatures to gain insight into how this tropical, freshwater, non-teleost fish responds to changes in temperature. This was accomplished by accurate determination of the RRM, adequate acclimation to temperature for each individual in order to remove the effects of any short-term compensatory mechanisms in response to temperature, and a review of those studies from the literature that followed these criteria. Protopterus aethiopicus exhibited a strong, positive increase in the resting rate of metabolism (RRM) with an increase in temperature across an ecologically relevant temperature range. To our knowledge, this study is the first to quantify the relationship between temperature and the resting rate of metabolism for a non-teleost fish, and the first study to examine this relationship in an air-breathing fish. Previous studies that examined the effect of temperature on metabolic rate in fishes are limited and in most instances focused on deviations from Kroghs normal curve. Clarke and Johnston (1999) sought to refine this relationship for teleost fishes utilizing literature-derived data that met a specific set of criteria, similar to those listed above (i.e. accurate determination of RRM, adequate acclimation time, etc. See Clarke and Johnston 1999 for full description of criteria). Because they focused on describing the effects of temperature as a general relationship among as many species as possible, they included measurements made on the same species from different studies at different temperatures. Using an Arrhenius model to compare resting metabolic rates adjusted for a

PAGE 53

42 50 g fish across a temperature range, they found that tropical fish require 6.2 times more oxygen at 30 C than do polar fish at 0 C (Clarke and Johnston 1999). This relationship explained 59% of the variance among 69 species of teleosts, with the remaining variance concluded to be the combined effect of phylogeny, ecology, and/or lifestyle (Clarke and Johnston 1999). Clarke and Johnston (1999) also examined taxonomic variation within this relationship. Using the same species data pooled by taxonomic order, they detected no difference in slopes of the relationship between temperature and RRM among orders. Clarke and Johnston (1999) also derived a Q 10 value of 1.83 from their curve based on 69 species of teleosts, and reported a median Q 10 value of 2.40 from 14 published values on individual species that were exposed to different temperature regimes. However, they did not explicitly explore Q 10 values for individual species in different thermal habitats. Studies that measure the resting rate of metabolism within the same species across a gradient of temperatures directly examine the effects of temperature upon metabolism by attempting to remove compensatory short-term mechanisms that occur in non-acclimated organisms. It has been shown that animals that are exposed to abrupt temperature changes without adequate acclimation, respond metabolically in a fashion that tends to inflate Q 10 values when compared with acclimated animals (Holeton 1974). Thus, an accurate determination of Q 10 values on temperature acclimated fish provides a means to compare different species from different environmental temperature regimes. Studies of Q 10 values and comparisons of these values across species are useful in illustrating (a) if the rate of change follows the standard derived relationships calculated from interspecific data, and (b) whether the rate of change varies among species, or groups of ecologically or phylogenetically related taxa. We found 10 published studies

PAGE 54

43 that met our criteria for comparing effects of temperature on the resting rate of metabolism within the same species (Table 3-1). Eight of these studies involved fishes that occur in temperate zones, while this study and the remaining two studies examined tropical species. Mean masses were available from many of these published studies (tropical n=3 and temperate n=8). We tested the effect of mass as a covariate against Q 10 by running an ANCOVA. The effect of mass was not significant (p=0.170). Therefore we removed mass from our model and used a t-test to test for a difference between the Q 10 values between the two groups. Removal of the unusually high Q 10 value for Anguilla rostrata leads to a mean Q 10 value for temperate species of 2.43, a value significantly lower than the mean value for the three tropical species (Q 10 =2.95, t=3.85, p=0.027). When the A. rostrata value is included, the mean Q 10 for the tropical fish group does not differ from the temperate group (mean Q 10 for temperature species=2.54, t=1.39, p=0.191). Sample size was insufficient to perform a phylogenetic contrast analysis. However, both tropical and temperate species represent a broad range of lineages. Assuming that the A. rostrata value is not representative of the group, we can speculate on the apparent differences in Q 10 values between the sample of temperate and tropical species available. It may reflect the lower degree of thermal variability that is experienced throughout the year in tropical species. For P. aethiopicus from Lake Nabugabo, daytime water temperatures from the swamp where juveniles are exclusively found range between 21.3C and 27.2C (Chapman et al. 2002, based upon 1 year of monthly samples). This is far smaller than the ranges of temperatures experienced seasonally in temperate climates. It is possible that temperate species might possess a greater ability to maintain resting metabolism across their natural temperature gradient.

PAGE 55

44 Table 3-1. Q 10 values calculated for various species of teleost fishes. Resting rates of metabolism were provided by Andrew Clarke (British Antarctic Survey) adjusted to a 47 g fish. Q 10 values were calculated across temperature ranges listed. Citations indicate initial source of raw metabolic data. All studies included measured resting rate of metabolism (mg O 2 l -1 hr -1 ) and acclimated individuals for sufficient periods of time prior to measurement at different temperatures. Means are for tropical and temperate Q 10 values and are significantly different (p=0.027). *In calculating the mean Q 10 for temperate species, Anguilla rostrata was not included as described in the text. Species Temp range (C) Q 10 Mean Study Tropical Colossoma macropomum 20-30 2.97 Saint-Paul et al. 1988 Oreochromis mossambicas 19-28 2.74 Caulton 1978 Protopterus aethiopicus 20-30 3.14 this study 2.95 Temperate Scophthalmus maximus 6.0-22.0 3.06 Mallekh and Lagardere 2002 Gambusia affinis 10.0-30 2.24 Cech et al. 1985 Anguilla anguilla 20.0-27.0 2.48 Degani et al. 1989 Anguilla rostrata 15.0-25.0 3.67 Degani & Gallagher 1985 Cyprinus carpio 10.0-30.0 2.60 Beamish 1964 Carassius auratus 10.0-30.0 2.29 Beamish and Mookherji 1964 Catostomus commersonii 10.0-30.0 2.44 Beamish 1964 Oncorhynchus nerka 5.0-20.0 2.02 Brett & Glass 1973 Salvelinus fontinalis 10.0-20.0 2.84 Beamish 1964 Pleuronectes platessa 2.0-22.0 2.06 Fonds et al. 1992 Platichthys flesus 2.0-22.0 2.24 Fonds et al. 1992 2.43* Comparing our derived Q 10 values for both temperate and tropical species to the Q 10 value of 1.83 derived by Clarke and Johnston (1999) for all teleosts (whether polar, temperate or tropical), highlights the problems associated with using published Q 10 values to adjust metabolic rates in the absence of species-specific values. It is a common practice to adjust metabolic rates using Winbergs Q 10 coefficients derived from temperate species based on Kroghs normal curve, and it has been argued that a Q 10 of 2

PAGE 56

45 is an appropriate adjustment for this relationship among diverse species (Fry and Hochachka 1970, Miller and Mann 1973). The derived Q 10 values presented here clearly show how a generalized application of between-species Q 10 values can lead to erroneous determinations when adjusting metabolic rates. Furthermore, while within-species Q 10 values show variation, a lack of appropriate data for a diverse array of fishes confounds understanding the source of this variation. Although the number of fish species for which Q 10 data are available and their phylogenetic diversity is small, this comparison highlights the need for further examination of phylogenetic and geographic trends in Q 10 data. With the exception of this study, all previous data represent teleost fishes. Preliminary examination of resting metabolic rates in non-teleost fishes suggests a low resting rate of metabolism to be a primitive character (A. Seifert, unpublished data). It will be of interest to investigate if a lower thermal sensitivity in terms of Q 10 is correlated with taxonomic position or if it is primarily a function of habitat. In addition, future work that examines effects of temperature outside of the range experienced in nature will help to address whether metabolic compensation approaches the higher and lower bounds of thermal tolerance. A few studies suggest that there is metabolic depression above some critical temperature, but mechanisms for these observations are not well understood (Saint-Paul 1983). Further investigations will shed light on the contributions of protein turnover, membrane homeostasis, and cellular ion balance to metabolism. Whereas these processes presumably play an important role, it is unclear what the individual contributions of these processes are to maintenance metabolism or their relative relationship to temperature and lifestyle (Clarke and Johnston 1999). A more complete picture of the relationship

PAGE 57

46 between temperature and its effect on resting metabolic rate across all groups of fishes will yield a deeper understanding of the effect of environmental factors upon physiological processes in ectotherms.

PAGE 58

CHAPTER 4 SUMMARY AND CONCLUSIONS This study quantified respiratory allocation and the resting rate of metabolism (RRM) in the African lungfish, Protopterus aethiopicus, under ecologically relevant levels of dissolved oxygen. Simultaneous measurements of oxygen consumed from both air and water were used to measure RRM of juvenile lungfish across the size range typically found in hypoxic swamps of the Lake Victoria basin. I also measured the effects of temperature on the resting rate of metabolism and compared calculated Q 10 values to those calculated for temperate and tropical fishes. The following section summarizes the major findings of this study. 1. Aerial oxygen consumption measured at hypoxic DO 2 levels reflective of the natural habitat averaged 98% (range = 94% to 100%) of total respiratory allocation for juvenile lungfish between 2 and 221 g (including externally-gilled individuals). This finding contradicts the long-held belief that small juvenile lungfish primarily depend on aquatic respiration to meet their aerobic metabolic demand, but is supported by ecological data that suggest individuals < 40 cm TL primarily are found in hypoxic wetlands. 2. Total metabolic rate was positively correlated with body mass with a scaling exponent of 0.78. 3. Measurements of oxygen consumption made across a gradient of dissolved oxygen from normoxia to anoxia showed that P. aethiopicus generally maintains its RRM despite a change in respiratory allocation between water and air. 4. The relationship between aquatic gas exchange and DO 2 is described by a curvilinear relationship with the degree of curvilinearity increasing as a function of body mass. There is as yet no explanation for this phenomenon. 5. In the process of evaluating the RRM in P. aethiopicus, two methodological issues were discovered that influence metabolic rate measurements in this species: chamber size and acclimation time. Both a change in chamber size with body mass and the short acclimation time used in the field led to differences between our field 47

PAGE 59

48 lab and laboratory measurements of RRM taken at UF in P. aethiopicus. A decrease in chamber size led to an increase in RRM. An increase in acclimation time from 3 h to 16 h led to a two-fold decrease in RRM. 6. Protopterus aethiopicus exhibited a strong, positive increase in the resting rate of metabolism (RRM) with an increase in temperature across an ecologically-relevant temperature range. The Q 10 value for the temperature interval of 20C and 30C was 3.14. To our knowledge, this study is the first to quantify the relationship between temperature and the resting rate of metabolism for a primitive, non-teleost fish, and the first study to examine this relationship in an air-breathing fish. 7. A comparison of literature-derived Q 10 values for both tropical and temperate fishes indicated a higher Q 10 in tropical species (2.95, n=3) than in temperate species (2.43, n=10) across an ecologically relevant thermal range. This difference may reflect the lower degree of thermal variability that is experienced throughout the year in tropical species.

PAGE 60

LIST OF REFERENCES Abe A.S., and J.F. Steffensen. 1996. Lung and cutaneous respiration in awake and estivating South American lungfish, Lepidosiren paradoxa. Revista Brasileira de Biologia 56:485-489. Babiker M.M. 1979. Respiratory behaviour, oxygen consumption and relative dependence on aerial respiration in the African lungfish (Protopterus annectens, Owen) and an air-breathing teleost (Clarias lazera, C.). Hydrobiologia 65:177-187. Beamish, F.W.H. 1964. Respiration of fishes with special emphasis on standard oxygen consumption. II. Influence of weight and temperature on respiration of several species. Canadian Journal of Zoology 42:177-188. Beamish, F.W.H., and P.S. Mookherjii. 1964. Respiration of fishes with special emphasis on standard oxygen consumption. I. Influence of weight and temperature on respiration of goldfish, Carassius auratus L. Canadian Journal of Zoology 42:161-175. Brett, J.R., and N.R. Glass. 1973. Metabolic rates and critical swimming speeds of sockeye salmon (Oncorhynchus nerka) in relation to size and temperature. Journal of the Fisheries Research Board of Canada 30:379-387. Burggren W.W., and K. Johansen. 1987. Circulation and respiration in lungfishes (Dipnoi). Pp. 217-236 in W.E. Bemis, W.W. Burggren, and N.E. Kemp, eds. The Biology and Evolution of Lungfishes. Alan R. Liss, Inc., New York. Burggren W.W., K. Johansen, and B. McMahon. 1986. Respiration in phyletically ancient fishes. Pp. 217 in R.E. Foreman, ed. Evolutionary Biology of Primitive Fish. Plenum Press, New York. Caulton, M.S. 1978. The effect of temperature and mass on routine metabolism in Sarotherodon (Tilapia) mossambicus (Peters). Journal of Fish Biology 13:195-201. Cech, L.J. Jr., Massingill, M.J., Vondracek, B., and A.L. Linden. 1985. Respiratory metabolism of the mosquitofish, Gambusia affinis: effects of temperature, dissolved oxygen, and sex difference. Environmental Biology of Fishes 13:297-307. Clarke, A. 1991. What is cold adaptation and how should we measure it? American Zoologist 31:81-92. 49

PAGE 61

50 Clarke, A. 1993. Seasonal acclimatization and latitudinal compensation in metabolism: do they exist? Functional Ecology 7:139-149. Clarke, A., and N.M. Johnston. 1999. Scaling of metabolic rate with body mass and temperature in teleost fish. Journal of Animal Ecology 68:893-905. Chapman L.J., F.G. Nordlie, and A.W. Seifert. 2002. Respiratory oxygen consumption among groups of Pseudocrenilabrus multicolor victoriae subjected to different oxygen concentrations during development. Journal of Fish Biology 61:242-251. Coates M.I., and J.A. Clack. 1991. Fish-like gills and the breathing in the earliest known tetrapod. Nature 352:234-236. Degani, G., and M.L. Gallagher. 1985. The relationship between growth, food consumption and oxygen consumption in the developed and underdeveloped American eels, Anguilla rostrata Lesueur. Journal of Fish Biology 27:635-641. Degani, G., Gallagher, M.L., and A. Meltzer. 1989. The influence of body size and temperature on the oxygen consumption of the European eel, Anguilla anguilla. Journal of Fish Biology 34:19-24. Ege, R., and A. Krogh. 1914. On the relation between the temperature and the respiratory exchange in fishes. Internationale Revue der Gesamten Hydrobiologie und Hydrographie 1:48-55. Farber J. and H. Rahn. 1970. Gas exchange between air and water and the ventilation pattern in the electric eel. Respiration Physiology 9:151-161. Fonds, M., Cronie, R., Vethaak, A. D., Vanderpuyl, P. 1992. Metabolism, food consumption and growth of Plaice (Pleuronectes platessa) and flounder (Platichthys flesus) in relation to fish size and temperature. Netherlands Journal of Sea Research 29: 127-143. Fry, F. E. J., and Hochachka, P.W. 1970. Fish. Pp. 79-134 in G.C. Whittow, ed. Comparative Physiology of Thermoregulation. Vol. I. Invertebrates and Nonmammalian Vertebrates. Academic Press, New York. Gardiner B. 1980. Tetrapod ancestry: A reappraisal. Pp. 177-185 in A.L. Panchen, ed. The Terrestrial Environment. Academic Press, London. Gordon M.S., and E.C. Olson. 1994. The Invasions of the Land. Columbia University Press, New York. Goudswaard K.P.C., F. Witte, and L.J. Chapman. 2001. Decline of the African lungfish (Protopterus aethiopicus) in Lake Victoria (East Africa). African Journal of Ecology 39:1-11.

PAGE 62

51 Graham J.B. 1997. Air Breathing Fishes: Evolution, Diversity and Adaptation. Academic Press, San Diego, California. Greenwood P.H. 1958. Reproduction in the East African lungfish Protopterus aethiopicus Heckel. Proceedings of the Zoological Society of London 130:547-567. Greenwood P.H. 1965. The cichlid fishes of Lake Nabugabo, Uganda. Bulletin of the Museum of Natural History (Zoology) 12:315-357. Greenwood P.H. 1987. The natural history of African lungfishes. Pp.163-179 in W.E. Bemis, W.W. Burggren, and N.E. Kemp, eds. The Biology and Evolution of Lungfishes. Alan R. Liss, Inc., New York. Grigg G.C. 1965. Studies on the Queensland lungfish Neoceratodus forsteri (Krefft). III. Aerial respiration in relation to habitats. Australian Journal of Zoology 13:413-421. Holeton, G.F. 1974. Metabolic adaptation of polar fish: fact or artefact? Physiological Zoology 47:137-152. Jesse M.J., C. Shub, and A.P. Fishman. 1967. Lung and gill ventilation of the African lungfish. Respiration Physiology 3:267-287. Johansen K., and C. Lenfant. 1967. Respiratory function in the South American lungfish, Lepidosiren paradox (Fitz). Journal of Experimental Biology 46:205-218. Johansen K., and C. Lenfant. 1968. Respiration in the African lungfish Protopterus aethiopicus. II. Control of breathing. Journal of Experimental Biology 49:453-468. Johansen K., J.P. Lomholt, and G.M.O. Maloiy. 1976. Importance of air and water breathing in relation to size of the African lungfish Protopterus amphibius (Peters). Journal of Experimental Biology 65:395-399. Johnels A.G., and G.S.O. Svensson. 1954. On the biology of Protopterus annectens (Owen). Arkiva Zoology 7:131-158. Johnston, I.A., Clarke, A., and P. Ward. 1991. Temperature and metabolic rate in sedentary fish from the Antarctic, North Sea and Indo-West Pacific Ocean. Marine Biology 109:191-195. Krihak M.K., and M.R. Shahriari. 1996. Oxygen sensor based on the sol-gel coating technique. Electronics Letters 32:3. Krogh, A. 1914. The quantitative relation between temperature and standard metabolism in animals. Internationale Zeitschrift fur Physikalisch-Chemische Biologie 1:491-508. Krogh, A. 1916. The Respiratory Exchange of Animals and Man. Longmans, London.

PAGE 63

52 Lahiri S., J.P. Szidon, and A.P. Fishman. 1970. Potential respiratory and circulatory adjustments to hypoxia in the African lungfish. Federation Proceedings 29:1141-1148. Lenfant C., and K. Johansen. 1968. Respiration in the African lungfish Protopterus aethiopicus. I. Respiratory properties of blood and normal patterns of breathing and gas exchange. Journal of Experimental Biology 49:437-452. Lenfant C., K. Johansen, and D. Hanson. 1970. Bimodal gas exchange and ventilation-perfusion relationship in lower vertebrates. Federation Proceedings 29:1124-1129. Little C. 1990. The Terrestrial Invasion. Cambridge University Press, Cambridge. Long J. A. 1995. The Rise of Fishes: 500 Million Years of Evolution. Johns Hopkins University Press, Baltimore. Mallekh, R., and J.P. Lagardere. 2002. Effect of temperature and dissolved oxygen concentration on the metabolic rate of the turbot and the relationship between metabolic scope and the feeding demand. Journal of Fish Biology 60:1105-1115. Marshall, C. R. 1987. Lungfish: Phylogeny and Parsimony. Pp.151-162 in W.E. Bemis, W.W. Burggren, and N.E. Kemp, eds. The Biology and Evolution of Lungfishes. Alan R. Liss, Inc., New York. McMahon B. R. 1970. The relative efficiency of gaseous exchange across the lungs and gills of an African lungfish Protopterus aethiopicus. Journal of Experimental Biology 52:1-15. McNab B.K. 2002. The Physiological Ecology of Vertebrates. Cornell University Press, London. Miller, R.J., and K. H. Mann. 1973. Ecological energetics of the seaweed zone in a marine bay on the Atlantic coast of Canada. III. Energy transformations by sea urchins. Marine Biology 18:99-114. Oduleye S.O. 1977. Unidirectional water and sodium fluxes and respiratory metabolism in the African lungfish, Protopterus annectens. Journal of Comparative Physiology 119:127-139. Ogutu-Ohwayo R. 1993. The effect of predation by Nile Perch, Lates niloticus L., on the fish of Lake Nabugabo, with suggestions for conservation of endangered endemic cichlids. Conservation Biology 7:701-711. Potvin, C., M. J. Lechowicz and S. Tardif. 1990. The statistical analysis of ecophysiological response curves obtained from experiments involving repeated measures. Ecology 7:1389-1400.

PAGE 64

53 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. Sacca R. and W. Burggren. 1982. Oxygen uptake in air and water in the air-breathing reedfish Calamoichthys calabaricus: Role of skin, gills and lungs. Journal of Experimental Biology 97:179-186. Saint-Paul, U. 1983. Investigations on the respiration of the Neotropical fish, Colossoma macropomum (Serrasalmidae). The influence of weight and temperature on the routine oxygen consumption. Amazonia 4:433-443. Sanchez A., and Glass, M.L. 2001. Effects of environmental hypercapnia on pulmonary ventilation of the Southern American lungfish, Lepidosiren paradoxa (Fitz). Journal of Fish Biology 58: 1181-1189. Sanchez A., R. Sonncini, T. Wang, P. Koldkjaer, E.W. Taylor, and M.L. Glass. 2001. The differential cardio-respiratory responses to ambient and systemic hypoxaemia in the South American lungfish, Lepidosiren paradoxa. Comparative Biochemistry and Physiology Part A 130: 677-687. Sawaya P. 1946. Sobr a biologia de alguns peixes de respiraco aerea (Lepidosiren paradoxa Fitzinger e Arapaima gigas Cuvier). Boletim da Faculdade de Filosofia, Ciencias e Letras da Universidade de So Paulo. Geologia (Zool.) 11:255-285. Singh B.N., and G.M. Hughes.1971. Respiration of an air-breathing catfish Clarias batrachus (Linn.). Journal of Experimental Biology 55:421-434. Smith H.W. 1930. Metabolism of the lungfish Protopterus aethiopicus. Journal of Biological Chemistry 88:97-130. Smith H.W. 1931. Observations on the African lungfish, Protopterus aethiopicus, and the evolution from water to land environments. Ecology 12:164-181. Ultsch G.R. 1995. On adjusting metabolic rates for body size. Florida Scientist. 58: 270-273. Ultsch G.R. 1996. Gas exchange, hypercarbia and acid-base balance, paleoecology, and the evolutionary transition from water-breathing to air-breathing among vertebrates. Palaeogeography, Palaeoclimatology, Palaeoecology 123:1-27. Worthington E.B. 1932. Scientific results of the Cambridge Expedition to the East African lakes, 1930-1. Fishes other than Cichlidae. Journal of the Linnean Society (Zool.) 38:121-134.

PAGE 65

BIOGRAPHICAL SKETCH Ashley Winn Seifert was born on April 6 th 1976, in Syosset, New York. He is one of three children of Randolph Winn Seifert and Janine Epstein Seifert. Raised in a liberal-minded atmosphere, he was taught to search for his own answers to the mysteries of life. Following a whim, he attended Bowdoin College in Brunswick, Maine, where he earned an A.B. in biology with a minor in philosophy. After graduating, his interests led to a now coincidental meeting and consequent friendship with Dr. Karel Liem who suggested an avenue for potential study with Dr. Lauren Chapman. After working as her research assistant for one year he enrolled in the graduate program in the Department of Zoology at the University of Florida where he completed his Master of Science degree. He is still seeking answers and still following whims. 54


Permanent Link: http://ufdc.ufl.edu/UFE0004562/00001

Material Information

Title: Respiratory Allocation and the Resting Rate of Metabolism in the African Lungfish Protopterus aethiopicus
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0004562:00001

Permanent Link: http://ufdc.ufl.edu/UFE0004562/00001

Material Information

Title: Respiratory Allocation and the Resting Rate of Metabolism in the African Lungfish Protopterus aethiopicus
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0004562:00001


This item has the following downloads:


Full Text












RESPIRATORY ALLOCATION AND THE RESTING RATE
OF METABOLISM IN THE AFRICAN LUNGFISH
Protopterus aethiopicus















By

ASHLEY W. SEIFERT


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2004

































Copyright 2004

by

Ashley W. Seifert




























I dedicate this thesis to my parents and my family. More than anyone else, their
undying encouragement has helped keep me sane. Their love and belief in my potential
pushed me to believe in myself.
I also dedicate this work to my grandparents, Evelyn and Nathan Epstein and
Louise and Randolph Seifert. While Lou has had the opportunity to share in my triumphs
along the way, I know that the others are smiling from somewhere in the cycle of life. I
only hope that one day I can fulfill the destiny that all of you have seen for me. It is often
I recall your memories and they keep me smiling one day at a time.















ACKNOWLEDGMENTS

I am sure that this thesis would not have been possible without the support and

commitment of my advisor, Dr. Lauren Chapman. Her knowledge and guidance as both a

scientist and mentor have been truly inspiring. Her passion towards science served as a

daily reminder of what it means to truly love one's work. It is a lasting impression that

will stay with me forever. I would also like to thank the other members of my committee,

Dr. Brian McNab, Dr. Steven Walsh, and Dr. James Albert. I appreciate their sacrifice of

both time and energy to serve on my committee. The combined wisdom of these men has

helped me shape both my science and myself. My conversations with each of them

challenged me to rigorously question my work, and have deeply affected my application

of the scientific method. Lastly, I would like to thank Dr. Frank Nordlie. From the very

beginning of my tenure here at the University of Florida he has served as both scientific

advisor and sage. Our trips to Cedar Key and our time together in Africa are burned in the

annuls of my memory. Perhaps more than anyone, his carefree being reminded me that I

might be in the right place after all.

In addition, I would also like to thank all those at Lake Nabugabo, Uganda, for

their assistance and friendship. Kiberu Jackson and Mutebi Jackson, in particular, treated

me as a brother. Together we bridged cultural boundaries that ingrained in me the belief

that we are all brothers and sisters on this tiny blue sphere. I am indebted to Dr. Colin

Chapman for his patience in entertaining my seemingly endless questions about anything

and everything. How we seemed to have free spots of time that always overlapped is still









a mystery. Of course, it goes without saying that his passion for science should stand as a

pillar for us all. I would like to thank Mac Kobza, Tom Gillespie, Chrissy Apodaca and

Nat Seavy for making my time in Kibale National Park both enjoyable and a wee bit

nerve-racking.

Last, I would like to acknowledge the support of Sigma Xi and the Raney Fund for

contributing funds to help support my research. This work was also supported by the

National Science Foundation (IBN 0094393), the Wildlife Conservation Society, and

Ocean Optics support to Dr. Lauren J. Chapman
















TABLE OF CONTENTS

page

ACKN OW LED GM EN T S ..................................................... iv

LIST OF TABLES .................... .......... ......... ................ viii

LIST OF FIGURES ........................ ........................... ix

ABSTRAC T ............. ............ ................. ........ ........x

CHAPTER

1 GENERAL INTRODUCTION ....................................................... 1

2 RESPIRATORY ALLOCATION AND THE RESTING RATE OF METABOLISM
IN Protopterus aethiopicus.............................................7

Introduction ...................................... .......... .......... ...7
M materials an d M eth o d s .......................................................................................... 10
Study Site................................................ ........10
Collection and M maintenance of P. aethiopicus ........................ ..................12
Respirometry ........................................ .........13
Results................... .............................................. ................ ........ 18
Field Laboratory Data.............. .... .. ................. 18
UF Laboratory Data............. ... .................. 18
Chamber Effect...................................... .......................................23
Resting rate of m etabolism ............................................................. ... ....24
Discussion......................................................... ......................... 26
M etabolic Profile ............... ............ .. .........................28
M ethodological Issues ................... ................................ ........ .... .3 1

3 THE RELATIONSHIP BETWEEN TEMPERATURE AND THE RESTING RATE
OF METABOLISM IN Protopterus aethiopicus............... ...............35

Introduction......... ......... .... ......... .... ....... ............... 35
M materials and M methods ............................................................37
Statistics ............. ......... .... ......... ... ...... ......... 38
R results ...................... .... ........... .............................. 39
Discussion .............. ................ .................. 41









4 SUMMARY AND CONCLUSIONS................................................................... 47

L IST O F R E FE R E N C E S ............................................................................. 49

BIOGRAPHICAL SKETCH .................................................. ............... 54















LIST OF TABLES


Table page

2-1 Respiratory allocation of Protopterus aethiopicus from the Lwamunda Swamp,
Uganda, between water and air at high levels...................................................19

2-2 Respiratory allocation of Protopterus aethiopicus from the Lwamunda Swamp,
Uganda, from air at the aquatic respiration plateau (representing the level of DO2 at
which individuals respired almost completely aerially)................ ............ 23

3-1 Qio values calculated for various species of teleost fishes...............................44















LIST OF FIGURES


Figure page

2-1 Lwamunda Swamp, Lake Nabugabo, Uganda. Red stars indicate two collection
sites where specimens of Protopterus aethiopicus were captured.......................... 11

2-2 Respirometry system used to measure the resting rate of metabolism. The entire
system is submerged in a water bath to maintain temperature. The system is
connected to a labtop that records the data........................................................14

2-3 Bilogarithmic relationship between resting rate of metabolism (RRM) and body
mass of Protopterus aethiopicus from the Lwamunda Swamp, Uganda, measured at
high levels of D O 2 in the field ............................................................................ 20

2-4 The decline in oxygen partial pressure over time for Protopterus aethiopicus from
the Lwamunda Swamp, Uganda, in a two-phase, closed respirometer, where fish
had access to both air and water simultaneously.................................21

2-5 The relationship between the degree of curvilinearity (bl from the equation
y=bieb0*t) and body mass (g) for Protopterus aethiopicus from the Lwamunda
Swamp, Uganda. ................................................. ........ 22

2-6 The bilogarithmic relationship between RRM (mg 02 hf1) and body mass (g) for
the African lungfish, Protopterus aethiopicus, collected from the Lwamunda
Swamp, Uganda. Data represent RRM at low (ecologically relevant) DO2 levels ..25

2-7 The bilogarithmic relationship between resting rate of metabolism (RRM) (mg 02
hr-) and body mass (g) of Protopterus aethiopicus from the Lwamunda Swamp,
Uganda, measured at levels of DO2 prior to the plateau in aquatic respiration .......27

3-1 Bilogarithmic relationship between resting metabolic rate (RRM) and body mass
for six individuals of Protopterus aethiopicus from the Lwamunda Swamp,
Uganda, measured at three different temperatures: 30 'C, 25 'C, and 20 C. .........40

3-2 Mean resting rate of metabolism adjusted to 110.5 g fish (mean for six individuals)
for three temperatures: 30 'C, 25 'C, and 20 C. ..................................... 40
















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

RESPIRATORY ALLOCATION AND THE RESTING RATE
OF METABOLISM IN THE AFRICAN LUNGFISH
Protopterus aethiopicus

By

Ashley W. Seifert

May 2004

Chair: Lauren J. Chapman
Major Department: Zoology

Air breathing in fishes has a very rich evolutionary history, having evolved

independently many times in several different phylogenetic lineages. The ancient air-

breathing lungfishes are of particular interest because of their proximate relationship to

the ancestor of tetrapods. The use of lungs as the primary site of oxygen exchange and

the seemingly increased physiological dependence upon aerial respiration across

ontogeny provide an excellent opportunity to study the physiology of these fishes along

an evolutionary vector similar to that which the tetrapods might have experienced as they

began to invade terrestrial environments. However, there remain some key gaps in our

understanding of the respiratory profile of lungfishes, in particular respiratory allocation,

and metabolic demand of juveniles in the context of ecologically relevant aquatic oxygen

levels.

This objectives of this study were to (1) investigate respiratory partitioning and

quantify the resting rate of metabolism among juvenile Protopterus aethiopicus that vary









in body mass across a dissolved oxygen (DO2) range typical of that found within hypoxic

wetlands and (2) quantify the effects of temperature on resting rates of metabolism of

juvenile P. aethiopicus through calculation of Qio values across a range of temperatures

representative of those found within their natural habitat.

Chapter 1 quantifies the relationship between respiratory allocation (air versus

water) and the resting rate of metabolism (RRM) in the primitive air-breathing lungfish,

P. aethiopicus. Simultaneous measurements of oxygen consumed from both air and water

were made to determine the RRM at ecologically relevant levels of dissolved oxygen for

juveniles 2 to 221 g. Aerial oxygen consumption averaged 98% (range = 94% to 100%)

of total respiratory allocation. This finding contradicts the long-held belief that small

juvenile lungfish primarily depend on aquatic respiration to meet their aerobic metabolic

demand. Measurements of oxygen consumption made across a gradient of dissolved

oxygen from normoxia to anoxia showed that P. aethiopicus maintains a stable RRM

despite a change in respiratory allocation between water and air.

Chapter 2 examines the effect of temperature on the resting rate of metabolism in

P. aethiopicus. We present a Qio value of 3.14 for six individuals ranging in body mass

from 42 to 222 g across an ecologically relevant 10 'C temperature range. This is the first

study to our knowledge to determine a Qio value for a non-teleost fish and the first to

describe this relationship for an air-breathing fish. A comparison of literature-derived Qio

values for both tropical and temperate fishes indicated a higher Qio in tropical species

(n=3) than in temperate species (n=10) across an ecologically relevant thermal range.














CHAPTER 1
GENERAL INTRODUCTION

The transition from water to land stands as a pivotal stage in vertebrate evolution.

Any study of this transitional period undoubtedly alludes to the importance of the

atmosphere as the new prime source of metabolic oxygen and fishes were the first

vertebrates to rise to the surface and begin to break free of their bonds to aquatic

existence.

Air breathing in fishes has a rich evolutionary history, having evolved

independently numerous times in several different phylogenetic lineages and with various

structures for gas exchange (Randall et al. 1981; Graham 1997). Air-breathing fishes are

thought to have had their origins in the Late Silurian, based on fossil traces of lung-like

structures in placoderms and ostracoderms (Gardiner 1980; Little 1990, Coates and Clack

1991; Gordon and Olson 1994; Long 1995; Graham 1997). The early evolution of lungs

and their proposed widespread occurrence among ancient fishes are also strengthened by

the observation that extant representatives of mostly extinct groups are bimodal breathers,

who use either paired lungs (ancestral condition) or a single lung as the primary air-

breathing organ (Randall et al. 1981; Burggren et al. 1986; Ultsch 1996). The phyletically

ancient air breathers are of particular interest because of their proximate relationship to

the ancestor of tetrapods. Their use of lungs as the primary site of oxygen exchange and

their seemingly increased physiological dependence upon aerial respiration across

ontogeny provide an excellent opportunity to study the physiology of these ancient fishes









along an evolutionary vector similar to that which the tetrapods might have experienced

as they began to invade the land.

Surely the most infamous of air-breathing fishes are the lungfishes. Aside from the

discovery of the coelacanth, Latimeria chalumnae, in the early part of the 19th century,

the near simultaneous discovery of both Lepidosirenparadoxa and Protopterus

annectens in the 1830's and the discovery of Neoceratodusforsteri in 1870 stand as three

of the most important zoological discoveries of the last two centuries (Burggren and

Johansen 1987). Their exquisite fossil record provides a good picture of the interspecific

relationships of both extinct and extant forms. However, ever since their discovery, their

taxonomic position in relation to the tetrapods has remained a source of rich scientific

debate. Lungfishes belong to the Ceratodontimorpha and, along with Latimeria (the

coelacanth), remain the only extant representatives of the Sarcopterygii, a group from

which tetrapods later emerged. Within the Ceratodontimorpha, the Lepidosireniformes,

comprised of the African family Protoperidae (four species) and the South American

family Lepidosirenidae (one species), appear largely unchanged from the ancestral

Dipterus of the Carboniferous period (Long 1995; Graham 1997). Neoceratodusforsteri

is the lone species representative of the Ceratodontidae and is considered the most

primitive of the three genera based on several phylogenetic characters (Marshall 1987).

Of the three genera of lungfishes, two are obligate air-breathers, while the Australian

species, Neoceratodusforsteri, uses aerial respiration as a supplementary means of gas

exchange (Grigg 1965; Graham 1997). Fossil evidence and their relative morphological

stability through evolutionary time allow for the conjecture that modem day









Lepidosireniformes may be physiologically similar to a tetrapod ancestor (Burggren et al.

1986).

Are these interesting creatures the closest extant relative of the tetrapods?

Regardless of when or whether scientists will come to answer this question, the fact

remains that lungfish represent a very special organism for evolutionary biologists

working from an ecological and physiological perspective. The respiratory and

cardiovascular similarities that both Lepidosiren paradox and Protopterus species share

with terrestrial amphibians make them interesting transitional species to investigate these

systems in the context of vertebrate evolution.

Beginning with Homer Smith in the 1930's, the majority of early research on

Protopterus aethiopicus, and lungfishes in general, centered on cardio-respiratory

adaptations and the physiological effects of aestivation. Given that P. aethiopicus shifts

from having strongly developed external gills as small juveniles to poorly developed

internal gills and well-developed paired lungs as adults, researchers hypothesized a strong

shift in respiratory allocation between air and water. Indeed, this is precisely what early

research uncovered. A comparison of early studies that measured oxygen consumption in

lungfishes yields a profile whereby juveniles use primarily aquatic respiration to meet

metabolic demands, and adults rely almost exclusively on aerial respiration (Smith 1930;

Sawaya 1946; Johansen and Lenfant 1967; Lenfant and Johansen 1968; Lenfant et al.

1970; McMahon 1970; Johansen et al. 1976; Babiker 1979). Most of these investigations

centered exclusively on either juveniles or adults. However, a study done by Johansen et

al. (1976) examined changes in respiratory allocation between water and air across a wide

range in body size from 4 g to 500 g. Their experiments established a two-phase









respiratory profile for P. amphibious based on these experiments where the smallest

juveniles relied on dissolved oxygen to meet metabolic demands, while larger animals

acquired their oxygen aerially. Based on their findings they concluded that "the altered

dependence on water and air-breathing during the life-cycle of P. amphibious will most

likely set territorial limits to the distribution of larvae and young fish within the

distribution range of the species" (Johansen et al. 1976. p. 399). This statement is striking

in that it predicted the opposite of what is known of juvenile and adult African lungfish

based on dissolved oxygen levels within their natural habitat.

A careful consideration of the ecology of juvenile lungfish highlights this

interesting paradox. Juvenile species of African lungfish are born within nests that are

found within swamp regions surrounding larger bodies of water or in portions of rivers

where current is slow. Measurements of dissolved oxygen concentrations made by

Greenwood (1958) at the surface of swamps surrounding Lake Victoria ranged from 1.01

to 1.71 mg 1- and within the breeding nests from 0.98 to 2.66 mg 1- Research from the

Lwamunda Swamp surrounding Lake Nabugabo (Uganda) found that juveniles remain

within fringing swamps surrounding the main lake until they reach a body size of

approximately 40 cm TL (Goudswaard et al. 2001). Individuals smaller than 40 cm TL

are rarely found in the open waters of Lake Nabugabo where larger lungfish supported a

thriving fishery before the introduction of Nile Perch in the 1960's. Dissolved oxygen

levels in the Lwamunda Swamp are extremely low, and it is the younger juveniles

(nestlings to juveniles of 40 cm TL) that are found under these hypoxic conditions

(Goudswaard et al. 2001). With these ecological parameters in mind, it is curious that the

youngest juveniles, who seemingly experience the greatest respiratory stress, would rely









mainly upon aquatic respiration, while adults who move into open water habitats would

suddenly transform into animals dependant on aerial respiration. These previous findings

set up an interesting paradox created by the seemingly opposing forces of dissolved

oxygen levels and respiratory physiology of these animals.

If the transition from water to land in early tetrapods was mediated by hypoxic

stress within ancient freshwater habitats, it was of great interest to me to determine if

small, juvenile lungfish were primarily air-breathers and if their respiratory physiology

was determined by strong environmental gradients. Given the ecological characteristics

of Lake Nabugabo with a fringing hypoxic swamp surrounding a main lake of much

higher DO2, and the findings of Johansen et al. (1976), I sought to resolve the above

stated paradox and set out to examine the respiratory allocation of Protopterus

aethiopicus. My objectives were to (1) investigate respiratory partitioning and quantify

the resting rate of metabolism among juvenile Protopterus aethiopicus that vary in body

mass across a dissolved oxygen (DO2) range typical of that found within hypoxic

wetlands and (2) quantify the effects of temperature on resting rates of metabolism of

juvenile P. aethiopicus through calculation of Qio values across a range of temperatures

representative of those found within their natural habitat.

Chapter 2 explores respiratory allocation and the resting rate of metabolism in

juveniles of the African lungfish (Protopterus aethiopicus). The objective of this chapter

is to quantify respiratory allocation between water and air and the rate of oxygen

consumption over the size range typically found in marginal hypoxic wetlands in the

Lake Victoria basin (nestlings to 40 cm TL, Goudswaard et al. 2001). To meet this

objective, the rate of oxygen consumption and relative allocation to air and water









breathing was measured across a broad range of dissolved oxygen concentrations

including those reflective of the natural habitat. I also explored the effect of chamber

acclimation time and chamber size on respiratory allocation and rate of oxygen

consumption.

In Chapter 3, I quantify the relationship between temperature and the resting rate of

metabolism in Protopterus aethiopicus through a temperature acclimation study. Fishes,

as the most species-rich group of chordates, lack a comprehensive synthesis of metabolic

data that includes both phyletically ancient (non-teleost) and advanced groups. In

particular, studies examining metabolic rates and the effects of temperature in non-teleost

fishes are completely absent from the literature. This chapter determined the Qio value for

P. aethiopicus. This Qio value is discussed in relation to those discerned from the

literature for both temperate and tropical teleosts.














CHAPTER 2
RESPIRATORY ALLOCATION AND THE RESTING RATE OF METABOLISM IN
Protopterus aethiopicus

Introduction

Air breathing in fishes has a rich evolutionary history, having evolved

independently numerous times in several different phylogenetic lineages and with various

structures for gas exchange (Randall et al. 1981; Graham 1997). Air-breathing fishes are

thought to have had their origins in the Late Silurian, based on fossil traces of lung-like

structures in placoderms and ostracoderms (Gardiner 1980; Little 1990; Coates and Clack

1991; Gordon and Olson 1994; Long 1995; Graham 1997). The early evolution of lungs

and evidence of their widespread occurrence among ancient fishes is supported by the

observation that extant representatives of mostly extinct groups are bimodal breathers,

using either paired lungs (ancestral condition) or a single lung as the primary air-

breathing organ (Randall et al. 1981; Burggren et al. 1986; Ultsch 1996). Despite air

breathing having evolved independently several times in more derived fishes, primitive

air breathers are of interest because of their proximate relationship to the ancestors of

tetrapods. Their use of lungs as the primary site of oxygen exchange and apparent

increased physiological dependence upon aerial respiration across ontogenetic stages

provides an excellent opportunity to study the physiology of these ancient fishes along an

evolutionary vector similar to that which tetrapods might have initially experienced

during the transition to a terrestrial existence. However, very little is known about

energetic of primitive air-breathers under ecologically relevant oxygen levels and the









relationship between respiratory allocation (water vs. air) and body size. In this study, I

examined the energetic of juvenile African lungfish (Protopterus aethiopicus) across a

range of body sizes to detect ontogenetic shifts in respiratory allocation and to measure

resting rate of metabolism.

Of the three families of lungfishes, the Lepidosirenidae (comprised of one species,

Lepidosiren paradox) and the Protopteridae (comprised of four species of Protopterus)

are obligate air-breathers, whereas the Ceratodontidae (comprised of the Australian

species, Neoceratodusforsteri), uses facultative air breathing as a supplementary means

of gas exchange (Grigg 1965; Graham 1997). Obligate air-breathing lungfishes possess

highly partitioned, paired lungs that originate from the esophagus, have separate

pulmonary circuits, and circulatory systems with a high degree of septation in the heart,

resembling that of extant amphibians (Burggren and Johansen 1987).

The majority of early research on P. aethiopicus, and lungfishes in general, focused

on cardio-respiratory adaptations in larger fish (Sawaya 1946; Johansen and Lenfant

1968; Lahiri et al. 1970; Lenfant et al. 1970; McMahon 1970). The few studies that

explored ontogenetic variation in respiratory strategies reported shifts in respiratory

allocation with body size. For example, Johansen et al. (1976) found that small juveniles

(3.7 g) of Protopterus amphibious relied on aquatic respiration to meet 70% of their

oxygen requirement, whereas aquatic respiration accounted for as little as 10% of the

oxygen uptake in larger specimens (300 g to 500 g). Similarly, Lenfant et al. (1970)

found that aerial respiration accounted for nearly 97% of the total oxygen demand in

large individuals of Lepidosiren paradox, but only 26% for small juveniles. The

combination of these studies created a physiological portrait of the Lepidosirenidae and









Protopteridae as animals with two distinct respiratory life stages (Jesse et al. 1967;

Lenfant et al. 1970; McMahon 1970; Johansen et al. 1976; Babiker 1979). Small, juvenile

lungfish were thought to be primarily dependent on aquatic respiration for survival, while

larger individuals became increasingly dependent on aerial respiration as they grew,

ultimately relying on aerial respiration for over 90% of their metabolic demand.

This scenario, however, seems inconsistent within the ecological context of their

habitats. In the field, nestling P. aethiopicus have been observed to begin air breathing at

a length of 23 to 27 mm while still possessing external gills (Smith 1931; Greenwood

1987). At this size, they are associated with either floating mats of papyrus roots or

amongst heavily vegetated patches within the swamp (Greenwood 1987). In lakes

Victoria and Nabugabo (East Africa), Goudswaard et al. (2001) reported that P.

aethiopicus smaller than 40 cm were very rare in open waters and almost exclusively

associated with the fringing swamp. There, dissolved oxygen concentration can be

extremely low (Chapman et al. 2002). Both Greenwood (1958) and Johnels and Svensson

(1954) suggested that the shallowness of P. aethiopicus nests may be connected with the

air-breathing behavior of the nestlings. In these nests, young have been observed resting

on the roots near the surface (Greenwood 1987). Dissolved oxygen measurements made

by Greenwood (1958) at the surface and at the bottom of the nests ranged between 1.01

and 1.71 mg 1-1, and between 0.98 and 2.66 mg 1-1, respectively. Although the smallest

size of aestivating individuals is unknown, Protopterus dolloi, a West African lungfish

species, has been found in cocoons at lengths of 40-50 mm (Johnels and Svensson 1954).

The fact that individuals this small can aestivate suggests that they are capable of

sustaining themselves solely on aerial respiration. In conditions such as these, and with









the capacity to acquire oxygen aerially, it would appear that small, juvenile lungfish

might use aerial respiration in their natural environment to an extent previously

unrecognized.

In earlier studies, measurements of oxygen consumption in small, juvenile P.

aethiopicus were made under conditions of near-saturated dissolved oxygen, which do

not reflect the observed natural ambient conditions at this life-history stage. The objective

of this study was to quantify respiratory allocation between water and air and the rate of

oxygen consumption in juvenile African lungfish (Protopterus aethiopicus) over the size

range typically found in marginal hypoxic wetlands in the Lake Victoria basin (nestlings

to 40 cm TL, Goudswaard et al. 2001). To meet this objective, we measured rate of

oxygen consumption and relative allocation to air and water breathing across a range of

dissolved oxygen concentrations reflective of the natural habitat. We also explored the

effect of chamber acclimation time and chamber size on respiratory allocation and rate of

oxygen consumption.

Materials and Methods

Study Site

Specimens of the African lungfish, Protopterus aethiopicus were obtained from a

natural population within the Lwamunda Swamp that surrounds Lake Nabugabo, Uganda

(31o 50' E; 31o 56' S; to 0o20' E; 0o25' S, Fig. 2-1). Lake Nabugabo lies just south of the

equator and is a small satellite lake (surface area = 24 km2, mean depth = 4.5 m) that was

isolated from Lake Victoria approximately 4000 years ago (Greenwood 1965). The lake

lies within the extensive Lwamunda Swamp that was formerly a bay on the western shore





































Figure 2-1. Lwamunda Swamp, Lake Nabugabo, Uganda. Red stars indicate two
collection sites where specimens of Protopterus aethiopicus were captured.

of Lake Victoria (Worthington 1932; Greenwood 1965; Ogutu-Ohwayo 1993). The lake

margin is dominated by hippo grass (Vossia cuspidatum), Miscanthidium violaceum,

water lilies (Nymphaea lotus and N. caerulea), and small stands of papyrus (Cyperus

papyrus). The dense interior of the swamp (dominated by M. violaceum) is characterized

by numerous small lagoons and channels that serve as important habitats for P.

aethiopicus (<40 cm total length, Goudswaard et al. 2001). Dissolved oxygen

concentrations (DO2) at several sites within a small swamp lagoon were measured over a

1-year period. Monthly DO2 measurements in the upper 20 cm of water averaged 1.5 mg

1- (&26.8 mm Hg) in the morning and 1.8 mg 1- (&32.9 mm Hg) during the afternoon









(Chapman et al. 2002). Water temperatures at the sites averaged 22.3" C in the morning

and 23.8' C in the early afternoon (Chapman et al. 2002). DO2 values in the Lwamunda

swamp are particularly low in the dry season. In a survey of 12 small lagoons within the

Lwamunda Swamp in the early dry season Chapman et al. (2002) reported an average

DO2 level of only 0.6 mg 1-1 (,11 mm Hg, upper 20 cm of water) in the early morning

and 1.3 mg 1- (&24 mm Hg) in the mid-day period. In samples taken just above the

sediments, DO2 averaged 0.3 mg 1-1 (&5 mm Hg) in the early morning and 0.4 mg 1-1 (&7

mm Hg) in the mid-day (Chapman et al. 2002).

Collection and Maintenance of P. aethiopicus

Individuals were live-captured using baited minnow traps set at various places

within the lagoons and left overnight. Traps were collected the following morning and

each lungfish placed in its own separate holding container. This method was appropriate

for catching individuals ranging in mass from 10 g to ~250 g. Smaller individuals (<5 g)

were obtained by dip netting vegetated areas within the lagoons and in all cases were

caught in dense vegetation mats.

Individual lungfish were held in 1.2-1 buckets with lids while in Uganda and then

transported to the University of Florida. Once at the University, individual lungfish were

held in partitioned 208.2-L tanks with no two lungfish occupying the same space within a

tank. Water temperature was maintained at 23 0.5" C, with a 12/12 photoperiod.

Dissolved oxygen concentration averaged 8.16 0.05 mg 1-1. Temperature, dissolved

oxygen, pH, conductivity, ammonia, and nitrite levels were measured weekly to maintain

appropriate water quality. Individual fish were held under these conditions at least 1 year

before experimental trials. Fish were fed once a week on fresh chicken or beef liver and,









in the case of the smallest individuals, blood worms. Our experimental protocols for P.

aethiopicus were approved by the University of Florida (IACUC project #Z090).

Respirometry

Oxygen consumption was measured using a two-phase respirometer (Fig. 2-2). The

respirometer was constructed using PVC, oxygen impermeable vinyl tubing and brass

compression fittings (Swagelokc). Several chambers of different sizes were built and

were interchangeable within the system. The chamber where fish were placed consisted

of a long length of PVC connected to a T-junction that served as an air chamber. The T-

junction was constructed from clear PVC to allow periodic observation of fish during

respiratory trials, while the air chamber itself was constructed from opaque PVC. The

length of PVC that constituted the bulk of the water chamber was opaque. A 1.5 amp

Mag-drive water pump, flow meter, oxygen-sensing probe, and thermistor were all placed

in series for the water phase of the system. A stopper, with an oxygen probe, thermistor,

and small fan inserted through it, was used to seal the air portion of the system. Thus,

each phase of the respirometer contained an associated oxygen sensing probe and

thermistor that provided for simultaneous measurement of oxygen consumption and

temperature from both aerial and aquatic phases of the respirometer.

The respirometer utilized Ocean Opticso Oxygen Sensing technology to measure

oxygen concentration in both water and air (Dunedin, FL; http://www.oceanoptics.com).

Oxygen detection using these probes is based on excitation of a ruthenium complex

suspended in a sol-gel matrix at the ends of the probes. Light is emitted from a blue LED

source at -470 nm that travels via fiber optic cables to the tip of the probes.







































Figure 2-2. Respirometry system used to measure the resting rate of metabolism. The entire system is submerged in a water bath to
maintain temperature. The system is connected to a labtop that records the data. (a) Water chamber; (b) air chamber; (c) 2.5
amp Mag Drive pump; (d) flowmeter; (e) three stop-cock compression fittings that allows the system to convert from a
closed system to a flow-through system; (f) Ocean Optics oxygen sensing probe to measure DO2 in water phase; (g)
thermistor measuring water temperature; (h) thermistor and Ocean Optics oxygen sensing probe to measure temperature
and DO2 in air phase; (i) spectrophotometer.









The ruthenium is excited when exposed to the light, and emits energy at -600 nm

that is returned via the fiber optic cables to a spectrophotometer. As oxygen molecules

come in contact with the tip of the probe, the energy associated with the excitation via the

blue LED is quenched. This quenching (a non radiative transfer) results in a decrease of

intensity. An inverse relationship therefore exists between the intensity of excitation

associated with the ruthenium and the presence of oxygen in a given medium. This

information in intensity units is converted to percent saturation of oxygen in the medium

based on probe calibration (Krihak and Shahriari 1996).

The probes were calibrated to 0% DO2 in water containing sodium sulfite. The

probes were calibrated to 100% saturation for the water chamber in air-saturated water

and for the air chamber in air. The probes were calibrated on a scale of percent saturation

in the medium. Percent saturation was then converted to units of mg I1 with corrections

for ambient air pressure and water vapor pressure. The respirometer was designed as a

closed system with the capacity to flush out and replace the contents of either phase. This

allowed the removal of any metabolic waste products that accumulated during the

acclimation period.

Probes were calibrated prior to each run. Individual fish were placed in the system

containing a mixture of aerated tap water and filtered tank water. Controls were run on

the system with no fish present to determine biological activity of the water. In all cases

microbial 02 consumption in the water was determined to be negligible. Controls were

also run to determine the diffusion of oxygen between the air and water phases under

various levels of dissolved oxygen. These controls showed that the transfer of oxygen

between the two phases was undetectably small.









The resting rate of metabolism (RRM) was measured both at the Makerere

University Biological Field Station in Uganda and in our laboratory at the University of

Florida (UF). Resting rate of metabolism was measured as a function of oxygen

consumed over time. We refer to our measurements of oxygen consumption as RRM,

although we believe our final estimates represent minimum rates for these fish (measured

in the dark). Movement of the fish during trials was almost exclusively associated with

respiration and therefore considered a portion of maintenance metabolism since it is

necessary for survival. The protocols differed slightly as described below. For those fish

measured in the field (N=16), individual fish were acclimated for 3-h periods inside the

chamber with free access to air. After the acclimation period the air chamber was sealed

and the run was initiated. Oxygen consumption was recorded over a DO2 range from 7.21

( 130 mm Hg) to 5.77 mg Y1 ( 104 mm Hg) at 230 C. Measurements were made within 3

wks of capture, on post-absorptive fish, and we refer to these as our "field" estimates.

For those fish measured in the laboratory (N=14), individuals were acclimated

overnight to the system with free access to air. The DO2 concentration in the water

chamber was allowed to drop to near 0.0 mg 1-1, the approximate level at which aquatic

oxygen consumption ceased. An overnight acclimation was used because oxygen

consumption measurements revealed that the RRM measured after a 3-h acclimation

dropped approximately two-fold when compared to measurements made after a 16-h

acclimation. The effect of a short acclimation time on the RRM only became apparent

after laboratory tests were run at the UF where temperature control over an extended

period could be maintained. Prior to each run in the morning the system was flushed with

air-saturated water until the DO2 concentration in the system was greater than 2.57 mg 1-









(&46.4 mm Hg). This served to remove any metabolic wastes that may have accumulated

overnight and to raise aquatic DO2. The air chamber was then closed and the run

initiated. Oxygen measurements were made every 10 sec over the course of the run and

recorded automatically by the computer software running the system. Each run was

monitored and the percent saturation of oxygen in the air chamber was never allowed to

drop below 608.0 mm Hg. When oxygen concentration within the air chamber

approached 608.0 mm Hg, the stopper was removed and fresh air allowed to mix into the

air chamber to return the oxygen concentration to 100% saturation. This mixing did not

increase DO2 within the water chamber. Once the DO2 concentration within the water

chamber reached a plateau representing zero uptake from water, the run was allowed to

continue for 1 h and was then terminated. Oxygen consumption was recorded over a DO2

range between 4.29 mg 1- (&77.4 mm Hg) and 0.00 mg 1- at 230 C ( 0.5C).

Temperature was maintained throughout the run by placing the entire system inside a

Rubbermaid cooler that served as a temperature bath. Most of the system was submerged

in the cooler and the entire apparatus was covered with a black sheet to shield the fish

from any movements of the investigators. Individuals were starved for 7 d prior to each

experiment to ensure that they were post absorptive. Runs lasted an average of 12 h. At

the end of each run, individuals were drip dried and weighed (0.0 g).

An additional experiment was run in the laboratory to investigate any changes in

the RRM at near saturated levels of DO2 compared to those at ecologically relevant DO2

levels. We used the overnight laboratory acclimation period to provide comparability

with our runs at lower DO2. Following acclimation, the entire chamber was flushed and

the DO2 level in the chamber returned to near saturation. These runs lasted 6 h or less and









were terminated when DO2 concentration in the water phase was less than 6.0 mg 1-

(=108.4 mm Hg).

Analysis of covariance was used to compare the bilogarithmic relationships of

RRM and body mass among size groups and DO treatments. A t-test was used to

compare means when body size was not required as a covariate.

Results

Field Laboratory Data

Measurements of oxygen consumption on lungfish taken within 3 wks of capture

and measured at our field laboratory (N=16) at high DO2 concentrations revealed a strong

shift in respiratory allocation with body mass (Table 2-1). Individuals less than 55 g were

almost exclusively water breathers (percent allocation from water 95.7 0.9%). The

allocation to air breathing increased from an average of 4.3 0.9% in individuals less

than 55 g to 50.7 9.0% in individuals 55 to 111 g, and to 79.7 2.7% in individuals

greater than 111 g (Table 2-1). Measurements of oxygen consumption also revealed a

shift in the bilogarithmic relationship of RRM and body mass (Fig. 2-3). There was no

difference in the slopes of the bilogarithmic relationship of RRM (mg 02 hr-) and body

mass (g) between individuals < 83 g and > 83 g (F = 4.187, p = 0.063). However, the

intercepts (representing adjusted mean RRM) differed (F = 20.743, p = 0.016).

Individuals < 83 g had a 5.4-fold higher adjusted mean RRM of 10.26 mg 02 hr-

compared to individuals > 83 g (1.89 mg 02 hr- RRM adjusted to the mean mass of

82.0 g).

UF Laboratory Data

The RRM of P. aethiopicus (N=14) measured in the laboratory comprised both

aerial and aquatic oxygen consumption: Figs. 2-4 A,B illustrates these separate










components as oxygen consumed (mm Hg) over time (min) for two individuals. While

aerial consumption of oxygen fit a linear function, aquatic oxygen consumption fit an

exponential function (Figs. 2-4 A,B). To examine if the degree of curvilinearity within

the water phase varied with body size, we fit an exponential function (y=boebl*t) to

individual curves and compared the exponents (bi) among individuals as a function of

body mass. A comparison of the exponents indicated the degree to which the relationship

was curved or departed from a linear relationship. The degree of curvilinearity was

positively correlated with body mass (r = 0.598, p = 0.031, Fig. 2-5). The smallest

individual (2 g) was excluded from this analysis.

Lungfish reduced aquatic respiration to less than 1% of total oxygen consumed at

low aquatic P02 represented by the plateau point where aquatic respiration was nearly

Table 2-1. Respiratory allocation of Protopterus aethiopicus from the Lwamunda
Swamp, Uganda, between water and air at high (130 mm Hg 104 mm Hg)
levels of DO2 and measured at an average temperature of 23' C.
Measurements were made within 3 wks of capture at the Makerere University
Biological Field Station, Uganda.

Mass (g) Percent 02 from Water Percent 02 from Air
17 96 4
21 97 3
27 94 6
55 68 32
77 67 33
83 19 81
89 65 35
107 61 39
108 19 81
111 46 54
120 25 75
123 31 69
149 16 84
160 14 86
169 20 80
180 16 84










2.0





N 1.0- .





0.0- E]
0




-1.0
0.0 1.0 2.0 3.0
Logio body mass (g)

Figure 2-3. Bilogarithmic relationship between resting rate of metabolism (RRM) and
body mass of Protopterus aethiopicus from the Lwamunda Swamp, Uganda,
measured at high levels of DO2 in the field [(7.21 (=130 mm Hg) to 5.77 mg Y
1 (&104 mm Hg) at 230 C]. Individuals were acclimated for 3 h prior to trials.
Closed squares represent individuals < 81 g, and open squares represent
individuals > 83 g. Individuals < 81 g were measured in 2.5 cm chambers, and
fish > 83 g were measured in 5.1 cm chambers.










800
A
700

600 -



100-

80 -

60 -
Water
40 -

20 -
22 g

0 50 100 150 200 250 300 350 400
Time (min)


0 100 200 300 400 500 600 700 800
Time (min)


Figures 2-4. The decline in oxygen partial pressure over time for Protopterus aethiopicus
from the Lwamunda Swamp, Uganda, in a two-phase, closed respirometer,
where fish had access to both air and water simultaneously. The curvilinear
function y=boebl*t for aquatic respiration is indicated for two representative
individuals A)22 g B)221 g).









0.0035


0.0030-


0.0025-


0.0020-


0.0015-


0.0010-


0.0005-


0 i i I I
0 50 100 150 200 250

Body mass (g)

Figure 2-5. The relationship between the degree of curvilinearity (bl from the equation
y=bieb0*t) and body mass (g) for Protopterus aethiopicus from the Lwamunda
Swamp, Uganda. Curvilinearity describes the decline in oxygen partial
pressure over time in the aquatic phase of a two-phase, closed respirometer
where fish had access to both air and water simultaneously.

zero on each graph (Table 2-2). Whereas some variation among individuals occurred, fish

> 87 g reached this plateau at a significantly lower DO2 level than individuals < 81 g (t-

test, t = 3.182, p=0.030). Active aquatic respiration ceased at a mean DO2 of 0.48 0.2

mg 1- (&8.7 mm Hg) for fish > 87 g and 2.58 0.7 mg 1- (&46.5 mm Hg) for fish < 81 g.










Table 2-2. Respiratory allocation of Protopterus aethiopicus from the Lwamunda
Swamp, Uganda, from air at the aquatic respiration plateau (representing the
level of DO2 at which individuals respired almost completely aerially). This
plateau point was considered representative of ecologically relevant levels of
DO2. Water temperature was approximately 230 C.
Mass Percent 02 from air
2.0 100.0
9.0 99.0
22.0 100.0
41.5 98.2
50.0 97.7
81.0 99.6
87.0 100.0
112.0 100.0
134.0 97.7
139.5 94.0
144.0 100.0
206.5 97.0
208.0 98.2
221.8 100.0

Chamber Effect

In the field, chamber size was altered among specimens to maintain a similar ratio

between chamber volume and fish mass following previously published protocols for

other fishes. In the laboratory, three individuals (2, 9, and 22 g) were measured within a

2.5 cm (1-inch) PVC chamber (total water volume=266 ml), while the remaining

individuals were measured in 5.1 cm (2-inch) PVC chambers (total water volume=l 175

ml). To explore potential effects of chamber volume, we tested the 9- and 22-g fish in

both 2.5 cm and 5.1 cm chambers to compare RRM. For these two individuals we

replicated each chamber size twice at two different levels of DO2 corresponding to two

different ratios of respiratory allocation (pre-plateau and plateau). The RRM was

calculated as two rates for each chamber size: from the portion of the curve describing

the water phase before it plateaus when lungfish used both aquatic and aerial respiration

(pre-plateau), and from the plateau of the curve when the lungfish consumed primarily









atmospheric oxygen (plateau). We adjusted RRM for each DO2 regime by removing

effects of body mass based on the equation described by Ultsch (1995), [VO2

(adjusted)=(mean weight)b-1(observed weight)1-b(observed VO2)]. This allowed us to

compare the adjusted mean RRM for each chamber size for both the pre-plateau and

plateau. At pre-plateau DO2 levels, individuals run in the 2.5 cm chamber had a 2.4-fold

higher RRM when compared to the same individuals run in the 5.1 cm chamber (RRM,

2.5 cm: 0.147 0.02, 5.1 cm: 0.06 0.01). At plateau DO2 levels, RRM values were

similar (RRM, 2.5 cm: 0.046 0.01, 5.1 cm: 0.045 0.01). When these smaller lungfish

were respiring from both media (which generally occurs above field-relevant DO2 levels)

there was a distinct chamber effect, with higher RRM values in the smaller chamber.

When respiratory allocation was primarily aerial (at ecologically relevant levels of DO2),

there was no detectable chamber effect.

Resting rate of metabolism

The main objective of this study was to collect metabolic data for Protopterus

aethiopicus at ecologically relevant DO2 levels to provide a more realistic physiological

portrait of how animals survive under natural conditions in their respective environments.

The RRM was calculated as total oxygen consumed (mg 02 hfr1) from both the air phase

and aquatic phase of the respirometer. We used the aquatic phase of the curve at its

plateau where the lungfish is respiring almost completely aerially because this is a good

representation of the level of DO2 found in habitats used by small-sized lungfish in the

field, and chamber effects were not detectable (Goudswaard et al. 2001; Chapman et al.

2002). With the exception of the 2-g individual, we used data for individuals run in the

5.1 cm chambers. Total metabolic rate was positively correlated with body mass










(r2=0.9392, p=0.031, Fig. 2-6), with a scaling exponent of 0.78. Individuals showed a

mean percent allocation to air breathing of 98.7 0.48%.


0.0.







-1.0


0
C


0.0


Log 10 body mass (g)


Figure 2-6. The bilogarithmic relationship between RRM (mg 02 hf1) and body mass (g)
for the African lungfish, Protopterus aethiopicus, collected from the
Lwamunda Swamp, Uganda. Data represent RRM at low (ecologically
relevant) DO2 levels (measurements made at plateau point where oxygen
consumption in the aquatic phase was almost 0%; this was taken to represent
ecological conditions because DO2 levels would either not drop any lower or
were in the range found in the field). Measured at 230 C.


-L~II .









Individuals were also run at high (near saturated) levels of DO2 to test for an effect

of DO2 on RRM. No difference in slopes of the bilogarithmic relationship between RRM

(mg 02 hr1) and body mass (g) occurred between high and low levels of DO2 (F = 3.177,

p = 0.094). In addition, the intercepts (representing the adjusted mean RRM) did not

differ between the two oxygen levels (F = 0.895, p = 0.357). However, respiratory

allocation differed with body size and percent respiratory allocation from water decreased

with an increase in body size for fish tested in both the "field" (r2=0.822, p<0.001) and

the laboratory (r2=0.547, p=0.058).

We also measured pre-plateau RRM approximating bimodal respiration before

oxygen acquisition from the aquatic phase plateaued. The level of dissolved oxygen at

which the pre-plateau RRM was measured differed with each individual according to the

plateau level. No difference in slopes of the bilogarithmic relationship between RRM (mg

02 hr1) and body mass (g) was found between pre-plateau levels and low levels of DO2

(F= 0.265, p= 0.612). No difference in the intercepts (representing the adjusted mean

RRM) occurred between groups (F= 0.242, p=0.626, Fig. 2-7). For the above two

comparisons the 2-g individual was not included, because it had to be run in a different

chamber size.

Discussion

The results of this study clearly indicate the importance of considering ecological

conditions when measuring physiological parameters. Juveniles between 2 and 221 g

(including externally- gilled individuals) respired almost exclusively aerially at low











Pre-plateau









0
0 Low













-0.5 -

1 1.5 2 2.5

Log 1o body mass (g)

Figure 2-7. The bilogarithmic relationship between resting rate of metabolism (RRM)
(mg 02 hr-') and body mass (g) of Protopterus aethiopicus from the
Lwamunda Swamp, Uganda, measured at levels of DO2 prior to the plateau in
aquatic respiration versus low levels (plateau point at which fish is respiring
almost completely aerially; ecologically relevant). Measured at 230 C.

concentrations of DO2, which is contrary to earlier studies on small juveniles at high

oxygen levels. Protopterus aethiopicus also showed a remarkable ability to maintain its

RRM over the entire range of DO2 from 100% saturation to near anoxic conditions. The

relationship between aquatic gas exchange and DO2 is described by a curvilinear

relationship with the degree of curvilinearity increasing as a function of body mass. This

result, along with both the findings related to acclimation time and chamber size, stress









the importance of protocol in determining the RRM and respiratory allocation to water

and air in air-breathing fishes. These results differ from previous studies where both total

oxygen consumption and respiratory allocation were measured at high DO2 levels. To

better understand the importance of these results, they need be examined in light of what

is known of respiration in lungfish.

Metabolic Profile

A comparison of previous studies that measured oxygen consumption in lungfishes,

yielded a profile whereby juveniles used primarily aquatic respiration to meet metabolic

demands, and adults relied almost exclusively on aerial respiration to meet metabolic

oxygen demands (Smith 1930; Sawaya 1946; Johansen and Lenfant 1967; Lenfant and

Johansen 1968; Lenfant et al. 1970; McMahon 1970; Johansen et al. 1976; Babiker

1979). Johansen et al. (1976) reported that immediate post larval and early juvenile

Protopterus amphibious depended on aquatic respiration for 70% of their oxygen demand,

whereas fish between 300 and 500 g obtained as little as 10-15% of their oxygen from

water (Johansen et al. 1976). When measuring respiratory allocation between water and

air in Protopterus annectens Babiker (1979) found that small specimens < 200 g utilized

less than 10% aerial respiration to meet metabolic demands, whereas individuals > 400 g

utilized aerial respiration for 83% of their total oxygen uptake. In the only other

published study to measure juvenile lungfish and report body mass, Johansen and Lenfant

(1967) found that one juvenile L. paradox (150 g) used aquatic respiration to meet 64%

of its metabolic demand. Oduleye (1977) found that 4 to 150 g juvenile P. annectens

acquired about 94% of their oxygen aerially, however variation among size classes was

not reported in this study.









These previous findings mirror RRM measures that we made on lungfish within 3

wks of capture under high levels of DO2, when juvenile P. aethiopicus (< 83 g) used

aquatic respiration to account for 84 7% of their total oxygen uptake. Similarly,

measurements in our laboratory at UF after long-term acclimation to normoxia and taken

at high DO2 indicated higher allocation to aquatic respiration with juveniles < 83 g

acquiring approximately 66 10% oxygen uptake from water. These results differed

from laboratory measures at low aquatic oxygen levels that indicated a mean allocation

towards aerial respiration for individuals between 2 and 221 g of 98.7 0.46%.

Our laboratory measurements of oxygen consumption made across a gradient of

DO2 from normoxia to anoxia showed that P. aethiopicus maintains its RRM despite a

change in respiratory allocation between water and air. While respiratory allocation

differed depending on size, with smaller individuals acquiring more oxygen from the

aquatic phase at higher levels of DO2, P. aethiopicus displays a remarkable ability to

maintain its RRM despite these drastic changes in respiratory allocation. Larger

individuals still satisfy the bulk of their oxygen demand aerially, while the smallest

individuals apparently rely more on aquatic respiration at higher levels of dissolved

oxygen. This difference may represent the ability of the gills to efficiently extract oxygen

from water in smaller individuals. Based on the curvilinear relationship between aquatic

oxygen consumption and increasing body size, it would appear as if smaller individuals

are more capable of satisfying their oxygen demand via the gills and or skin. The

curvilinear pattern of oxygen uptake observed in P. aethiopicus in the aquatic phase of

respiration may indicate a passive adjustment to falling levels of DO2 that would support

active adjustments to aerial respiration as the primary mechanism to maintain a relatively









constant RRM. For a metabolic regulator breathing from one medium, the relationship

between oxygen consumed as a function of time would be linear on a given ample

oxygen to meet the metabolic demand. Measurements made on both terrestrial air

breathers and aquatic water breathers yield this result (McNab 2002). The exponential

nature of the relationship shown here suggests that the acquisition of oxygen from the

aquatic phase may be fueled by a diffusion gradient, maintained across the gills, skin, or

both, and related to total respiratory surface area. Smaller individuals have a larger

surface area to body mass ratio than do larger individuals and thus a passive diffusion of

oxygen across multiple gas exchange surfaces may be ample to satisfy the oxygen

demand at higher levels of dissolved oxygen. A large diffusion gradient that would help

to facilitate oxygen uptake in hypoxic water would cause increased rates of oxygen

diffusion under higher levels of DO2 due to the passive nature of the diffusion gradient.

Although aquatic respiration in the lungfish has been assumed to occur mainly at

the gills, recent studies on Lepidosirenparadoxa have indicated that the skin may play a

more prominent roll in the aquatic exchange of gases (Abe and Steffensen 1996; Sanchez

and Glass 2001). No direct experiment has been performed on lungfishes to determine the

contributions of both skin and gills to aquatic respiration. However, Sacca and Burggren

(1982) investigated this possibility in Cul/iun ,,i/1niy\ calabaricus, a primitive air-

breathing polypterid and determined that the skin and gills contributed 32% and 28%,

respectively, to oxygen uptake. Spaces between the large, thick ganoid scales of C.

calabaricus therefore permitted aquatic gas exchange suggesting the relatively smaller

scales of P. aethiopicus would pose little resistance to oxygen uptake. In fact, in similar

shaped, elongate fishes, such as Electrophorus (Farber and Rahn 1970), and









Saccobronchus (Singh and Hughes 1971), which are all capable of air breathing, the skin

has been shown to be an equally important gas exchange organ. The frequency of gill

ventilation in lungfishes is known to remain constant with increasing hypoxia, and thus it

seems reasonable to suggest that the increase in aquatic respiration can be attributed to

diffusion across the skin (Johansen and Lenfant 1968; Sanchez et al. 2001). Furthermore,

when burrowing in soft mud, the gills of P. aethiopicus would be severely compromised,

although burrowing individuals efficiently excrete both CO2 and metabolic wastes. Under

conditions of extreme hypoxia, it is possible that the skin plays an even greater role in

oxygen exchange. Our observations of individuals before and after trials indicated that P.

aethiopicus might actively alter vascularization patterns to the skin. The junction of both

the pelvic and pectoral fins with the body were always very red, and the color intensified

in some individuals over the course of a trial. This red color also extended along portions

of the appendages. In some cases, lighter-colored individuals showed an increase in

vascularization over extensive portions of the body. Lenfant and Johansen (1968)

documented a progressive vasodilation of the skin in response to prolonged air exposure.

An increased vascularization of the skin could facilitate increased oxygen uptake and/or

increased offloading of carbon dioxide. Future investigations comparing total respiratory

surface area to body mass in the lungfish along with direct measurements of gas

exchange across the skin will be needed to determine more clearly the role of the skin in

oxygen uptake in P. aethiopicus.

Methodological Issues

In the process of evaluating the RRM in P. aethiopicus, two methodological issues

were discovered that influence metabolic rate measurements in this species: chamber size

and acclimation time. Both a change in chamber size with body mass and the short









acclimation time used in the field led to differences between our field and laboratory

measurements of RRM in P. aethiopicus. At our field lab, small juveniles exhibited

elevated metabolic rates (when total metabolic rate was graphed as a function of body

size) compared to larger individuals, with small juveniles having a 5.9-fold higher RRM

than those individuals > 83 g. However, the shift was coincident with our shift from a

small to large chamber size. When we evaluated chamber size effects in the laboratory,

we found that use of a smaller chamber size for fish < 22 g coincided with an elevation in

the RRM at high DO2, but not when fish were breathing primarily atmospheric air

(plateau level). Thus, we attribute the shift in adjusted mean RRM that we observed

between small and large individuals in our field lab to reflect, at least in part, the change

in chamber size. The smaller chambers may be perceived by smaller individuals as

confining and illicit a stress response in these smaller individuals. Similar results were

found, in both Protopterus amphibious (Johansen et al 1976) and in Protopterus annectens

(Babiker 1980) with elevated RRM in smaller individuals. Methods used in these studies

indicated that alternate chamber sizes were also used. Many previous studies also report a

short acclimation time in the chamber. Our acclimation comparisons in the laboratory

revealed at least a two-fold decrease in RRM between a 3-h and 16-h acclimation period.

This suggests that chamber acclimation time should be carefully considered in any design

or meta-analysis of fish metabolic rates.

Both methodological issues may reflect behavioral responses. A short acclimation

time potentially increases stress levels. Within the various chambers the smallest

individuals were seen to hover at the water/air interface, whereas large individuals spent

the majority of their time hidden within the opaque portion of the water chamber except









when moving to breathe air. Prodding of smaller individuals both in the chamber and in

holding tanks elicited a burst of activity. This behavior resembled an escape mechanism

or predator avoidance. While all juveniles measured in this experiment were from the

swamp, the smallest fish, particularly the gilled individuals, were often found associated

with dense vegetation in the field. When disturbed in the field, they behaved in a similar

manner. Their reflex is to retreat rapidly to the bottom of the water column, and this

behavior occurs when an animal is confined or threatened. The close confines of the

smaller chamber could have increased the stress level of these individuals enough to

account for the increase in metabolic rate. The effect disappeared when individuals

switched to an almost complete dependence on aerial respiration. This suggests that the

increase in RRM may be attributed to an increase in aquatic respiration brought about by

a stress response to a confined chamber. This conclusion is reinforced by the finding that

chamber size apparently had no effect on aerial respiration.

It is clear from this study that Protopterus aethiopicus is able to maintain a

relatively constant RRM over a range of DO2 levels from normoxia to hypoxia.

Measurements of the RRM at normoxic conditions, pre-plateau levels, and with

individuals respiring almost completely aerially did not differ. Although respiratory

allocation differs across the range of dissolved oxygen from normoxia to extreme

hypoxia, adjustments in oxygen uptake serve to balance the overall uptake of oxygen to

stabilize RRM. Since smaller juvenile specimens of Protopterus aethiopicus are clearly

taking in higher amounts of dissolved oxygen at near saturated levels of DO2, it will be

interesting to determine if these adjustments are passive or active and contribute to

surplus energy uptake, or if they are related to both carbon dioxide excretion and






34


nitrogenous waste removal. Future experiments will be necessary to define the role of the

skin in oxygen uptake and how this relates to changing levels of DO2.














CHAPTER 3
THE RELATIONSHIP BETWEEN TEMPERATURE AND THE RESTING RATE OF
METABOLISM IN Protopterus aethiopicus

Introduction

Few of all environmental factors known to affect metabolism are as significant as

temperature. The study of this relationship, from both a molecular and whole organism

perspective, yields insight into the intimate connection between animal energetic and the

environment. A derivation from first principles of the association between temperature

and the resting rate of metabolism has been a topic of rich conversation ever since August

Krogh and his co-workers attempted to formulate such a relationship for poikilothermic

organisms (Krogh 1914, 1916). His quantitative expression of this relationship for a

single species, the goldfish (Cassius auratus) led to his derivation of what came to be

known as "Krogh's normal curve" (Ege and Krogh 1914). The normal curve expressed an

exponential increase in metabolic rate as a function of increasing temperature.

Winberg (1956), examining resting metabolic rates in fishes, compared values

drawn from the literature to Krogh's normal curve in an effort to compare metabolic rates

from different species measured at different temperatures. Using the normal curve for

orientation, Winberg (1956) derived a series of coefficients that could be used to adjust

metabolic rates of temperate fishes by applying the Van't Hoff equation,

[Qio=(K2/K)(10 t2-1t)]

where Ki and K2 are the metabolic rates measured at two different temperatures ti

and t2, for changes in temperature across intervals of 10 'C. This permitted adjustment of









metabolic rates measured at a specific temperature to a second specific temperature of

biological interest and a description of how an individual's resting metabolism responded

to an increase or decrease in temperature. As long as an ectotherm is acclimated for an

adequate length of time to each new temperature, this rate reflects the long-term

compensatory effects associated with metabolism affecting a homeostatic balance within

the individual despite a change in environmental conditions.

The Qio coefficients for fishes derived by Winberg were based largely on data

collected for temperate fishes. The degree to which these coefficients apply to metabolic

data collected for fishes that inhabit tropical or polar waters and that experience a smaller

range of natural temperature fluctuation is largely unknown. Researchers investigating

species from these systems are faced with the decision to calculate their own Qlo values

or to use published values measured on other species. Based on the experiments of

Johnston et al. (1991) and Clarke and Johnston (1999), utilizing Qio relationships for

organisms with different lifestyles and/or from unrelated taxa can clearly lead to

erroneous adjustments to calculated metabolic rates.

Fishes, as the most species-rich group of chordates, lack a comprehensive

synthesis of metabolic data that includes both ancient relic fishes (non-teleost) and

teleosts, and that adequately represents tropical, temperate, and polar regions. In

particular, studies examining metabolic rates and the effects of temperature in non-teleost

fishes are completely absent from the literature. The objective of this study was to

determine the Qio value for the African lungfish, Protopterus aethiopicus, an ancient air-

breathing fish from a tropical freshwater habitat. Here we present the Qio values for six

individuals ranging in body mass from 42 to 222 g evaluated across a biologically









relevant 10 'C temperature range. These Qio values are discussed in relation to values

drawn from the literature for both temperate and tropical teleosts.

Materials and Methods

Six juvenile lungfish ranging in size from 44 to 222 g were used in this study.

Individuals had been housed at the University of Florida (UF) since their collection in

June 2001 from Lake Nabugabo, Uganda. Lake Nabugabo lies just south of the equator

and is a small satellite lake (surface area = 24 km2, mean depth = 4.5 m) that was isolated

from Lake Victoria approximately 4000 years ago (Greenwood 1965). The lake lies

within the extensive Lwamunda Swamp that was formerly a bay on the western shore of

Lake Victoria (Worthington 1932; Greenwood 1965; Ogutu-Ohwayo 1993). In the Lake

Nabugabo system, Goudswaard et al. (2001) found that 90% of the lungfish captured in

minnow traps in the swamp were between 20 cm and 45 cm TL. Larger lungfish in the

system are found in more open waters (L. Chapman, personal observations). Water

temperatures measured throughout the year and within the swamp averaged between 21.3

'C and 25.2 'C in the early morning and between 21.3 'C and 27.2 'C in the early

afternoon (Chapman et al. 2002).

Prior to metabolic trials, individuals were held at 23 'C 0.5 'C (12/12

photoperiod) in four partitioned 20-gallon tanks with no two lungfish occupying the same

space within a tank. Tank temperatures were controlled with aquarium heaters, and

temperatures were measured daily. Dissolved oxygen, pH, conductivity, ammonia, and

nitrite levels were measured weekly to maintain appropriate water quality, and tanks were

maintained on a 12/12 photoperiod. Individuals were fed fresh chicken liver once a week.

Our experimental protocols for P. aethiopicus were approved by the University of Florida

(IACUC project #Z090).









Six individuals were acclimated across the temperature range of 20 'C to 30 OC

over a 5-month period beginning in January 2003. Individuals initially at 23 'C were

gradually increased to 30 OC over a 1-wk period. Once at 300C, individuals were

acclimated for at least 2 wks with a daily fluctuation in temperature of less than 0.5 'C.

Each lungfish was assigned a number, and a random number table was used to select the

order in which individuals were measured for their resting rate of metabolism (RRM).

For fishes the resting rate of metabolism is approximately the same as the standard rate of

metabolism (SRM) and is "the respiration rate of an unfed fish resting quietly in the

experimental chamber" (Clarke 1991, 1993). Here we use the (RRM), which we believe

accurately reflects the standard rate of metabolism for Protopterus aethiopicus.

Individuals were starved for 1 wk prior to their metabolic run. The RRM was measured

via the measurement of oxygen consumption in a two-phase respirometer following the

protocol described in Chapter 2. Oxygen consumed from both water and air was

measured simultaneously. Respirometer temperature was maintained to within 0.5 'C

during each experimental run.

Once all six individuals were measured for the RRM, individuals were lowered to

the next temperature regime over a 3-d period and again acclimated for at least a 2-wk

period. In this way, the above metabolic experiments were performed at 30 OC, 25 'C,

and 20 'C.

Statistics

The resting rate of metabolism (RRM) for each individual was calculated as an

adjusted mean by controlling for the effects of body mass based on the equation

described by Ultsch (1995) [V02 (adjusted)=(mean weight)b-1(observed weight)1'

b(observed V02)]. We used the scaling exponent b=0.78 from Chapter 2 that describes









the bilogarithmic relationship between metabolic rate and body mass for this population

of lungfish across a broad size range of 14 juvenile fish. Repeated measures analysis of

variance was used to calculate the adjusted mean resting rate of metabolism at each water

temperature. Following Potvin et al. (1990), Mauchly's criterion was used to test for the

compound symmetry of the variance-covariance matrix. The Qio value relating RRM to

water temperature over a 100C range was calculated using the equation: Qio=(K2/K1)(10/t2-

tl), where K2 and Ki are metabolic rates at temperatures t2 and ti. The determination of

the Qio value indicates an adjustment in RRM to temperature changes across a known

temperature gradient (Winberg 1956) and was determined for Protopterus aethiopicus

over the range of 20 'C to 30 OC. All errors reported are standard errors.

Results

Protopterus aethiopicus showed a positive bilogarithmic relationship between the

rate of oxygen consumption and body mass (range=44 to 222 g) at all three water

temperatures (20 'C: r2=0.798, P=0.016; 25 'C: r2=0.847 P=0.009; 30 'C: r2=0.862,

P=0.007, Figure 3-1). Because of the relatively small sample size for which we could

repeatedly measure metabolic rates across a temperature gradient, we used the slope of

the bilogarithmic relationship described in Chapter 2 (0.78) to derive adjusted mean

resting metabolic rates at each of the three temperatures. Adjusted mean RRM ranged

from 0.00833 mg 02 hf1 at 200C to 0.01567 mg 02 hf1 at 250C and 0.02617 mg 02 hr1 at

30C (adjusted to a mean body mass of 110.5 g, Fig 3-2).

The mean RRM at each temperature interval was significantly different from the

preceding interval (20 'C vs 25 'C: p=0.006, and 25 'C vs 30 'C: p=0.019, repeated

contrasts, Fig. 3-2).









1.00 O


0.50-




0.00-


"'&

0"







o


-0.504-
1.50


2.00


2.50


Logio mass (g)
Figure 3-1. Bilogarithmic relationship between resting metabolic rate (RRM) and body
mass for six individuals of Protopterus aethiopicus from the Lwamunda
Swamp, Uganda, measured at three different temperatures: 30 'C, 25 'C, and
20 'C. Lungfish were acclimated for at least 2 wks prior to measurement.
Triangles represent 30 'C, squares represent 25 'C, and closed circles
represent 20 'C.


11 411


0.025-

0.020-

0.015-

0.010-

0.005-


0 -I--


25
Water temperature (oC)


Figure 3-2. Mean resting rate of metabolism adjusted to 110.5 g fish (mean for six
individuals) for three temperatures: 30 'C, 25 'C, and 20 'C. A within-subject
contrast found the mean RRM at each temperature interval to differ
significantly from the preceding interval: 20 'C vs 25 'C p=0.006, and 25 'C
vs 30 'C p=0.019. Error bars are less than 0.002.


U









The relationship between RRM and temperature was calculated as the Qio value

for the temperature interval of 100C between 200C and 300C. The Qio value was 3.14.

Discussion

The primary focus of this study was to examine the response in RRM of

Protopterus aethiopicus within a biologically relevant range of water temperatures to

gain insight into how this tropical, freshwater, non-teleost fish responds to changes in

temperature. This was accomplished by accurate determination of the RRM, adequate

acclimation to temperature for each individual in order to remove the effects of any short-

term compensatory mechanisms in response to temperature, and a review of those studies

from the literature that followed these criteria. Protopterus aethiopicus exhibited a

strong, positive increase in the resting rate of metabolism (RRM) with an increase in

temperature across an ecologically relevant temperature range. To our knowledge, this

study is the first to quantify the relationship between temperature and the resting rate of

metabolism for a non-teleost fish, and the first study to examine this relationship in an

air-breathing fish.

Previous studies that examined the effect of temperature on metabolic rate in

fishes are limited and in most instances focused on deviations from Krogh's normal

curve. Clarke and Johnston (1999) sought to refine this relationship for teleost fishes

utilizing literature-derived data that met a specific set of criteria, similar to those listed

above (i.e. accurate determination of RRM, adequate acclimation time, etc. See Clarke

and Johnston 1999 for full description of criteria). Because they focused on describing

the effects of temperature as a general relationship among as many species as possible,

they included measurements made on the same species from different studies at different

temperatures. Using an Arrhenius model to compare resting metabolic rates adjusted for a









50 g fish across a temperature range, they found that tropical fish require 6.2 times more

oxygen at 30 OC than do polar fish at 0 OC (Clarke and Johnston 1999). This relationship

explained 59% of the variance among 69 species of teleosts, with the remaining variance

concluded to be the combined effect of phylogeny, ecology, and/or lifestyle (Clarke and

Johnston 1999). Clarke and Johnston (1999) also examined taxonomic variation within

this relationship. Using the same species data pooled by taxonomic order, they detected

no difference in slopes of the relationship between temperature and RRM among orders.

Clarke and Johnston (1999) also derived a Qio value of 1.83 from their curve based on 69

species of teleosts, and reported a median Qlo value of 2.40 from 14 published values on

individual species that were exposed to different temperature regimes. However, they did

not explicitly explore Qio values for individual species in different thermal habitats.

Studies that measure the resting rate of metabolism within the same species across

a gradient of temperatures directly examine the effects of temperature upon metabolism

by attempting to remove compensatory short-term mechanisms that occur in non-

acclimated organisms. It has been shown that animals that are exposed to abrupt

temperature changes without adequate acclimation, respond metabolically in a fashion

that tends to inflate Qio values when compared with acclimated animals (Holeton 1974).

Thus, an accurate determination of Qio values on temperature acclimated fish provides a

means to compare different species from different environmental temperature regimes.

Studies of Qio values and comparisons of these values across species are useful in

illustrating (a) if the rate of change follows the standard derived relationships calculated

from interspecific data, and (b) whether the rate of change varies among species, or

groups of ecologically or phylogenetically related taxa. We found 10 published studies









that met our criteria for comparing effects of temperature on the resting rate of

metabolism within the same species (Table 3-1). Eight of these studies involved fishes

that occur in temperate zones, while this study and the remaining two studies examined

tropical species. Mean masses were available from many of these published studies

(tropical n=3 and temperate n=8). We tested the effect of mass as a covariate against Qio

by running an ANCOVA. The effect of mass was not significant (p=0.170). Therefore we

removed mass from our model and used a t-test to test for a difference between the Qio

values between the two groups. Removal of the unusually high Qio value for Anguilla

rostrata leads to a mean Qio value for temperate species of 2.43, a value significantly

lower than the mean value for the three tropical species (Q1o=2.95, t=3.85, p=0.027).

When the A. rostrata value is included, the mean Qio for the tropical fish group does not

differ from the temperate group (mean Qio for temperature species=2.54, t=1.39,

p=0.191). Sample size was insufficient to perform a phylogenetic contrast analysis.

However, both tropical and temperate species represent a broad range of lineages.

Assuming that the A. rostrata value is not representative of the group, we can speculate

on the apparent differences in Qlo values between the sample of temperate and tropical

species available. It may reflect the lower degree of thermal variability that is

experienced throughout the year in tropical species. For P. aethiopicus from Lake

Nabugabo, daytime water temperatures from the swamp where juveniles are exclusively

found range between 21.30C and 27.20C (Chapman et al. 2002, based upon 1 year of

monthly samples). This is far smaller than the ranges of temperatures experienced

seasonally in temperate climates. It is possible that temperate species might possess a

greater ability to maintain resting metabolism across their natural temperature gradient.









Table 3-1. Qio values calculated for various species of teleost fishes. Resting rates of
metabolism were provided by Andrew Clarke (British Antarctic Survey)
adjusted to a 47 g fish. Qio values were calculated across temperature ranges
listed. Citations indicate initial source of raw metabolic data. All studies
included measured resting rate of metabolism (mg 02 11 hr-) and acclimated
individuals for sufficient periods of time prior to measurement at different
temperatures. Means are for tropical and temperate Qio values and are
significantly different (p=0.027). *In calculating the mean Qio for temperate
species, Anguilla rostrata was not included as described in the text.


Species
Tropical
Colossoma macropomum
Oreochromis mossambicas
Protopterus aethiopicus



Temperate
',p e/1//hiliahm1% maximus
Gambusia affinis
Anguilla anguilla
Anguilla rostrata
Cyprinus carpio
Carassius auratus
Catostomus commersonii
Oncorhynchus nerka
Salvelinusfontinalis
Pleuronectes platessa
Ph i /ii/ij \ flesus


Temp range (C)

20-30
19-28
20-30


6.0-22.0
10.0-30
20.0-27.0
15.0-25.0
10.0-30.0
10.0-30.0
10.0-30.0
5.0-20.0
10.0-20.0
2.0-22.0
2.0-22.0


Qio Mean

2.97
2.74
3.14
2.95


3.06
2.24
2.48
3.67
2.60
2.29
2.44
2.02
2.84
2.06
2.24


Study

Saint-Paul et al. 1988
Caulton 1978
this study


Mallekh and Lagardere 2002
Cech et al. 1985
Degani et al. 1989
Degani & Gallagher 1985
Beamish 1964
Beamish and Mookherji 1964
Beamish 1964
Brett & Glass 1973
Beamish 1964
Fonds et al. 1992
Fonds et al. 1992


2.43*


Comparing our derived Qio values for both temperate and tropical species to the

Qio value of 1.83 derived by Clarke and Johnston (1999) for all teleosts (whether polar,

temperate or tropical), highlights the problems associated with using published Qio values

to adjust metabolic rates in the absence of species-specific values. It is a common

practice to adjust metabolic rates using Winberg's Qio coefficients derived from

temperate species based on Krogh's normal curve, and it has been argued that a Qio of 2









is an appropriate adjustment for this relationship among diverse species (Fry and

Hochachka 1970, Miller and Mann 1973). The derived Qio values presented here clearly

show how a generalized application of between-species Qio values can lead to erroneous

determinations when adjusting metabolic rates. Furthermore, while within-species Qio

values show variation, a lack of appropriate data for a diverse array of fishes confounds

understanding the source of this variation.

Although the number of fish species for which Qio data are available and their

phylogenetic diversity is small, this comparison highlights the need for further

examination of phylogenetic and geographic trends in Qlo data. With the exception of this

study, all previous data represent teleost fishes. Preliminary examination of resting

metabolic rates in non-teleost fishes suggests a low resting rate of metabolism to be a

primitive character (A. Seifert, unpublished data). It will be of interest to investigate if a

lower thermal sensitivity in terms of Qio is correlated with taxonomic position or if it is

primarily a function of habitat. In addition, future work that examines effects of

temperature outside of the range experienced in nature will help to address whether

metabolic compensation approaches the higher and lower bounds of thermal tolerance. A

few studies suggest that there is metabolic depression above some critical temperature,

but mechanisms for these observations are not well understood (Saint-Paul 1983). Further

investigations will shed light on the contributions of protein turnover, membrane

homeostasis, and cellular ion balance to metabolism. Whereas these processes

presumably play an important role, it is unclear what the individual contributions of these

processes are to maintenance metabolism or their relative relationship to temperature and

lifestyle (Clarke and Johnston 1999). A more complete picture of the relationship






46


between temperature and its effect on resting metabolic rate across all groups of fishes

will yield a deeper understanding of the effect of environmental factors upon

physiological processes in ectotherms.














CHAPTER 4
SUMMARY AND CONCLUSIONS

This study quantified respiratory allocation and the resting rate of metabolism

(RRM) in the African lungfish, Protopterus aethiopicus, under ecologically relevant

levels of dissolved oxygen. Simultaneous measurements of oxygen consumed from both

air and water were used to measure RRM of juvenile lungfish across the size range

typically found in hypoxic swamps of the Lake Victoria basin. I also measured the effects

of temperature on the resting rate of metabolism and compared calculated Qio values to

those calculated for temperate and tropical fishes. The following section summarizes the

major findings of this study.

1. Aerial oxygen consumption measured at hypoxic DO2 levels reflective of the
natural habitat averaged 98% (range = 94% to 100%) of total respiratory allocation
for juvenile lungfish between 2 and 221 g (including externally-gilled individuals).
This finding contradicts the long-held belief that small juvenile lungfish primarily
depend on aquatic respiration to meet their aerobic metabolic demand, but is
supported by ecological data that suggest individuals < 40 cm TL primarily are
found in hypoxic wetlands.

2. Total metabolic rate was positively correlated with body mass with a scaling
exponent of 0.78.

3. Measurements of oxygen consumption made across a gradient of dissolved oxygen
from normoxia to anoxia showed that P. aethiopicus generally maintains its RRM
despite a change in respiratory allocation between water and air.

4. The relationship between aquatic gas exchange and DO2 is described by a
curvilinear relationship with the degree of curvilinearity increasing as a function of
body mass. There is as yet no explanation for this phenomenon.

5. In the process of evaluating the RRM in P. aethiopicus, two methodological issues
were discovered that influence metabolic rate measurements in this species:
chamber size and acclimation time. Both a change in chamber size with body mass
and the short acclimation time used in the field led to differences between our field









lab and laboratory measurements of RRM taken at UF in P. aethiopicus. A
decrease in chamber size led to an increase in RRM. An increase in acclimation
time from 3 h to 16 h led to a two-fold decrease in RRM.

6. Protopterus aethiopicus exhibited a strong, positive increase in the resting rate of
metabolism (RRM) with an increase in temperature across an ecologically-relevant
temperature range. The Qio value for the temperature interval of 200C and 300C
was 3.14. To our knowledge, this study is the first to quantify the relationship
between temperature and the resting rate of metabolism for a primitive, non-teleost
fish, and the first study to examine this relationship in an air-breathing fish.

7. A comparison of literature-derived Qio values for both tropical and temperate fishes
indicated a higher Qio in tropical species (2.95, n=3) than in temperate species
(2.43, n=10) across an ecologically relevant thermal range. This difference may
reflect the lower degree of thermal variability that is experienced throughout the
year in tropical species.
















LIST OF REFERENCES


Abe A.S., and J.F. Steffensen. 1996. Lung and cutaneous respiration in awake and
estivating South American lungfish, Lepidosirenparadoxa. Revista Brasileira de
Biologia 56:485-489.

Babiker M.M. 1979. Respiratory behaviour, oxygen consumption and relative
dependence on aerial respiration in the African lungfish (Protopterus annectens,
Owen) and an air-breathing teleost (Clarias lazera, C.). Hydrobiologia 65:177-187.

Beamish, F.W.H. 1964. Respiration of fishes with special emphasis on standard oxygen
consumption. II. Influence of weight and temperature on respiration of several
species. Canadian Journal of Zoology 42:177-188.

Beamish, F.W.H., and P.S. Mookherjii. 1964. Respiration of fishes with special emphasis
on standard oxygen consumption. I. Influence of weight and temperature on
respiration of goldfish, Carassius auratus L. Canadian Journal of Zoology 42:161-
175.

Brett, J.R., and N.R. Glass. 1973. Metabolic rates and critical swimming speeds of
sockeye salmon (Oncorhynchus nerka) in relation to size and temperature. Journal
of the Fisheries Research Board of Canada 30:379-387.

Burggren W.W., and K. Johansen. 1987. Circulation and respiration in lungfishes
(Dipnoi). Pp. 217-236 in W.E. Bemis, W.W. Burggren, and N.E. Kemp, eds. The
Biology and Evolution of Lungfishes. Alan R. Liss, Inc., New York.

Burggren W.W., K. Johansen, and B. McMahon. 1986. Respiration in phyletically
ancient fishes. Pp. 217-252 in R.E. Foreman, ed. Evolutionary Biology of Primitive
Fish. Plenum Press, New York.

Caulton, M.S. 1978. The effect of temperature and mass on routine metabolism in
Sarotherodon (Tilapia) mossambicus (Peters). Journal of Fish Biology 13:195-201.

Cech, L.J. Jr., Massingill, M.J., Vondracek, B., and A.L. Linden. 1985. Respiratory
metabolism of the mosquitofish, Gambusia affinis: effects of temperature,
dissolved oxygen, and sex difference. Environmental Biology of Fishes 13:297-
307.

Clarke, A. 1991. What is cold adaptation and how should we measure it? American
Zoologist 31:81-92.









Clarke, A. 1993. Seasonal acclimatization and latitudinal compensation in metabolism:
do they exist? Functional Ecology 7:139-149.

Clarke, A., and N.M. Johnston. 1999. Scaling of metabolic rate with body mass and
temperature in teleost fish. Journal of Animal Ecology 68:893-905.

Chapman L.J., F.G. Nordlie, and A.W. Seifert. 2002. Respiratory oxygen consumption
among groups of Pseudocrenilabrus multicolor victoria subjected to different
oxygen concentrations during development. Journal of Fish Biology 61:242-251.

Coates M.I., and J.A. Clack. 1991. Fish-like gills and the breathing in the earliest known
tetrapod. Nature 352:234-236.

Degani, G., and M.L. Gallagher. 1985. The relationship between growth, food
consumption and oxygen consumption in the developed and underdeveloped
American eels, Anguilla rostrata Lesueur. Journal of Fish Biology 27:635-641.

Degani, G., Gallagher, M.L., and A. Meltzer. 1989. The influence of body size and
temperature on the oxygen consumption of the European eel, Anguilla anguilla.
Journal of Fish Biology 34:19-24.

Ege, R., and A. Krogh. 1914. On the relation between the temperature and the respiratory
exchange in fishes. International Revue der Gesamten Hydrobiologie und
Hydrographie 1:48-55.

Farber J. and H. Rahn. 1970. Gas exchange between air and water and the ventilation
pattern in the electric eel. Respiration Physiology 9:151-161.

Fonds, M., Cronie, R., Vethaak, A. D., Vanderpuyl, P. 1992. Metabolism, food
consumption and growth of Plaice (Pleuronectes platessa) and flounder
(Phlaiin //i -\flesus) in relation to fish size and temperature. Netherlands Journal of
Sea Research 29: 127-143.

Fry, F. E. J., and Hochachka, P.W. 1970. Fish. Pp. 79-134 in G.C. Whittow, ed.
Comparative Physiology of Thermoregulation. Vol. I. Invertebrates and
Nonmammalian Vertebrates. Academic Press, New York.

Gardiner B. 1980. Tetrapod ancestry: A reappraisal. Pp. 177-185 in A.L. Panchen, ed.
The Terrestrial Environment. Academic Press, London.

Gordon M.S., and E.C. Olson. 1994. The Invasions of the Land. Columbia University
Press, New York.

Goudswaard K.P.C., F. Witte, and L.J. Chapman. 2001. Decline of the African lungfish
(Protopterus aethiopicus) in Lake Victoria (East Africa). African Journal of
Ecology 39:1-11.






51


Graham J.B. 1997. Air Breathing Fishes: Evolution, Diversity and Adaptation. Academic
Press, San Diego, California.

Greenwood P.H. 1958. Reproduction in the East African lungfish Protopterus
aethiopicus Heckel. Proceedings of the Zoological Society of London 130:547-567.

Greenwood P.H. 1965. The cichlid fishes of Lake Nabugabo, Uganda. Bulletin of the
Museum of Natural History (Zoology) 12:315-357.

Greenwood P.H. 1987. The natural history of African lungfishes. Pp.163-179 in W.E.
Bemis, W.W. Burggren, and N.E. Kemp, eds. The Biology and Evolution of
Lungfishes. Alan R. Liss, Inc., New York.

Grigg G.C. 1965. Studies on the Queensland lungfish Neoceratodusforsteri (Krefft). III.
Aerial respiration in relation to habitats. Australian Journal of Zoology 13:413-421.

Holeton, G.F. 1974. Metabolic adaptation of polar fish: fact or artefact? Physiological
Zoology 47:137-152.

Jesse M.J., C. Shub, and A.P. Fishman. 1967. Lung and gill ventilation of the African
lungfish. Respiration Physiology 3:267-287.

Johansen K., and C. Lenfant. 1967. Respiratory function in the South American lungfish,
Lepidosiren paradox (Fitz). Journal of Experimental Biology 46:205-218.

Johansen K., and C. Lenfant. 1968. Respiration in the African lungfish Protopterus
aethiopicus. II. Control of breathing. Journal of Experimental Biology 49:453-468.

Johansen K., J.P. Lomholt, and G.M.O. Maloiy. 1976. Importance of air and water
breathing in relation to size of the African lungfish Protopterus amphibious (Peters).
Journal of Experimental Biology 65:395-399.

Johnels A.G., and G.S.O. Svensson. 1954. On the biology of Protopterus annectens
(Owen). Arkiva Zoology 7:131-158.

Johnston, I.A., Clarke, A., and P. Ward. 1991. Temperature and metabolic rate in
sedentary fish from the Antarctic, North Sea and Indo-West Pacific Ocean. Marine
Biology 109:191-195.

Krihak M.K., and M.R. Shahriari. 1996. Oxygen sensor based on the sol-gel coating
technique. Electronics Letters 32:3.

Krogh, A. 1914. The quantitative relation between temperature and standard metabolism
in animals. International Zeitschrift fur Physikalisch-Chemische Biologie 1:491-
508.

Krogh, A. 1916. The Respiratory Exchange of Animals and Man. Longmans, London.









Lahiri S., J.P. Szidon, and A.P. Fishman. 1970. Potential respiratory and circulatory
adjustments to hypoxia in the African lungfish. Federation Proceedings 29:1141-
1148.

Lenfant C., and K. Johansen. 1968. Respiration in the African lungfish Protopterus
aethiopicus. I. Respiratory properties of blood and normal patterns of breathing and
gas exchange. Journal of Experimental Biology 49:437-452.

Lenfant C., K. Johansen, and D. Hanson. 1970. Bimodal gas exchange and ventilation-
perfusion relationship in lower vertebrates. Federation Proceedings 29:1124-1129.

Little C. 1990. The Terrestrial Invasion. Cambridge University Press, Cambridge.

Long J. A. 1995. The Rise of Fishes: 500 Million Years of Evolution. Johns Hopkins
University Press, Baltimore.

Mallekh, R., and J.P. Lagardere. 2002. Effect of temperature and dissolved oxygen
concentration on the metabolic rate of the turbot and the relationship between
metabolic scope and the feeding demand. Journal of Fish Biology 60:1105-1115.

Marshall, C. R. 1987. Lungfish: Phylogeny and Parsimony. Pp.151-162 in W.E. Bemis,
W.W. Burggren, and N.E. Kemp, eds. The Biology and Evolution ofLungfishes.
Alan R. Liss, Inc., New York.

McMahon B. R. 1970. The relative efficiency of gaseous exchange across the lungs and
gills of an African lungfish Protopterus aethiopicus. Journal of Experimental
Biology 52:1-15.

McNab B.K. 2002. The Physiological Ecology of Vertebrates. Cornell University Press,
London.

Miller, R.J., and K. H. Mann. 1973. Ecological energetic of the seaweed zone in a
marine bay on the Atlantic coast of Canada. III. Energy transformations by sea
urchins. Marine Biology 18:99-114.

Oduleye S.O. 1977. Unidirectional water and sodium fluxes and respiratory metabolism
in the African lungfish, Protopterus annectens. Journal of Comparative Physiology
119:127-139.

Ogutu-Ohwayo R. 1993. The effect of predation by Nile Perch, Lates niloticus L., on the
fish of Lake Nabugabo, with suggestions for conservation of endangered endemic
cichlids. Conservation Biology 7:701-711.

Potvin, C., M. J. Lechowicz and S. Tardif. 1990. The statistical analysis of
ecophysiological response curves obtained from experiments involving repeated
measures. Ecology 7:1389-1400.









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.

Sacca R. and W. Burggren. 1982. Oxygen uptake in air and water in the air-breathing
reedfish C11i,11n i111iy1 calabaricus: Role of skin, gills and lungs. Journal of
Experimental Biology 97:179-186.

Saint-Paul, U. 1983. Investigations on the respiration of the Neotropical fish, Colossoma
macropomum (Serrasalmidae). The influence of weight and temperature on the
routine oxygen consumption. Amazonia 4:433-443.

Sanchez A., and Glass, M.L. 2001. Effects of environmental hypercapnia on pulmonary
ventilation of the Southern American lungfish, Lepidosirenparadoxa (Fitz).
Journal of Fish Biology 58: 1181-1189.

Sanchez A., R. Sonncini, T. Wang, P. Koldkjaer, E.W. Taylor, and M.L. Glass. 2001.
The differential cardio-respiratory responses to ambient and systemic hypoxaemia
in the South American lungfish, Lepidosiren paradox. Comparative Biochemistry
and Physiology Part A 130: 677-687.

Sawaya P. 1946. SobrE a biologia de alguns peixes de respiracdo aerea (Lepidosiren
paradox Fitzinger e Arapaima gigas Cuvier). Boletim da Faculdade de Filosofia,
Ciencias e Letras da Universidade de Sdo Paulo. Geologia (Zool.) 11:255-285.

Singh B.N., and G.M. Hughes.1971. Respiration of an air-breathing catfish Clarias
batrachus (Linn.). Journal of Experimental Biology 55:421-434.

Smith H.W. 1930. Metabolism of the lungfish Protopterus aethiopicus. Journal of
Biological Chemistry 88:97-130.

Smith H.W. 1931. Observations on the African lungfish, Protopterus aethiopicus, and the
evolution from water to land environments. Ecology 12:164-181.

Ultsch G.R. 1995. On adjusting metabolic rates for body size. Florida Scientist. 58: 270-
273.

Ultsch G.R. 1996. Gas exchange, hypercarbia and acid-base balance, paleoecology, and
the evolutionary transition from water-breathing to air-breathing among
vertebrates. Palaeogeography, Palaeoclimatology, Palaeoecology 123:1-27.

Worthington E.B. 1932. Scientific results of the Cambridge Expedition to the East
African lakes, 1930-1. Fishes other than Cichlidae. Journal of the Linnean Society
(Zool.) 38:121-134.















BIOGRAPHICAL SKETCH

Ashley Winn Seifert was born on April 6th, 1976, in Syosset, New York. He is one

of three children of Randolph Winn Seifert and Janine Epstein Seifert. Raised in a liberal-

minded atmosphere, he was taught to search for his own answers to the mysteries of life.

Following a whim, he attended Bowdoin College in Brunswick, Maine, where he earned

an A.B. in biology with a minor in philosophy. After graduating, his interests led to a

now coincidental meeting and consequent friendship with Dr. Karel Liem who suggested

an avenue for potential study with Dr. Lauren Chapman. After working as her research

assistant for one year he enrolled in the graduate program in the Department of Zoology

at the University of Florida where he completed his Master of Science degree. He is still

seeking answers and still following whims.