Gas Exchange and Metabolism in the Sirenidae (Amphibia, Caudata) By GORDON RICHARD ULTSCH A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1972 This work is dedicated to my wife, Sandra, for the many years of work to support a student husband, and to my daughter, Julie, who must have wondered at times if she actually had a father. ACKNOWLEDGMENTS I gratefully acknowledge the many people, both students and instructors, who helped with various phases of this study. The list is much too long to give here, but to any who may read this work and who have contributed either mind or muscle, you have my warmest thanks. I am indebted to Drs. Brian McNab, Frank Nordlie, Hugh Popenoe, John Anderson, Henry Prange and Carmine Lanciani for reading and critiqueing the manuscript and for use of space and materials. Dr. Stephen Zam also provided space and equipment necessary for the study. Dr. Robert DeWitt and Mrs. Ruth Smith deserve special credit for supplying my sometimes exorbitant demands, which I hope will prove justified. Mr. Paul Laessle of the Department of Zoology and Mr. Gary Hellermann of the Center for Aquatic Sciences did most of the figures. This work would not have been completed by the deadline without their capable aid. Mrs. Donna Gillis typed the manuscript, and aided in completing the final details of preparation. iii TABLE OF CONTENTS Page Acknowledgments j^ List of Tables v List of Figures vi Abstract ix Introduction 1 Materials and Methods 4 Field Work 4 Maintenance of Animals 5 Surface Area Determinations 5 Metabolic Rates of Submerged Animals 6 Gas Exchange Partitioning Experiments with S_. lacertina 9 Results and Discussion of Field Studies 11 Results of Laboratory Studies 28 Discussion of Laboratory Studies 42 Metabolic Rate, Gas Exchange and Body Weight 42 Cas Exchange Partitioning 65 Summary 74 Appendix 76 Literature Cited 109 Biographical Sketch 112 IV LIST OF TABLES Page 1. Comparisons of pH, temperature, dissolved 0„ and dissolved C0„ in hyacinth and "open" water areas. ... 20 2. Survival of submerged S^. lacertina at 23-24°C in air-equilibrated water (p0„ approximately 155 mmHg) 32 3. Conditions of gas exchange partitioning experiments for two large and one small _S. lacertina 33 4. Gas exchange partitioning between air and water for large and small S^. lacertina under conditions of high 0„ and low C0„ in the water . 34 5. Conditions of gas exchange partitioning experiments in a single, large S^. lacertina 38 6. Gas exchange partitioning between air and water for an individual, large S^. lacertina (1453-1489 g) as a function of concentrations of dissolved 0„ and CO- in the water phase 39 7. Relationships between metabolic rate and weight and 0„ exchange capacity and weight for various groups of animals 51 8. Skin vascularization and epidermal thickness in ranid frogs (data from Czopek, 1965) 55 9. Standard metabolic rate, 0~ exchange capacity, and critical oxygen tension of JS. lacertina of various body sizes (for submerged animals) 60 LIST OF FIGURES Page 1. Comparison of aquatic Vq2 of S^. lacertina submerged for approximately two hours (enclosed area) and for 145 hours 8 2. Temperature and pH at the surface and bottom in "open" water and hyacinths 13 3. CO2 and O2 concentrations at the surface and bottom in "open" water and hyacinths 15 4. CO2 and O2 profiles for different thicknesses of hyacinth cover, times of day, and times of year 17 5. Concentrations of dissolved oxygen and carbon dioxide as a function of thickness of hyacinth cover, time of day, and time of year 19 6. Location of £. lacertina and S^. intermedia within the hyacinth area of the pond during the hyacinth growing season (March-October), as a function of concentrations of dissolved respiratory gases 23 7. Location of P_. striatus within the hyacinth area of the pond during the growing season of Eichhornia (March-October) , as a function of concentrations of dissolved respiratory gases 25 8. Location of Siren and Pseudobranchus within the hyacinth area of the pond during the non-growing season of Eichhornia (November-February) , as a function of concentrations of dissolved respiratory gases 27 9. Surface area of the skin in the thr e species of Sirenidae as a function of body weight 30 10. Aerial-aquatic gas exchange partitioning in a high 02-low CO2 aquatic phase as a function of body size 36 11. Aerial-aquatic gas exchange partitioning in a large S^. lacertina (1453-1489 g) under various conditions of dissolved O2 and CO2 41 vi 12. Predicted maximum size attainable by a 1 g organism whose surface area of the gas exchanger if (f) wO-67 and whose metabolic rate is either (f) Wl.O or (f) WO.75 47 13. Permeability of the skin to oxygen in the Sirenidae as a function of body size 58 14. Metabolic rate and 0„ exchange capacity of submerged _S. iacertina as a function of body weight 62 15. Per cent utilization of oxygen exchange capacity as a function of body weight in the Sirenidae 64 16. Metabolic rate and 0„ exchange capacity as a function of body weight for submerged P_. striatus and S. intermedia 67 17. Aquatic Vq„ of a 44 g S^. Iacertina breathing air and water 71 1A. Metabolic rate of submerged P_. striatus (0.51 and 1.58 g) as a function of oxygen tension 79 2A. Metabolic rate of submerged P_. striatus (2.88 g) as a function of oxygen tension 81 3A. Metabolic rate of submerged JS. intermedia (3.3 and 7.0 g) as a function of oxygen tension 83 4A. Metabolic rate of submerged S^. intermedia (13.7 and 29.6 g) as a function of oxygen tension 85 5A. Metabolic rate of submerged _S. Iacertina (0.36 and 0.56 g) as a function of oxygen tension 87 6A. Metabolic rate of submerged S_. Iacertina (3.0 and 6.5 g) as a function of oxygen tension 89 7A. Metabolic rate of submerged jS. Iacertina (13.7 and 42.7 g) as a function of oxygen tension 91 8A. Metabolic rate of submerged S^. Iacertina (73 and 103 g) as a function of oxygen tension 93 9A. Metabolic rate of submerged S^. Iacertina (178 and 269 g) as a function of oxygen tension 95 10A. Metabolic rate of submerged S^. Iacertina (357 and 541 g) as a function of oxygen tension 97 11A. Metabolic rate of submerged JS. Iacertina (825 and 1310 g) as a function of oxygen tension 99 vii IB. Predicted relationship between total weight and skeletal of mammals required to give M = 70 W "*'5 when metabolism of active tissues does not change with body size (black circles) 103 2B. Metabolic rate of mammals as a function of body size. 105 3B. Relationship between weight of active tissues (W^) and metabolic rate in mammals 108 vm Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy GAS EXCHANGE AND METABOLISM IN THE SIRENIDAE (AMPHIBIA, CAUDATA) By Gordon Richard Ultsch August, 1972 Chairman: B. K. McNab Major Department: Zoology All members of the family Sirenidae (Amphibia, Caudata) are usually found in areas of aquatic vegetation. They are particularly abundant in water hyacinth communities, which were found to have low concentrations of dissolved oxygen and high concentrations of dissolved carbon dioxide. Field studies indicated that the Sirenidae do not select micro- habitats on the basis of dissolved respiratory gases, and that they choose hyacinth areas over open water within the same pond. Sirenids above 800 g are obligate air-breathers at 25°C in air saturated water. Theory and experiments had predicted this weight closely when body size, metabolic rate, and permeability of the skin to oxygen were considered. Siren lacertina was found to adapt to the hyacinth community by breathing air in response to low levels of dissolved oxygen and by eliminating carbon dioxide to the air and tolerating a high blood pC0? in response to elevated levels of carbon dioxide in the water. ix A model is presented that shows the advantage in attaining a large size of a decreasing weight-specific metabolic rate with in- creasing size. It is suggested that metabolic rate is limited by the Op exchange capacity of an organism. INTRODUCTION Animals that utilize both aerial and aquatic gas exchange are particularly interesting subjects for studies of respiratory adapta- tions. Differences within the two fluids in gas concentrations, medium density, types of exchange organs, and diffusion and solubility constants of the respiratory gases contribute to a complex set of alternatives for meeting various respiratory demands. If the organisms exhibit a large range in body sizes and inhabit waters with varying concentrations of dissolved oxygen and carbon dioxide, one can expect to observe various types of adaptations for metabolism and gas exchange. A family of aquatic salamanders, the Sirenidae, is such a group of animals. It is comprised of three species, Siren lacertina, S^. intermedia, and Pseudobranchus striatus. The wide range of body sizes is indicated by the maximum weights of adults collected: S. lacertina, 1698 g, S. intermedia, 45 g, and P_. striatus, 5 g. Considerable attention will be paid to S_. lacertina in this paper, which was studied over a weight range of greater than three orders of magnitude. The Sirenidae are effective air and water breathers (Czopek, 1962; Freeman, 1963; Goin, 1941; Guimond, 1970). They possess highly vascu- larized lungs (Guimond, 1970), and surface to breath air regularly. Although external gills are present, the aquatic mode of respiration is almost entirely cutaneous (Czopek, 1962; Guimond, 1970). Therefore no effort was made in this study to evaluate the role of the gills as separate from that of the skin in aquatic gas exchange. These salamanders are found in waters that offer some degree of refuge, usually in the form of aquatic vegetation. Pseudobranchus has become particularly abundant in the water hyacinth (Eichhornia crassipes) community and is usually associated with some form of dense vegetation. Both species of Siren are also usualiy associated with relatively dense vegetation, although S^. lacertina appears to venture into more open waters frequently, perhaps because its large size makes it less susceptible to predation. Aquatic animals that live in such vegetated areas are usually subject to unfavorable concentrations of dissolved oxygen and carbon dioxide. Data from Lynch et al. (1947), for water samples from areas with various types of aquatic vegetation, indicate that oxygen is usually below saturation levels and carbon dioxide concentrations are elevated. They found that this was especially true for water covered by water hyacinths, which makes this community a particularly hostile environment for aquatic gas exchange. The effects of Eichhornia on dissolved gases was one of the factors considered in the choice of a field study site. The pond used was on the edge of Payne's Prairie, Gainesville, Alachua County, Florida. Approximately 95% of the water surface was covered with water hyacinths, leaving an area of about 1/2 acre of "open" water (meaning the surface was not covered by Eichhornia, although the submerged aquatic Ceratophyllum was often dense) . All three of the species of Sirenidae were present in large numbers, especially P_. striatus. The two available aquatic micro- habitats provided an excellent field laboratory for studies of behavioral adaptations associated with habitat selection. The purpose of this study was to investigate some of the be- havioral and physiological adaptations of the Sirenidae that enable them to cope with the respiratory stress placed on them by the low levels of dissolved oxygen and high levels of dissolved carbon dioxide associated with water hyacinth communities. As the extent of the effects of Eichhornia on the aquatic microenvironment were rather poorly documented, it was also necessary to study certain environ- mental parameters in detail in order to properly plan and evaluate the laboratory studies. Temperature, PH, dissolved oxygen and dissolved carbon dioxide in both the "open" and hyacinth-covered portions of the pond were the factors chosen for investigation. MATERIALS AND METHODS Field Work Temperature, pH, dissolved 02 and dissolved CO2 were measured in the hyacinth mat and "open" water at the surface and bottom. Temperature was measured to the nearest 1/2°C. Water samples were drawn through a water-saturated cloth, which excluded most of the detritus, into a glass tube that was stoppered at both ends with rubber stoppers. The samples were placed in an ice chest and re- turned to the laboratory, where they were warmed to 25 °C. A Radio- meter PHA 27 pH meter and Gas Monitor, calibrated at 25°C, was used to determine pH, pCG^ and p0£ for each sample. The pC02 scale could only be read down to 7 mmHg; any readings beloxj this value were assigned a value of 3.5 mmHg (7 ppm) . Partial pressure readings were converted to parts per million by weight (ppm) by multiplying the pC02 readings by 1.96 ppm/mmHg and the pC>2 readings by 0.053 ppm/ mmHg. It was found that some gas was being exchanged between the rubber stoppers and the water samples between the times of collect- ing and measuring. This change was predictable and was corrected for by a table of correction factors derived from observing changes in water samples of known concentrations treated in the same manner as the field samples. The environmental parameters were measured throughout the year as a function of time of day, depth of the water column, season, and thickness of vegetation. All measurements intended for use in describing annual cycles were taken on sunny days to prevent the effect of cloud cover from masking the effect of season. In order to determine the location of the animals in the pond, extensive collections were made with a dredge in the hyacinths and with seines in the "open" water. Water samples were taken prior to all collections. In addition to animals collected at the Payne's Prairie site, experimental animals were also collected from the River Styx in Alachua County and from a culvert passing under SR-121 near Biven's Arm in Gainesville. Maintenance of Animals All animals used in the laboratory work were maintained at least two weeks at 25 + 2°C with a 12-12 light-dark photoperiod. Animals were not fed within three days of the start of any experiment. If fecal material was visible in an experimental chamber at the end of an experiment, the results were discarded because of the possibility of oxygen consumption by decomposition of the feces. Surface Area Determinations Surface area determinations were made initially by anesthetizing animals with MS-222 (tricaine methanesulfonate) and wrapping them in aluminum foil. The foil was cut to fit the body contours, unrolled 2 and the outline traced on graph paper ruled in mm . Surface area was determined by counting squares. It was later found that a good approx- imation could be made by treating the animal as a cylinder from the head to the vent and then considering the remainder as a triangle with a base equal to the distance from the vent to the tip of the tail. Six com- parisons of the techniques were made, with animals of 21 - 690 g. Con- sidering the square-counting method to represent the actual surface area, the geometrical approximation averaged an error of only + 1.8%. Most of the additional determinations utilized the geometrical technique. Metabolic Rates of Submerged Animals Animals used in these experiments were placed in the experimental containers with access to air the night before use to become accustomed to the chamber. The type of container used varied with the size of the animal, but was usually a jar or Erlenmeyer flask with a volume that would result in a drop of p0~ of about 10 mmHg/hr when the animal was in a relatively inactive state. The water was changed the next day to eliminate any accumulated fecal material, skin, etc. Well water was used rather than distilled water, to alleviate osmo- regulatory stresses. Controls indicated the biochemical oxygen demand of the well water was insignificant (decreases in p0„ of 0.0 - 0.5 mmHg/hr) . At the start of an experiment the container was filled with water supersaturated with 0„ at 25°C. The animal was then submerged in the filled and sealed chamber and allowed to respire for about two hours before an initial oxygen determination was made, in order to deplete the lungs of oxygen. Figure 1 shows a test of this assumption. The enclosed areas represent metabolic rates measured during experiments that were preceded by the two-hour submergence period. The scores indicate metabolic rates of submerged animals after six days of submergence. Presumably, ail available 0^ in the Figure 1. Comparison of aquatic Vq2 of jS. lacertina submerged for approximately two hours (enclosed area) and for 145 hours. Scores represent metabolic rates after 145 hours of sub- mergence. 6CH 50 40- 30 o> ^m 20- O O 10- X= I3.3g * = l2.9g o= I |.2g S.lacertina ifr -i 1 i 1 — — i 1 1 r- — i p i 1 1 1 CO o o z LU O 30H > X o ?Q- 10- X = 40.0g S.lacerfina y/i 1 r 15 35 -i 1 1 1 1 1 1 r 55 75 95 p02(mm Hg) 115 135 155 lungs would have been utilized by this time. The metabolic rates after 145 hours of submergence are similar to those after only two hours of submergence, indicating that the animals are functioning only as water breathers in the submerged metabolism experiments. Metabolic rate was determined by allowing the animal to consume oxygen from the water for at least one hour, and converting the decrease in p0„ to oxygen consumption in yl CL/g*hr. The animal was considered to have exhibited that metabolic rate at an oxygen tension equal to the average tension for the interval. Determinations of metabolic rate as a function of oxygen tension were made for S^. lacertina of average weight groups from 0.36 g to 1310 g, for S^. intermedia from 3.3 to 29.6 g, and for .P. striatus from 0.51 to 2.88 g. Gas Exchange Partitioning Experiments With S_. lacertina Gas exchange partitioning experiments were conducted in two- section plexiglas chambers. The lower portion of a chamber contained the animal and was filled with water; the upper portion was filled with air. A small opening connected the two sections to allow the animal to air breathe. The surface of the water at the opening was covered with a 1 cm layer of heavy duty paraffin oil to minimize gas exchange between the air and water. The chamber was painted a dull gray except for the area above the breathing hole. Observation showed that the animals located the breathing hole much more readily when it was the only source of light. In practice, the animals usually stayed near the opening, and air breathing merely necessitated raising the upper third of the body to be able to protrude the mouth into the aerial chamber. 10 Gas concentrations were set at the desired levels by bubbling CC>2> O2, or N2 through the water. Gas exchange between the air and water in controls was nil (0.006 vol%/hr for CO2 and 0.022 vol%/hr for O2) • The animals were acclimated to the experimental conditions for at least 24 hours before the start of a series of experiments by placing them in the chamber at 25°C and with the appropriate set of concentrations of dissolved gases and leaving the aerial portion open to the atmosphere. Immediately prior to an experiment, the water was changed and the desired gas concen- trations re-established in the water. One to two hours were allowed before the aerial portion was sealed from the atmosphere, after which the tank was completely submerged in the water bath at 25°C. Generally, two experiments were run per day, after which the gas concentrations would be reset to the desired levels, and the animals would be left in the chamber overnight. The same procedure was followed for the next day and thereafter until a series of experiments for a particular set of conditions was complete. Any acclimation was apparently complete at the end of 24 hours, since there was no significant change in metabolic rates or partitioning between first and final days of experimentation for a particular set of conditions. Changes in the vol% of CO2 and O2 in the aerial phase were measured with a Scholander 1/2 cc gas analyzer. Dissolved O2 was measured with the Radiometer PHA 2.1 and Gas Monitor. Dissolved CO2 was determined by calculation from an RQ of 0.91 (SE = 0.03) reported by Guimond (1970) for S^. lacertina at 25°C. All gas volumes are reduced to STPD. RESULTS AND DISCUSSION OF FIELD STUDIES Curves were constructed for pH, CO™, 0„ and temperature for a 24-hour sunny day of each month- Measurements in the hyacinth area were made during September and October of 1969, and for all other months in 1970. "Open" water measurements were made monthly from February, 1970, to January, 1971. Each curve was averaged over the 24-hour period, and these average values were plotted as single points. Figure 2 shows the results for temperature and pH, and Figure 3 for dissolved 0,- and C0_. Figure 4 gives profiles for 0? and C0„ in summer and winter, and as a function of time of day. The profiles reveal that depth is not an important factor in selection of a favor- able respiratory microenvironment for a sirenid in the hyacinths, since once the depth reaches only 20 cm, dissolved gases remain rather constant. Therefore, the bottom measurements in the hyacinth area of the pond may be considered to be roughly equivalent to the general con- ditions existing anywhere within the water column under a hyacinth mat. However, Figure 5 clearly indicates that horizontal movements from hyacinth areas to open water can have a marked effect on the respira- tory microenvironment. The growing season for water hyacinths in the Gainesville area is March-October. It is during this period that conditions become partic- ularly unfavorable for aquatic gas exchange in the water covered by hyacinths. Table 1 shows the average 0„ for this period to be only 11 Figure 2. Temperature and pH at the surface and bottom in "open" water and hyacinths. Each point is the average value for a 24-hr sunny day for a given month. 30- 25- 20- 15- »-6.2 6.0 ^5.8 5.6 •5.4 ' h5.2 5.0 h4.3 Q. ,--o O SURFACE TEMP. @ BOTTOM TEMP. A SURFACE pH A BOTTOM pH -r / ' » / // x A / 6/ HYACINTHS \s \J & 6.0 5.8 5.6 •^ 4 L-5.2 •5.0 -4.8 a. ~i 1 1 ! n ; 1 1 1 1 i i i- JFMAMJJAS-ONDJ MONTH Figure 3. CO2 and O2 concentrations at the surface and bottom in "open" water and hyacinths. Each point is the average value for a 24-hr sunny day for a given month. IIO-i 100- 90- 80- 70- 60- 50 40- 30- — 20- E gioH O SURFACE Og BOTTOM 02 G SURFACE C09 m BOTTOM CO, CM 10 9 8 7 ■6 5 4 3 -2 ■I •0 HYACINTHS s £ Q. Q. ****** CO o MONTH o U cd 01 <1) . CO (0 l*-t -. <4-l rt •H T) TJ U-i U O O <4-l CO 0) to B a) •H rH 4-1 •H iw •» o M u o CM o O X TJ 4-1 C 0 cd •H O CM C8 O >> C_> ^3 u ton •H P>4 OUJ o < u. ce 3 in (wo) Hld3(J O 5 CD £ o CO Figure 5. Concentrations of dissolved oxygen and carbon dioxide as a function of thickness of hyacinth cover, time of day, and time of year. Station 1 is "open" (no hyacinths, but considerable Ceratophyllum in summer), Station 2 is "thin1 hyacinths (no accumulation of detritus in root mass) , and Station 3 is "thick" hyacinths (larger plants character- istic of a mature mat, with considerable accumulation of detritus in the root mass) . 100 STAIION NUMBER 20 CM 00 O C O -a 0) > •H O 00 O 0) CD X! T3 CO s n « a) cmo rS O -H td O U CD I O > X. e cj cu & 4-1 rl > CO Q) CU cd rd M a) s-i P P O o ,c u-< = o •n co c c CO o CO JS o w 14-1 o u c c td'H o P CJ CO S « (fl o >-> cu U X CO Pi w H S3 w o C/3 S3 H 53 M C_) o 53 R) CU 4J o o I X! o Rl a ■O O !3 u RJ CU m CM CO CM m m in VO m CM co o O o en oo m CN i — CO in A CM o CM vO in m in -a- CO cm in CM co cm CM m m CO o Ol CO CO vO in co o CM in m in in CM VO CM m CM o C7\ m co m CM -3- co m x> cd H o /•"» B ^^ >^ B p u p •— v p o • ^■s P B N-X N_X P. S B P N-^ p p c • CU Pd P CM v-^ O P E-i p \-^ o c_> fi CM CU B IS B CM B o B H o p. o o o o o ■u 4-1 4-1 4-1 p 4-1 p ■U p 4-J p 4-1 o o o o o O o o H w H M H « H pq 21 0.6 ppm under the hyacinths, while the CO is 63 ppm. For an organism dependent upon aquatic respiration, the "open" water is a refuge during this period. In fact, larger fish such as the centrarchids are found only in the open water during the summer, while they are also in the hyacinth areas in winter. Some fish, therefore, migrate horizontally due to respiratory stress. Extensive collections were made in both the "open" water and hyacinth areas to determine where sirenids were located. Regardless of the time of year, no P_. striatus and only one Siren (probably lacertina) were collected in "open" water. The hyacinth areas are definitely the preferred micro environment. To determine if the animals were selecting areas within the hyacinth portion of the pond, successful collecting attempts were paired with the corresponding gas concentrations at the site of collection and compared to the set of all available gas concentrations i>rithin the hyacinth portion of the pond. These results are shown in Figures 6-8 as a function of the size of the animal (for Siren) and the time of the year. In all cases, the animals are not avoiding any particular gas concentrations, regardless of species, size, or time of year . The conclusion drawn from the field work is that aquatic gas exchange considerations are not limiting in habitat selection in the Sirenidae. This evidence is in agreement with the results of a pre- vious study for P_. striatus (Ultsch, 1971). A probable corollary is that the Sirenidae must depend heavily upon aerial respiration as a source of oxygen, and perhaps as a method of carbon dioxide elimination. ti "d o •rl X •rl r-l QJ 4J 4-1 o ,c •rl 14-1 o CO 4-1 M s O c 0) 3 ?►. 4-1 X> >. cd "4-1 X> cd O rH 0) 4-1 4-1 !-i cd TJ CNJ CJ ti cd cu o O a> CO CO c_) 1 1-4 £i cd o Xi u 4-> i-i TJ CJ 3 ti M o c M cj •H •"N ti cd cd G U JH a) S O cd a) CN CJ >> -Q CO o rH ^ O cd rH QJ 4-1 a) •d cd TJ OJ CJ n a) cd JS O <: > r< B 4-1 1 r-l o Xi O 4H CO ti O • CO C •H M CO CO »* o * cd a) •H >> •H 4-1 s CO td rH 4-J •H v-^ cd a) cd s 00 CM > C c o •H •H cd o >. 4-1 B •H CO u CO CJ !-l *d cd o ti CD QJ OJ eg 4-1 o CX 4-J B CO cd •H co OJ U H 4-1 0) t> <0 to •H cd M 4-1 C Pi j-i 14-1 c •H co 4-1 •t o ■H & a) ti /-~N O n a) • QJ • M CJ ■U > W 00 td ti O •H 0) o cq CO -a JC > o «*-/ 3 ti 4J rH r-l cd ti o 4-1 6 CJ •rl CO ti o X cd O CO > -a a) o 4-1 Xi J-4 X> CO rl <4-4 a. C 0J a) o OJ tJ o o -C M ti •I-I cd 4-1 CO cd 4-1 rH C CO cd 60 o 0) •-s C • a •H C • •rl 03 •H 4-1 •rl 4-4 B • m cd rH 1-1 U s~\ m 3 M 3 OJ to o T3 4-1 -a C/3 4-1 c c a) v^> QJ o c T3 QJ U 60 •H ■o en O O ro i O m en O O rO A t- <^J -o =3 O =J O co|co co co co b a a n -a n <\1 ■ o £ a. a. >- X o (uidd) 3QIX0IQ N08dV0 Figure 7. Location of jP. striatus within the hyacinth area of the pond during the growing season of Eichhornia (March- October) , as a function of concentrations of dissolved respiratory gases. Enclosed areas are as in Figure 6. Rstriatus e Surface o Bottom OXYGEN (pprn) Figure 8. Location of Siren and Pseudobranchus within the hyacinth area of the pond during the non-growing season of Eichhornia (November-February) , as a function of concentrations of dissolved respiratory gases. Enclosed areas are as in Figure 6, except for determinations being for November- February. E ex o X 60- 50- 40- D ,' \ A "™ "" "■» A\ \ A Ai. A S.lacertina A Surf.i cn _- ABot. p5°-30°S ■ Surf.i . ' n a , ) >300g 3 0- 20- 10- \ / £\ v •» v> \l A o Bot. y u \ A ) Oht- 1 1 i 1 1 1— — 1 1 1 1 1 1 1 1 1 0 60 50- 40 10 S.lacertina ^ o Surf. o Bot. S. intermedia ><50g Q 20- Z O J£ 03 OFF U-4 O U) a) •H O a) P. CO > 4-1 T3 O 4-1 JO O m cd o o z z> UJ 2 CC UJ t- z t- < »- V) UJ o _l a V) 10 CO en en d d Q UJ 2 DC u fO 10 1^ q lO o O lO q JO in cvi q lO in d o d o i ( 2 W 3 ) o o V3dv aovddns ocn I- x a] >- Q O CO iD O 31 merit of regulation. Some comparative use might be derived from con- sidering the minimal metabolic rates given by the 10/90 line, but since the animals were relatively inactive in the metabolism chambers, the mean metabolic rate while the animal is regulating is probably the most realistic estimate of the standard metabolic rate while submerged and is the value that will be used in all further discussions and calcu- lations. Pseudobranchus and small Siren can survive submerged in air- equilibrated water indefinitely. Siren lacertina of various body weights were submerged at 23-24°C to determine if there was an upper limit to the size at which a submerged Siren could survive. Table 2 shows that animals 800 g or larger cannot survive without access to air, while smaller ones can function as water breathers only. This observation led to an investigation of the effect of body size on gas partitioning in S_. lacertina. Two large individuals (the largest one very near the maximum size for the species: Conant, 1958) and one relatively small individual ware studied. The conditions for these experiments are given in Table 3, and the results in terms of absolute values and percentages in Table 4. It is possible for a particular partition to be significantly different between large and small animals without the absolute values being different, depending on the absolute value of the total gas exchange. Figure 10 presents a statistical interpretation of the absolute values for each partition for comparison to Table 4, which gives a statistical presentation of the percentage of the total gas exchange represented by each partition. 32 Table 2. Survival of submerged S^. lacertina at 23-24°C in air- equilibrated water (p0„ approximately 155 mmHg) . Animals removed at 336 hours were in no difficulty. Other times indicate the first observance of death, and are therefore maximum estimates of survival time. Weight (g) Survival time (hr) 83 336 + 191 336 + 277 336 + 463 336 + 670 336 + 805 114 998 30 1046 48 33 Table 3. Conditions of gas exchange partitioning experiments for two large and one small S^. lacertina. The duration of the runs was 2.75-5.00 hrs (aerial) and 3.08-5.50 hrs (aquatic) for large animals, and 3.25-6.25 hrs (aerial) and 2.00-6.33 hrs (aquatic) for the small animal. All experiments were at 25°C. Values are given as 3c + 2 S.E. Dates of Weight Final Air Mean Water Mean Water Experiment (g) 02 (vol%) p02 (mmHg) pC02 (mmHg) 2/26/72- 1489 15.6 145 less than 7 3/4/72 ± 26 + 1.0 ± 8 2/22/72- 1182 15.2 144 less than 7 3/2/72 ± 9 ± i-4 + 3 3/28/72- 89.6 18.3 143 less than 7 4/3/72 + 0.7 + 0.8 + 4 34 cd c •H ■U CO y* 4J H rH -* CO <3 CN oo 00 CO 3 H O cd a\ a\ CTi rH r~~ CT> o rH O 4H -H • * • • * • s~\ • • ca o u u ►J 60 r-- •->> rH -K ,o .O /I B) Cd O + vo + H + «>• + |rH + 1 +1 •,< a) II ^j a\ 1 rH I CO 60 V4 CO 00 CN rH 1 u-) vO in r^ CO r^ ttj S3 O VO CN VO i-H CN C c T3 J3 Cu U cd cu C cd CO •O cd H CO 4J CO •H Vj C 4-> 4-1 co co cd i-i .h c 60 H 11 O 4J .•* .or* .&> .r^ 0) QJ 0) J3 O CN + cr> + CN + CN + av + + 1 rH u x -a CO I 1 1 I J) tJ BIN C r-l in vo rH CN 00 rH CTi ro CN r^ B-« £ CO IT) Cd r-l rH in a\ rH vO CN -3- vO CN r~» in C 3o\ • n_^ • v^ • s-^ • N.^ • • er> "O 'H H O O LO -sT O VO LO o\ o r^ C 4J Xi O cd cd 4J >-l C_> ••> 3 w •H ^-s C a* O 4-1 cd & • cd cd 1 cd o W co C .-I • g ^ ►J CO CO CO o o 00 CN 00 CN ■vf rH co t) co u-i o 00 vO CN O r^. cd & CjCM£ • • • • • • 0 4-1 Cd 4J VO 60 CO •-s r-l ^-n O /-s rH ^-n CO CN •H CO + .vT* ."1 r-. .rH C J2 CN CO II CTi + 00 + CO + |-"« rH rH VO CTv rH CN CN rH VO CO o 60 •H 60 CO »• • Vw/ • -^ • v— ' • v^ • • S-i C -H r-l CO 60 cti m m vo VO to r-^ -h 4J •HH OWH C CO 4-> 3. -H co 60 -r-l 4— > II }_l u C TJ CO -H O CO cd C V4 4-1 13 4-4 4-t ,c o cd n 4-1 o u cd •• co CN CN CN •H a co a. 60 4-> CM o CN o CN O 13 CO U CO iH O o O u O U U) 3Xoi 3 • > •> •> •> •> • > >. CO -O H O CO CO rH cd c cd cd fflrl ^ • •H <■ l-\ O ■H 4-1 •H c/d| -u • CM CO Q) C o 3. 60 -H O o H •H Cd CM s 4J rH O o •H O rH -3 •i 1 3 tH id CM O cd 3 o CJ 3 cd T3 ja 1-1 •H CM 60 co > O •H r3 •H x: 4-1 13 13 O 3 a> rH •H > rH rH rH cd rH 6 ° cd cd to 4-) co cd U r4 -H o O T3 00 4-1 14-1 *H II vO !-i o CD S II 4-1 CO cd 3 &s & o • •H 4-1 •r4 Cd cu cd 4-1 rJ CJ 1-1 CO cd •H 0) cd cd 4-1 i-< 3. CO cd CU •~N CO 3 S 0) 60 3 O4 60 rH cd CO 3 ON cd 4-) cd oo > u 3 .3 vO vO rH r- o o CO •— N +|rr O CO rH r^ • v_ ' on uo iH r^ CM o o oo CM y\ vO <* r-. co « V— ' "0 00 d ^-n + |CM O CO O CN m vjo CM CM O 00 CO r^- co r-» O oo oo + |co CO vO CTl rH un vo CM CM O V4 CO O vO IT) | r^ cm • ^— * o ~* CM d vO I to m rH 00 O 00 O -~N + | + 1 o o 00 -d- CM +1 m rH CN CM o CO +1 CM CO m CM CM o CN 00 CO +1 rH oo 00 in + 1 CM CO rH vO CN CM + 1 O VO CT> rH CN + 1 ON o CO CM CM O CJ •H 4-1 cd 3 cd 4J o Figure 11. Aerial-aquatic gas exchange partitioning in a large S^. lacertina (1453-1489 g) under various conditions of dissolved 0„ and C0„ (see Table 5) . Results are shown as x and the 95% confidence interval. Non- overlap of confidence intervals yields statistical difference in the mean values at the 95% level. Numbers in parentheses are sample sizes. o •> < U < o •> < O < A B C D A B C D -".3) -J (6) -U6) 14 16 13 H(6) 20 • h I 1 1(6) 4(7) 5 24 4(8) t o •> < H O A B C D 4(3) 4(6) 1(6) -I (6) 15 19 23 i 27 i o o L'J A B C D 4(8) 4(6) (6) 8 10 12 -i vo> 14 o ->° < o A B C D f— + 4(3) 4(6) K6) 4(6) 6 12 18 21 o o •> < A B C D 4(6) 4(0) (6) 4 (6) !3 17 2! i— 25 A = HIGH 02,L0W C02 B «=HIGH 02,HIGH C02 C = LOW 02,HI6H C02 D » LOW 02 , LOW C02 02 CONSUMPTION OR C02 ELIMINATION (//l/g-hr) DISCUSSION OF LABORATORY STUDIES Metabolic Rate, Gas Exchange and Body Weight Hemraingsen (1960) has shown that the metabolic rate as a function of body size for a large number of organisms can be expressed by (1) M = k W0,75 where M is the rate of oxygen consumption and W is the body weight. This equation was the result of interspecific comparisons, and Kleiber (1961) has argued that the relationship also holds for large intra- specific ranges of body weight, although supporting data are scarce. Why is the power function of weight in Eq. (1) 0.75 rather than 1.0? One possibility often mentioned is a disproportionate increase of non-metabolizing (or low-level metabolizing) tissues with an in- crease in body size. The W in Eq. (1) is actually W , the whole body weight, which is composed of W (active tissue) and W (relatively A. J- inactive tissue) . Even without any actual change in the metabolic rate of the active tissues, a disproportionate increase in W with increasing W would result in an apparent decrease in weight-sepcif ic metabolic rate. Appendix B shows calculations that indicate that only a minor percentage (at least for mammals) of the decrease in weight- specific metabolic rate with size can be accounted for as apparent and that the power of W in this relationship is close to 0.75 (about 0.77), even when disproportionate increases in W are taken into account. 42 43 Another explanation has been offered by Kleiber (1961) which considers heat exchange. He points out that a 60 g mouse with the same weight-specific metabolic rate as a steer would need a 20 cm thick fur coat to maintain its body temperature if the air tempera- ture were 3°C. He uses this example to point out why it is advantageous for mammals to have an increasing weight-specific metabolic rate with decreasing size. However, this cannot be considered the cause in the general case, since poikilotherms, which do not regulate their body temperature, and small aquatic organisms, which would obviously have no problem in dissipating heat, also have metabolic rates proportional to W . Therefore it would appear that insulation is adapted to U * , rather than vice versa, and some other factor is responsible P M 1 IT0'75 for M = k W Hemmingsen also showed that unicellular organisms, poikilotherms and homeotherms have rates of metabolism that are quite different for a given weight. Poikilotherms had metabolic rates about eight times those that would be predicted from considerations of the metabolic rates of unicellular organisms, and homeotherms had metabolic rates about 28.6 times those of poikilotherms of the same size. However, since the poikilotherm data was for 20°C, and the homeotherm data for 39°C, we can say that the poikilotherm values (considering Q _ = 2 and the temperature correction to be about 20°C) should have been some four times higher than shown. This means that homeotherms generally have metabolic rates about seven times greater than poikilotherms of the same size. Hemmingsen then argued at considerable length that higher levels can be directly related to the increase in respiratory 44 surface area per unit weight that occurs as one follows the phylo- genetic sequence from unicellular organisms to homeotherms. This viewpoint has been supported by several workers. Anderson (1970) showed that for two species of spiders of similar body weight, there was a close correlation between metabolic rate and book lung surface area. Tenney and Remmers (1963) compared the manatee with the porpoise at approximately the same body weights. The porpoise had the greater metabolic rate and the greater lung diffusing area. Whitford and Hutchison (1967) demonstrated that lungless salamanders have lower metabolic rates than salamanders of the same size with lungs (see Table 7 for a re-interpretation of their data) . These observations suggest that the metabolic rate of an organism is correlated with its life style, and that the surface area of the gas exchanger will be adapted to help meet the metabolic demands for oxygen. I suggest that there is a limitation to the ability of an organism to obtain oxygen, and that this limitation is a function of body size and is responsible for organisms showing a decrease in weight- specific metabolic rate with increasing body size. The surface area of the gas exchanger is one of the two factors determining this 0. exchange capacity; permeability to CL is the other. A model will illustrate this point. Assume that the surface area (SA) of the external gas exchanger is the only factor determining the rate of oxygen delivery to the tissues. The equation relating surface area to volume for objects of similar shape but varying sizes is (2) SA = kx W°-67 where W has been substituted for the volume by assuming the average 45 density of an organism to be 1.0, and where k depends on the shape of the object. The exponent of W in Eq. (2), which relates weight to SA and therefore to the ability to supply 02, is less than the exponent of W in Eq. (1), which relates weight to oxygen demand. Therefore, as an animal increases in size, the demand for 0„ will increase more than the ability to supply it. At the theoretical maximum size, the demand and supply of oxygen would be equal. The model is illustrated in Figure 12. For an immature (small) animal, the ability to supply 0_ under the conditions set above must be greater than the demand expressed by the standard metabolic rate, since it does grow to a larger size, and must also maintain some scope for activity regardless of size. For some given differential in p0„ across the external gas exchanger, assume the permeability to 2 oxygen to be 1 cc 0„/hr*cm and constant with body size. This makes the 0„ exchange capacity (E ) a function of surface area of the gas exchanger only. By choosing a starting weight of 1 g, the values of k and k.. in Eq. (1) and (2) will be the antilogs of the y- intercepts of these equations when they are plotted in logarithmic form. For convenience, points A and B are chosen to give values of 1.0 cc 0„/hr«g 2 for k and 1.74 cm /g for k.. . If the power of W on M in Eq. (1) were 2 1.0, then for a 1 g animal, M = 1.0 cc 0„/hr and SA = 1.74 cm . The same values are obtained for a 1 g animal if M is proportional to w0-75. The capacity of the external gas exchanger to take up 0„ for a given differential in p0„ is (3) Ec = SA (P) CO e £ cd u ■u cu 6 cu o. • CO 6 •-N •H T3 4J C CU c X cd co cd CU 00 o •P VI ja PQ o & cu TJ CU 00 TJ c co d cd *~s T-i cd < CO cd r^ a vo co M :>> « w (U ,£> o C 4J 12 ■H CU O CJ rH /^N PL. t-l > cd v^ •H F! • t>0 •H CO U") td •H r~- CU +j • U j-j VI o cd cd CU & 60 CN CU c #"\ o N cd •4-1 •H .C >*-s O CO o 4-> X u 6 cu o J-l 3 cu a CO o 60 •H cd • C « 00 iH cd cd S J= B (U o .13 --^ X -a •P m cu cu ^^ 4J cw CO O o Vi cd •H cu 60 -a cd X a) cu •p or combining constants, (6) rL = k W0,75. This means that if the increase in the product of radius and length is greater than the increase in metabolic rate, a cylindrical animal can grow indefinitely large. When a sirenid is submerged, it is essentially a cutaneous breather. Figure 9 indicates that the larger sirenids do not change their skin surface area with increasing body weight in a manner different from the relation predicted in Eq. 2. A more general statement regarding the relationship between the power functions of weight on surface area and on metabolism can be made: there will be no limitation to body size due to limitations of oxygen supply if the power of weight against 0_ exchange capacity is greater than the power of weight against metabolic rate. Since the surface area of the gas exchanger is an important determinant of 0„ exchange capacity, one might expect the relationship to hold for 50 it also. Table / presents a number of values for these two exponents from the literature for various groups of animals. In all cases except fresh-water turtles, the exponent of weight against surface area is equal to or greater than that of weight against metabolism. While this does not prove a cause and effect relationship between E and M, it certainly supports the conclusion that the two are closely correlated. One might ask why the exponent for weight vs. the exponent for the E -related factor (usually surface area of the gas exchanger) is usually greater than that of weight vs. metabolism. It is possible that the permeability of the external gas exchanger to oxygen de- creases with body size in many organisms, especially those using the skin as a respiratory organ. Table 8 indicates that this is the case for lar^.e ranid frogs, due to an increase in the thickness of the epidermis without a compensatory increase in the vascularization of the skin. Since E is both a function of surface area and permeability, a decreasing permeability with increasing body size could be partly compensated for by the increase in 0„ exchange area being greater than the increase in metabolic rate as body size increases. The permeability to oxygen of the gas exchanger in a living animal is difficult to measure directly, but it can be inferred for the skin of submerged sirenids from the data of Figures 1A-11A. When 0„ tensions are high and an animal is regulating oxygen consumption, permeability will vary according to oxygen needs. But as the 0. tension in the water falls toward the critical 0„ tension, whatever adjustments can be made to increase the effective permeability of the skin to 0„ will come into play. Such changes could include 51 0 cd >. 4-1 •H CJ cd ft cd o 0) 60 c cd .C y x cu CM O *3 C « CO •H 0) a ca qj to 4-1 rH rt cd y< e •H a a •H Cd iH O >4H X> O cd (1) 0 to ft 3 O a u CU 60 ai & 03 4J 3 QJ O -Q -H IH co cd ft > •H X M CO O d ih o •H 4J cd 60 X! 60 •rl CU & flj O Ol, a W QJ CJ Vl 3 o co "3 a) w td rH ^ QJ O Pi 4J I CJ Oct to W in > B I « Cd -rl 4-1 rH CU O X) 4-1 CU CJ 4-) 0) 4-J cd a 0) w •H QJ CJ •H CI) O P. qj CO a. to p a) X > CO •H CD 4-J CO CO C a) B • • B o oo o r^ vO u a\ o\ cd rH CD Pi 1 ocd W CU g 4J a) cd oo B C •H cd 60 X Pi B o l-i o o cd o UH ,0 r4 3 VH CD O to U cu -a •H ft CO P o CO U ' cu c_> p o <; cn o 1° o CN CO CM m CM 30 VO CTi .3 O • o c cd U O H-l CM I to tu •H u cu P CO c o to to •H O -H 4J K 3 cd K > o oo o I CO cd u 3 3 CO CU •H CJ cu CO m cyi P o CO • •H O > ° cd in Q cm vO 60 O m I CM A' CD P. O N CJ uh o cd j-i cd • T3 CO 0) B -H O CJ U CU m ft to CU CT> ■U rH cd rH 3 • o m rH VO cd cr. O H i m oo l 1 o I I I I >> u cd . cd u o AJ CO cd cu u o ft U-l cd o a H 4J -H cd cd 60 ft u-i 4-1 C CO Vl o cu cu 3 4-1 r-H U CO 60 CM I m VO CT> rH m M • cu r-. a. vo o CT> N ^4. u « HH C o o CO • cd •H CJ 4J JS o cd cj in TJ 4J rH 3 B X 4J o cd r4 •3 14H • a co CO cd cu ■3 QJ •H CU ■H X) o 4-1 a r4 0) cd QJ o a. rH p. 14-1 CO 3 CO 4J a •H X rH a rG -H cd QJ |3 CO c_> H * ■ M CJ S S I co 1 o td -h 1 Pi 4-1 rH 1 QJ o a *> XI 0) 4-1 QJ E QJ Vl 3 co o nj W aJ a QJ I X) QJ c •H 4-1 c o o r-- QJ rH X) fO H 4-J OJ 00 CO OC 0 C vO •H CO « X Pi CO O rH l-i OJ 1 ft N m ft-H • <; to o ft 0 S-i (.0 AS OJ o N U CO CO 4-1 QJ tO •H X> O QJ | P. O OJ u 4-1 C QJ XI > QJ QJ 4J CO CO rH 3 • QJ CO % o r- OS rH A >s 0) C c Q) H X) C cO • m >> QJ QJ •H C O C 01 QJ a. H OJ 00 OS ffl CO 00 >. QJ U U O CO 4-1 CO 0) u a -H CO ft.4-1 CO J-i Ai QJ 3 CO U CO a •H CO •H 43 •H I N Vi C 00 »-n u u rH CO CO 43 OJ •H Vl rH 4-4 3 •H 5 4-1 Vl CJ CO s-* O •> 4-1 4C o CO CO 43 OS fH > CO o S s c rH •* VD QJ Vl rH QJ 0) o VI CM 43 CO CO •H ^ or> • > 4-1 4-1 QJ 3 cu • rH /-n M CO O Vl o >> CJ ■H CO >> 4-1 4-4 •H 4J •H QJ •* *0 B rfl e • Vl u 3 Vl ON O CO CO XI o c_> CO XI Ph •H QJ rH Vl QJ a qj & o > QJ 6 C • M-l Vl 4-1 H O 4-1 B o CO •> u ^x V|— | CJ m a o rH 0) • o . OJ 1 « QJ u o 3 rH m vO 14-4 VO Vl • CO vO Vi 4-1 -J PC CO CN in o in vi 00 CM "I Vi ^^ OS • CT> o vO as o CD X) 4-1 4-1 rH CO u rH CJ OS 4-1 rH u as 3 C 0) cO OJ o rH CO \D Vi cO •> -H CM * «t » n •» • as QJ CO t>0 CJ CM co co 00 n CO 00 CO CJ rH 4-J Vi -H 0) Vi QJ 1 Vl QJ Vi tO a VI QJ 0 ft QJ CO •H oj ex. oo 0) -H QJ Vi •H 0) •H o M o CO CO CJ J3 CO rH -a ° 13 QJ 43 CO CJ 43 O CN Vl Vi >. 3 QJ pi c PC a) c QJ •H o CO 4-> ft •H CN 4-1 •H ft •H — ft •rl a. o 3 ■H W w CO & CN CO & CO |3 o CO s DO 4-J a a 00 43 10 •H 1 00 1 r^ 1 • 1 o OS o rH rH r^ OS 1 r~- 00 oo • r^. 1 ■ • • o • 1 o o o 1 rH O o 1 0) o 4-1 Vi 1 1 1 1 1 3 1 1 1 1 1 1 1 1 1 gill s area o o • • • • 1 cn o o o rH o rH 53 CO •H ^ IH CO O > u Oi o P-. o u -a a cd a) o 3 o a) B) rH rl 01 O a! u I o o cd W 4-1 6 I ca cd t-i 4-1 rH a o o 0) rH H 3 o m o\ cd O 6 o n • <4-l CO v_y a) iH o CU ON p. t-l co n a cd IH l-l 3 CO rH cd tH cu •H n 60 cd 60 on CO I CM m o 3 O) H ON vO ON CO a 60 3 3 cd 3 ON o 01 o cd <4-4 1-1 3 to ■H rH •rl o Oi >n 0) 3 3 N U cd 0) o cd cd gin 01 rH U U =1 3 60 cd a. co 60 O O 60 O O O o o o rH rH 1 1 CM rH CM vO on u cu o 3 • H U o T3 O 3 CO cd & 01 6 o rH o 4-1 u cd cq cd u 01 3 0) 60 CM vO 60 O O CO rH I rH O O O CO CO cu •H CJ 0) u. CO m CM on ■H B Q O vO 60 U in CM VO ON 0) 3 CO 01 3 W TJ 3 cd u cu co >> cd vO oo CM O m 60 M CM O CM o 1 o rH rH in 1 CO CO CO 0) ■H o • 01 CO O. vO CO o> rH >. 3 ■k cd CO S t-l 0) rH 01 v£> Pi CT> rH TJ 3 ■I cd u cu >> ,0 01 •H 3 0) 3 rH 0) W H 1 m 1 r» 1 1 o 60 60 O O O O O o o o o o o VO rH I I rH O CM rH u a) cd 4J •H CO cd co CO rH Tj CO S cu rH •H U CU X >-i cd +J cd M CO 4-> g (3u N cd cu u £ CU •rl 3 u 3 cd Pi ^J in PH 4-1 X o\ rH m. >> TJ o M PQ & Q u <4-l N-*' • r- r~ vO vO o\ a\ rH rH M 41 rH rH .3 •e cd cd o> o\ 60 vO O n Z3 1 1 1 cd 1 1 1 1 1 1 1 1 4J 01 4-1 •rl M Xi O cd 60 cd u •iH P. cd 0) CU cd 1 rH o & o 1 O cd 1 CU HH 60 60 1 > f-i 3 3 rH 3 3 3 cd CO rH rH 60 o O o o o o *) A o o o o CO 1 00 1 1 o 1 o rH CJ 54 4J rS oo •H CU 13 o > u cu o TJ CU 3 3 •H 4J 3 o o OJ rH rO rj H CO 4-1 3 TJ 3 cd •H ,><; CO & a) •H CO cd ►J 00 m CN rH I m o o o\ U CU TJ rH Cd CJ TJ 3 cd •H CO & 0) •H CO cd 1-4 CTi 4-1 a) j3 e 00 3 •H rH a) O & > 00 00 s 3 3 3 oo CO 00 I vO O o o ON lt-1 CJ rH O W MH Q) U-l O U O 3 CO 0J Cd 4-1 cu cd 6 cd cd TJ co cd 6 •H CO CO CU u cu cu cu a > O -H N 00 4J U 4J .s a) 00 4-1 ,£} 3 oo •H c cd CO cu 3 rH CU U >> >> o cd > « 4J 4J W -H CO 00 CJ CU 3 . 'H . oo co cd 3 3 O •» cu U 00 cu 3 , > cd <4H CU ,3 & CJ O M o rH K CU VO TJ 3 CM cd cu u cd cd TJ cu oo cd 3 m 3 r4 rH 3 CO CU CU CJ CO CJ r4 o 4-1 CTi cd ex cu >> 4J n o CO 4-1 CCJ en •H u •H CO .3 cu cu o 01 5 •H -3 u o Tj 4-1 •> CU 3 UNO r4 VO -H o a\ 4j O. rH cu u TJ 3 cd to •H M co 3 cd CU rH 3 •> 3 rH rH CJ 3 VO CO >CH B) ■X H > 55 Table 8. Species Skin vascularization and epidermal thickness in ranid frogs (data from Czopek, 1965) . W (g) Epidermal Thickness (u) 9 meshes capillary net per mm skin Rana escuelenta Rana escuelenta 1.7 53.0 35.0 39.1 204 220 Rana escuelenta 250.0 62.3 225 Rana grylio Rana grylio Rana grylio 4.5 24.0 291.0 24.5 32.3 103 106 107 Rana pipiens sphenocephala Rana pipiens sphenocephala 0.6 127.0 12.7 37.0 132 100 56 vasodilation of the peripheral arterioles and shunting of blood to the skin. At P , the permeability of the skin to oxygen should be maximal, and at this time exchange capacity (E ) will equal metabolic rate. Therefore the 0 consumption at P is equal to (maximal permeability) (surface area) , and permeability can be solved for 2 directly in terms of yl 0„/cm -hr. To express permeability in terms 2 of yl 0„/cm •hr»mmHg, one need only to know the p0„ of the blood, since the pO of the water is known (i.e., P ). A reasonable estimate can be made by assuming that the 0„ tension of the blood is close to the p50 when the blood returns to the skin to pick up oxygen. The p50 values for the large aquatic salamanders Necturus, Amphiuma and Cryptobranchus range from 14.5 - 29 (Lenfant and Johansen, 1967; McCutcheo and Hall, 1937; Scott, 1931). Even small (0.36 - 0.56 g) submerged S_. lacertina die at p0„ tensions between 10 and 20, indicating that such tensions are not enough to saturate the blood to a degree per- mitting survival and are probably below the p50. Therefore, the loading tension of the blood of a Siren will be considered to be 25 mmHg for purposes of calculation. For example, for S_. lacertina 2 with W = 0.36 g, M = 35.3 yl 0 /hr, SA = 4.886 cm and P = 80 mmHg, 2 permeability would have units of yl 0^/hr-cm •mmHg or M/SA(P -25). We are calculating maximal permeability (P ) and have P =35.3 2 2 yl 02/hr-4.886 cm -55 mmHg or 0.1313 yl 02/hr-cm -mmHg. Utilizing this method, permeabilities were calculated for each size of sirenid for which submerged metabolism data were taken, and the results are shown in Figure 13. cO o 4-> CO t3 cu cu 4-1 td 4-i T3 •H •H M-l C cu CO u •H •H Cfl CU •H .C ^ 4-> c c O •H •H cn C CO CU cu 00 M f^ 60 • X CU >. O u O CU O 4-1 X H CO a c •H •H AS • +J co (U S-i N CU cu •H O x: CD cO 4-1 >> rH M-l 13 • O O X O) >. J-l 4-1 M-l o •H o 4-4 r-l ■H C CO .0 o 4-1 cd •H cO cu 4-1 t) B a u a CU CU s X PLI 4H 4-1 CO cu u 00 Pn a "O O a> to c F a £3 a 2 S d O d m 1 r i i i i o CO OJ CO «- i d d .o d o d o (6H UJ uj-2ujo.jm/2o |^) AinigV3!AIU3d 59 Oxygen exchange capacity can now be calculated as (7) E (pi 0o/hr) = P (pi 0o/cm2-tir.mmHg) SA (cm') ApO^(ranHg) c / max / z where ApO„ is the differential between whatever the oxygen tension of the water happens to be and the 25 mmHg loading tension assumed for the blood. All _S. lacertina were compared at 155 mmHg of 0„ in the water, which makes Ap0„ = 130 mmHg. In the example this means 2 2 that E = (0.1313 pi 02/hr-cm -mmHg) (4.886 cm ) (130 mmHg) or 83.4 pi 0_/hr. This is the E for a 0.36 g S. lacertina submerged in air- equilibrated water (p0„ = 155 mmHg of oxygen) at 25°C. Table 9 gives E , M and P values for 13 sizes of S. lacertina. No P could ° c c — c be determined for the 17S g size class, therefore no E could be calculated. Note that 0_ exchange capacity fell behind metabolic demand for the largest size class. Regression lines are fitted to log-log plots of the data in Figure 14. The slope of the metabolism plot is 0.65, which is less than the 0.67 found for the surface area of the skin in Figure 9, but the decrease in permeability with size caused the slope of the E line to become 0.54, and, as predicted by the model in Figure 12, there is a theoretical maximum size, which is 2908 g. This would mean that a S_. lacertina of up to 2908 g could survive submerged in water at 25 °C and p0_ - 155 if it continually maintained only his standard metabolic rate. However, some scope for activity must be maintained by retaining a reserve 0» exchange capacity. Thus we would expect death to occur in submerged animals at a size smaller than 2908 g. Table 9 indicates that large animals have standard metabolic rates close to their 0„ exchange capacities. Using the data from that table, a plot is shown in Figure 15 of the 60 Table 9. Standard metabolic rate, 0~ exchange capacity, and critical oxygen tension of S^. lacertina of various body sizes (for submerged animals) . Body weight Cx and range) Average Metabolic Rate (yl 02/hr) 0„ exchange capacity at 25°C, 155 mmHg (yl 02/hr) Pc (mmHg) 0.36(0.32-0.42) 35.3 83.4 80 0.56(0.49-0.63) 47.6 137 70 3.0 (2.7-3.6) 153 331 85 6.5 (5.0-8.5) 280 485 100 13.7(11.0-17.1) 438 759 100 42.7(40-45) 769 1334 100 73(67-76) 1080 1756 105 103(93-112) 1710 2616 110 178(170-191) 2083 269(233-280) 3040 4652 110 357(346-375) 3034 3432 140 541(472-590) 5031 6543 125 825(770-857) 5692 6165 145 1310(1090-1451) 7467 6696 170 0) u 60 •H o u 0 o 3 <+-( M-l 0) n) CO w to to 0) ^ c 4-1 •H •H tH O cd C p« O rt •H o CO CO Q) 0) 60 5-1 c 60 rt 0) rC Pd o X 0) Q) C ca CD •n 60 •rH >> c X CD o 5-1 •H <4-l C/3 o CD c! X o 4J •H 4-> G ct) •H N * •H 4J >> tH Si iH •H 60 C 4J •H O 2 C e :>. •H 01 -a 4-> o o 5-1 Xi CU u CJ CD 4-J Cfl P4 o r-t m cu 3 60 •r-l o -a CD CO £ D i— O . — i_ c 1/3 c/)| « I o M- O CM o in cvi q CM m CD C9 O _J m d o d in d i AllOVdVO 39NVH0X3 20 JO NOIlVZIIlin % 65 per cent utilization of E for each body weight. It is obvious that there is a more rapid increase in the per cent utilization at larger body weights. This approach predicts a maximum body weight of 955 g for submerged survival, and shows that animals of 825 g are using 92% of their 0„ exchange capacity just to maintain their standard metabolic rate. Actual survival was tested by submerging animals and checking them periodically for up to 336 hours. If they were still alive at that time, they were removed and considered to be able to survive in- definitely. The results were given in Table 2, and it is clear that animals of 800 g were in trouble, and that larger ones could not survive. The smaller size ranges available for F_. striatus and !S. intermedia make similar calculations more tenuous. The results are given in Figure 16, and qualitatively agree with the observation that all sizes of these sirenids can survive in air-saturated water at 25°C. It is interesting that although the predicted maximum size for S^. intermedia is essentially identical to that of S^. lacertina, S^. intermedia reaches a much smaller maximum size than S_. lacertina where the two are sympatric than it does where S^. lacertina does not occur. For the few sizes of _P. striatus studied, the 02 exchange capacity is increasing faster than the metabolic rate with increasing body size. Gas Exchange Partitioning The previous discussion has centered on _S. lacertina as a water breather, but it must be borne in mind that this organism also uses <4-t • O cd •H a T3 o . cd ■u •H en o 3 cd ■m Pu cd OS •H a M 4-» (U w 60 c • cd p4 .a o Id K 0J » ■u X) a) o S .£> vO ai u 00 •H in CO CD CD Is- CD CD o o II 11 1— i— > — * ^— ^ CO O "- CO o 0 $: ^ m m ■3" CD rO — CD — 11 11 o ro o ro' CO CM q cvi — r- m ro q ro' 10 cvi O CO o d o 0 10 0 1 o _J (JM/^I^) 39NVH0X3 30 901 68 aerial gas exchange. Even small individuals in an aerated tank will breathe air if allowed to surface, and certainly large in- dividuals must do so to survive. Therefore to gain a more complete understanding of what the animal is actually doing in his natural habitat, the partitioning of gas exchange between the air and water must be considered. This will, however, still be a function of body size, regardless of the environmental conditions. The results of experiments dealing with the effect of body size on gas partitioning were given in Table 4 and Figure 10. Large animals are obligate air breathers, due mainly to their inability to obtain sufficient 0„ from the water. Most aquatic amphibians eliminate the majority of their CO into the water, and Siren is no exception. The increase from 6% to 26% in aerial C0„ elimination associated with large size is probably caused largely by the increase in aerial 0„ consumption from 42% to 75%. Assumably, Siren smaller than the 90 g individual studied obtain even less than 42% of their 0„ from the air. Guimond (1970) studied S_. lacertina at 25°C and found the per cent aerial V for three individuals to be 56% (1200 g) , 53% (907 g) , and 2 42% (482 g) . For 11 S^. lacertina averaging 628 g he found the aerial V to be 50%, and the per cent aerial V to be 78%. This body size lies between the 90 g and 1489 g used here, and the percent- ages for the aerial partition of each gas lie between the correspond- ing results for this study, even though the absolute values given by Guimond are much lower. The total V for animals allowed to breathe air was about 3-5 2 times greater in _S. lacertina than the average metabolic rate for the 69 same size animals submerged at the same aquatic pO„. This is par- ticularly relevant in the case of the 90 g Siren, which can live submerged indefinitely. Some of this increase in oxygen consumption can be associated with an increase in activity connected with air breathing, but it has already been mentioned that. the increase in activity was of no obviously great importance. It appears that the Siren is functioning essentially like the lunged and lungless salamanders studied by Whitford and Hutchison (where lunged salamanders had higher metabolic rates) , and that at least part of the increase in metabolic rate can be attributed directly to the presence in the air and water breathing Siren of the additional surface area for gas exchange added by the lungs. This interpretation fits with the original suggestion made earlier that the exchange capacity for 0 may limit the level of metabolic rate of an organism. It was assumed that the submerged animals were extracting all the 0„ they needed from the water while they were regulating metabolic rate. To determine if air breathing effects the level of aquatic V , a 44 e animal was placed in a container similar to that used for o2' B v the submergence experiments and allowed to breathe air through a layer of paraffin oil. The results (Figure 17) show that the aquatic V was unchanged. Therefore all of the increase in total V in an °2 2 air-breathing animal is due to pulmonary respiration. The results appear to strongly suggest that the presence of an additional amount of respiratory surface area resulted in an elevation of the standard metabolic rate. However, metabolic rates of animals breathing both air and water over a large range of body sizes will have to be determined CD A H # w| M 01 X) ■u a) ed 00 3 a) T3 e a .£> tO 3 to n •H u nJ o M a to •H a) .c 3 ■U rH co CO ai > S-i • .o CN ^^ O <; rt •> r^ C •rl CJ cfl 13 c o 0) •H •H to 4-1 ■U o rl CO iH CD 3 CJ CJ c C! cO j i Z) o o 80 e> >" 60 X o 40^ 20- ** *jf^> X I0/9Q MR 59 46 PC 65 60 R R, striatus (l.40-l.82g, X = l.58) 04 , 1 1 1 j ? — -t r 1 1 1 1 — -t 1 i "J" > 0 20 40 60 80 100 120 140 160 p02 (mm Hg) to 60 00 oo CM to 3 ■u td •H u u to T3 • 0) C 60 o H •H 0) to S c x> a) 3 4J to c m % 0) K •u o nj M '+4 o o •H 3 .H o O •H 43 •U td o 4J c - X O 30-1 15- (5.0-8.5g,*«=7.0) S IC^Q MR 40 30 PC 75 70 °"V/-'- ' -'- ' 60 ' 80 ' 100 ' 120 ' 140 ' 160 20 40 p02(mm Hg) Figure 4A. Metabolic rate of submerged S^. intermedia (13.7 and 29.6 g) as a function of oxygen tension. 60-t 50- d1 40- _S. intermedia (N.4-l7.lg,X = l3.7) • • • • ^ 30- o jZ 20H ^ lo- co z o o o-y/i i r • • • *-'' .*9jT t R X io/sc MR 33 21 PC 110 65 T 3 J \ j ! ! i 1 r U 40i O >- X o 30-^ S_. intermedia (26-33g, X = 29.6) 20- ^ • » o 10- A R X |]0/90 MR 24 17.4 PC I05| 55 0-y/r r 20 — t r 40 60 — i — 80 t r 100 —3 j- 120 140 Iso p02(mm Hg) Figure 5A. Metabolic rate of submerged S^. lacertina (0.36 and 0.56 g) as a function of oxygen tension. 120 100 o> 60 d* ^40- "Z. o 20 h- Q- ^E Z3 n en ;z o o 120 2: uj 100 o >- X o 80 60 40- 0 V* R S.. locertina (0.32-0.42g, X»0.36) X 10/gr MR 98 82 PC 80 60 J— S. lacertina (0.49-0.63g, X = 0.56) 1 5? I0/M MRl 85 71 PC 1 70 70 20 ~40~ 60 80 100 120 140 160 Teo p02(mm Hg) CO 60 in a cd CO cd C o CO w a) C 60 O n •H -. cu X 4J o CO U CM o o •rl e i-l o O •H 43 •P cd a 4-J C! to cci i q in o 0) «-o to lO IX o o o a. n* o to o CVJ O o CO _T_ o m o o fO ~i — — V o o o \.\J o CO Lo <4- o {jq-&/0 ^) NOlldlAinSNOO M39AX0 03 X S o a Figure 7A. Metabolic rate of submerged S^. lacertina (.13.7 and 42.7 g) as a function of oxygen tension. 60 n 50 40- en 5? l0/90 MR. 32 21 PC 100 75 CM 30H O 20- O h- 10- CL .^;' t ZD CO "Z o o 30- UJ CD X 20H O 10- /i j r i r | S. lacertino r ( Il.0-I7.lg. X« 13.7) T — — i J r 7 ! 1 1 S. lacertina (40-45g, X = 42.7) • • 2--<- t R X l0/90 MR 18 13 PC 100 75 o-Va- 15 T I i ! j 1 1 1 1 1 r , , , , 35 55 75 95 115 135 155 p02(mm Hg) Figure 8A. Metabolic rate of submerged S_. lacertina (73 and 103 g) as a function of oxygen tension. C? 2 O CO O O LU O >- X o 24 n 20- 16H 12- 8- 4- 0-V/t- S. lacertina (67-76g,X = 73) .• ••• . X l0/90 MR 14.8 11.9 PC 105 95 t j 1 1 r 32i 23 24 20-] 16 12 8 4 S. locertina (93-H2gt X=I03) •- ^-* YA 1 1 r 30 50 t R • V l0/90 MR 16.6 10.4 PC 110 85 70 SO p02(mm Hg) — i 1 1 1 1 r— — > 110 130 150 170 cd w to \£> CM C cd oo cd G •H 4J U 0) u cd "O. • (U c M) o >-i •H CD CO e C -Q . 0) X ■U o cd u m o o vH £ rH o O •H rO 4-1 cd O ■u C '. X ,0/90 MR 8.5 8.6 PC 140 125 R S. locertino (472-590g, X*54l) • • • X l0/90 MR 9.3 6.2 PC 125 110 60 80 !00 120 140 ISO ISO p02(mrn Hg) 200 6C o H CO rH •d a cd IO r\i 00 ^ to C •H V n O CJ cd rH • CJ • O W •H eg -d C cj ai to 4J U 0) fl e a) 43 00 3 >, CO X o t+H o -w o GJ 4-1 a nj o u •H ■U u O vi c H 3 O •4-1 43 c3 d 4J CD w s cd QJ U 00 •H D O O o CO CI ■PI « O ( v a> lO 1* t£> t a: o -~ a. C£ to co tl IX to CO I O r- C\J — T" o i CO CO — r •3- CvJ O o r cvJ O to •— • CI IX c CO O 75 O rO 7] I o — ;r-o c!s co °-| O — :hco o co o I hcM o HO o CO CO i —J— CVJ o > ct) •H cfl 4J 6 o i4 0) CO s-^ iH •H ^5 QJ tH CI CO -M O cfl rH CO .£> rH Cfl cfl ,o E • TJ 4-1 *— ' ^ CO e a Cfl 5 cfl B QJ N B QJ U 4J 3 •H m J3 X! cu W o CO bO ,3 •H & >S CO H OJ •a ■u o & m o .3 14-1 r^- rQ 00 iH • •H 4J R) o x: QJ 3 +J H j-> 3 •H o & •H O u O 3 Cfl P. 3 l-~ 0) 4J cfl QJ 00 QJ 0) II 3 rH CO £ Cfl QJ •H 4J g .3 ,« QJ o CO X ^Q 0) > 4-) •H TJ (X •H o Cfl 3 •H 00 3 3 cfl ,3 4J CO O CO CJ n 3 4J QJ Cfl — s O o vi- •H T3 T3 a) vo 4J CO U CO cfl ■H t-I •H QJ M co •u CO •H a) o rH QJ CJ 3 •H CtJ > « T3 £ •H 3 0) § 4J OJ T) U R) O & 3 PM B cfl o cfl « QJ 3 00 — r- to - <* to CO II II 3? =>w ^ \ "\ -o C9 O _J -T- T- O T T" CM I to i t- CVJ u to I (6>I)SM 901 u o > J3 -a 00 o •rl rQ CU M-l o cu 3 e CO o CO •H •H 4-1 4-1 CJ C (U 3 > CH m a H •H cu cd cd tH CO o ,£> CO CO ,a O cd td cd ■w «-^ ■**% s^\ cu «3 m c_> S s_x \— ' \-^ • CN CO 5-1 3 to •H pH ■■n ?- o * '(Adp/idomJW u,901 o o 6 . o o o o go o o o - ■ . d d o (top/!DD>i)^ 106 and k^ is 72.6. Line A on Figure 2B is the actual relationship between metabolic rate and body size as given by Eq. (IB). Line C, representing metabolic rate as (f) W ' is a much better approxi- mation of the actual results than line B, which represents metabolic rate as (f) W * . Note that both lines B and C are corrected for increasing skeletal weights which are disproportionate to increasing total weights. It is obvious from Figure 2B that most of the 0.25 units difference from 1.0 in the weight exponent is actually due to a decrease in weight- specific metabolic rate of active tissues. However, curves B and C in that figure are not exactly linear, since W. was calculated from W = A A 1.13 W - 0.089 W ' ' , which is not a linear function. Therefore, C cannot be fitted perfectly to A merely by increasing the exponent of W . To find what the actual difference between 0.75 and some slightly higher exponent of active weight should be to account for M = 70 W , one need only plot the actual metabolic rate against W . This is done in Figure 3B, where the slope of the line represents the exponent of W , and gives a power of 0.767. This can be interpreted as meaning that approximately 0.017 units of the 0.25 units, or only about 7%, of the decrease in weight-specific metabolic rate with increasing size can be accounted for by a disproportionate increase in skeletal weight . T3 a cd CO •u 0) ^3 3 toO CO •H CO cu •H & •U iH 0) cd > •U •H o 4J ■u O cd II H-l E- O ES ■M ,Ei • toO CO •H rH CD cd fs c cd CD 6 (1) & C ■P •H a) rO cu •u P. cd •H M ,c co a a •H o rH •H o 4-> ,Q cd cd rH ■u CU CU Pi s • « en a) m 3 toO •H Pn o r*t -ro r- m at oo o 6 i ii 3? ■O O N- i — T- ro (-0 O ii CO -CvJ 0* •O o \ -CvJ -i hro O ' (/fop/|D0><) |A|0I90"1 LITERATURE CITED Anderson, J. 1968. Metabolic rates in spiders. Ph.D. Dissertation. Univ. of Fla., Gainesville, Fla. Anderson, J. 1970. Metabolic rates of spiders. Comp. Biochem. Physiol. 33(1): 51-72. Bartholomew, G. and V. Tucker. 1964. Size, body temperature, thermal conductance, oxygen consumption, and heart rate in Australian varanid lizards. Physiol. Zool. 37: 341-354. Brody, S. 1945. Bioenergetics and growth. Reinhold Pub. Corp., New York. Conant, R. 1958. A field guide to reptiles and amphibians. Houghton Mifflin Co., Boston. Czopek, J. 1960. Unaczynienie powierzchni oddechowych u salamandry plamistej (Salamandra salamandra L.) w rozwoju ontoneretyxznym. Zessyty Nauk. UMK, Torun. Biol. 5: 3-16. Czopek, J. 1961. Vascularization of respiratory surfaces in some Plethodontidae. Zool. Pol. 11: 131-150. Czopek, J. 1962. Vascularization of respiratory surfaces in some Caudata. Copeia, 1962(3): 576-587. Czopek, J. 1965. Quantitative studies on the morphology of respira- tory surfaces in amphibians. Acta anat. 62: 296-323. Davison, J. 1955. Body weight, cell surface, and metabolic rate in anur an amphibia. Biol. Bull. 109(3): 407-419. Dmi'el, R. 1972. Relation of metabolism to body weight in snakes. Copeia 1972(1): 179-181. Freeman, J. 1963. Studies of respiratory mechanisms of the salamander Pseudobranchus striatus. Ph.D. Dissertation. Univ. of Fla., Gainesville, Fla. Goin, C. 1941. "The striped siren, Pseudobranchus striatus (LeConte)." Master's Thesis. Univ. of Fla., Gainesville, Fla. Gray, I. 1954. Comparative study of the gill area of marine fishes. Biol. Bull. 107: 219-225. Guimond, R. 1970. Aerial and aquatic respiration in four species of paedomorphic salamanders: Amphiuma means means , Cryptobranchus alleganiensis alleganiensis , Necturus maculosus maculosus, and Siren lacertina. Ph.D. Dissertation. Univ. of Rhode Island, Kingston, Rhode Island. 109 110 Hemmingsen, A. 1960. Energy metabolism as related to body size and respiratory surfaces, and its evolution. Rept. Steno Mem. Hosp. Nord. Insulin Lab. 9 (Part II). Hutchison, V., W. Whitford and M. Kohl. 1968. Relation of body size and surface area to gas exchange in anurans. Physiol. Zool. 41(1): 65-85. Ivlev, V. 1954. The relation of metabolic rate to size in fish. Fiziologich. zhurnal 40: 717-721. Kayser, C. and A. Heusner. 1964. £"tude comparative du metabolisme energetique dans la serie animale. J. Physiol. (Paris) 56: 489-524. Kleiber, M. 1961. The fire of life. John Wiley & Sons, Inc., New York. Lasiewski, R. and W. Calder. 1971. A preliminary allometric analysis of respiratory variables in resting birds. Resp. Physiol. 11: 152-166. Lenfant, C. and K. Johansen. 1967. Respiratory adaptations in selected amphibians. Resp. Physiol. (1967)2: 247-260. Lynch, J., J. King, T. Chamberlain and A. Smith. 1947. Effects of aquatic weed infestations on the fish and wildlife of the Gulf States. U. S. Dept. Int. Spec. Sci. Rept. 39: 1-71. McCutcheon, F. and F. Hall. 1937. Hemoglobin in the amphibia. J. Cell and Comp. Physiol. 9: 191-197. Muir, B. 1969. Gill dimensions as a function of fish size. J. Fish Res. Bd. Canada. 26: 165-170. Muir, B. and G. Hughes. 1969. Gill dimensions for three species of tunny. J. Exp. Biol. 51: 271-285. O'Hara, J. 1968. The influence of weight and temperature on the metabolic rate of sunfish. Ecology 49(1): 159-161. Price, J. 1931. Growth and gill development in the smallmouthed black bass, Micropterus dolomieu, Lacepede. Studies, Ohio State Univ. 4: 46. Scott, W. 1931. Oxygen and carbon dioxide transport by the blood of the urodele, Amphiuma tridactyla. Biol. Bull. 61(2): 211-222. , Stahl, W. 1967. Scaling of respiratory variables in mammals. J. / Appl. Physiol. 22(3): 453-460. Stussi, T., A. Huesner, C. Kayser, C. Marx and M. Harmelin. 1963. Relation entre le poids et la consommation d'oxygene. III. Etude interspecif ique chez les poissons. C. R. Soc. Biol., Paris, 157: 667-669. Ill Tenney, S. and J. Remmers. 1963. Comparative morphology of the mammalian lung: diffusing area. Nature 197: 54-56. Tenney, S. and J. Tenney. 1970. Quantitative morphology of cold- blooded lungs: amphibia and reptilia. Resp. Physiol. 9: 197-215. Ultsch, G. 1971. The relationship of dissolved carbon dioxide and / oxygen to microhabitat selection in Pseudobranchus striatus. Copeia, 1971(2): 247-252. Ursin, E. 1967. A mathematical of some aspects of fish growth, respiration, and mortality. J. Fish. Res. Bd. Canada. 24: 2355-2453. Whitford, W. and V. Hutchison. 1967. Body size and metabolic rate in salamanders. Physiol. Zool. 40(2): 127-133. Winberg, G. 1956. Rate of metabolism and food requirements of fishes. (Original in Russian) Translation Series No. 194 (J. Fish. Res. Bd . Canada) . BIOGRAPHICAL SKETCH Gordon Richard Ultsch was born in Waltham, Massachusetts on June 30, 1942. He graduated from Fitch Senior High School in Groton, Connecticut in 1960, and attended the University of Connecticut from 1960 to 1963, when he was married to Sandra Lawrence. He returned to the University of Connecticut in 1966 and graduated with a B.A. in 1957. That year he entered the University of Florida as a graduate student and has since pursued work leading to the Ph.D. Gordon and Sandra have one child, Julie Anne. 112 I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. / B. K. McNab, Chairman Professor of Zoology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Associate Professor of Zoology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. '^JZ's^JZr12- 'L. Popenoe 3rofessor of Soils u This dissertation was submitted to the Department of Zoology in the College of Arts and Sciences and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August, 1972 Dean, Graduate School At ^\V ft***-*