MOVEMENT OF INERT GASES INTO THE SWIMBLADDER OF THE TELEOST, LAGODON RHOMBOIDES (LINNAEUS) APPROVED: MOVEMENT OF INERT GASES INTO THE SWIMBLADDER OF THE TELEOST, LAGODON RHOMBOIDES (LINNAEUS) By CHARLIE ROY POWELL, B.S. THESIS Presented, to the Faculty of the Graduate School of The University of Texas in Partial Fulfillment of the Requirements For the Degree of MASTER OF ARTS THE UNIVERSITY OF TEXAS AUGUST 1961 ACKNOWLEDGMENTS I wish to thank Dr. William N. McFarland for suggesting and supervising this thesis, and Dr. Clark Hubbs for reviewing it. Experiments were performed at the Institute of Marine Science at Port Aransas, Texas. The work was supported by a National Science Foundation research grant, number NSF-G10690. TABLE 0F CONTENTS ACKNOWLEDGEMENTS CHAPTER PAGE I. INTRODUCTION 1 11. MATERIALS AND METHODS 6 Capture and Care of Animals 6 Experimental Apparatus and Methods .... 6 Analytical Procedures for Gases 9 Experimental Error ....... 10 111. RESULTS 12 Control Experiments ... 12 Experiments with Argon ..... 16 Experiments with Neon .......... 18 Experiments with Gas Mixtures 20 Solubilities . 24 IV. DISCUSSION 30 V. SUMMARY 32 VI. REFERENCES . . . . .33 LIST OF TABLES TABLE PAGE 1. Effect of the Trisma 8.3 Buffered water on the gas composition of the swimbladder of resecreting and nonresecreting specimens of Lagodon rhomboides 13 2. Normal gas composition from swimbladder of Lagodon rhomboides obtained from 15 to 18 foot depth 14 3. The effect of high ambient argon content on the diffusion of argon into the swimbladder of nonresecreting Lagodon rhomboides. ...... 17 4. Effect of high ambient neon content on the diffusion of neon into the swimbladder of nonresecreting Lagodon rhomboides 19 5. Effect of an artificial gas mixture on diffusion of argon, helium, and hydrogen into the swimbladder of nonresecreting Lagodon rhomboides . 21 6. The effect of an artificial ambient gas mixture on the diffusion of inert gases into the swimbladder of nonresecreting Lagodon rhomboides 22 7. The effect of an artificial ambient gas mixture on the diffusion of inert gases into the sVimbladder of resecreting Lagodon rhomboides 23 LIST OF FIGURES FIGURE PAGE 1. Experimental apparatus 8 2. Relationship of diffusion of inert gases into the swimbladder of Lagodon rhombodies and molecular weight 26 3. Absolute solubility of various gases as a function of molecular weight . . 28 INTRODUCTION The swimbladders of fishes function primarily as buoyancy organs to maintain equilibrium at a specific water depth (Hall, 1924; Copeland, 1952 b). This is achieved by secretion and absorption of gases found in the swimbladder. Most physostomous fish that live near the water surface and have access to the atmosphere gulp air (Saunders, 1953), whereas other physostomes and all physoclistic fishes secrete and absorb gases to regulate buoyancy. Oxygen was found to constitute as much as 90 to 100 per cent of the gas content of the swimbladder of teleosts living in deep water (Biot, 1807; Richard, 1895; Schloesing and Richard, 1896; Scholander and van Dam ; 1953). In contrast, Hufner (1892) reported 103 per cent nitrogen in the swimbladder of Coregonus captured from 80 meters depth. Subsequent to this Saunders (1953) confirmed that nitrogen contents are characteristic of physostomous forms, while high oxygen contents are characteristic of physoclists Copeland. (1952 a demonstrated, that during reinflation oxygen and carbon dioxide were released into the swimbladder of Fundulus heteroclitus. He attributed this secretion of gases to a physiological mechanism involving the blood and a pH shift. This species, however, was unable to inflate its swimbladder against five atmospheres pressure . Nitrogen was found to enter against a diffusion gradient and, if the swimbladder was stimulated to reabsorb the gases by overinflation, nitrogen would constitute 100 per cent of the gas' retained in the swimbladder. This result is analogous to the high nitrogen contents of physostomes living at depth. Saunders (1953) and others would consider high nitrogen content an effect attributable to differential reabsorption of gases from the swimbladder. In Ambloplites rupestris (Rafinesque), Rostorfer (1942) found that under several different pressures oxygen and carbon dioxide were initially secreted, the latter of which rapidly diffused out after equilibrium was attained. Oxygen deposition compensated for the loss of carbon dioxide and nitrogen was apparently only passively important in the adjustment to changes in hydrostatic pressure. As indicated, in fishes which live at great depths the swimbladder gas composition varies according to the type of swimbladder present (Saunders, 1953). In physostomous fish living near the surface the nitrogen portion of the swimbladder gas mixture is high and increases to nearly 100 per cent for fishes dwelling at great depth. The physoclistic fishes show the opposite trend. Near the surface the swimbladder gas has nearly the same composition as the atmosphere, but at great depths the oxygen increases to nearly 100 per cent and nitrogen decreases to a fraction of a per cent (Saunders, 1953). In either type the partial pressure of nitrogen at depth greatly exceeds the partial pressure of the nitrogen dissolved in the surrounding water (Scholander and van Dam, 1953). Although the evidence is not conclusive, the secretion of gases into the swimbladder by dissociation of oxyhemoglobin cannot be explained by the "Root Effect" for fish living at great depths (Scholander and van Dam, 1954). Copeland (1952 a found that glycogen was normally found in the cells of the secretory epithelium. It disappeared during secretion of gases and was stored during absorbtion. This indicates an active mechanism for the transport of oxygen and carbon dioxide across the swimbladder membrane • A cellular mechanism for the secretion of oxygen into the swimbladder was postulated by Scholander (1954) rather than one that splits off oxygen from the blood. In the barracuda, Sphryaena barracuda, there was but slight loss in the oxygen content of the blood leaving the active gas gland (Scholander, 1956). However, in experiments with the isotope ols0 ls Scholander, van Dam and Enns (1956) demonstrated that oxygen secreted into the swimbladder of the codfish, Gadus collarias, was derived from oxygen dissolved in the surrounding water, Wittenberg (1961) demonstrated that molecular oxygen is secreted into the swimbladder of the toadfish, Opsanus tau, and the scup, Stenotomus versicolor (Mitchell). Using a gas mixture containing a nonequilibrium mixture of the three molecular isotopic species of Os: ols-0 ls -0 18 , 0 16 -0 16 , and 0 18 -0 18 he found no isotopic rearrangement of the secreted gas. By the use of CO and 0 2 evidence for the presence of an intracellular hemoglobin in the active transport and secretion of oxygen was found (Wittenberg and Wittenberg, 1961). The precise secretory mechanism for oxygen and carbon dioxide, however, remains undetermined. The high nitrogen partial pressures that exceed expected partial pressures in the swimbladders of fishes from greater depths has resulted in controversy. Scholander and van Dam (1953, 1956), Scholander (1954, 1956), Scholander et al. (1956), and Sundes, Enns and Scholander (1958) conclude that the high partial pressures of nitrogen are attributable to an active secretory process. At an earlier date, Powers (1932), suggested that nitrogen entered the swimbladder passively by entrapment in oxygen bubbles being secreted into the swimbladder. This postulate was extended and supported by the work of Wittenberg (1958). Most previous workers have maintained that inert gases such as nitrogen and argon are passively involved (Hall, 1924; Rostorfer, 1942; Copeland, 1952 a; and Saunders, 1953). With the exception of Wittenberg’s work (1958) direct experimental investigation of inert gas "secretion" into the swimbladder of fishes has not been accomplished. In this paper the entry of several inert gases into the swimbladder of the physoclistic teleost, Lagodon rhomboi des (Linnaeus), are reported. MATERIALS AND METHODS Capture and Care of Animals Experimental animals were collected in the marine waters of the bays and ship channels near Port Aransas, Texas, by trawl and by hook and line The latter method resulted in better survival of the fish and was therefore adopted. Damage to the epidermis of the fish by the trawl led to severe skin infections and rapid death. No infections were noted in fish caught by hook and line. Fish were maintained in shallow tanks with running sea water for at least twenty-four hours prior to experimental use. Experimental Apparatus and Methods: Experiments were run using the apparatus shown in Figure 1. The five-gallon battery jar was filled to within two or three inches of the top with sea water and enough trisma 8.3 buffer [Tris(hydroxymethyl)ami nomethane with its hydrochloride] added to give a concentration of five grams per gallon. This prevented any appreciable rise in pH due to metabolic C0 2 released by the fish during the course of the experiment (McFarland and Norris, 1958). The experimental animals were placed in the battery jar and the lid clamped over the top sealing the chamber. Recirculation and subsequent aeration of the fish was begun as soon as possible. The experimental gases were added by water displacement. In the case of krypton and xenon a vacuum was created in the air hose by closing off the input stopcock to the pump at the experimental chamber. The gases were added directly into the air hose by means of a glass "T" and a stopcock. Stopcocks were provided to prevent loss of the air phase during addition of gases. The experimental chamber was then submerged in running sea water to control the temperature. However, the temperature did vary from one to five degrees centigrade per experiment. The manometer was used as an indication of the oxygen present in the air phase. Oxygen was added as it was used by the fish. FIGURE 1 EXPERIMENTAL APPARATUS Analytical Procedures for Gases: Gas samples were withdrawn with hypodermic syringes. Dilution of the sample with air was prevented by using vaseline on all joints or areas where gas leakage might occur. The swimbladders were punctured with small hypodermic needles and the gas removed for analysis. Samples of the air'phase were taken through the syringe cap provided on the experimental chamber. Samples for mass spec trometer analyses were withdrawn into evacuated mass spectrometer tubes using hypodermic needles to puncture the swi mbla dder. Analyses of the earlier experiments were made on the gas chromatograph. Argon and oxygen peaks occur at the same point. Therefore; oxygen was absorbed with basic pyrogallol in order to separate it from argon. An additional sample was analysed to obtain a complete analysis of a sample containing argon. Mixtures of gases containing krypton and xenon were analysed on the mass spectrometer since the gas chromatograph had not been calibrated for these gases. Experimental Error Replicate calibration of the gas chromatograph for a specific gas resulted in an error of not more than one per cent. Analyses of traces of gases gave a high degree of error, a result attributed to the use of small sample volumes . Analyses of air by introduction of. a 2cc sample, into the gas chromatograph gave a value for COg ten times the actual value. In many cases the sample taken from the fish was only 3cc., consequently, the sample introduced into the machine was only l/2cc. Gases exceeding one per cent of sample volume gave an analytical error of less than two per cent. Experimental error due to sampling technique was as high as ten per cent. Dead space in the needles and syringes reached ten per cent of the total volume of the sample where sample volume was less than 3.5 cc. Thus, sampling error most often exceeded analytical error. RESULTS Control Experiments: Control experiments were run to determine the normal swimbladder gas composition, resecreted gas composition, and the effects of buffering (Table 1). Analyses were made of the initial swimbladder gases of six fish acclimated in from ten to fifteen inches of water (initial measurements on resecreting fish). Nitrogen was found to constitute from 74.38 to 84.75 per cent of the gas in the swimbladder with an average of 80.32 per cent. Oxygen ranged from 15.02 to 25.62 per cent averaging 19.15 per cent. Carbon dioxide made up from 0.00 to 1.46 per cent and averaged 0.53 per cent. The control fish (unbuffered) at the end of the experiment had Ng and concentrations very close to these but had COg concentrations that were higher. Fish taken from a depth of 15 to 18 feet had quite different swimbladder gas compositions (Table 2). Nitrogen averaged 64.07 per cent, oxygen averaged 34.80 per cent and carbon dioxide aver aged 1.12 per cent of the swimbladder gas. This is in agree ment with previous observations (Saunders, 1953 and others) that oxygen increases in the swimbladder with depth and nitrogen decreases . In resecreting fish the nitrogen content of the swimbladder showed a decrease of 15 per cent or more from the normal value after 24 hours. Oxygen content increased a minimum of 15 per cent. These nitrogen and oxygen changes indicate that oxygen is the principle gas involved in inflation of the pinfish swimbladder. Carbon dioxide concentration in the swimbladders of both resecreting and control fish kept in unbuffered experimental chambers for 24 hours was at least two times greater than the highest value found for normal fish acclimated to 10 to 15 inches of water. Carbon dioxide also accumulated in the unbuffered air phase. Fish, on the other hand, kept in the buffered experimental chamber had a low concentration of CO2 in the swimbladder. The concentration of COg in the buffered air phase was too low to analyse quantitatively. Trisma buffer was found by McFarland and Norris (1961) to prevent any accumulation of carbon dioxide in the air phase in experimental apparatus identical to that shown in Figure 1. The fact that fish easily lived for 44 hours in water buffered with trisma 8.3 buffer, the lack of high concentrations of COg in the swimbladder, and the lack of COg in the air phase indicate that the buffer is removing C0 9 . Zu Oxygen and nitrogen in the swimbladders of the controls remained at a normal level during the experiments. For all fish the nitrogen concentration in the swimbladder was greater than in the air phase. Probably outward diffusion of inert gases was prevented or slowed down by the swim bladder membrane. Temperature ranged from 28 to 32°C. All values are reported in % Exp. No. Gas Final Airphase Buffered Unbuffered Resecreting Fish Buffered Unbuffered Initial Final Initial Final Nonresecreting Control Fish Buffered Unbuffered N 2 34.25 68.95 82.81 59.97 84.75 39.44 1 °2 65.75 29.16 15.73 38.23 15.02 56.91 oo 2 1.89 1.46 1.80 0.23 3.65 N 2 63.10 72.40 77.65 60.69 83.31 58.76 78.07 81.40 2 °2 36.90 25.02 22.35 38.73 15.21 35.48 20.86 15.65 co 2 Trace 2.58 Trace 0.58 1.48 5.76 1.07 2.95 N 2 61.93 77.38 79.03 23.81 74.38 29.16 83. 50 79.91 3 °2 38.07 19.45 20.97 76.19 25.62 62.61 16.03 16.63 C° 2 Trace 3.17 Trace Trace Trace 8.23 0.47 3.46 Note: Duration of experiment 24 hours TABLE I EFFECT OF THE TRIZMA 8.3 BUFFERED WATER ON THE GAS COMPOSITION OF THE SWIMBLADDER OF RESECRETING AND NONRESECRETING SPECIMENS OF LAGODON RHOMBOIDES. Fish * N 2 £0 2 C0 2 1 61.82 37.06 1.12 2 65.76 33.15 1.10 3 64.89 34.06 1.05 4 60.86 38.09 1.05 5 6 7.04 31.66 1.30 Average 64.07 34.80 1.12 TABLE 2 NORMAL GAS COMPOSITION FROM SWIMBLADDER OF LAGODON RHOMBOIDES OBTAINED FROM 15 TO 18 FOOT DEPTH Experiments with Argon: Fish were placed in an experimental chamber containing an artificial atmosphere with a high argon content. Analyses were made of the swimbladder gases and the air phase (Table 3). Nitrogen was about normal in the swimbladder whereas the air phases had lower than normal concentrations of Ng indicating that diffusion of Ng out of the swimbladder is slow. Argon diffused into the swimbladder in varying concentrations. The concentration of argon in the swimbladder would depend upon the length of time of the experiment, its solubility, and the temperature at which the experiment was run. Also, whether the fish happened to be secreting or not would have an effect. Experiments one, two, and three were run for approximately the same length of time (38 to 42 hours). Swimbladder concentrations varied a great deal. This is particularly apparent when the per cent diffusion of the theoretical maximum gas in the swimbladder divided by % gas in final air phase times 100) is calculated for argon. The temperatures during the course of the experiment are not known and could explain the variation In experiment one a hose came loose and fell into the sea water in which the experimental chamber was submerged. The air phase was lost and sea water was pumped into the experimental chamber by the recirculation pump diluting the buffered water. When this was discovered the air phase was replaced and the chamber resealed. The composition of the original replacement air phase is not known. Since 0 2 percentages are low in all three fish, utilization of the 0 2 during the disrupted period could in part account for these high argon values. Temperature was not known Exp. No. Time of Exp. Gas Initial Air- phase Final Air- phase Fish 1 Fish 2 Fish 3 $ Diffusion of Maximum into Swimbladder Fish 1 Fish 2 Fish 3 n 2 55.25 59.43 77 .17 71.01 75.39 1 38j °2 21.89 17.66 7 .71 8.44 8.19 Hrs. A 22.86 22.42 13.76 19.14 15.20 61.4 86.4 67 .8 co 2 0.49 1.36 1.41 1.22 n 2 64.29 67.85 83.31 73.17 2 41J °2 23.68 21.00 11.99 17.13 Hrs. A 12.03 10.81 4.70 9.70 43 ,5 89.8 co 2 Trace 0.34 n 2 70.25 81.58 82.85 83.94 3 591 °2 15.67 15.06 13.39 12.99 Hrs. A 13.81 3.35 3.75 3.07 24.3 27.2 22.3 co 2 0.27 Trace Trace Trace n 2 76.66 80.86 83.73 85.81 4 24 °2 12.63 16.55 13.61 11.98 Hrs. A 10.71 2.59 2.65 2.21 24.2 24.8 20.7 co n 2 TABLE 3 THE EFFECT OF HIGH AMBIENT ARGON CONTENT ON THE DIFFUSION OF ARGON INTO THE SWIMBLADDER OF NONRESECRETING lAGODON RHOMBOIDES. ALL VALUES REPORTED IN Experiments with Neon: Neon experiments indicate a relationship between the per cent diffusion of theoretical maximum and the duration of the experiment (Table 4). The longer the experiment was run the greater the diffusion with one exception. However, the rate of diffusion of neon was much slower than of argon. It took 64 hours for neon to attain the same per cent diffusion of theoretical maximum that argon^attained in 24 hours. The air phase composition did not change sig nificantly during the experiments. Temperature varied from 28 to 32°C. All values are reported in Exp. Time Initial Final % Diffusion of Maximum No. of Gas Air- Air- into Swimbladder Exp. phase phase Fish 1 Fish 2 Fish 3 Fish 1 Fish 2 Fish 3 N 2 35.15 67.76 75.46 73.45 02 48.77 27.46 19.77 24.35 1 64 Hrs. Ne 14.81 3.35 3.50 1.32 ’22.6 23.6 8.9 co 2 1.27 1.43 1.63 0.88 »2 57.84 57.43 78.33 83.42 °2 28.25 26.73 19.58 14.56 2 40- 5/4 Ne 13.91 15.46 1.51 2.02 9.8 13.1 Hrs. CO? 0.36 0.57 ' -.. TABLE 4 EFFECT OF HIGH AMBIENT NEON CONTENT ON THE DIFFUSION OF.NEON INTO THE SWIMBLADDER OF NONRESECRETING LAGODQN RHOMpOIDES. Experiments with Gas Mixtures: A mixture of gases containing argon, hydrogen, and helium was used initially and their analyses performed on the mass spectrometer (Table 5). Hydrogen diffused com pletely out of the air phase . In other experiments using hydrogen it was found to diffuse out of the air phase in six hours. Hydrogen never reached a significantly measurable concentration in the swimbladder. Of the three gases introduced, argon had the highest per cent diffusion of theoretical maximum. The nitrogen and oxygen percentages of the swimbladder gas remained about normal during the ex periment. The next two experiments were run using mixtures of gases containing argon, neon, krypton, and xenon. One was for nonsecreting animals (Table 6) and the other was for resecreting animals (Table 7). Fish one of the nonsecreting animals was obviously secreting gas during the experiment for its nitrogen concentration was low and its oxygen concentration was high. The other two fish had normal oxygen and nitrogen concentrations in the swimbladder. Oxygen values for resecreting fish were high and nitrogen values were low (Table 7). The per cent diffusion of theoretical maximum into the swimbladder was calculated for the introduced gases and nitrogen for the gas mixture experiments (Tables 5, 6 and 7). These values were plotted against the molecular weights of the gases (Figure 2). The graph in Figure 2 shows that the relative diffusion rates are related to the molecular weight. Nitrogen did not follow this pattern. It was higher than expected in all cases but one. This could be explained by the presence of a very high concentration of nitrogen in the swimbladder at the beginning of the experiment. Unfortunately, only two resecreting fish were used and one had low nitrogen and one high. Gas N 2 °2 co 2 Ar He H 2 Initial Airphase 58.7 20.3 0.16 3.4 5. 8 10.2 Final Airphase 63.7 27.6 0.27 2.7 4.1 .025 Fish' 1 70.4 25.2 0.43 1.25 0.54 0.44 Fish 2 77.6 19.13 0.17 1.10 0.20 0.15 Fish 3 77 .04 19.56 0.20 1.18 0.26 0.15 % Diffusion Fish 1 110.5 46.3 13.2 of Maximum Fish 2 121.8 40.7 4.87 into Swimbladder • Fish 3. 121.0 43.7 6.35 TABLE 5 EFFECT OF AN ARTIFICIAL GAS MIXTURE ON DIFFUSION OF ARGON, HELIUM, AND HYDROGEN INTO THE SWIMBLADDER OF NONRESECRETING LAGODON RHOMBOIDES. ALL VALUES REPORTED IN Temperature varied from 17.5 to 20.0°C. Duration of Experiment was 25 hours Gas N 2 °2 C0 2 Kr Xe Ar Ne Final Airphase 66.77 12.79 0.76 2.85 2.34 7.05 7.44 Fish 1 56.04 34.91 5.10 0.76 1.33 1.59 0.28 Fish 2 78.08 10.09 1.07 2.12 2.20 4.12 1.32 Fish 3 80.50 10.84 1.00 1.62 1.94 3.26 0.83 Diffusion Fish 1 84.5 26.6 57.0 22.6 3.70 of Maximum Fi sh 2 118.6 74.4 94.0 58.5, 17.8 into Swimbladder Fish 3 120.5 57.0 83.0 46.2 11.2 TABLE 6 THE EFFECT OF AN ARTIFICIAL AMBIENT GAS MIXTURE ON THE DIFFUSION OF INERT GASES INTO THE SWIMBLADDER OF NONSECRETING LAGODON RHOMBOIDES. ALL VALUES REPOR Temperature varied from 25 to 26°C. Duration of Experiment was 25 hours Gas K 2 °2 co 2 Kr Xe Ar Ne Final Airphase 36.78 35.53 0.30 3.18 2.62 9.24 11.87 Fish 1 33.13_ 47.90 3.37 2.18 2.08 6.13 5.21 Fish 2 7.31 83.81 1.69 1.25 2.05 2.38 1.52 % Diffusion of Maximum into Swimbladder Fish Fish 1 2 90.0 19.9 68.5 39.3 79.4 78.2 66.4 25.8 43.8 12.8 TABLE 7 THE EFFECT OF AN ARTIFICIAL AMBIENT GAS MIXTURE ON THE DIFFUSION OF INERT GASES INTO THE SWIMBLADDER OF . RESECRETING lAGODON RHOMBOIDES. ALL VALUES REPORTED IN Solubilities: Plotting the Bunsen Coefficient (Seidell, 194=0; Seidell and Linke, 1952) of the inert gases against the molecular weight (Figure 3) shows that solubility is proportional to molecular weight, the heaviest gases being most soluble. Hydrogen has a higher solubility than expected which is possibly the result of a high degree of hydration, while argon also shows a higher solubility than expected. FIGURE 2 RELATIONSHIP OF DIFFUSION OF INERT GASES INTO THE SWIMBLADDER OF LAGODON RHOMBOIDES AND MOLECULAR WEIGHT. High nitrogen values are attr ibutable to their high initial value. FIGURE 3 ABSOLUTE SOLUBILITY OF VARIOUS GASES AS A FUNCTION OF MOLECULAR WEIGHT DISCUSSION Comparison of the plot for the per cent diffusion of theoretical maximum for individual gases against molecular weight (Figure 2) with the graph of the Bunsen Coefficient against molecular weight (Figure 3) shows that the entry of inert gases into the swimbladder is directly related to solubilities and as a result their respective molecular weights. The groups of curves, as well as individual curves, show a general higher diffusion tendency with increased molecular weight, and these at least roughly follow the solubility of each gas shown in Figure 3. The 100 X ratio, for steady state gases found in the swimbladders of fishes, is essentially that found in the atmosphere (Scholander, 1954; Scholander and van Dam, 1956; Tait, 1956; and Sundes et al., 1958). Using argon, neon and helium, Wittenberg (1958) showed that the 100 X a/n 2 ratio of newly secreted gases closely approaches the value in air saturated water, but is far larger than that for air. He found that Ne and He being less soluble than N 2 were relatively low in the swimbladder gas. He concluded that for the goldfish, Carassius auratus (Linnaeus), the presence of pure Ng in the swimbladder was accomplished by secretion of oxygen and entrapment of Ng and a differential reabsorption of oxygen. Copeland (1952 a and Safford (1940) have shown that 0g is removed preferentially from the swimbladder of Fundulus and several species of marine fish. Hall (1924) found that for bass, perch, and carp, the swimbladder could act as a reservoir of 0g which could be drawn on in times of metabolic need. Since argon and nitrogen at equilibrium exist in a definite ratio in the swimbladder gases, it may be postulated that nitrogen enters the swimbladder via the same mechanism as argon. The exact mechanism for the transport of inert gases into the swimbladder is unknown. Nevertheless, the results of Wittenberg (1958) for helium, argon and neon and the data reported here for the same gases plus krypton and xenon, which relate entry to gas solubility, appear to lend credence to a passive mechanism of inert gas "secretion," like that postulated by Powers (1932). SUMMARY 1. Normal gas composition in the swimbladders of pinfish, Lagodon rhomboides, acclimated to 10 to 15 inches of water is essentially the same as that for the atmosphere Oxygen concentrations in the swimbladder increase with depth and nitrogen concentrations decrease. 2. Active secretion of oxygen is involved in reinflation of the swimbladder. 3. Buffering with trisma 8.3 buffer removes carbon dioxide from the water. 4. Nitrogen diffusion out of the swimbladder is negligible over a 24 to 44 hour period. 5. Entry of inert gases into the swimbladder is a function of their solubilities and is related to molecular weight The more soluble gases enter more rapidly. 6 . All inert gases enter the swimbladder most probably via a common mechanism. They possibly are "secreted" passively via scavenging by a mechanism involving active secretion of tiny bubbles of oxygen (Powers, 1932; Wittenberg, 1958 ) . REFERENCES Biot, M. 1807 Sur la nature de l’air contenue dans la vessie natatoire des poissons. Mem. Phys. Chim. Soc. d'Arcuiel, 1, 252. Copeland, D. E. 1952 a The histophysiology of the teleostean physoclistous swimbladder. J. Cell. Comp. Physiol., 40, 317-334. Copeland, D. E. 1952 c The stimulus of the swimbladder reflex' in physoclistous teleosts. J. Exptl. Zool., 120, 203-212. Hall, F. G. 1924 The function of the swimbladder of fishes. Biol. Bull., 47, Hlifner, G. 1892 Zur physikalischen chemie der Schwimmblas en gase. Arch. Anal. u. Physiol., Physiol. Abt., 59. McFarland, W. N., and K. S. Norris. 1958 The control of pH by buffers in fish transport. Calif. Fish and Game, 44, 291-310. McFarland, W. N., and K. S. Norris. 1961 The use of amine buffers in the transport of fish. Conf. Proc. N.Y. Acad. Sci. (in press). Powers, E. B. 1932 The relation of respiration of fishes to environment. X. Mechanism of the deposition of gases into the swim-bladder . Ecol . Monographs, 2, 443-465. Richard, J. 1895 Sur les gas de la vessie natatoire des poissons. Compt. rend. Acad. Sci., 120, 745. Rostorfer, H. M. 1942 The gas content of the swimbladder of the rock bass, Amblipotes rupestris, in relation to hydrostatic pressure. Biol. Bull., 82, 138-153. Safford, V. 1940 Asphyxiation of marine fish with and with out carbon dioxide and its effect on the gas content of the swimbladder. J. Cell. Comp. Physiol., 16, 165-173. Saunders, R. L. 1953 The swimbladder gas content of some fresh water fish with particular reference to the physostomes. Canadian J. Zool., 31, 547 -560. Schloesing, T., and J. Richard. 1896 Recherche de l’argon dans le gay de la vessie natatoire de poissons et des physalies. Compt. rend. Acad. Sci ~ 122, 615 . Scholander, P. F» 1954 Secretion of gases against high pressures in the swimbladder of deep sea fishes. 11. The rete mirable. Biol. Bull., 107, 260-277. Scholander, P. F. 1956 Observations on the gas gland in living fish. J. Cell. Comp. Physiol., 48, 523-528 Scholander, P. F., and L. van Dam. 1953 Composition of the swimbladder gas in deep sea fishes. Biol. Bull., 104, 75-86. Scholander, P. F., and L. van Dam. 1954 Secretion of gases against high pressures in the swimbladder of deep sea fishes. I. Oxygen dissociation in blood. Biol. Bull., 107, 247-259. Scholander, P. F., and L. van Bam. 1956 Nitrogen secretion in the swimbladder of whitefish. Science, 123, 59-60. Scholander, P. F., L„ van Dam, and T. Enns. 1956 The source of oxygen secreted into the swimbladder of cod. J. Cell. Comp. Physiol., 48, 517-522. Seidell, A. 1940 Solubilities of Inorganic and Metal Organic Compounds, Vol. I, 3rd Ed. D. van Nostrand Co., New York. Sei,dell, A., and W. F. Linke. 1952 Solubilities of Inor ganic and Organic Compounds, Supplement to the 3rd Ed., D. van Nostrand Co., New York. Sundes, G., T. Enns, and P. F. Scholander. 1958 Gas secretion in fishes lacking rete mirabile. J. Exptl . Biol., 35, 671-676. Tait, J. S. 1956 Nitrogen and argon in salmonoid swimbladders. J. Zool . , 34, 58-62. Wittenberg, J. B. 1958 The secretion of inert gas into the swim-bladder of fish. J. Gen. Physiol., 41, 783- 804 Wittenberg, J. B. 1961 The secretion of oxygen into the swimbladder of fish. I. The transport of molecular oxygen. J. Gen. Physiol., 44, 521-526 . Wittenberg, J. 8., and B. A. Wittenberg. 1961 The secretion of oxygen into the swim-bladder of fish. 11. The simultaneous transport of carbon monoxide and oxygen. J. Gen. Physiol., 44, 527-542. The vita has been removed from the digitized version of this document.