Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Influence of sea water on the osmoregulatory mechanisms of the rainbow trout (Salmo Gairdneri) Milne, Robert Stephen 1974

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-UBC_1974_A6_7 M54.pdf [ 3.26MB ]
JSON: 831-1.0099906.json
JSON-LD: 831-1.0099906-ld.json
RDF/XML (Pretty): 831-1.0099906-rdf.xml
RDF/JSON: 831-1.0099906-rdf.json
Turtle: 831-1.0099906-turtle.txt
N-Triples: 831-1.0099906-rdf-ntriples.txt
Original Record: 831-1.0099906-source.json
Full Text

Full Text

* 95 THE INFLUENCE OF SEA WATER ON THE OSMOREGULATORY MECHANISMS OF THE RAINBOW TROUT (SALMO GAIRDNERI) by ROBERT STEPHEN MILNE' B.Sc. (Hons.), University of Br i t ish Columbia, 1971 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in the Department of Zoology We accept this thesis as conforming to the required standard THE-UNIVERSITY OF BRITISH COLUMBIA January 1974 In p r e s e n t i n g t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the r e q u i r e m e n t s f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h Co lumb ia , I a g ree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s tudy . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the Head o f my Department o r by h i s r e p r e s e n t a t i v e s . It i s u n d e r s t o o d tha t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i thou t my w r i t t e n p e r m i s s i o n . Department o f Zoology The U n i v e r s i t y o f B r i t i s h Co lumbia Vancouver 8, Canada Date I a January 1974 i i ABSTRACT The effect of S.W. on the osmoregulatory processes of Rainbow trout was studied. Rainbow trout were monitered for various and respiratory parameters during F.W. - S.W. transfer. Plasma Cl~, total COg and HCOg concentrations were measured, as well as blood pH, Pa02 and PC02- V0 2 > VG, VC02 > and VH were calculated from the measured parameters. During the exposure to S.W. for 3 hours none of the measured variables changed s ign i f icent ly , indicating that the Rainbow trout is indeed euryhaline. It appears that unlike the goldfish the trout has a branchial C1~/HC0~ exchange diffusion pump that is of small capacity^ because changes in the external concentrations of CI and HCO^  did not affect internal levels of these ions. Raising external C l " levels would cause a lowering in plasma HCO^  i f there was a CWHCO^ exchange diffusion pump in the g i l l s . A second group of experiments involved the use of Cl to measure C l " fluxes in the g i l l s of intact trout. Injections of HCO^  into the f i sh did not stimulate C l " ef f lux , indicating that i f there is a C1"/HC03 exchange diffusion pump, i t is of small capacity. G i l l carbonic anhydrase levels were measured in the g i l l s of F.W. and S.W. trout, S.W. and F.W. coho, and goldfish. Goldfish have a high level as does the S.W. coho. Carbonic anhydrase catalyses the formation of HCO" from C02 for the Cl'/HCO^ pump, and is thus indicative of the importance of the pump in C l " and HCO^  fluxes.. The significance of the differences in the ionic pumps of trout, coho, and goldfish are discussed in relation to their l i f e histor ies. i n Trout having a f lex ib le osmoregulatory system (being euryhaline) while the coho and goldfish are more specialised being essentially stenohaline. iv TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS • iv LIST OF TABLES v LIST OF FIGURES vi ACKNOWLEDGMENT vi i i INTRODUCTION 1 METHODS 13 RESULTS . 20 DISCUSSION 42 LITERATURE CITED . 60 APPENDIX 1 67 APPENDIX 2 74 APPENDIX 3 ' 80 I Table LIST OF TABLES Page I. Branchial C l" Fluxes in F.W. Rainbow Trout 35 II. Branchial C l " fluxes in S.W. Rainbow Trout 37 III. Concentrations of Carbonic Anhydrase in G i l l s of Experimental Fish IV. Stat is t ica l Analysis of Differences in Carbonic Anhydrase Levels of G i l l s . . V.. Comparison of G i l l Carbonic Anhydrase Levels In Various Fish VI. Various Respiratory and Acid-Base Parameters in S.W. Trout 40 41 50 73 vi LIST OF FIGURES Figure Page 1. Pathways of water and salt movement in F.W. and S.W. f ish 3a 2. Model of F.W, goldfish g i l l 7a _ 3. Model of S.W. f ish g i l l 10a 4. Arterial oxygen tension during exposure to S.W 21a 5. Blood Cl" concentration during exposure to S.W 22a 6. Total CO2 of the blood during exposure to S.W. . . . . 23a 7. Plasma PCOg during exposure to S.W 24a 8. Plasma HCO" concentration during exposure to S.W. . . . 25a 9. Blood pH during exposure to S.W. 26a 10. % change in blood pH during exposure to S.W 27a 11. Ventilation volume during exposure to S.W 29a 12. H + ion excretion during exposure to S.E. 30a 13. CO2 excretion during exposure to S.W .. . 31a 14. O2 uptake during exposure to S.E 32a 15. Blood pH during long term exposure to S.W 33a 16. C l " fluxes in F.W. trout 36a 17. C l" fluxes in HC0~ stimulated F.W. trout . . . . . . . . . . 38a 18. Kerstetter's data on C l " fluxes in HCO^  stimulated Trout 47a 19. Model of F.W. and S.W. f i sh g i l l s . 52a 20. Model of trout g i l l in fresh water . 53a v i i Figure Page 21. Model of trout g i l l in S.W. 54a 22. Heart rate during long-term exposure to S.W 68a 23. Or, uptake during long-term exposure to S.W 69a 24. Ventilation rate during long term exposure to S.W. . . 70a 25. Arterial Or, tension during long-term exposure to S.W. . 71a 26. Ventilation rate in control F.W. group 75a 27. Heart rate in control F.W. group 76a 28. Blood pH in control F.W. group . 77a 29. Arterial tension in control F.W. group . . . . . . . 78a 30. consumption in control f ish 79a 31. Mortality curve for goldfish in S.W 81a ACKNOWLEDGEMENTS I wish to express my thanks and appreciation to Dr. D.J. Randall for his fortitude and advice during my tenure as his student. I also wish to thank Dr. R. Lester for introducing me to those strange creatures of the sea; f i s h . Thanks are due to my mother for her help in translating this thesis into English. 1 INTRODUCTION Animals l i v ing in water can be divided into two types, osmoregulators and osmoconformers. The former must keep their body f lu id ion concentrations at a constant level to permit their ce l ls to function normally. Any deviation from this constant internal medium results in the death of the c e l l s , leading ultimately to the death o f ~ the whole organism. Osmo-conformers, on the other hand, permit their internal medium to assume the same osmotic content as the external medium, thus showing that, to survive, either the ce l ls of the organism must be capable of functioning in a variety of bathing solutions, or the . animal must be restricted to a single environment (stenohaline). Teleost f ish are osmoregulators since they regulate their internal body f lu id composition. The osmolarity of the blood shows l i t t l e v a r i a t i o n . The composition of sea water (S.W.) and fresh water (F.W.) are so dissimilar that the methods of ion regulation employed by f ish w i l l be decidedly different according to the medium in which they l i v e . S.W. has a high concentration of Na , Cl and C 0 2 (most_of which is in the form of HC0~, since the reaction C02•+ H20 t H 2 C0 3 t H + + HCO3 is favoured to the right at the pH of S.W. so that the f ish faces a problem of water exit and ion inf lux. In F.W.., on the other hand, there is a low concentration of these ions, resulting in a loss of ions and a gain of water by the f i s h . Because of the different mechanisms of ionic and osmotic regulation which are required, most f ish are restricted to either F.W. or S.W., that is they are stenohaline. There are, however, some f i s h , known as euryhaline, which l i ve in both media. Salmonids are one of the better known examples of a euryhaline f i s h . The g i l l s and the kidneys are the main sites of ion exchange is f ish since they are the only parts of the integument which are readily permeable to ions [Gordon, 1963]. The g i l l s are permeable because they are-also the s i te of gas exchange in the animal, and one of the conse-quences of a surface being permeable'to gases is that i t is also permeable to ions. This is why the f ish cannot solve i ts ion regulation problem by surrounding i t s e l f with an impermeable integument. Due to their large amount of infolding, the g i l l s are about 10 to 60 times larger than the animal's surface area, so that there is an appreciable amount of ion movement due to diffusion between the f ish and the surrounding water. F.W. f ish tend to gain water and lose ions via the g i l l s as their body f luids are more,concentrated than the water. To counteract this they take up ions via the g i l l s and expel 1 the water via the kidneys by producing a very di lute urine. S.W. teleost f ish face an influx of ions and a loss of water. To compensate for this they drink water, extract the ions from the water via the intestine, and then excrete these ions, together with those that have entered the f i sh via the g i l l s , by active transport across the g i l l or via a urine isoosmotic to S.W. (see Figure 1). The g i l l s are arranged in a series of branchial arches sub-divided into filaments bearing secondary lamellae. They are served by Figure 1 Pathways of water (w) and salt (s) movement in freshwater and seawater f ish OCL f:W. HYPOTONIC URINE - j i 4 a blood supply that runs from the ventral aorta, through the g i l l s and then via the dorsal aorta and the rest of the circulatory system back to the heart. The secondary lamellae contain two dist inct ce l l types; the specialised ce l ls in the outer layer of the epithelium which have desmosomes and interdigitating membranes; and the less specialised ce l ls of the inner epithelium layer, the p i lar ce l ls [Berridge and Oschman, 1972], Keys [1932] identif ied a particular cel l type in the interlamellar spaces which appeared to be connected with ion transport. More recently, i t has been shown that these ce l ls are extremely rich in mitochondria (a characteristic of ce l ls concerned with active transport) and that they contain various enzymes, Na -' K activated ATPases. and perhaps carbonic anhydrase [see Review of Conte, 1970]. This latter enzyme is present in the g i l l s of f ish [Maren, 1967] where i t catalyses the reaction HgO + C02 j h^CO .^ Its main function, however, is to speed up the formation of HCOg in the red blood ce l ls from the C02 that has entered from the tissues. At the gas exchange surface i t then acts to catalyse the formation of C02 which leaves the red blood c e l l s . The level of carbonic anhydrase ( c a . ) in f ish blood is about 2% of that of mammals. The significance of this fact is not c lear , but i t could be related to the low level of HCOg in f ish blood. The role of the carbonic anhydrase enzyme found in the g i l l s appears to be twofold; gas exchange and ion transport [Maren, 1967]. In goldfish, diamox, an inhibitor of carbonic anhydrase, inhibited C l " uptake as well as Na+ uptake [Maetz and Romeau, 1964]. This suggests a role for carbonic anhydrase in ion transport as well as a respiratory function. The two functions appear to be connected in that the c a . catalyses HCO" formation in the g i l l , with the HCO^  excretion linked to C l " uptake. The precise location of c a . in the g i l l s is important in determining i t s role in CG^  excretion. If there is no c a . in the lamellae then most of the COg lost by venous blood passing over the lamellae must be in the formof free CC^. If there is c a . in the lamellae, most of the CC^ could be excreted in the form of HCOg •• Unfortunately i t is not known exactly where the c a . is located in the g i l l s , although the experiments outlined below indicate that i t is involved in ion regulation in some f i s h . It is by no means certain that the ce l ls involved in ion transport are chloride ce l ls [Conte, 1970]. The mechanism of ion .transport in the g i l l s of f i sh has been the subject of much investigation since the discovery of radioactive isotopes. Maetz and Romeau, using the stenohaline F.W. goldfish, showed that the Na uptake and the Cl. uptake are independent of each other and that their movement is linked with that of other ions. It had been suggested earl ier by Krogh [1939] that Cl~ was exchanged with HCO^  and Na+with NH .^ To verify this Maetz and Romeau examined the effects of changing blood and water HCO" and C l " levels in g i l l Na+ and C l " fluxes. + + It was found that raising the external NH^  concentration slowed the Na inf lux, while raising the blood NH^  level; stimulated Na + influx via the g i l l s . S imilar ly , Cl~ influx was slowed by the addition of HCO^  to the water and was stimulated by the addition of HCO^  to the f i s h . Another finding which supports the concept of linked Na+/NH^ and Cl'/HCO^ exchanges is that acetotazolamide, a carbonic anhydrase inhib i tor , when administered to goldfish slows down Na and Cl" influxes. Maetz and Romeau subsequently proposed the following model for ion excretion in F.W. f ish (see Figure 2, p.7a ). Although the model appears to imply that a l l the CC»2 is excreted via the HC0~/C1~ exchange system, Maetz himself states "the principal route of excretion of excess HCO^  is a question necessitating further study . . . a further question concerns the form in which this excess HCO^  being excreted crosses the g i l l epithelium" [Maetz and Romeau, 1964, p. 1223], It appears that in the goldfish most of the HC0~ being excreted is exchanged with C l " . This finding is supported by the work of Dejours' [1969] who found that C02 excretion in goldfish is dependent on the Cl~content of the water. When the f ish is transferred from a solution containing C l " to one with no Cl~, there is a sudden reduction of g i l l C02 output. Upon restoration of the high Cl~ medium, output is resumed and considerably increased. These observations are best explained on the basis of a Cl'/HCO^ exchange in the g i l l s : HCOg entry against C l " loss after supression of external C l" and massive HCOg exit upon restoration of the Cl~ solution as a result of complementary C l " uptake. This interpretation suggests that much of the C02 released by goldfish is in the form of HCO^  • Other experiments indicate that there may be similar exchange mechanisms in the euryhaline trout. Kerstetter et al_. [1970] found a progressive downward shi f t of the pH of the surrounding medium as the Na+ influx of the f ish increased. In addition, diamox (acetalazolamide) injections prevented the downward s h i f t , although this might have been due to the diamox inhibit ing the R.B.C. carbonic anhydrase and preventing Figure 2 Model of fresh water f ish g i l l showing the ionic pumps involved and movement of ions. From Maetz [1971], p. 229 8 C0"2 exit from the R.B.C. It appears that in the trout Na +is exchanged with H rather than NH^  as in the goldfish. However i t has been found recently [Maetz, 1973] that in the goldfish both NH^  and H + are exchanged with Na+in the g i l l . Another paper [Kerstetter, 1972] pur-ports to show that C l " is exchanged with HCO" in the trout. Injections of HCO" into the trout stimulated C l " influx via the g i l l s . However there are serious objections to this conclusion, not the least being that the f ish were anaesthatised during the experiment. Lloyd and White [1967] showed that trout exposed to high concentrations of free C02 in the surrounding medium, exhibited a drop in their plasma Cl" concentration. Although this might appear to show that the high C02 in the water, by preventing C02 excretion, is stopping C l " uptake (meaning that C l " and HCO^  movements are linked) one must remember that only C l " plasma concentration, not C l " fluxes were measured. It is entirely possible that the C l " was excreted via the kidneys to produce the lowered plasma concentrations. Although the transport of C l" and Na+ is presumably linked with that of other ions in F.W., the influx is nevertheless by active transport. To account for the recorded rates of influx by passive diffusion alone would require a transepithelial potential across the g i l l of as high as -190 mv. Such a high biological potential has never been recorded [Maetz, 1971]. In the few instance where the g i l l transepithelial potential has been recorded, i t appears to be about -20 mv in F.W. The active intake of Cl and Na is linked with the passive efflux of HCO ,^ H + and NH^. Such a linkage is an example of exchange dif fusion, which is the coupling of the "uphi l l " transport 9 of one ionic species with the "downhill" transport of another ionic species. This allows the expenditure of minimum energy by the animal [Ussing, 1952]. One important corollary of this process is that the change in concentration of one ion affects the flux rate of the other ion. This we have seen in the goldfish, where changing levels of HCO^  affect Cl" f lux , but this is not necessarily true in the trout or other f ish species. Maetz and his associates have presented us with a picture of the ion processes occurring in the g i l l s of S.W. f i s h . On the outer border of the g i l l there is a Na+ active transport pump which extrudes + + . + + + Na . This Na efflux may occur in conjunction with K inf lux. A Na /Na exchange diffusion pump may also be present, but i t s function is purely + + passive as the inward flux of Na drives the efflux of Na . Thus the Na+ entering the g i l l s is promptly excreted with a minimum of energy + expenditure. There is also a small diffusionable component of Na entry, as not a l l the Na+ entering the f ish via the g i l l s enters by exchange diffusion. This entry of Na together with the Na absorbed by the intestine is excreted via the active Na+ pump.- C l " ion movements across the g i l l s of marine f ish are similar to Na movements. There is a C1~/C1.~ exchange diffusion mechanism, active transport of C l " and a passive influx of C l " (see Figure 3, p.lOa). A number of rapid transfer experiments have been performed on stenohaline S.W. f ish as well as on euryhaline f ish [Maetz, 1971]. Stenohaline S.W. f i s h , when suddenly transferred to F.W. usually do not reduce their rapid efflux of C l " and Na + , so that they eventually die from excessive sal t loss. Other stenohaline f i s h , once in F.W., slow down Na+ and C l " ef f lux , but Figure 3 Model of seawater f ish g i l showing the ionic pumps involved and the ions involved. From Maetz [1971], p. 229 .5W. GILL BLOOD 11 nevertheless s t i l l die. Euryhaline f i s h , however, show a second or delayed response of slowing down Na+ and C l " efflux by gradual decrease in the permeability of the g i l l to ions. This prevents sal t loss and gives the f ish time to adjust the ionic pump to respond physiologically to the new environment. The stenohaline f ish dies because i t is using up a large amount of energy to prevent sal t loss and i t cannot continue to do th is , while the euryhaline f ish prevents salt loss by means of reduced g i l l permeability. Another difference between marine stenohaline and euryhaline f ish is that, while both reduce active efflux of Na+ and C l " , the stenohaline f ish w i l l not reduce i t s efflux i f the osmolarity of S.W. is maintained by the addition of mannitol. It therefore is responding to an osmotic effect rather than to an ionic effect as the euryhaline f ish does. The question of CC^  excretion in marine teleost f ish has not been researched' extensively. Hodler et al_. [1958] determined that in the dogfish the principal route of C02 excretion is the g i l l s and that the rate of CO^  excretion was affected by acetazolamide. This suggests that C02 is excreted in the form of HCO". Whether this occurs in conjunction with another ion is a matter of conjecture, there being no other information on C02 excretion in S.W. elasmobranchs or teleosts. It can be seen that the mechanisms for ion regulation of f ish d i f f e r , depending on the medium and the type of f i s h . Most of the studies of euryhaline f ish involve f i sh which normally l i ve in the sea and the transfer has therefore been from S.W. to F.W. I decided to examine the ionic regulation of a euryhaline f ish normally resident in 12 F.W., the rainbow trout. The question I set out to answer was whether a Cl~/HC0~ exchange diffusion mechanism was present in the trout, and, i f so, was this exchange also present when the f ish was in S.W. If this exchange mechanism continued to function in S.W. i t would lead to salt loading and the death of the f i s h . The easiest way to study this question was to observe the changes in the acid base parameters of the blood during the transfer of the f ish from F.W. to S.W. If a Cl~/HC0g linkage existed in trout, there would be an increase in HCO^  excretion and a f a l l in plasma HCO^  upon the entry of the f ish into S.W. A second group of experiments were performed, in which HCO^  was injected into trout and the Cl" influx measured. The C l " influx should increase i f a HCO^/Cl" exchange diffusion mechanism is present. The reason why the experiments were performed was that I suspected that, since the trout is euryhaline, .it would not have the same ionic control system as the stenohaline goldfish. There is no evidence that the ionic system of the goldfish is in any way pre-adapted to l i f e in S.W., whereas one would hypothesize that that of the trout is so adapted. F inal ly , an attempt was made to examine the level and the location and concentration of carbonic anhydrase in the g i l l s of various f i s h , as this enzyme appears to be important for ionic regulation. 13 MATERIAL AND METHODS Part 1 Rainbow trout of approximately 250-325 gms were cannulated in the dorsal aorta in the manner described by Smith and Bell [1964], and then placed in venti lation boxes similar to those used by Davis and Cameron [1970]. The rubber membrane which is sewn around the mouth of the f ish separates the water flowing from the g i l l s from the water enter-ing the g i l l s via the mouth, thus permitting separate sampling of inspired and expired water. After the operation the f ish were allowed to recover for at least 18 hours before the start of experiments, as Houston [1971] has shown that the disturbing effects of the anaesthetic (MS 222) on blood ion levels persist for several hours. The experimental procedure was as follows: .a number of parameters were measured when the f ish was in F.W. and also when i t was exposed to S.W. at 45, 90, 135 and 180 minutes respectively. The parameters measured were: pH of blood, inspired and expired water plasma Cl~ concentration (mM/1) total C02 of plasma (mM/1) arter ia l 0 2 tension (Pa02 mmHg) venti lation volume (VG mls/min) from which the below were calculated. 14 C02 excretion (VC02 mM/min/Kg) 0 2 uptake (V02 ml/min/Kg) + + H ion excretion (VH mM/min/Kg) pH and P0 2 were measured using a Radiometer PHM27 Gas Analyser and micro pH and P0 2 electrodes at the same temperature as the f ish (10°C). Plasma Cl was measured by a Buchler-Cotlove chlorideometer following the procedure of Siggaard-Anderson [1967]. Total CG*2 of the plasma was measured by the method of Cameron [1971], which u t i l i ses a PC02 electrode to measure the change in PC02 of a HC1 solution caused by the addition of a known amount of plasma. This change is compared to the change caused by a known standard of HCO .^ The PC02 and the,HCO^ concentration of the plasma were calculated from plasma pH and total C02 via the Henderson-Hasselblach equation with the so lub i l i t y of C02 in f ish plasma and the pK of C02 taken from Albers [1970]. VG was measured direct ly from the outflow of the boxes using a measuring cylinder. In F.W. the C02 excretion (VC02) was measured by collecting inspired and expired water samples of approximately 500 mis under Hexane in separatory funnels. 10 mis of 6M NaOH and 100 mis of IM BaOH were added and s t i r red , forming a precipitate (BaCO )^ which was f i l tered onto pre-weighed f i l t e r papers. The amounts of C02 was calculated as follows: 15 wt of BaC03/mlH20 x 30.4% MW of BaC03 wt of BaC03/mlH20 x 30.4% MW of BaC02 moles of C09/mlH90 x VG vco2 = - -weight of f ish In S.W. the VC02 was computed by measuring the difference between the total C02 of the inspired and expired water samples by the method of Cameron [1971]. Hydrogen ion excretion was calculated from the difference in pH of the inspired and the expired water samples. A curve for S.W. and F.W. was constructed to determine the change in the pH of the water caused by the addition of H +ions. From this curve the number of moles of H + excreted could be determined. V02 was calculated from the difference in the P0 2 of the inspired and the expired water samples and VG. Part 2 The second group of experiments performed involved the use of radioisotopes to determine C l " flux rates in rainbow trout. The dorsal aorta was cannulated, the urinary pappillae blocked, and the f ish were then placed in plexiglass boxes each containing 1400 mis of aerated water. Moles of C09/mlH90 = 16 Vrocedwe for F. W. The f ish was l e f t in the box for approximately 1 hour unti l 36 i t had recovered from the stress of handling. 1.26 uC of H Cl was then added to the water. The f ish was l e f t to equil ibr iate for half an hour, after which a 1 ml water sample was taken. Half of the sample was placed in a s c i n t i l l a t i o n vial containing 10 mis of Bray's solution, and the other half was frozen for later total C l " analysis. Further 1 ml samples of water were taken every hour, in most cases up to ten hours in a l l . At the end of the experiment a blood sample was taken via the cannula and pippeted into a s c i n t i l l a t i o n v i a l . A second group of f ish were injected with NaHC03 (Itfmillimole), the experimental protocol being the same as that described above, except that after 2 1/2 hours the f ish received the injection of NaHC03 intravenously. The water samples were taken every 15 minutes during the f i r s t hour and then every half hour unti l the experiment was terminated at the end of 4 hours. Fluxes were determined by the method of Maetz [1956a]. Br ie f l y , the equations used were: , f _ Cl 'ext '• n t 17 3. A_Q A T y q Cl" int AC1 ext A T Cl" int Cl ext Q0 " Q f Cl int = y 1 _ s . a . of plasma where Cl~ext = total quantity of C l" in exterior compartment at any time t Cl~int = the total quantity of Cl~ contained in the interior compartment, exchangeable with the external C l " QQ = the quantity of radio chloride introduced into the C l " ext at a time tg Q = the amount of radio chloride present -in the Cl"ext at any time t = the quantity of radio chloride in the CText at the end of the experiment f . = the C l " flux entering the f ish at any time t f = the C l " flux leaving the f ish at any time t f = net Cl~ flux across g i l l s at any time t . 18 Procedure for S. W. The above procedure cannot be used for S.W. f ish because of the high level of C l " in salt water. It is d i f f i c u l t i f not impossible to obtain f . It is therefore only possible to measure f . The cannulated f ish was placed in the box and allowed to recover from handling. Approximately 10 uC of H Cl was then injected into the f ish via the cannula. After half an hour a blood sample was taken together with a 1/2 ml sample of the external medium. 200 ml of the plasma and the water sample were then put into separate sc in t i l l a t ion v ia ls containing 10 mis of Bray's solution. Another plasma sample was frozen for total C l" analysis. The experiment lasted 6 hours, samples being taken at 0, 1/2, 1, 2, 4, and 6 hours. The f of C l" was calculated using the method of Chan et al_. [1967]: AQet f e 3 6C1/Clpt where Qet is the rate of change of the radioactivity in the water at a 36 time t and Cl/Clpt is the specif ic gravity of the plasma at time t . Total chloride was measured on a Bachler-Chloridometer using the modifi-cation of Siggaard-Anderson [1967]. Urinary cannulae were not needed as the urine flow is extremely small in marine f i s h . Part 3 - Carbonic Anhydrase Measurements The f ish were cannulated in the dorsal aorta so that blood could be extracted and replaced by heparenised saline until the blood was very plae red in colour. To complete the flushing a cannula was 19 implanted in the ventricle of the heart and saline allowed to perfuse the f ish unti l the blood had been flushed from the g i l l s . The g i l l s were excised and homogenized in d i s t i l l e d water before the analysis for carbonic anhydrase was performed following the method of Maren [I960]. C02 is bubbled into a reaction vessel and the change in pH of the solution monitored by an indicator, phenol red. The rate of colour change of phenol red in the g i l l homogenates is compared with i ts rate of change in a control of d i s t i l l e d water. The g i l l homogenate was examined for blood contamination by analysis for haemoglobin, using the method of Crosby and Furth [1956]. The amount of contamination by the blood was found to be insignif icant. Five types of f ish were analysed for g i l l carbonic anhydrase, F.W. and S.W. coho, F.W. and S.W. rainbow trout and goldfish. 20 RESULTS Part 1 - F.W. to S.W. Transfer The results of transferring rainbow trout to S.W. from F.W. are presented below. In Figure 4 the arterial 0 2 tension shows an initial drop to 88 mm Hg after 90 minutes exposure to S.W. However, the Pa02 soon returns to the value found in F.W. There is a definite trend for the plasma Cl" concentration to rise (Figure 5) but this rise is not significant. It is caused by an increase in the influx of Cl" ions due to the large diffusion gradient favouring entry of ions into the fish, but i t can be seen that this increase is compensated for by an increase in Cl" efflux, as the plasma Cl concentration does not rise to any appreciable extent. The three acid-base aparameters total CO2, plasma PCO2 and plasma HCO" concentration (Figures 6, 7, 8) show no trends when the fish is exposed to S.W. although the plasma HCO^  concentration tends to decrease with time after the transfer. This is,however,probably an artifact caused by the process of blood sampling and replacement of the blood with saline. The saline has a lower HCO^  concentration than the blood and therefore when i t is added to the blood tends to lower the plasma HCO^  concentration. The pH of the blood shows a dramatic and significant rise ( .01) after the fish has been exposed to S.W. for 90 minutes (Figure 9). As .there was no accompanying rise in plasma HCO^  levels, the rise in blood pH is suspect. When i t is plotted on the basis of a % change (Figure 1 0 ) , Figure 4 Arterial 0 2 tension of trout (n = 12) during short term exposure to S.W. Error expressed as ± one standard deviation . Pa02 measured in F.W. and then f ish transferred to S.W. at zero time Pa mmHg Figure 5 Blood Cl" concentrations during exposure of trout (n = 12) to S.W. Error expressed as ± one S.D.. C l" concentration measured in F.W. and then f ish transferred to S.W. at 0 time. P LAS MA Cf mM I i ter" o o l 3 5' 00 o Figure 6 Total CG^  i f the plasma during short term exposure of trout (n = 11) to S.W. Error expressed as ± one S.D. Total plasma C 0 2 measured in F.W. and then f ish transferred to S.W. at 0 time. Figure 7 PC02 of plasma during short term exposure of trout (n = 12) to S.W. Error expressed as ± one S.D. PC02 of plasma calculated in F.W. and then f ish transferred to S.W. at 0 time. Figure 8 Plasma HCO^  concentration of trout (n = 12) during short term exposure to S.W. Error expressed as ± one S.D. Plasma HCO "^ concentration calculated in F.W. and then f ish transferred to S.W. at 0 time. Figure 9 Blood pH during short term exposure of trout (n = 12) to S.W. Error expressed as ± one S.D. Blood pH measured in F.W. and then f ish transferred to'S.W. at 0 time. BLOOD pH o o •o o 3 3 8 o oo o Figure 10 Blood pH expressed as % of F.W. value during exposure of trout (n = 11) to S.W. Error expressed as ± one S.D. Blood pH % change measured in F.W. and then f ish transferred to S.W. at 0 time. 28 there is no change in blood pH after the f ish has been in S.W. for 90 minutes. This tends to confirm the suspicion that the original r ise was spurious. We should expect any change in the blood pH to be accompanied by a change in plasma HCO^  levels since the two are inter -related, the hydration of CO2 forming H+ and HCO^  ions. Thus a r ise in HCO^  of the plasma would buffer H + ions, leading to a r ise in pH. During the F.W. to S.W. transfer there was no change in the venti lation rate of the f ish (Figure 11) although the values for VG are extremely variable. This is almost certainly due to differences between individual f i s h , as i t is well known that ventilation rates of f ish are extremely variable [Davis and Cameron, 1970]. There are no definite trends in H + excretion, C02 excretion or 0 2 uptake during the exposure of trout to S.W. (Figures 12, 13, 14). This is to be expected as there is no dist inct pattern in the concentration 1 of these molecules in the blood. However, both VH and V02 are s ignif icantly different from the F.W. value after the f ish has been in S.W. for 90 minutes. V02 is up to 0.75 ml/min/Kg, while VH drops to 0.19 mM/min/Kg. These changes occur at the same time as the Pa02 of the blood drops. The fact that V02 can increase while VH decreases serves to emphasise the point that ion regulation and respiratory process are separated in the g i l l . The f inal figure (15) shows the long term effects of S.W. on blood pH, which remains relat ively constant for the four days of exposure to S.W. The long term experiments were done primarily to determine whether the experimental protocol of confining the f ish by venti lation boxes had any adverse effect of their ab i l i t y to osmoregulate. The experiments were Figure 11 Ventilation rate of trout (n = 12) during short term exposure to S.W. Error expressed as ± one S.D. Ventilation rate measured in F.W. and then f ish transferred to S.W. at 0 time. %1a VG ML MIN"' Figure 12 H ion excretion rate during exposure of trout (n = 12) to S.W. Error expressed as ± one S.D. H+ ion excretion rate measured in F.W. and then f ish transferred to S.W. 30a. J . -1 .1 VH mM mm kg x 10 1 Figure 13 Rate of C02 excretion during short term exposure of trout (n = 10) to S.W. Error expressed as ± one S.D. C02 excretion rate was measured in F.W. and then fish.transferred to S.W. at time 0. Figure 14 uptake during short term exposure of trout (n = 12) to S.W. Error expressed as ± one S.D. 0^ uptake was measured in F.W. and then f ish was transferred to S.W. at time 0. Figure 15 Blood pH during long term (4 days) exposure of trout (n = 14) to S.W. Error expressed as ± one S.D. Blood pH measured in F.W. and then f ish were transferred to S.W. at time 0. o C O 34 carried out also to determine the cause of death of the f ish in S.W. as invariably some of the f ish died soon after exposure to S.W. The findings of the long term experiments are found in Appendices 1 and 2. Part 2 - C l " fluxes Table I shows the g i l l C l " fluxes for F.W. f i s h . It can be seen that 3 out of the 4 f ish showed a positive net f lux , indicating that they were gaining ions, while the other f ish showed a negative net flux indicating that i t was losing ions. However i t was observed during the course of the experiment that a f ish could show net fluxes of either sign within the space of 2 hours. This is i l lustrated in Figure 16. The animals appear to be able to control the intensity of their fluxes readily , although at the start of the experiment this a b i l i t y was more pronounced. This point w i l l be discussed more fu l l y later . . Figure 17 shows the C l " fluxes for the f ish injected with HCO .^ It w i l l be noted that there was a r ise in fQ almost immediately after the injection which was accompanied by a similar r ise in f^. F^  osci l lated at f i r s t but became more or less constant. Table II shows the f of S.W. adapted trout. The efflux is extremely variable with a large standard deviation. The plasma C l " concentrations remained constant. The large S.D. of the plasma Cl concentration is a ref lection of the variation between different f ish and not of variation in the same f i s h . In f i sh no. 8, for example, plasma C l " concentration ranged from 144 Meq/1 to 154 Meq/1, an extremely small range. 35 TABLE I Cl Fluxes in F.W. Fish (Mean Values: 8 Samples From Each Fish) Fish Fluxes in Meq/hr/Kg 1 f n = 1.082 f i = .603 f e = 2 ' f n = 1.15 f i = 4.35 f e = 3.2 3 f n = .27 = 14.8" f e = 15.08 4 f n .399 f i .802 f e = .402 Chloride.concentration of the water 2-3 Meq/l. Figure 16 Branchial C l" fluxes in trout (n = 4) in F.W. Means of f ish are plotted. a. 4 a * " * " " * ^ ^ ;II a a a ^ TIME hrs. 37 TABLE II Summary: C l " Fluxes of S.W. Adapted Fish f = 6.93 ± 3.9 Meq/hour/Kg n = 7 f for the individual f i sh ranges from; . " .123 to 12.95 Meq/hour/fish Plasma C l " concentrations mean = 128.8 Meq/T ±21.3 Figure 17 Branchial Cl fluxes in F.W. trout after injection of NaHCO • N = 5. Means of fish are plotted. Time expressed as hours after the injection at 0 time. I ; 39 Part 3 - Carbonic anhydrase Studies of G i l l s The carbonic anhydrase concentrations in the g i l l s of various f ish are presented in Table III . There is no correlation between levels of carbonic anhydrase and the medium to which the f ish is adapted. S.W. cohoe and goldfish have the highest values, while the F.W. and S.W. trout have the lowest. F.W. coho have a g i l l carbonic anhydrase level s igni f icant ly lower (Student's T test) than that of S.W. coho (see Table IV), while there is no difference between the levels in S.W. and j F.W. trout. The goldfish g i l l has a s ignif icantly higher level of carbonic anhydrase than that of the F.W. trout, S.W. trout and F.W. coho, but i t s level is similar to that of the S.W. coho g i l l . An attempt was made to stain g i l l for carbonic anhydrase by the method of Hansson [1967] with a view to. examining them for local isation of the stain . However i t was found that there was no consistent staining, except for the carti lage rods of the g i l l . Both Maren [1967] and Pearse [1961] have raised doubts about the speci f ic i ty of the staining method, stating that i t stains chondritoen, a component of cart i lage. In view of my results and this opinion, i t was decided not to continue this work. 40 TABLE III Concentrations of G i l l Carbonic Anhydrase Fish Carbonic anhydrase e.u./gm. wet tissue ± l .S.D. n S.W. coho F.W. coho F.W. goldfish F.W. trout S.W. trout 12.65 ± 1.75 8.86 ± 1.5 13.12 ± 2.86 6.76 ±1.15 6.27 ± 0.98 13 23 23 23 13 41 TABLE IV Significant Differences in G i l l Carbonic Anhydrase Levels of Various Types of Fish as Determined by T-test Fish F.W. coho S.W. coho F.W. trout S.W. trout goldfish F.W. coho - .01 .01 .01 .01 S.W. coho - .01 .01 not significant F.W. trout - - not significant .01 S.W. trout - - - - .01 goldfish - - - - -42 DISCUSSION It can be seen that the Rainbow trout has no apparent problem with pH or ion regulation when moving from F.W. to S.W., as the f ish readily adapts to the new medium, where the CT7 Na+ and HCO^  concen-trations are higher than those of F.W. Trout are indeed euryhaline, as they can l i ve in S.W. for at least a month after being transferred from F.W. (see Appendix 2). During the time when the trout were adapting to S.W., blood pH and plasma Cl" concentrations did not change s igni f icant ly . A similar result was found by Rao [1969], who showed that, when Rainbow trout were subjected to media of high sa l in i t y , the plasma Cl" concen-trations of the f ish were on the whole independent of external Cl concentrations. As mentioned in the Introduction, Maetz and Romeau postulated Na+/Nrf|j and Cl"/HCO^ exchange diffusion mechanisms in the goldfish g i l l s , while Kerstetter came to similar conclusions about the Rainbow trout; i . e . that C l" entered and HCO^  le f t the body across the g i l l epithelium. Placing the trout in S.W. should therefore increase Cl~ inf lux , which in turn w i l l stimulate HCO^  eff lux , (assuming that the exchange diffusion pumps are not saturated in F.W.), thus leading to a higher level of plasma C l " and a lower level of plasma HCO .^ Similar ly , the rise in external Na+ caused by the transfer to S.W. would be expected to raise, plasma Na and increase H excretion. The pH of the plasma wi l l depend on the rate of H+ production and removal from the plasma. Variations of total plasma CO,, w i l l affect arter ial pH according to the following reaction: 43 C02 + H20 ^ HCO3 + H + (1) . However, plasma C0"2 levels did not change during FW •> SW transfer, as total C02 production and total plasma C02 remained constant during this period. An increase in HCO^  excretion wi l l result in reaction (1) moving to the right and in the absence of any increase in H+ excretion, this w i l l cause a f a l l in pH, this also did not occur. The tendency for the plasma C l " to rise during the transfer from F.W. to S.W. was due to the inabi l i t y of the f ish to match the rate of C l" efflux to the rate of C l" inf lux. In S.W. Cl" influx wi l l be high, owing to the fact that the diffusion gradient between the f ish and the water favours entry of C l " into the f i s h . As efflux was therefore lower than inf lux , the plasma concentration of C l " rose, but the rate of C l " efflux soon matched that of C l " influx as the f ish adapted to S.W. and plasma Cl" concentrations consequently stabi l ised (see Tables I and II showing that fe was higher in S.W. trout than in F.W. trout). This i n i t i a l r ise in plasma C l " level for f ish adapting to S.W. has been observed by other investigators. Gordon (1963) found that Rainbow trout placed in S.W. momentarily raised their plasma Cl" levels. Parry (1961) showed that Salmo salar adults increased their plasma C l " concentration from 156 Meq/1 to 172 Meq/1 when entering S.W. As elevating the external ion levels did not e l i c i t a change in either the pH or the HCO^  and C l " concentrations of the f ish 's plasma, we can conclude either that there is no CWHCO^ exchange diffusion similar to that described by Maetz for the goldfish, or that, i f one 44 does exist , i t is operating at a low capacity. ' Holeton and Randall [1967] measured the 0^ consumption of trout at 12°C to be 1.04 mis/Kg/min. If one assumes a R.Q. of 0.8, this gives a value for C02 excretion of 498 mM/Kg/hr. This is similar to the value of 600 mM/Kg/hr given for F.W. trout in Figure 13. Assuming that CO^  is the source of the HCO^  and H+ excreted via the Cl'/HCO^ pump and given that there is a 1:1 ratio for the ions in the pump, then, i f a l l the C02 is excreted in the form of HCO ,^ C02 excretion should equal C l " inf lux. However, as Table I shows, C l " influx was much lower than C02 excretion, so, either the excretion of C02 as HCO^  was only a small proportion of the total C02 excretion, or the ratio of the two ions in the pump was not 1 : 1 . -As mentioned in the Introduction, there is some evidence for + + - -the existence of a Na /H and CT /HCO^  exchange diffusion pump in trout as well as in goldfish. Kerstetter et aj_., [1970] found that a downward sh i f t occurred the pH of the external water when the Na+ influx of the trout rose. This apparent linkage of Na and H ion movement only occurred i f the water was buffered. In unbuffered water the pH shif ts were erratic and sometimes upwards instead of downwards. The excretion of C02 to the water wi l l also change the pH of the external medium. C02 w i l l make the water more acidic i f i t enters as C02 gas, but w i l l cause the water to be more basic i f i t enters in the form of HCO .^ It is therefore extremely hard to correlate pH shifts in the water with changes in Na+ inf lux, as there is more than one ionic species contributing to the change in pH of the water. In my own estimations of H+ excretion by 45 trout I used a t i t ra t ion curve of S.W. and F.W. with HC1 to estimate H ion excretion, hoping thereby to minimise the role of CO^ , in affecting water pH. Nevertheless, some of the change in the water pH which was used to calculate H + excretion was undoubtedly caused by the excretion of CG^ to the water by the f i s h . There are several factors contributing to the untidy nature of the experiments performed by Kerstetter. During the experiments the . f ish were anaesthetised with tricainemethanesulphonate (TMS) although Houston et al_. [1971] have shown that anaesthetics, and TMS in part icular, lower plasma ion levels in trout. The anaesthetic lowers the blood haematocrit and haemoglobin content, impairing the buffering capacity of the blood and leaving the f ish more susceptible to acidosis. Moreover, during his experiments Kerstetter perfused the g i l l s of the trout by a pump at the rate of 100 ml a minute, a rate so low as to result in a blood pH of 7.52, which substantially below the normal value of 7.9 [Randall and Cameron, 1973] for the given temperature of the f i s h . Acidosis was probably caused by fa i lure to adequately remove CO^ as the flow rate of water over the g i l l s was too low, and also by the reduced buffering capacity of the blood due to anaesthesia discussed above. To compensate for the acidosis the f ish may excrete H + or take on HCO .^ The f ish therefore is not in a steady state with regard to ion regulation, and consequently conclusions drawn from Kerstetter's experiments must be considered suspect. Wood [1971] found that in the Rainbow trout, Na+ influx via the g i l l s followed a Michaelis-Menton curve which exhibited a low 46 V , indicating that the Na+ influx was by some transport mechanism max 3 ^ which is fu l l y saturated in F.W. This finding suggests that i f there is a Na+/H+ exchange diffusion mechanism in the Rainbow trout, i ts capacity is smal1. A r ise in the free CO^  of the water in which Rainbow trou.t were acclimated was found to cause a decrease in plasma Cl" concentration and a r ise in plasma HCO^  concentration [Lloyd and White, 1967]. Although this appears to demonstrate that a r ise in the HCO^  of the water caused by the increase in C02 reverses the HCO^/Cl" pump in the g i l l (HC0~ entering and Cl" leaving), this explanation is not the only one com-patible with the results . The r ise in plasma HCO^  is not exactly equal to the decrease in plasma C l " , the latter being 12 Meq/1 more than the r ise in plasma HCO .^ It is possible that the C02 from the water may enter the f i sh via the g i l l by diffusion' and be converted to HCO^  in the blood. The resulting upset in the electr ical neutrality of the blood may lead to excretion of Cl" by the kidneys to restore the electr ical balance. Kerstetter & Kirschner [1972] concluded that in the Rainbow trout injections of HCO^  stimulated C l " influx in F.W. An injection of isosmotic saline did not e l i c i t any change in C l" inf lux. Kerstetter's data has been replotted in Figure 18. The rate of C l" influx at time 0 is the influx before injection of either saline or HCO .^ It is very d i f f i c u l t to determine from the data whether HCO^  does in fact stimulate Cl" inf lux , as the points on the graph are well within 2 s .d . of one another. Secondly, the control rate for C l " influx (in the saline Figure 18 Plot of Kerstetters & Kirschner [1972] results on effect of HCO3 injections on Cl" fluxes of Rainbow Trout. Arrows indicate injections of HCO .^ # fluxes in saline injected f i sh J^ f l u x e s in HCO^  injected f ish ± one S.D. . 48 injections) is higher- than the rate when the pump is supposedly stimu-lated by HCO .^ The injection of HCO" into the f ish did , however, raise blood HCO" as indicated by the increase in the pH of the blood, and the HCO^  was then excreted as blood pH returned to the i n i t i a l value 3 hours after the injection. The HCO^  may have been converted to COg by the c a . in the g i l l epithelium or in the red blood c e l l s , and may then have been excreted as molecular C02 rather than by a HCO^/Cl exchange mechanism. I performed experiments similar to Kerstetter 1 s, in loading trout with HCO^  and observing C l " fluxes. They were performed before the publication of Kerstetter's results . Unlike Kerstetter 1 s, my f ish were not anaesthetised and were unrestrained, and I therefore f e l t that they were in a more normal state during the experiments. The results which I obtained do not show a r ise in f i immediately after the inject ion, although they do show a definite increase in f i 45 minutes after the injection. It has been observed in other experiments on Rainbow trout that almost a l l the HCO" is excreted in the f i r s t 30 minutes after the injection [R. Janssen, P.C.] . As the r ise in C l " influx ( f i ) did not occur during the maximum efflux of HCO", i t appears that the two did not coincide. In other words, i t appears that in the Rainbow trout Cl'/HCO^ exchange is not obligatory. It is possible that during the f i r s t half hour the injected HCO3 escapes as C 0 2 via the g i l l , and that the residual HCO^  is then exchanged with C l " , producing the observed peak in f i . There may be a lag period in altering the a f f in i t y of the carr ier of the Cl~/HC0~ : • 4 9 system, assuming that the carr ier is operating at fu l l capacity under conditions of normal HCO" excretion. I observed an increase in both f i and fe during the course of the experiment. However, an increase in C l " efflux (fe) is observed only i f HCO^  is added to the water, thus reversing the Cl~/HC0g pump [Maetz and Romeau, 1964]. The changes in f i and fe are caused not by any ion exchange process but probably by the trauma of the injection. Unfortunately no experiments were performed to ascertain the effect of the injection on C l " fluxes. We can conlcude that, in the trout, unlike the goldfish, the Cl'/HCO^ exchange diffusion mechanism plays a minor part in Cl~ and HCO^  exchange. The role of c a . in the g i l l s of f ish and in particular in the g i l l s of trout remains to be considered. The low level of c a . in the trout in both S.W. and F.W. provides further evidence that the Cl /HCOg pump is of low capacity, as c a . catalyses the formation of HCOg. Goldfish, on the other hand, have high levels of c a . in the g i l l s , indicating that the Cl'/HCO^ pump is operating at a high capacity. S.W. coho have more c a . in the g i l l s than F.W. coho, and a similar difference between S.W. and F.W. perch was found by Maetz [1956]. However, some F.W. f ish do have high g i l l c a . , so that we cannot con-clude that there is a clear case for S.W. f ish having more c a . in the g i l l that F.W. f ish (see Table V). We are now in a position to present several models of the g i l l s of trout, goldfish and coho. It must be noted that these models are tentative and undoubtedly w i l l be altered in the l ight of new knowledge. However, they serve a purpose in helping one to understand the problem 50 TABLE V Concentrations of Carbonic Anhydrase in G i l l s of Various Species of Fish Fish Media * Tissue Reference Dogfish S.W. 13 - 22 Hodler et al_., 195E Dogfish S.W. 37 Maren, 1958 Anebulosus F.W. 55 Maren, 1958 Saurus griseus S.W. 4. 43 Leiner, 1938 Sargus annulrns S.W. 11. 3 Leiner, 1938 Sargus rondetti S.W. 22. 40 Leiner, 1938 *eu = enzyme units - 51 and in stimulating new experiments. Figure 19 show the models suggested by Maetz [1971] for the g i l l s of S.W. and F.W. f i s h . For F.W. f ish a Na /K pump is placed on the inner border of the g i l l , as a Na-K ATPase is found in the g i l l s and in goldfish high external levels of K+do not affect Na+ eff lux. Cl'/HCO^ and Na+/H+ pumps are located on the outer borders of the g i l l , as changing the external levels of Cl~ and Na+ affect HCO3 and H + fluxes (see Introduction). A number of other assumptions have been made, including one as to the role of c a . in the g i l l s . The S.W. model is essentially the F.W. model, but the g i l l is inverted, with the inner side out: Maetz's model for F.W. f ish is speci f ical ly for -the goldfish and in my view cannot be construed as a general model for F.W. f i s h . Figure 20 is a model of the F.W. Rainbow trout g i l l . A Na+/K+ pump is located on the outer membrane of the g i l l , but i t is not operative due to the low level .of Na+ in the g i l l . The pump is + + stimulated by internal Na rather than by external levels of K , so i f i t is indeed operating, i t is operating at a low rate. The Na /H and CWHCOg exchange diffusion pumps are operating at a low capacity, requiring a low concentration of c a . in the g i l l s . Most of the C02 exists by diffusion'. There are active C l " and Na+ pumps on the inner membrane of the g i l l which transfer these ions to ithe blood. When the trout enters S.W. the Cl'/HCO^ pump reverses direction, since S.W. has a high concentration of Cl~ and HCO^  (Figure 21). Dejours [1967] found that in the goldfish the Cl'/HCO^ pump is indeed reversible. The Cl~/HC0l pump operates at a low capacity in S.W., and, as Cl~ influx Figure 19 Model of F.W. and S.W. f ish g i l l s . From Maetz [1971] p. 229. Figure 20 Proposed model for F.W. f ish g i l l . COg diffuses out of g i l l as C02 or is converted to H and HCO^  of carbonic anhydrase ( c a . ) . The amount of CO2 leaving as CO2 depends on the species of f i s h . Two active pumps for Na+ and Cl~ are located on the inner border of the g i l l , while 2 exchange differs ion pumps Na /NH^, H and Cl /HCO^  as well as a Na /K pump are on the outer border of the g i l l . Figure 21 Proposed model for S.W. f ish G i l l . CO^  is excreted as molecular CO ,^ while carbonic anhydrase ( c a . ) is used to help catalyse the conversion of HCO^  to CG^. Active H + , C l" and Na+ pumps are on the inner g i l l border as well as 3 exchange diffusion pumps on the outer border. (Na+/H+CNHj), C1~/HC0~ and Na +/K +). 55 is restricted by the low permeability of the g i l l membrane to C l " , c a . concentration in the g i l l remains low. The influx of Na+ into the g i l l by diffusion is compensated for by the Na+/K+ pump which is now stimulated + • + • + and is expelling Na . The Na /H pump is working in opposition to the Na+/K pump by pumping Na+ i n , but i t s capacity is so low that there is no danger of the Na /K pump being overloaded and therefore handle the Na+ inf lux. The reversal of the Cl~/HC0^ pump causes build-up of HCO^  in the g i l l s , but this is converted to CG^  by the g i l l c a . and , subsequently escapes out of the g i l l . The two active pumps on the inner border of the g i l l now reverse their direct ion, as the f ish attempts to excrete ions to compensate for passive influx of ions. The model of the goldfish g i l l is similar to that of the trout, except that the Na+/H+ and Cl~/HC0~ pumps are operating at a high rate in F.W. as evidenced by the large amount of c a . in the g i l l s and the relat ively high rates of Cl~ and Na+ fluxes in the goldfish [Maetz, 1971]. When the f ish .enters S.W. the Na /K pump excretes Na to compen-sate for Na inf lux , but i t also has to cope with the large Na influx + + via the Na /H pump. The amount of Na entering is too large for the + + + Na /K pump to handle, and the Na builds up in the f i s h , eventually caus-ing death. As the Cl"/HCO^ pump reverses in S.W., there is no danger of Cl~ loading and there is moreover suff icient c a . in the g i l l s to convert the incoming HCO^  to CG^. The S.W. coho has a g i l l similar to that of the S.W. trout. However, the Cl'/HCO^ pump, which has reversed i t s direction from the one taken in F.W., is operating at a high capacity, as is shown by the large 56 amount of c a . in the g i l l . This implies that the g i l l is readily permeable to C l " , necessitating a large Cl'/HCO^ pump. It is this large capacity Cl'/HCO^ pump which enables the f ish to alter rapidly plasma HCGg and therefore buffering capacity. When the f ish transfers to F.W. the Cl'/HCO^ pump is in i ts original configuration (Cl~ in and HCO^  out), but the rate of Cl~ inf lux cannot balance the massive C l " loss via the permeable membrane and the f ish eventually succumbs to C l" loss. Upon entry into S.W. from F.W. several changes other than those in the g i l l s occur within the f i s h . The f ish drinks in S.W., the perme-a b i l i t y of the skin changes and the kidneys excrete divalent ions. The f ish drinks in S.W. to offset.the loss of water resulting from the increase in permeability to water of the skin and g i l l s [Shehadeh and Gordon 1967, Potts e_t al_., 1970]. In the trout 80% of the ingested ions are passively absorbed by the intestine and excreted by the kidney. In fact the main role of the kidney in S.W. appears to be excretion of ions absorbed as a result of drinking. In the F.W. coho there is a low concentration of C l" and Na+ in the urine (12 - 13 Meq/1) which is expelled rapidly because the f ish is attempting to lose water. S.W. coho have a high concentration of + ++ Cl , Na and Mg in their urine (55 - 100 Meq/1). The urine flow is low because the f ish is conserving water. The high concentration of ions in the urine indicate that the ions absorbed by the intestine are f i l tered into the kidneys [Miles, 1971]. Rainbow trout show a similar r ise in C l " concentration of urine in S.W. [Holmes, 1961; Shehadeh and Gordon, 1967], There i s , however, a structural difference in the kidneys of Rainbow trout and S.W. coho, in that the coho has lost the distal tube of 57 the kidney. It is the possession of this distal tube which gives the trout the greater reabsorbative powers essential for F.W. survival [Miles, 1971]. The loss of the distal tube of the kidney is in fact a characteristic of stenohaline marine f ish [Hickman and Trump, 1969] and the loss of this tube in the coho is therefore an indication that they are more at home in S.W. than in F.W. The difference in ion regulation between Rainbow trout, goldfish and coho which have been discussed are related to the l i f e histories of the f i s h . Salmo gairdneri (Rainbow trout) and the Atlantic salmon (Salmo Salar) are members of the genus Salmo, whose main characteristic is that they spawn more than once during their l i f e . After transferring to S.W. from F.W. as a smolt, the f ish spends a year in the ocean and then returns to F.W. to spawn, after which i t returns to the ocean. This cycle may be repeated up to three times. Coho (Oncorhynchus kitsuch) are of the genus Oncorhynchus or the Pacif ic Salmon, who go to sea when young and return to F.W. to spawn and die. Goldfish are s t r i c t l y F.W. f ish [Clemens and Wilby, 1961]. Salmo and Oncorhynchus d i f fer in the age at which they i n i t i a l l y migrate to the sea, Salmo migrating as two year old smolts, and Oncorhynchus as one year olds [Parry, 1961; Conte et aj_., 1966]. There is a definite time at which Oncorhynchus must transfer to S.W. If they f a i l to enter S.W. at this time they must remain in F.W. Attempts to enter S.W. at another time w i l l result in death. F.W. restricted f i s h , known as kokanee, are f ish which have become landlocked due to their inabi l i t y to enter the sea. Houston [1959] concluded that S. Salar and S. gairdneri showed a preadaptation to S.W. transfer as smolts, while Conte et al_., [1966] showed 58 that in 0. kitsuch no such adaptation took place. Although the two genera are related, they show signif icant differences in their l i f e histories. Salmo have evolved an ion regulation system which enables them to tolerate changes in sa l in i t y . This they accomplish by means of a small capacity branchial HCO^/Cl" exchange diffusion system. Oncorhynchus, however, having only to tolerate F.W. for a short time during spawning, possess a system with a high Cl'/HCO^ exchange diffusion mechanism. The goldf ish, endowed with a Cl'/HCO^ exchange diffusion system of high capacity, can cope with pH shifts in the water. It can alter rapidly the buffering capacity of i t s blood by changing the rate of excretion of HCOg to suit the pH of the external medium. This enables the goldfish to l i ve in stagnant waters which have a high PCC^ and are thus acidic . The trout, on the other hand, with a low capacity Cl'/HCO^ system, has a minimal a b i l i t y to change the blood buffering capacity, and is there-fore restricted to water with a low PCG^. However, unlike the goldfish, i t is able to enter S.W. The buffering capacity capacity of f ish blood is dependent HC0-on the ratio ^g—pgg— . Blood PCO^ is only a few mm Hg above ambient. HCO^ concentration in the blood is low (10 mM/1) in order to achieve the correct ratio between HCO" and PC02 that is compatable with the pH of the blood. A large capacity C1~/HC03 pump can rapidly and simply regulate plasma pH, by altering the HCOyPCO^ rat io . The capacity of the pump, as we have seen, depends on the f ish and i t s " l i f e s ty le . " In conclusion, i t appears that the genus salmo and particularly S. gairdneri possess a regulation system that enables them to l ive in S.W. and F.W. In the trout this a b i l i t y is related the fact that the f ish 59 possesses l i t t l e or no Cl'/HCO^ exchange diffusion capacity. This observation is supported by the fact that trout g i l l s show a low concen-tration of ca., an enzyme believed to have an ion regulatory function, and that they also have a low permeability to C l " . In contrast, the goldfish and the coho possess l i t t l e or no euryhaline capabi l i t ies , but have large capacity HC0~/C1~ exchange diffusion mechanisms. 60 LITERATURE CITED ALBERS, C. 1970. Acid-Base Balance. In Fish Physiology. Vol. IV. Eds. W.S. Hoar and D.J. Randall. Academic Press. N.Y. BERRIDGE, M and OSCHMAN, J . 1972. Transport Epithelia. Academic Press. N.Y. CAMERON, J . 1971. Rapid Method of Determining Total Carbon Dioxide In Small Blood Samples. J . Appl. Physiol. 31:632-634. CHAN, D.K.O., PHILIPS, J . G . , and CHESTER JONES, I. 1967. Studies on Electrolyte Changes in the Lip Shark Hemisoyllium Plagiosium (Bennett) With Special Reference to Hormonal Influence on Rectal, Gland. Comp. Biochem. Physiol. 23: 185-198. CLEMENS, W. and WILBY, G. 1961. Fishes of the Pacif ic Coast of Canada. Bu l l . 68. Of Fisheries Research Board of Canada. CONTE, F.P. 1970. Salt Secretion. In Fish Physiology. Vol. 1. Eds. W.S. Hoar and D.J. Randall. Academic Press. N.Y. CONTE, F .P. , WAGNER, H., FESSLER, J . , and GNOSE, C. Development of Osmotic and Ionic Regulation in Juvenile Coho Salmon 0. Kisutch. Comp. Biochem. Physiol. 18: 1-15. 61 CROSBY, W. and FURTH, F. 1956, A Modification of the Benzidine Method for Measurement of Haemoglobin in Plasma and Urine. Blood. 11: 380-383. . DAVIS, J .C. and CAMERON, J . 1970. Waterflow and Gas Exchange at the G i l l s of Rainbow Trout. S. Gairdneri. J . Exp. B i o l . 54: 1-18. DEJOURS, P. 1969. Variations of C02Output of a F.W. Teleost Upon Change of the Ionic Composition of the Water. J . Physiol. 113p. GARCIA-ROMEAU, F. and MAETZ. 1964. The Mechanisim of Sodium and Chloride Uptake by the G i l l s of A Fresh Water Fish Carassius. Auratus. I. Evidence for An Independent Uptake of Na and Cl ions. J . Gen. Physiol. 47: 1195-1208. GORDON, M. 1963. Chloride, Exchanges in Rainbow Trout. (S. Gairdneri). Apapted to Different Sa l in i t ies . B i o l . Bu l l . 124: 45-54. HANSSON, H.P. 1967. Histochemical Demonstration of Carbonic Anhydrase Act iv i ty . Histochemie. 11: 112-128. . HICKMAN, C. and TRUMP, B. 1969. The Kidney. In Fish Physiology. Vol. I. Eds. W.S. Hoar, and D.J. Randall. Academic Press, N.Y. 62 HODLER, J . , HEINEMAN, H.O., FISHMAN, A . P f , and SMITH, H.W. 1955. Urine pH and Canbonic Anhydrase Act iv i ty in Marine Dogfish. Am. J . Physiol. 183: 155. HOLETON, G. and RANDALL, D.J. 1967. The Effect of Hypoxia Upon Partial Pressure of Gases in the Blood and Water Afferent and Efferent to the G i l l of Rainbow Trout. J . Exp. B i o l . 46: 317-327. HOLMES, R.M. 1961. Kidney Function in Migrating Salmonids. Rept. Challanger Soc. ( Cambridge ), 3. No. 13: 23. HOUSTON, A. 1959. Osmoregulatory Adaptation of Steel head Trout {S. Gairdneri). To Sea Water. Can. J . Zool. 37: 729. HOUSTON, A . , MADDEN, J . A . , WOODS, R . J . , and MILES, H.M. 1971. Some Physiological Effects of Handling and T.M.S. Anesthetization Upon Brook Trout {Salvelinis Fontunalis), J . Fisheries Research Board. Canada. 28: 625-633. JONES, D.R., RANDALL, D.J . , and JARMAM, CM. 1970. A Graphical Analysis of 0 2 Transfer. In Fish. Resp. Physiol. 10: 285-298. KERSTETTER, T.H. , KIRSCHNER, L.B. and RAFUSE, D.R., 1970. On the Mechanism of Sodium Ion Transport of the Irrigated G i l l s ' of Rainbow Trout. (S. Gairdneri), J . Gen'. Physiol. 56:342-358. 63 KERSTETTER, T.H. and KIRSCHNER, L.B. 1972. Active Chloride Transport By the G i l l s of Rainbow Trout {S. Gairdneri), J . Exp. B i o l . 56: 263-272. KEYS, A. 1931. Chloride and Water Secretion and Absorption by the G i l l s of the Eel. Z. Vergl. Physiol. 15: 364-388. KROGH, A. 1939. Osmotic Regulation In Aquatic Animals. Cambridge University Press. LEINER, M. 1938. Die Augenkiemend Ruse (Pseudobranchie) Der Knochenfische Z. Vergle. Physiol. 26: 416. LLOYD, R. and WHITE, fi. 1962. Effect of High Concentration of C02 on Ionic Composition of Rainbow Trout Blood. Nature. 216: 1341-1342. MAETZ, J . 1956a\ • Les Exchanges De. Sodium Chez le Poisson Carassius Auratus. "Action D'un Inhibiteur d l ' Anhydrase Carbonique. J . Physiol-. (Paris) , 48: 1085-99. MAETZ, J . 1956b. Bu l l . B io l . Fr. Belg. (suppl.). 40:1-129. Quoted in Maetz, J . 1971. Fish G i l l s . Mechanisms of Salt Transfer in Fresh Water and Sea Water. P h i l . Trans. Roy. Soc. Lond. B. 262: 209-249. . 64 MAETZ, J . 1971. Fish G i l l s : Mechanisms of Salt Transfer In Fresh Water and Sea Water. P h i l . Trans. Roy. Soc. Lond. B. 262: 209-249. MAETZ, J . 1973. Na+/NH*, Na+/H+ Exchanges and NH3 Movement Across G i l l s of Carissius Auratus. J . Exp. B i o l . 58: 255-275. MAETZ, J . and GARCIA-ROMEAU, F. 1964. The Mechanism of Na+ and C l " Uptake by the G i l l s of A.F.W. Fish Carassius Auratus. II. Evidence for NH^/Na+ and HCOg/Cl" Exchanges. J . Gen. Physiol. 47: 1209-1226. MAREN, T. 1958. Distribution of Carbonic Anhydrase in Several Non-Mammalian Species with a Few Notes on Function. Bu l l . Mt. Desert Island. B i o l . Lab. 3: 72. MAREN, T. 1960. A Simplified Micro Method for Determining Carbonic' Anhydrase and Its Inhibiters. J . Pharm. and Eptl . Therap. 130: 26-29. MAREN, T. 1967. Carbonic Anhydrase: Chemistry Physiology and Inhibition. Physiol Review. 47: 595. ' ^. MILES, H. 1971. Renal Function in Migrating Adult. Coho Salmon. Comp. Biochem Physiol. 38A: 787-826. 65 PARRY,E. 1961. Osmotic and Ionic Changes in Blood and Muscle of Migrating Salmonids..J. Exp. B i o l . 38: 411-427. PEARSE, A.G. 1961. Histochemistry: Theoretical and Applied. 2nd Ed. J . and A. Churchi l l , London. POTTS, W.T.W., FOSTER, M., and STATTER, J . 1970. Salt and Water Balance in Salmon Smolts. J . Expl. B i o l . 52: 553-564. RAO, G. 1969. Effect of Act iv i ty Sal inity and Temperature on Plasma Concentrations of Rainbow Trout. (S. Gairdneri), Can. J . Zool. 47: 130-134. RANDALL, D.J. and CAMERON, J.D. 1973. Respiratory Control of Arterial pH as Temperature Changes in Rainbow Trout. Salmo Gairdneri. Am. J . Physiol. 225: 997-1002. SHEHADEH, Z. and GORDON, M. 1969. The Role of the Intestine In Sal ini ty Adaptation of the Rainbow Trout. S. Gairdneri. Comp. Biochem. Physiol. 30: 397-418. SIGGAARD-ANDERSON, 0. 1967. Ultra-Micro Determination of C l " In Biologic Fluids with Cotlove Chloride Titrater. Tech. B u l l , of Registary of Med. Tech. 39: 240. 66 SMITH, L.S. and BELL, G.R. 1964. A Technique for Prolonged Blood Sampling in Free Swimming Salmon. J . Fisheries Research Board. Canada. 21: 711-717. STEVENS, E.D. 1968. Cardiovascular Dynamics During Swimming In Fish, Particularly Rainbow Trout {Salmo Gairdneri), Ph.D. Thesis, U.B.C. USSING, H.H. 1949. Transport of Ions Across Cellular Membranes. Physiol. Review 229: 127-55. WOOD, C. 1971. The Influence of Swimming Activ ity on Sodium and Water Balance in Rainbow Trout. S. Gairdneri. MSC Thesis, U.B.C. 67 APPENDIX I Long Term Effects of S.W. on Trout Although i t was evident from the fact that they survived for at least three hours, that trout do adapt to S.W., i t was important to know what were the long term effects of salt water on trout ion regulation. Accordingly, f ish were placed in boxes similar to those mentioned previously, and VR, V0 2 , heart rate, and Pa02 were measured over a period of four days. In addition, several trout which had been in S.W. for a month were used to estimate V0 2 , pH, Pa0 2 , VC02 and total C0 2 . Rainbow trout of 200 - 300 gms. were cannulated in the dorsal aorta as described previously and placed in darkened plexiglass boxes which were s l ight ly larger than the f ish (3" x 3" x 14"). The f ish were able to move about, but at the same time were suff ic ient ly restrained to permit blood sampling. The f ish were allowed to recover overnight from the operation and were then sampled at 0, 10, 20, 30, 40 and 60 minutes after the introduction of S.W. into the boxes, as well as once a day on the 2nd, 3rd, and 4th, days of the experiment. Pa02 and V02 were measured as described in the section on methods, while VR was determined by counting the number of opercular movements in a given time. Heart rate was measured by connecting the dorsal aortic cannula to a Statham pressure transducer and counting the beats as they were displayed on a Beckman Dynograph recorder. Heart rate, V02 and VR a l l tended to remain constant during exposure of the f ish to S.W. for four days (Figures 22, 23 and 24). Figure 22 Heart rate of Rainbow trout (n = 12) during longterm exposure to S.W. (4 days). Variation expressed as ± one S.D. Heartrate measured in F.W. and then f i sh transferred to S.W. at time 0. Figure 23 0 2 consumption of Rainbow trout (n = 12) during long term exposure to S.W. (4 days). Variation expressed as ± one S.D. 0 2 consumption measured in F.W. and then f ish transferred to S.W. "So TIME MIN/DAY Figure 24 Ventilation rate of Rainbow trout (n = 11) during long term exposure to S.W. (4 days). Variation expressed as ± one S.D. Ventilation rate measured in F.W. and then f ish transferred to S.W. at time 0. 4 0 0 2 o 4 0 6 0 2 3 4 T I M E M I N D A Y Figure 25 Arterial 0^ tension in Rainbow trout (n = 12) during longterm exposure to S.W. (4 days). Variation expressed as ± one S.D. Pa02 measured in F.W. and then f ish transferred to S.W. at time 0. 1 2 0 TIME min/days 72 PaG^, however, dropped steadily and by the fourth day i t was at"45 mm Hg (Figure 24). The results for the f ish exposed to S.W. for an extended period are presented in Table VI. As f ish blood is s t i l l saturated with 0,, at an arter ia l tension of 90 mm Hg [Jones, et al_., 1970], i t is clear from Figure 25 that after four days in S.W. the f ish are suffering from a lack of 0 2 - However, i t must be remembered that the values in Figure 25 are an average and tha't some f ish have a Pa02 of about 60 mm Hg, while others have a Pa02 of 30 mm Hg (values ± 2SD of mean after four days in S.W.). Presumably those with the high Pa02 survived while the others died. It was observed that not a l l the f ish survived the adaptation to S.W., there being a mortal-i ty rate of around 30%. The lowering of Pa02 was due to the effect of S.W. and not to the boxes in which the f ish were confined, as Pa02 did not drop in the control f ish (Figure 29). The f ish which actually survived the transfer to S.W., did so successfully as is shown in Table VI. The values for the various acid base parameters are similar to those for F.W. trout. This shows that these trout have successfully completed the transfer to S.W. and is hardly surprising as trout are known to l ive in S.W. They are able to survive in S.W. by virtue of the fact that, unlike the goldfish, they either do not possess a Cl'/HCO^ exchange diffusion system, or, i f they do have one, i t is of low capacity. TABLE VI pH Blood 7.765 - 7.847 Pa02 9 3 - 1 0 9 mm Hg. V02 0.33 - 0.63 ml/min/Kg VC02 0.168 - .179 mm/min Kg Total plasma C02 6.8 - 9.09 mM/ n = 5. Acid Base Parameters for longterm S.W. adapted f i s h . 74 APPENDIX II Effect of Laboratory Conditions on Trout Stevens [1968] showed that the placing of trout in darkened chambers caused a drop in the heart rate as well as the respiration rate of the f i s h . After f ive days heart rate had dropped from 60 to 40 beats a minute, while respiration rate dropped from 100 to 78 beats a minute. During this time any disturbance of the f ish caused a r ise in the two variables. Consequently, in examining the effects of S.W. on the trout care must be taken to separate the possible effects of confining the f ish to their boxes from those effects caused by S.W. Accordingly, a series of control experiments were performed identical with those in Appendix 1, except that at the start F.W. was allowed to enter the boxes. There appeared to be no effect on any of the parameters except pH (Figures 26, 27, 28, 29), although there was some variation of values during the four days. The continued high level of heart rate during the four days was undoubtedly caused by disturbance of the f ish during sampling. Stevens [1968] in extended experiments found that by disturbing f ish whose heart rate had dropped he could raise the rate to i ts original leve l . Thus the disturbance caused by sampling was suff icient to prevent the f ish from spontaneously dropping i t s heart rate. The drop in pH parallels the drop found in S.W. adapted f ish (Figure 15) and presumably was caused by the restr ict ion of the boxes and not by exposure to S.W. It is conceivable that the drop in blood pH was caused by an inadequate water flow through the boxes inducing acidosis. As regards the other var i -ables, we can conclude that confinement in the boxes had no effect. Figure 26 Ventilation rate in Rainbow trout (n = 13) confined in ventilation 6-'o>'eSfor 4 days in F.W. variation expressed as ± one 5. 0. I i I I Figure 27 Heart rate in Rainbow trout (n = 13) confined in ventilation .boxes for 4 days in F.W. variation expressed as ± one S . D . Figure 28 Blood pH of Rainbow trout (n = 13) confined in venti lation boxes for 4 days in F.W. Variation expressed as ± one S. o... 7 2 pH Figure 29 Arterial 0^ tension of Rainbow trout (n = 13) confined in ventalit ion boxes for 4 days in F.W. Variation expressed as ± one S.D. 30 60 " 2 TIME M1N/DAY Figure 30 VO^  of f ish confined to venti lation boxes for 4 days ± one S.D. TIME MIN/DAY 80 APPENDIX 3 Effect of S.W. On Goldfish To establish that goldfish are indeed stenohaline 30 goldfish were placed directly into S.W. to assess their capabil ity for survival . Prior to death a l l the f ish showed a lack of nervous coordination and appeared unable to maintain neutral buoyancy as they repeatedly floated to the surface of the tank. After one hour in S.W. almost half the f ish were dead (Figure 30) and after 105 minutes a l l the f ish had expired. Goldfish are thus clearly stenohaline. The 0 2 consumption was measured in S.W. and in F.W. In F.W. the V02 was 54.2 ml/hr/Kg and in S.W. i t was 31.1 ml/hr/Kg. The difference between the two readings was signif icant to the .01 level by the T test. The drop in V02 can be compared with the results in Appendix 2, which showed in trout a drop in Pa02 but not in V0 2. The drop in V02 in goldfish exposed to S.W. appears to be a symptom of the fact that the f ish was dying as a result of i ts inab i l i t y to osmoregulate in S.W. As discussed previously, this inab i l i t y appears to be related to the existence of a large capacity branchial C1~/HC0~ pump in the goldfish. Figure 31 Histogram of death rate of goldfish in S.W. n = 35. %dead o o o o 00 o o o CO o o o PL • to o 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items