UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Air-breathing in the bowfin (Amia calva L.) Hedrick, Michael Scott 1991

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

Item Metadata

Download

Media
831-UBC_1991_A1 H42.pdf [ 7.01MB ]
Metadata
JSON: 831-1.0100451.json
JSON-LD: 831-1.0100451-ld.json
RDF/XML (Pretty): 831-1.0100451-rdf.xml
RDF/JSON: 831-1.0100451-rdf.json
Turtle: 831-1.0100451-turtle.txt
N-Triples: 831-1.0100451-rdf-ntriples.txt
Original Record: 831-1.0100451-source.json
Full Text
831-1.0100451-fulltext.txt
Citation
831-1.0100451.ris

Full Text

AIR-BREATHING IN THE BOWFIN (Amia calva L.) b y MICHAEL SCOTT HEDRICK B.Sc. Lewis and Clark College, 1980 M.Sc. Portland State University, 1985 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES (Department o f Zoology) We a c c e p t t h i s t h e s i s as c o n f o r m i n g t o t h e r e q u i r e d s t a n d a r d THE UNIVERSITY OF BRITISH COLUMBIA September 1991 © M i c h a e l S c o t t H e d r i c k , 1991 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ^ QO The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT The c o n t r o l of a i r - b r e a t h i n g i n the bowfin, Amia calva, was i n v e s t i g a t e d u s i n g experimental and a n a l y t i c a l approaches. The a i r - b r e a t h i n g p a t t e r n of conscious, u n d i s t u r b e d bowfin at 22+2 °C was c h a r a c t e r i z e d by t h e r e s p o n s e s t o changes i n r e s p i r a t o r y g ases i n t h e a q u a t i c and a e r i a l e n v i r o n m e n t s . P n e u m o t a c h o g r a p h i c measurements o f a i r f l o w s d u r i n g a i r -b r e a t h i n g e v e n t s r e v e a l e d two d i s t i n c t p a t t e r n s : i n t y p e I breaths e x h a l a t i o n was f o l l o w e d by i n h a l a t i o n ; i n type I I a i r breaths , which have not been d e s c r i b e d f o r t h i s s p e c i e s , only i n h a l a t i o n o c c u r r e d . Under normoxic c o n d i t i o n s both t y p e s of a i r b r e a t h s o c c u r r e d (60% t y p e 1:40% t y p e I I ) and t h e mean i n t e r - b r e a t h i n t e r v a l was 19.8+0.9 (95% C.I.) min. A q u a t i c or a e r i a l h y p o x i a s t i m u l a t e d a i r - b r e a t h i n g , IBI d e c r e a s e d t o about 13 min i n both c o n d i t i o n s , and t h e r e was a change i n a i r - b r e a t h i n g p a t t e r n t o p r e d o m i n a n t l y t y p e I a i r b r e a t h s (>80% o f t o t a l b r e a t h s ) . Maximum e x p i r e d volume f o r type I breaths averaged 25.1+6.2 ml k g - 1 . A i r bladder volume was 80 ml k g - 1 , so t h a t about 30% of t o t a l a i r b l a d d e r volume was exchanged d u r i n g a type I b r e a t h . Bowfin exposed t o 100% 0 2 i n t h e a e r i a l phase, r e g a r d l e s s of a q u a t i c P 0 2, sw i t c h e d t o t y p e I I a i r b r e a t h s a l m o s t e x c l u s i v e l y (>99% o f t o t a l b r e a t h s ) . A i r b l adder d e f l a t i o n i n conscious f i s h i n i t i a l l y r e s u l t -ed i n o n l y t y p e I I a i r b r e a t h s b e i n g t a k e n . The t i m e t o i n i t i a t e an a i r bre a t h and the number of a i r breaths f o l l o w i n g d e f l a t i o n were both s i g n i f i c a n t l y dependent upon the volume removed from the a i r bladder. The r e s u l t s suggest t h a t dynam-i c and s t a t i c c h a r a c t e r i s t i c s of a i r bladder mechanoreceptors are i n v o l v e d i n the a f f e r e n t limb of the type II breathing response and that type II breaths serve a buoyancy, rather than gas exchange, function. B r a n c h i a l denervation was used to t e s t the hypothesis that type I a i r breaths were stimulated by 0 2-chemoreceptors located on the g i l l s . Bowfin were e i t h e r sham-operated (SH), p a r t i a l l y - d e n e r v a t e d (PD) or t o t a l l y - d e n e r v a t e d (TD) and exposed to aquatic normoxia and aquatic hypoxia. Air-breath-i n g frequency, measured as t o t a l breaths, i n c r e a s e d from aguatic normoxia to hypoxia i n a l l three groups; air-breathing frequency was s i g n i f i c a n t l y higher i n the TD group. This was due, however, to large numbers of type II a i r breaths occur-r i n g between 0 and 1 min as a r e s u l t of excessive l o s s of i n s p i r e d gas during i n h a l a t i o n . There was no s i g n i f i c a n t d i f f e r e n c e i n the frequency of type I breaths f o r any group when analyzed s e p a r a t e l y from type II breaths; thus, the afferent limb of the air-breathing response to hypoxia was not i d e n t i f i e d , suggesting that e x t r a - b r a n c h i a l s i t e s f o r 0 2-chemorception may be involved. The re s u l t s also indicate that e i t h e r sensory or motor components of nerves IX and X to the g i l l arches are important for proper air-breathing function. The r o l e of c e n t r a l c h e m o s e n s i t i v i t y was examined by perfusing a mock extra-dural f l u i d e quilibrated with normoxic, hyperoxic, hypoxic or hypercapnic gas mixtures through the c r a n i a l space i n conscious f i s h . Air-breathing was only stimu-l a t e d by aquatic hypoxia, not changes i n e x t r a - d u r a l f l u i d composition, thus implicating p e r i p h e r a l . s i t e s for 0 2-mediated i i i e f f e c t s on a e r i a l v e n t i l a t i o n . Unfortunately, these r e s u l t s , along with g i l l denervation data, do not y i e l d any informa-t i o n about the location of 0 2-chemosensitive s i t e s or afferent pathways that modulate air-breathing i n bowfin. The temporal, int e r m i t t e n t pattern of a i r - b r e a t h i n g was examined by s p e c t r a l a n a l y s i s . The int e r m i t t e n t pattern was found to have s i g n i f i c a n t , non-random frequency components. A s i g n i f i c a n t low frequency component, corresponding with a 30 min period, was found i n the periodogram of 6 bowfin i n nor-moxic conditions. In aquatic or a e r i a l hypoxia, the dominant periods ranged between 5 and 10 min. The dominant p e r i o d i c i -t i e s i n normoxia, or either hypoxic condition, were correlated with the mean inter-breath i n t e r v a l for type I breaths. Since type I breaths were a f f e c t e d by changes i n e x t e r n a l and/or in t e r n a l P0 2, the r e s u l t s indicate that air-breathing behavior occurs p e r i o d i c a l l y and may be driven by 0 2 - s e n s i t i v e chemore-ceptors. A computer model was formulated to simulate the intermit-tent a i r - b r e a t h i n g pattern. The model used two independent t h r e s h o l d s f o r t r i g g e r i n g type I and type II a i r breaths. Type I a i r breaths were modeled as t h r e s h o l d responses to reductions i n i n t r a v a s c u l a r P0 2. Type II a i r breaths were simulated as feedback responses to reductions i n a i r bladder volume. Using empirical data from t h i s study and other pub-l i s h e d work, the model produced i n t e r m i t t e n t a i r - b r e a t h i n g s i m u l a t i o n s that c l o s e l y resembled the responses of bowfin exposed to a e r i a l normoxia, hypoxia and hyperoxia. Quantita-t i v e and q u a l i t a t i v e s i m i l a r i t i e s between the model and data iv from bowfin suggest the model i s r e a l i s t i c i n i t s assumptions r e g a r d i n g mechano- and chemoreceptive i n p u t s c o n t r o l l i n g intermittent air-breathing. The r e s u l t s from t h i s study indicate that bowfin normally use two types of respiratory behaviors that serve gas exchange and buoyancy functions. Intermittent breathing i n t h i s spe-c i e s i s shown to be p e r i o d i c , and the rhythmicity appears to be generated by feedback from 0 2 - s e n s i t i v e chemoreceptors l o c a t e d i n a p o s i t i o n to monitor i n t r a - v a s c u l a r changes i n P O 2 . v TABLE OF CONTENTS ABSTRACT i i LIST OF TABLES v i i i LIST OF FIGURES ix ACKNOWLEDGEMENTS x i INTRODUCTION 1 The Evolution of Air-Breathing 4 Mechanisms of A e r i a l V e n t i l a t i o n i n Air-Breathing Fish 6 Physiological Control of Air-Breathing i n Fish 9 Intermittent Breathing in Air-Breathing Fish 15 MATERIALS AND METHODS 2 0 Animals 2 0 Air-Breathing Behavior i n Undisturbed Fish 20 Measurement of A i r Flow and Expired Volume 24 A i r Bladder Deflation and I n f l a t i o n 29 Maximum A i r Bladder Volume Measurements 3 0 Branchial Nerve Denervation 31 Intra-Cranial Perfusion 33 Spectral Analysis of the Intermittent A i r -Breathing Pattern 3 8 Data Analysis and S t a t i s t i c s 42 RESULTS 4 3 Air-Breathing Rates and Behavior for Undisturbed f i s h 4 3 A i r Bladder Deflation and I n f l a t i o n 62 Maximum A i r Bladder Volume 72 Branchial Nerve Denervation 72 Intra-Cranial Perfusion 86 v i Spectral Analysis of the Intermittent A i r -Breathing Pattern 93 DISCUSSION 101 Air-Breathing Patterns and the Responses to Changes i n Aquatic and A e r i a l Gas Composition 101 A i r Bladder Mechanoreceptors i n the Control of Air-Breathing 110 Chemoreceptor Sites i n the Control of Air-Breathing 116 Peripheral Sites of Chemoreception 116 Central Sites of Chemoreception 123 Intermittent Air-Breathing i n Amia 12 6 A Computer Model of Intermittent Air-Breathing i n Amia 129 Results and Comparison with Data from Undisturbed Amia 137 Evolutionary Implications 152 LITERATURE CITED 154 APPENDIX 1 165 APPENDIX 2 169 v i i LIST OF TABLES Table I. Mean i n t e r - b r e a t h i n t e r v a l (min) f o r undisturbed Amia i n v a r i o u s c o m b i n a t i o n s of a q u a t i c and a e r i a l conditions 53 Table I I . Mean inter-breath i n t e r v a l s (min) for sham-operat-ed, p a r t i a l or t o t a l branchial denervate f i s h 83 Table I I I . Mean i n t e r - b r e a t h i n t e r v a l s (min) f o r type I breaths only for branchial denervate groups 85 Table IV. Mean g i l l v e n t i l a t i o n r ates (fg) f o r b r a n c h i a l denervate groups 88 Table V. Mean values f o r g i l l v e n t i l a t i o n (fg) and buccal pressure amplitude (Pb) and air-breathing during i n t r a - c r a n i a l perfusion with mock EDF e q u i l i b r a t e d with four d i f f e r e n t gas mixtures i n aquatic normoxia or hypoxia 90 Table VI. Dorsal a o r t i c P0 2, pH and PC0 2 and ai r - b r e a t h i n g r a t e s d u r i n g EDF p e r f u s i o n i n a q u a t i c normoxia and hypoxia 9 2 Table A l . Summary of i n i t i a l parameters used i n the model of intermittent breathing 165 v i i i L I S T OF F I G U R E S Figure 1. Cladogram showing the re l a t i o n s h i p of Amia calva to other vertebrate groups 3 Figure 2. Schematic diagram of the set-up used to record a i r -breathing behavior and frequency i n undisturbed Amia 23 Figure 3. T y p i c a l c a l i b r a t i o n records of manually-generated a i r flow measured with the pneumotachograph and pressure transducer system 27 Figure 4. Inter-breath i n t e r v a l as a function of time f o r a s i n g l e f i s h i n normoxia. Resolution of the time s e r i e s of ai r - b r e a t h i n g into p o s i t i v e and negative d e l t a functions for use i n spectral analysis 40 Figure 5. Recordings of expiratory a i r flow from two f i s h i n normoxic c o n d i t i o n s . Type I a i r breaths and corresponding type II a i r breaths are from the same f i s h 45 Figure 6. Diagram of the sequence of events involved i n a type I a i r breath 48 Figure 7. Diagram of the sequence of events involved i n a type II a i r breath 50 F i g u r e 8. Histogram of the d i s t r i b u t i o n of i n t e r - b r e a t h i n t e r v a l s (IBI; min) f o r eight f i s h i n aquatic and a e r i a l normoxia at 22+2 °C 55 Figure 9. Histogram of the d i s t r i b u t i o n of IBI (min) f o r eight f i s h i n aquatic normoxia and a e r i a l hypercapnia (5% C0 2 i n a i r) 57 Figure 10. Histogram of the d i s t r i b u t i o n of IBI (min) f o r eight f i s h i n aquatic hypoxia and a e r i a l normoxia 59 Figure 11. Histogram of the d i s t r i b u t i o n of IBI (min) f o r e i g h t f i s h i n a q u a t i c normoxia and a e r i a l h y p o x i a (8% 0 2) 61 Figure 12. Histogram of the d i s t r i b u t i o n of IBI (min) f o r ei g h t f i s h i n aquatic normoxia and a e r i a l hyperoxia (100% 0 2) 64 Figure 13. Histogram of the d i s t r i b u t i o n of IBI (min) f o r e i g h t f i s h i n a q u a t i c hypoxia and a e r i a l hyperoxia (100% 0 2) 66 Figure 14. Relationship between the time taken to i n i t i a t e an a i r breath (s) as a function of gas removed (ml kg - 1) from the a i r bladder i n aquatic normoxia, hypoxia and hyperoxia.... 69 Figure 15. Relationship between the number of ai r - b r e a t h i n g events i n 10 min afte r a i r bladder d e f l a t i o n as a function of i x volume removed (ml kg - 1) 71 Figure 16. Relationship between a i r bladder volume (ml) and body mass (g) determined in vitro 74 Figure 17. Histogram of the d i s t r i b u t i o n of IBI (min) f o r type I and type II breaths for sham-operated f i s h . A. Aquatic and a e r i a l normoxia. B. Aquatic hypoxia/aerial normoxia...77 Figure 18. Histogram of the d i s t r i b u t i o n of IBI (min) f o r p a r t i a l b r a n c h i a l denvervate f i s h . A. A q u a t i c and a e r i a l normoxia. B. Aquatic hypoxia/aerial normoxia 79 Figure 19. Histogram of the d i s t r i b u t i o n of IBI (min) f o r t o t a l branchial denervate f i s h . A. Aquatic and a e r i a l normox-i a . B. Aquatic hypoxia/aerial normoxia 81 Figure 20. Inter-breath i n t e r v a l s (IBI; min) plotted against cumulative time (min) f o r two f i s h i n a q u a t i c and a e r i a l normoxia (A) and aquatic hypoxia/aerial normoxia (B) 95 Figure 21. A. Spectrum-averaged periodogram f o r 6 f i s h i n aquatic and a e r i a l normoxia and aquatic hypoxia/aerial nor-moxia and B. Average periodogram before spectrum averaging in normoxia 97 Figure 22. Spectrum-averaged periodogram for 6 f i s h i n aquatic normoxia/aerial hypoxia (8% 0 2) 100 Figure 23. A schematic diagram of the e s s e n t i a l features of the model used to simulate air-breathing i n Amia 133 Figure 24. A. Simulated IBI (min) as a function of cumulative time (min) f o r the model with 0% e r r o r i n the breaths. B. Simulated changes i n e f f e r e n t blood PO, (Torr) as a function of cumulative time (min) from model r e s u l t s 139 Figure 25. A. Simulated IBI (min) p l o t t e d as a function of cumulative time (min) i n the model with ± 10% error i n both types of breaths. B. Simulated changes i n e f f e r e n t blood P0 2 (Torr) as a function of cumulative time (min) 142 Figure 26. Periodogram generated from 10 random data sets from t h e model w i t h + 12.5% e r r o r i n b o t h t y p e s of breaths 14 5 Figure 27. A. Plot of IBI (min) vs. cumulative time (min) for model data with simulated 100% 0 2 i n i n s p i r e d gas and + 15% error i n both types of breaths. B. Plot of IBI vs time for a singl e f i s h exposed to 100% 0 2 i n the a e r i a l phase i n aquatic normoxia 149 Figure 28. A. Plot of IBI (min) vs. cumulative time (min) for model data with simulated 8% 0 2 i n i n s p i r e d gas and + 10% e r r o r i n both types of breaths. B. IBI vs. cumulative time for a single f i s h exposed to 8% 0 2 i n the a e r i a l phase....151 x ACKNOWLEDGEMENTS There are several people I must acknowledge and thank for t h e i r support during my years i n Vancouver. F i r s t , I thank my s u p e r v i s o r , David Jones, f o r i n t r o d u c i n g me to the "bald f i s h . " Although the p r o j e c t took o f f i n a d i r e c t i o n that neither of us had anticipated, I think we can both be pleased with the outcome. I a l s o thank Dave f o r h i s comments on various d r a f t s of t h i s t h e s i s that were extremely h e l p f u l i n improving i t s q u a l i t y . Most importantly, however, I thank Dave fo r the many, many hours of great darts matches i n which I believe I held the edge. I am g r a t e f u l to Dr. Chris Wood and Steve Munger, Dept. of Biology, McMaster U n i v e r s i t y , f o r supplying most of the bowfin used i n t h i s study. I thank the members of my auto-da-fe 1, D.J. Randall, P.W. Hochachka, J.D. Steeves and W.K. Milsom, f o r not completely giving up on me when I gave them ample reason for doing so. I would a l s o l i k e to thank B i l l Milsom f o r commenting on the penultimate version of t h i s thesis. I am g r a t e f u l to s e v e r a l of my f r i e n d s , i n c l u d i n g Hugo Bergen, Barry and Joanie McKeon, Heather K i r k and Bernhard Weber, f o r many hours of enjoyment away from the department. A l s o , my h e a r t f e l t thanks f o r the support of f a m i l y and friends i n the Rose City whom I did not see often enough while here i n Vancouver: my parents, Gene and Judie Hedrick, Marcus "Farmer-of-the-Year" and M a r i l y n Simantel, the r e s t of the Hedrick and Simantel clans, Wayne Palioca, Debbie D u f f i e l d and Stan Hillman. Several of my past and present lab-mates must take r e -s p o n s i b i l i t y for making l i f e i n the Jones' lab at times actu-a l l y enjoyable: Richard Stephenson, Molly Lutcavage, Peter Bushnell, Geoff Gabbott, Agnes Lacombe and P h i l Davies. My s p e c i a l thanks to Claudia Kasserra for sharing o f f i c e space, many laughs and for helping to improve my English on occasion-including the Introduction to t h i s thesis. I am greatly indebted to my good friend, Steve Katz, for l i t e r a l l y hundreds of hours of intense d i s c u s s i o n s , b i t c h s e s s i o n s and infotainment, a l l over the approximately 1000 l i t e r s of c o f f e e we have drunk together over the past f i v e years. I also thank Steve for helping with various aspects of my work i n c l u d i n g the s p e c t r a l a n a l y s i s of bowfin breathing, and for discussions and assistance that led to the formulation of the model of intermittent a i r - b r e a t h i n g presented i n t h i s t h e s i s . Steve wrote the Turbo Pascal program fo r the i n t e r -mittent air-breathing simulations that appears i n Appendix 2. I also appreciate h i s valuable comments on parts of the Dis-cussion section of t h i s t h e s i s . x i I also g r a t e f u l l y acknowledge the collaborative e f f o r t of Mark Burleson on the i n t r a - c r a n i a l perfusion s e c t i o n of t h i s the-s i s . This work has been published i n the Journal of Experi-mental Biology (Hedrick, Burleson, Jones and Milsom, 1991); however, the views expressed i n t h i s thesis are my own and do not necessarily agree with any of the co-authors. Last - and c e r t a i n l y the most - I cannot thank my wife, Amy, enough fo r her love and support that has allowed me to complete t h i s d i s s e r t a t i o n . x i i INTRODUCTION The bowfin, Amia calva, i s the o n l y extant member of the s u b d i v i s i o n Halecomorphi, the most advanced A c t i n o p t e r y g i a n ( r a y - f i n n e d ) f i s h e s not i n c l u d e d among the t e l e o s t s ( C a r r o l l 1988). The e a r l i e s t known f o s s i l s of the genus Amia are from lower Cretaceous d e p o s i t s , and the f a m i l y Amiidae and c l o s e l y r e l a t e d forms a r e known from t h e upper J u r a s s i c (Boereske 1974). A n a t o m i c a l l y , A. calva and i t s f o s s i l congeners are c h a r a c t e r i z e d by a unique jaw a r t i c u l a t i o n : the s y m p l e c t i c and quadrate bones both a r t i c u l a t e with the lower jaw (Lauder and Liem 1983). The c u r r e n t g e o g r a p h i c d i s t r i b u t i o n of A. calva i s e n t i r e l y North American and i n c l u d e s f r e s h w a t e r l a k e s and streams i n the Great Lakes r e g i o n (except Lake S u p e r i o r ) , and most of the major r i v e r drainage systems i n the E a s t e r n U n i t e d S t a t e s (Boreske 1974). Amia has l o n g been known as a v o r a -c i o u s p r e d a t o r and was much d e s p i s e d by f i s h e r m e n around the t u r n of the century (Dean 1898), which probably accounted f o r i t s then pop u l a r name of "lake-lawyer." Amia i s i n an important e v o l u t i o n a r y p o s i t i o n s i n c e i t i s c o n s i d e r e d t o r e p r e s e n t the p r i m i t i v e s i s t e r l i n e a g e of modern t e l e o s t s (Lauder and Liem 1983; f i g . 1). A i r bladder v e n t i l a -t o r y mechanisms i n Amia are thought t o r e p r e s e n t the p r i m i t i v e c o n d i t i o n of the T e l e o s t e i (Liem 1989). Thus, d e t a i l e d knowl-edge of t h e mechanisms and c o n t r o l of a i r - b r e a t h i n g i n t h i s s p e c i e s may y i e l d f u r t h e r i n s i g h t i n t o the e v o l u t i o n of a e r i a l v e n t i l a t i o n i n t e l e o s t s and, perhaps, o t h e r A c t i n o p t e r y g i a n f i s h e s . 1 Figure 1. Cladogram showing the re l a t i o n s h i p of Amia calva to s e l e c t e d extant v e r t e b r a t e groups. Modified from C a r r o l l (1988) . 2 T E L E O S T S A m i a G a r P o l v p t e r u s A C T I N O P T E R Y G I I O S T E I C H T H Y E S S A R C O P T E R Y G I I Wilder (1877) f i r s t reported that Amia used i t s vascular-ized a i r bladder as a respiratory organ by demonstrating that Amia had the a b i l i t y to exhale and i n h a l e atmospheric a i r . There were few studies on Amia afte r Wilder's i n i t i a l observa-t i o n s u n t i l Reighard (1903) published the only d e t a i l e d ac-count of the l i f e h i s t o r y of t h i s species. Reighard (1903) pri m a r i l y described the breeding habits of Amia, but did note they often came to the surface for a i r . Later observations a l s o confirmed the use of a e r i a l r e s p i r a t i o n by bowfin i n nature (Doan 1938), but i t has not been demonstrated how the adaptation to air-breathing i n Amia contributes to i t s ecolog-i c a l or evolutionary success (see Endler 1986) . Other studies have reported i t s a b i l i t y to e s t i v a t e during c o n d i t i o n s of drought (Dence 1933, N e i l l 1950), with the g i l l s modified to prevent collapse in a i r during t h i s process (Bevelander 1938; Daxboeck et a l . 1981). Recent studies on Amia have focused p r i m a r i l y on the p h y s i o l o g i c a l aspects of g i l l v e n t i l a t i o n and air-breathing (Johansen et a l . 1970; Randall et a l . 1981; McKenzie 1990), and the morphological basis for a e r i a l v e n t i -l a t i o n (Deyst and Liem 1985; Liem 1988, 1989). The E v o l u t i o n of A i r - B r e a t h i n g Air-breathing i n fishes has evolved independently sever-a l times (Gans 1970) . I t i s g e n e r a l l y conceded that a i r -breathing o r i g i n a l l y evolved in a freshwater piscine ancestor (Romer 1972), but a marine o r i g i n for a i r - b r e a t h i n g has also been proposed (Packard 1974). Current theory also suggests that lungs are more p r i m i t i v e structures than swim bladders, 4 and the o r i g i n a l s e l e c t i v e force f o r the evolution of lungs was probably the need for gas exchange i n environments where seasonal droughts and hypoxia were prevalent (Romer 1972). In extant t e l e o s t s , the group from which most of the current knowledge of ai r - b r e a t h i n g mechanisms i s derived (Liem 1980, 1989), the d i v e r s i t y of structures used for air-breathing (see Carter 1957; Johansen 1970) r e f l e c t s secondary adaptations to hypoxic environments. The majority of teleo s t s , however, are not a i r breathers and have retained a swim bladder, a homo-logue of the o r i g i n a l lung, s t r i c t l y for buoyancy regulation i n the aquatic environment. In teleosts that have secondarily evolved a i r bladders for respiratory gas exchange, the buoyan-cy function of the a i r bladder has also remained (Liem 1989) . Liem (1988) suggested that of the two major functions of the o r i g i n a l p iscine lung, hydrostatic and respiratory, selec-t i o n has operated to i n t e n s i f y the hydrostatic function of the lung i n the Actinopterygii, one of the two subclasses of bony f i s h e s (Class Osteichthyes). In t h i s group there has been a s h i f t toward the need f o r buoyancy c o n t r o l i n the aquatic environment, which has led to the evolution of non-respiratory swim bladders i n t e l e o s t s . However, i n the extant p r i m i t i v e a i r - b r e a t h i n g actinopterygians, Amia, gar {Lepisosteus spp.) and the chondrostean polypterids, a i r bladders or lungs have r e t a i n e d the dual f u n c t i o n s of gas exchange and buoyancy. These groups a l l r e l y to some extent upon aquatic gas exchange through g i l l v e n t i l a t i o n , and the degree to which air-breath-ing i s used determines whether they are considered physiologi-c a l l y , but not taxonomically, as f a c u l a t i v e or obl i g a t e a i r -5 breathers (Shelton et a l . 1986). Amia and Lepisosteus are generally considered f a c u l a t i v e air-breathing f i s h since they are not completely dependent upon a e r i a l r e s p i r a t i o n f o r gas exchange. Some air - b r e a t h i n g t e l e o s t s , such as the e l e c t r i c e e l , Electrophorus electricus, are highly s p e c i a l i z e d obligate air-breathers (Johansen et a l . 1968). In the other subclass of bony fishes, the Sarcopterygii, however, the lineage from which a l l t e r r e s t r i a l vertebrates descended, the r e s p i r a t o r y gas exchange function of the lung has superceded the hydrostatic function (Liem 1988). There i s only one extant group of a i r - b r e a t h i n g sarcopterygians, the lungfishes (Order Dipnoi) , with representive species found i n A f r i c a , South America and A u s t r a l i a ( C a r r o l l 1988). Adult A f r i c a n and South American lungfishes, Protopterus spp. and Lepidosiren paradoxa, have reduced g i l l s , and are thus com-p l e t e l y dependent upon a e r i a l v e n t i l a t i o n f o r oxygen ac-q u i s i t i o n . The A u s t r a l i a n l u n g f i s h , Neoceratodus forsteri, however, i s a f a c u l a t i v e a i r - b r e a t h i n g f i s h with f u n c t i o n a l g i l l s able to meet metabolic demands in addition to occasional a i r - b r e a t h i n g (Grigg 1965b; Johansen et a l . 1967). The o v e r a l l view of the e v o l u t i o n of a i r - b r e a t h i n g i s that the primary s e l e c t i o n pressure, regardless of phylogenetic h i s t o -ry, was the need fo r oxygen a c q u i s i t i o n i n hypoxic environ-ments (Romer 1972; Randall et a l . 1981). Mechanisms of A e r i a l V e n t i l a t i o n i n A i r - B r e a t h i n g F i s h In a l l air-breathing fishes using a lung or a i r bladder, a positive-pressure buccal force pump i s used to v e n t i l a t e the 6 gas exchanger (Gans 1970). A notable exception i s the recent f i n d i n g that p o l y p t e r i d f i s h e s v e n t i l a t e t h e i r lungs with a unique aspiratory mechanism (Brainerd et a l . 1989). The a e r i a l v e n t i l a t i o n mechanics of the actinopterygians Amia, Lepisosteus and a number of air-breathing teleosts have been extensively examined (Johansen et a l . 1970/ Rahn et a l . 1970; Kramer 1978; Ishimatsu and Itazawa 1981; Greenwood and Liem 1984; Deyst and Liem 1985; Liem 1980, 1984, 1988, 1989). Although d e t a i l s concerning mechanisms vary among species, the fundamental breathing sequence appears to be c o n s i s t e n t : a double pulse buccal mechanism with exhalation preceding inha-l a t i o n , with the inhaled gas forced into the a i r bladder by the a c t i o n of b u c c a l musculature (Gans 1970; Liem 1989). E x h a l a t i o n and i n h a l a t i o n i n these s p e c i e s are t h e r e f o r e active processes involving intra-pulmonary and buccal pressure gradients. The action of the buccal pump forcing a i r into the gas bladder d i s t i n g u i s h e s t h i s mechanism as a p o s i t i v e - p r e s -sure pump, to contrast with aspiration breathing where a i r i s drawn into an expanding lung by negative pressure (Gans 197 0; Brainerd et a l . 1989). Recent studies of the a i r - b r e a t h i n g mechanism i n Amia (Randall et a l . 1981; Deyst and Liem 1985) have, therefore, confirmed Wilder's (1877) o r i g i n a l observa-t i o n s that e x h a l a t i o n preceded i n h a l a t i o n i n t h i s s p ecies. Liem (1989) has re-examined a i r - b r e a t h i n g i n Amia and has suggested that v e n t i l a t i o n may also occur as a passive event, without the a c t i v e f a c i l i t a t i o n of a i r flow by the buccal cavity. Thus, an accurate picture of air-breathing mechanisms in Amia appears incomplete. 7 A e r i a l v e n t i l a t i o n i n lungfishes, while using a s i m i l a r buccal force pump compared with actinopterygians, produces a somewhat d i f f e r e n t pattern of a i r t r a n s f e r . McMahon (1969) reported that Protopterus f i r s t i nhaled atmospheric gas by drawing a i r into the buccal cavity, exhaled by passive r e c o i l of the lung through an open g l o t t i s , then forced the buccal a i r into the lung. This sequence of a i r flow may p o t e n t i a l l y mix inspired and expired gases. McMahon (1969) also reported that Protopterus o c c a s i o n a l l y i n s p i r e d without e x p i r i n g , and t h i s o c curred when lung pressure was h e l d at atmospheric pressure. A s i m i l a r sequence of a i r flow events has been found f o r Lepidosiren (Bishop and Foxon 1968). The mechanics of a i r - b r e a t h i n g i n Neoceratodus have apparently not been examined i n d e t a i l (Grigg 1965a). The major d i f f e r e n c e be-tween the v e n t i l a t o r y patterns of actinopterygian and sarcop-terygian f i s h appears to be i n the sequence of a i r flow. In actinopterygians, exhalation nearly always precedes i n h a l a -t i o n , while i n sarcopterygians, inhalation may precede exhala-t i o n , with some mixing of i n s p i r e d and expired a i r i n the buccal cavity. The sequence of air-breathing events has been examined i n d e t a i l only i n Protopterus, but the v a r i a b i l i t y reported f o r breathing mechanisms i n l u n g f i s h would i n d i c a t e f u r t h e r examination i s probably necessary. The reasons f o r the basic differences i n a i r flow patterns between actinopter-ygians and sarcopterygians are unknown, but probably r e f l e c t the divergent phylogenetic h i s t o r i e s of these two o s t e i c h -thyean groups. 8 P h y s i o l o g i c a l C o n t r o l of A i r - B r e a t h i n g i n F i s h Water and a i r , c o n s i d e r e d as r e s p i r a t o r y media, have placed d i f f e r e n t demands and constr a i n t s on the evolution of gas exchange mechanisms and t h e i r c o n t r o l (Dejours 1975) . Water has a lower capacitance for 0 2 than i t does for C0 2 and f i s h using aquatic r e s p i r a t i o n must v e n t i l a t e large volumes of water to e x t r a c t enough 0 2 to meet metabolic demands. The r e l a t i v e l y high v e n t i l a t i o n rates for 0 2 e x t r a c t i o n , and the high s o l u b i l i t y of C0 2 i n water, r e s u l t i n low p a r t i a l pres-sures of C0 2 i n the blood of f i s h ; consequently, the co n t r o l of g i l l v e n t i l a t i o n i n f i s h i s dominated by 0 2 rather than C0 2 (Shelton et a l . 1986). Although some attention has been given to the e f f e c t s of C0 2 and pH on aquatic v e n t i l a t i o n i n f i s h (Janssen and Randall 1975; Wood et a l . 1990), a large body of work supports the hypothesis t h a t 0 2 e x e r t s the dominant e f f e c t on c o n t r o l of v e n t i l a t i o n i n f i s h e s (Shelton et a l . 1986). Almost without exception, a q u a t i c hypoxia s t i m u l a t e s a e r i a l v e n t i l a t i o n i n a l l air-breathing f i s h (see Carter 1957; Johansen 1970; Shelton et a l . 1986). This has been taken as strong evidence that the common selection pressure of environ-mental hypoxia was responsible for the independent development of air-breathing mechanisms in phylogenetically diverse groups of f i s h e s . F i s h that r e p o r t e d l y do not respond to aquatic oxygen l e v e l s are o b l i g a t e a i r - b r e a t h i n g f i s h , such as the t e l e o s t , e l e c t r i c e e l (Electrophorus; Johansen et a l . 1968; Farber and Rahn 1970) and the lungfishes, Lepidosiren (Johan-sen and Lenfant 1967) and Protopterus (Johansen and Lenfant 9 1968a), where metabolic demand i s met e n t i r e l y through a e r i a l v e n t i l a t i o n . In these species, the g i l l s are morphologically reduced, thereby l i m i t i n g branchial 0 2 extraction or loss, but C0 2 excretion continues through branchial routes owing to the much higher s o l u b i l i t y of C0 2 i n water. In f a c u l a t i v e a i r -breathers, such as Amia and Lepisosteus, that r e l y to varying extents upon branchial 0 2 uptake, C0 2 e l i m i n a t i o n also takes place almost completely at g i l l r e s p i r a t o r y surfaces; conse-quently, the lung or a i r bladder respiratory r a t i o (C0 2 e l i m i -nation/02 u P t a k e ) i n ai r - b r e a t h i n g f i s h i s u s u a l l y l e s s than 0.2, compared with a value of 1.0 i n the lungs of completely t e r r e s t r i a l vertebrates (Shelton et a l . 1986). Although there i s a vast l i t e r a t u r e documenting the p h y s i o l o g i c a l and behavioral (see Kramer 1987) responses of a i r - b r e a t h i n g f i s h to changes i n aquatic 0 2 concentrations, few studies have examined potential s i t e s of control for these r e f l e x e s . Considerably more study has been devoted to the control of g i l l v e n t i l a t i o n i n s t r i c t l y water-breathing f i s h . Several l i n e s of evidence from d i f f e r e n t species of water-breathing f i s h suggests that 0 2 - s e n s i t i v e chemoreceptors are located i n or near g i l l vasculature, where 0 2 may be sensed from both aquatic and i n t r a v a s c u l a r s i t e s (Shelton et a l . 1986). Aquatic hypoxia and chemical stimulants such as sodium cyanide (NaCN) applied to either or both locations stimulates g i l l v e n t i l a t i o n (Saunders and S u t t e r l i n 1971; Bamford 1974; Eclancher and Dejours 1975; Daxboeck and Holeton 1977; Smith and Jones 1982; Burleson and Smatresk 1990a). Neurophysiolog-i c a l evidence also supports the hypothesis of branchial loca-10 t i o n s f o r chemoreception s i n c e a f f e r e n t neural discharge, responding to changes i n P0 2 or pH has been recorded from c r a n i a l nerves IX (glossopharyngeal) and X (vagus) innervating the pseudobranch (Laurent and Rouzeau 1972), and the f i r s t g i l l arch i n t e l e o s t s (Milsom and B r i l l 1986; Burleson and Milsom 1990, Burleson 1991). Ablation or denervation of these s i t e s , however, has generally f a i l e d to abolish the v e n t i l a t o -ry responses to aquatic hypoxia (Hughes and Shelton 1962; Saunders and S u t t e r l i n 1971; Randall and Jones 1973). A recent study on c a t f i s h reports the a b o l i t i o n of g i l l v e n t i l a -tory responses to branchial denervation (Burleson and Smatresk 1990b): t r a n s e c t i o n of a l l b r a n c h i a l branches of c r a n i a l nerves IX and X were required to ab o l i s h v e n t i l a t o r y r e f l e x responses to aquatic hypoxia, in d i c a t i n g that there are sever-a l l o c a t i o n s on the g i l l s that convey 0 2 - s e n s i t i v e informa-t i o n . The equivocal evidence in support of branchial chemosen-s i t i v e s i t e s f o r mediating hypoxic v e n t i l a t o r y r e f l e x e s i n f i s h has l e d some authors to suggest that e x t r a - b r a n c h i a l l o c a t i o n s may a l s o be involved. These inc l u d e the venous va s c u l a t u r e , which may be b r a n c h i a l or non-branchial, f o r which there i s limited evidence (Barrett and Taylor 1984), and the central nervous system (CNS) (Saunders and S u t t e r l i n 1971; Bamford 1974; Jones 1983). Chemosensitive s i t e s i n the CNS have been proposed primarily on the basis of f a i l u r e to d e l i n -eate peripheral s i t e s , however, there i s no evidence support-ing t h i s hypothesis i n water-breathing f i s h (Graham et a l . 1990). 11 B r a n c h i a l r e f l e x responses to aquatic hypoxia vary be-tween air-breathing f i s h . For instance, i n Lepisosteus (gar) g i l l v e n t i l a t i o n increases i n i t i a l l y with aquatic hypoxia, but i s depressed with further reductions i n aquatic P0 2 (Smatresk and Cameron 1982a; Smatresk 1986). Johansen et a l . (1970) reported a s i m i l a r pattern f o r Amia, but a recent study sug-gests depression of branchial v e n t i l a t i o n does not occur with Amia i n aquatic hypoxia (McKenzie 1990). In the p o l y p t e r i d reedfish, Erpetoichthys (=Calamoichthys) calabaricus, branchi-a l v e n t i l a t i o n i s also i n h i b i t e d at low aquatic P0 2 ( P e t t i t and B e i t i n g e r 1985), s i m i l a r to the pattern i n gar. The g i l l c i r c u l a t i o n i n these f i s h i s d i s t a l to the lung or a i r blad-der, so the i n h i b i t i o n of g i l l v e n t i l a t i o n i n hypoxia has been interpreted as a mechanism to avoid branchial 0 2 loss through d i f f u s i o n (Johansen et a l . 1970; Smatresk and Cameron 1982a). In air-breathing f i s h , there i s some evidence that i n t e r -nal and external chemosensitive s i t e s modulate branchial and a e r i a l v e n t i l a t i o n . For instance, in the lungfish, Protopter-us , i n t r a - v a s c u l a r i n j e c t i o n of sodium cyanide (NaCN) or hypoxic blood increased air-breathing rates that were attenu-ated, but not abolished, by branchial denervation ( L a h i r i et a l . 1970). Recent evidence from spontaneously-breathing, anesthetized gar i n d i c a t e s that both external (aquatic) and i n t e r n a l (intravascular) chemosensitive s i t e s mediate branchi-a l and a e r i a l reflexes (Smatresk et a l . 1986). In gar, exter-nal receptors respond to reductions i n aquatic P0 2 by i n h i b i t -ing g i l l v e n t i l a t i o n and stimulating a i r - b r e a t h i n g ; i n t e r n a l chemoreceptors, however, appear to stimulate both g i l l and 12 a i r - b r e a t h i n g r e f l e x e s . Smatresk (1986) a l s o demonstrated s i m i l a r responses using NaCN in conscious gar. In s e v e r a l other a i r - b r e a t h i n g f i s h , e x t e r n a l versus i n t e r n a l chemosensitive s i t e s have been delimited non-inva-s i v e l y by independently manipulating aquatic and a e r i a l gas concentrations. Although allowing f i s h to breathe hypoxic, hyperoxic or hypercapnic gases from the a e r i a l environment has doubtful e c o l o g i c a l s i g n i f i c a n c e , these manipulations can be u s e f u l i n p o t e n t i a l l y separating aquatic and i n t r a v a s c u l a r s i t e s f o r the c o n t r o l of a i r - b r e a t h i n g . In general, a e r i a l hypoxia s t i m u l a t e s , and a e r i a l hyperoxia depresses, a e r i a l v e n t i l a t i o n i n most obligate and f a c u l a t i v e air-breathing f i s h ( L a h i r i et a l . 1970; Garey and Rahn 1970; Lomholt and Johansen 1974; Burggren 1979), suggesting that intravascular 0 2 chemo-receptors, at least, mediate some of the v e n t i l a t o r y responses to hypoxia. An exception was found with the a i r - b r e a t h i n g c a t f i s h , Brochis splendens, where ai r - b r e a t h i n g continued i n a e r i a l hyperoxia (100% 0 2) at rates s i m i l a r to those i n a e r i a l normoxia (Gee and Graham 1978). These r e s u l t s indicated that Brochis maintained air-breathing in response to reductions i n volume, and hence buoyancy, of the respiratory organ, owing to greater losses of 0 2 by d i f f u s i o n between breaths. Gee (1981) also found that the f a c u l a t i v e a i r - b r e a t h i n g t e l e o s t , Umbra limi, used a number of strategies associated with air-breathing to coordinate the respiratory and hydrostatic functions of i t s swim bladder. The two f u n c t i o n s of an a i r bladder, r e s p i r a t o r y and h y d r o s t a t i c , introduce p o t e n t i a l c o n f l i c t s (Gee and Graham 13 1978). Any a i r - f i l l e d c a v i t y used as a gas exchanger w i l l automatically change the density and, therefore, buoyancy of the animal (Alexander 1966). To be e f f e c t i v e as a gas exchang-er, an a i r bladder must permit oxygen to d i f f u s e through i t s th i n , vascularized membrane into the c i r c u l a t i o n . Oxygen l o s t through d i f f u s i o n i s generally not replaced by equal volumes of carbon dioxide (Johansen 1970), therefore, the volume of the organ decreases and the animal becomes negatively buoyant. In order to remain neutrally buoyant, a f i s h must replace the l o s t volume either by secretion of gases into the a i r bladder or, i f secretory mechanisms are too slow or not present, they may gulp a i r (Alexander 1966). Since e f f i c i e n t h y d r o s t a t i c organs have low p e r m e a b i l i t i e s to gas l o s s , they are not u s e f u l as gas exchangers. The swim bladders of t e l e o s t s are u s u a l l y l i n e d with guanine c r y s t a l s that prevent s i g n i f i c a n t amounts of oxygen d i f f u s i o n (Fange 1976). Secretory mecha-nisms, which regulate buoyancy in physoclistous fishes (those without connections between the swim bladder and gut), are usually slow or absent in physostomous (those with connections between the a i r bladder and gut) f i s h e s (Jones and Marshall 1953). In most physostomous teleosts, therefore, air-breathing i s e s s e n t i a l f o r buoyancy r e g u l a t i o n (Jones and M a r s h a l l 1953) . There i s evidence i n p r i m i t i v e f i s h e s that a i r bladders or lungs have a hydrostatic, i n addi t i o n to r e s p i r a t o r y gas exchange, f u n c t i o n . In Amia, Lepisosteus and Protopterus, lung d e f l a t i o n stimulates, and i n f l a t i o n i n h i b i t s , air-breath-ing reflexes (Johansen et a l . 1970; Babiker 1979; Smatresk and 14 Cameron 1982b; Pack et a l . 1984). The a i r bladders or lungs of these f i s h e s contain slowly-adapting and/or r a p i d l y adapting mechanoreceptors with afferents i n the vagus nerve (DeLaney et a l . 1983; Milsom and Jones 1985; Smatresk and A z i z i 1987). These mechanoreceptors respond to both dynamic and s t a t i c changes i n lung volume; a f f e r e n t discharge increases with i n f l a t i o n while d e f l a t i o n reduces a f f e r e n t discharge (Milsom 1990) . Increasing lung pressure, and presumably a f f e r e n t mechanoreceptor discharge, i n Protopterus has also been shown to inc r e a s e the i n t e r - b r e a t h i n t e r v a l (Pack et a l . 1990), sug g e s t i n g t h a t pulmonary mechanoreceptors i n f l u e n c e the breathing pattern. The mechanoreceptor-related r e f l e x e s i n air-breathing f i s h are sim i l a r to the i n s p i r a t o r y - i n h i b i t i n g , e x p i r a t o r y - f a c i l i t a t i n g Hering-Breuer r e f l e x e s i n mammals (Pack 1981). I n t e r m i t t e n t Breathing i n Air-Breathing Fish Air-breathing i n f i s h requires intermittent excursions to the water's surface for a e r i a l v e n t i l a t i o n . The control over the timing of i n t e r m i t t e n t a i r breaths i n f i s h , and other intermittently-breathing vertebrates, has received considera-ble a t t e n t i o n i n recent years (see Shelton et a l . 1986; Sma-tresk 1990; Milsom 1991, for reviews). I t has been suggested that i n t e r m i t t e n t a i r - b r e a t h i n g i n f i s h represents an "on-demand" phenomenon, dependent only upon per i p h e r a l a f f e r e n t feedback from receptors for i t s i n i t i a t i o n (Smatresk 1990). Afferent feedback from chemoreceptors and mechanoreceptors are therefore thought to play dominant r o l e s i n determining the 15 timing of a i r - b r e a t h i n g events i n i n t e r m i t t e n t l y - b r e a t h i n g vertebrates (Shelton et a l . 1986). The a b i l i t y of aquatic and a e r i a l 0 2 concentrations and lung volume manipulations to markedly a f f e c t inter-breath i n t e r v a l s , regardless of uncer-t a i n t i e s about receptor l o c a t i o n s , c l e a r l y i l l u s t r a t e s the r o l e of receptors i n modifying i n t e r m i t t e n t breathing pat-terns . Intermittent breathing contrasts with the near-continuous rhythmic v e n t i l a t o r y patterns of f i s h and mammals that occur under normal conditions (Milsom 1991). In both groups, stud-i e s have shown that rhythmic patterns can be generated from r e s p i r a t o r y - r e l a t e d neural discharges from brainstem s t r u c -tures in vitro, i n the complete absence of a f f e r e n t feedback (Rovainen 1974; Suzue 1984). Although the precise mechanisms accounting for rhythmogenesis i n these phylogenetically dispa-rate groups have not been determined, i t has been suggested that a c e n t r a l rhythm generator d r i v e n by pacemaker c e l l s (Rovainen 1977; B a l l i n t i j n 1982; Feldman et a l . 1990) or n e u r a l network i n t e r a c t i o n s ( R i c h t e r et a l . 1986) i n the brainstem are r e s p o n s i b l e . A key question with regard to i n t e r m i t t e n t l y - b r e a t h i n g v e r t e b r a t e s , then, i s whether the observed p a t t e r n s are i n i t i a t e d from the c e n t r a l nervous system, as i n water-breathing f i s h or mammals, or s o l e l y from feedback by p e r i p h e r a l chemo- and mechanoreceptors. Recent evidence from b u l l f r o g tadpole (Walker et a l . 1990) and t u r t l e (Douse and M i t c h e l l 1990) in vitro brainstem preparations has shown that intermittent, respiratory-related neural discharge can occur without sensory input, suggesting that intermittent, 16 or episodic, f i c t i v e breathing patterns can be endogenously generated by CNS structures i n the brainstem of these animals. Evidence from actinopterygian air-breathing f i s h , however, i s completely lacking, so i t i s not possible at present to deter-mine whether inte r m i t t e n t v e n t i l a t o r y patterns are c e n t r a l l y or p e r i p h e r a l l y generated in t h i s group. In humans, there i s considerable evidence that the con-tinuous breathing pattern contains several frequency compo-nents that are revealed by s p e c t r a l a n a l y s i s (see van den Aardweg and Karemaker 1991). Goodman (19 64) f i r s t demonstrat-ed s i g n i f i c a n t o s c i l l a t i o n s in the breathing pattern of humans with frequencies i n the range of 2-3 min to s e v e r a l hours. H l a s t a l a et a l . (1973) found s i g n i f i c a n t frequencies with periods up to 28 min i n the frequency spectra of r e s t i n g hu-mans; the peaks i n the frequency specta were c o r r e l a t e d with v a r i a t i o n s i n a number of respiratory parameters. Studies from humans i l l u s t r a t e that, i n s p i t e of near-continuous v e n t i l a -t i o n , u n d e r l y i n g r h y t h m i c i t i e s i n the v e n t i l a t o r y p a t t e r n occur under normal conditions. Although the intermittent patterns of air-breathing f i s h have often been termed i r r e g u l a r , a few studies have subjec-t i v e l y noted regular i n t e r v a l s between a i r breaths when'fish are undisturbed (see Milsom 1991). Disturbances, such as simulated predators, can markedly a l t e r any r e g u l a r i t i e s i n air-breathing rates (Gee 1980; Smith and Kramer 1986). Howev-er, there i s no q u a n t i t a t i v e evidence to show whether i n t e r -mittent breathing patterns are indeed regular. Most studies have observed a i r - b r e a t h i n g events over rather short (30 min to 2 h) time p e r i o d s , which i s probably i n s u f f i c i e n t f o r regular breathing patterns, i f they occur, to become estab-l i s h e d . The purpose of t h i s t h e s i s was to examine various as-pects of the c o n t r o l of a i r - b r e a t h i n g i n the bowfin. A com-bined approach of non-invasive, invasive and a n a l y t i c a l tech-niques were used i n t h i s investigation. In the f i r s t part of t h i s study, breathing patterns and the r e f l e x responses to changes i n r e s p i r a t o r y gases i n the aquatic and a e r i a l envi-ronments were examined using non-invasive techniques. This was done using long-term (8 h) videotaped recordings of a i r -breathing i n Amia. A i r flow during a i r - b r e a t h i n g events was measured by pneumotachography to resolve questions concerning a e r i a l v e n t i l a t o r y mechanisms. Spectral a n a l y s i s , which has been used e x t e n s i v e l y i n studies of human r e s p i r a t o r y pat-terns, was used to determine whether inte r m i t t e n t breathing patterns i n Amia are rhythmic or simply random. An invasive approach was used to examine po t e n t i a l s i t e s of c o n t r o l f o r a i r - b r e a t h i n g patterns. Both p e r i p h e r a l and central nervous system s i t e s for the control of air-breathing were examined. Peripheral control of air-breathing was exam-ined i n two ways. F i r s t , the c r a n i a l nerve innervation to the g i l l s of bowfin was e l i m i n a t e d to t e s t the hypothesis that v e n t i l a t o r y r e f l e x reponses to aquatic hypoxia are mediated by 0 2-chemoreceptor s i t e s on the g i l l s . Secondly, the r o l e of pulmonary mechanoreceptors i n a i r - b r e a t h i n g was examined by manipulating the a i r bladder volume i n conscious f i s h . Cen-18 t r a l chemical s i t e s for air-breathing were examined by chang-ing the chemical compostition of the e x t r a d u r a l f l u i d sur-rounding the brain. F i n a l l y , the empirical data obtained from bowfin i n t h i s study and from other sources were used to formulate a computer model to simulate intermittent air-breathing i n Amia. 1 9 MATERIALS AND METHODS Animals Bowfin were netted by commercial fishermen i n Lake Ontar-io and a i r - f r e i g h t e d to the U n i v e r s i t y of B r i t i s h Columbia. There was no m o r t a l i t y during shipment and the f i s h appeared to be healthy. The f i s h were kept in large c i r c u l a r f i b e r g l a s s tanks i n continuously running dechlorinated water at 6-15 °C on a 12:12 L:D c y c l e . The f i s h were not fed during winter months when ambient water temperature was low (4-6 °C) , but were o c c a s i o n a l l y fed l i v e g o l d f i s h during spring and summer when water temperature was higher and the f i s h were more active. Bowfin were not fed for at least two weeks before any experiments began. Air-Breathing Behavior i n Undisturbed Fish Eight bowfin, ranging i n s i z e from 246 g to 940 g (Mean Mass = 503 g), were brought i n t o the lab o r a t o r y and placed i n d i v i d u a l l y into 40 L p l a s t i c bins containing aerated water at the same temperature as the holding tank (T w= 6-10 °C). The f i s h were acclimated to room temperature (22 + 2 °C) by al l o w i n g the water to warm slowly overnight. The water was continuously aerated during a c c l i m a t i o n . Depending on the i n i t i a l temperature, e q u i l i b r a t i o n with room temperature took from 12 to 24 h. After 4-5 days at room temperature, the f i s h were transferred to a 68 L rectangular aquarium (60 cm X 3 0 cm X 38 cm deep) f i l l e d with water p r e - e q u i l i b r a t e d to room 20 temperature. The surface of the aquarium was covered with a perspex b a r r i e r containing several small (1 cm dia.) holes, and one large hole (either 14 cm dia. or 10 cm dia.) through which the f i s h could breathe gases. A diagram of the experimetal set-up i s shown in figure 2. An inverted funnel (vol.= 650 ml or 300 ml) with a pneumotachograph (Fleisch) attached at the top was used to record a i r flow changes during a i r - b r e a t h i n g events. The pressure drop across the pneumotachograph was measured with a d i f f e r e n t i a l pressure transducer (Validyne, model DP103-18). A constant gas flow of 200 ml m i n - 1 , regu-lated by p r e c i s i o n gas flow meters, was delivered through the funnel. The constant gas flow r e s u l t e d i n a voltage o f f s e t that was manually readjusted to zero on the c a r r i e r demodula-t o r of the pressure transducer. The d i f f e r e n t i a l pressure s i g n a l was fed through a voltage-frequency converter (A.C. Vetter, Inc., model 2D) and stored on the audio track of a JVC (model TU-S2U) video cassette recorder. Each animal's breath-ing behavior was recorded with a video camera (JVC) and also stored on the video tape. Each recording session was 8 h and most sessions were recorded between 1800h and 0600h to mini-mize numerous v i b r a t i o n a l disturbances that occurred during daytime. After a recording session, the tape was replayed and air-breathing rates were counted d i r e c t l y from the video tape. Inter-breath i n t e r v a l was recorded to the nearest minute using a d i g i t a l clock d i s p l a y i n g r e a l time i n view of the camera. Voltage signals from the pressure transducer were played back through the frequency-voltage c o n v e r t e r at t h i s time and displayed on a storage o s c i l l o s c o p e (Tektronix, model 5113). 21 Figure 2 . Schematic diagram of the set-up used to record a i r -breathing behavior and frequency i n undisturbed Amia. See Methods for d e t a i l s . Abbreviations: PT, Pressure Transducer; Freq.-Volt., Frequency to Voltage Converter; VCR, Video Cas-sette Recorder. 22 V C R 1 1 Freq . - V o l t . The a i r flow changes that occurred with breathing events were examined along with the breathing behavior recorded on the video tape. The voltage si g n a l s were t r a n s f e r r e d to a chart recorder (Gould, model 220) w r i t i n g on r e c t i l i n e a r c o o r d i -nates . Measurement of A i r Flow and E x p i r e d Volume The pneumotachograph was calibr a t e d by adjusting the a i r flow through the funnel from 0 to 9 L m i n - 1 and recording the voltage change. The voltage change was l i n e a r l y r e l a t e d to a i r flow over the range encountered i n t h i s study. A i r flow generated by the f i s h during breathing events was played back through the oscilloscope and the voltage converted to a i r flow (ml s - 1 ) . The volume of each breath was measured by d i r e c t integra-t i o n of the a i r flow t r a c e s , s i m i l a r to the techniques of Glass et a l . (1983), B o u t i l i e r (1984) and Funk et a l . (1986), but with some modifications. I t has been shown that measure-ment of breath volumes by i n t e g r a t i o n of pneumotachographic s i g n a l s can overestimate a c t u a l volumes, owing to i n e r t i a l e f f e c t s , e s p e c i a l l y at high a i r flow rates (Ohya et a l . 1988). T i d a l volumes were, t h e r e f o r e , c a l i b r a t e d i n the f o l l o w i n g manner: known volumes of a i r were injected manually through the funnel with a 30 ml p l a s t i c syringe to mimic the expected expired t i d a l volumes of the f i s h . The syringe b a r r e l was modified by cutting off the standard luer-lock t i p and replac-ing i t with a wide bore (0.5 cm) p l a s t i c connector. This arrangement minimized i n e r t i a l e f f e c t s r e s u l t i n g from i n j e c t -24 ing a i r through a narrow o r i f i c e . For a given volume, a i r was injected at d i f f e r e n t rates and the r e s u l t i n g pneumotachograph s i g n a l recorded on video tape as described above; thus, the c a l i b r a t i o n was performed using the same procedures as those used d u r i n g r e c o r d i n g of the f i s h e s ' b r e a t h i n g behavior. I n j e c t i n g a i r at d i f f e r e n t r a t e s r e s u l t e d i n v a r i a b l e time i n t e r v a l s ( T E) before a i r flow returned to baseline. A t y p i -c a l c a l i b r a t i o n series for 2 to 1 0 ml injected a i r i s shown in figure 3 . I t i s clear from the c a l i b r a t i o n traces that manu-a l l y i n j e c t i n g a i r r e s u l t e d i n o s c i l l a t i o n s that returned to baseline a f t e r approximately 2 0 0 ms. The o s c i l l a t o r y nature of the c a l i b r a t e d volumes was a f u n c t i o n of changes i n the water l e v e l a f t e r i n j e c t i n g a i r , since these o s c i l l a t i o n s were absent when the funnel-pneumotach arrangement was tested on a s o l i d s u r f a c e . Each flow p r o f i l e was i n t e g r a t e d from the recorder chart by measuring the area under the curve (shaded areas i n f i g . 3 ) with a d i g i t i z i n g tablet (Jandel S c i e n t i f i c ) and associated software (Sigma Scan). For each c a l i b r a t i o n volume, area under the flow curve (cm ) was p l o t t e d against T ' e (ms) to y i e l d a series of c a l i b r a t i o n curves. Expired a i r flow traces for each f i s h were integrated i n the same fashion (see Results, f i g . 5 ) , and the expired volume estimated to the nearest 0 . 5 ml by f i t t i n g the r e s u l t i n g area and expiratory i i n t e r v a l (T E) to the nearest value for area and T E . Each f i s h was given one day to become accustomed to the aquarium and the breathing funnel. On the second day, video taping of the a i r - b r e a t h i n g behavior began with the f i s h i n normoxic water and a i r passing through the funnel. The water 2 5 Figure 3. T y p i c a l c a l i b r a t i o n records of manually-generated a i r flow measured with the pneumotachograph and pressure transducer system. Known volumes (2 ml to 10 ml) of a i r were i n j e c t e d manually with a syringe and measured by i n t e g r a t i n g i the area under each flow curve (shaded areas) . T E was de-f i n e d as the i n i t i a l time f o r p o s i t i v e flow to r e t u r n to baseline. The baseline o s c i l l a t e s for approximately 200-300 ms a f t e r the i n j e c t i o n . Note the change in scale for the 8 ml and 10 ml volumes. 26 VOLUME (ml) 4 6 8 10 to 100 ms was continuously aerated to maintain a p a r t i a l pressure of oxygen above 140 Torr (normoxia). Water samples from the aquarium were i n j e c t e d onto a Radiometer oxygen e l e c t r o d e (E5046), maintained at the same temperature as the aquarium, and measured with a Radiometer (model PHM 71) oxygen meter. Water temperature and P0 2 were taken at the beginning and end of each r e c o r d i n g s e s s i o n . Water samples taken at various l o c a t i o n s i n the aquarium revealed no d i f f e r e n c e s i n P0 2 or temperature. There were no detectable l e v e l s of C0 2 i n the water. Several recordings were made for each f i s h i n a variety of combinations of aquatic and a e r i a l gas concentrations. The gas composition of the aquatic or a e r i a l phases was changed by mixing combinations of gases with the flow meters, or with pr e c i s i o n gas mixing pumps (Wosthoff) p r i o r to being delivered through e i t h e r the aquatic or a e r i a l phase. In a l l , the f i s h were exposed to seven combinations of a q u a t i c and a e r i a l conditions, and are r e f e r r e d to i n the text with the aquatic f o l l o w e d by t h e c o r r e s p o n d i n g a e r i a l c o n d i t i o n : (1) Normoxia/Air, (2) Normoxia/Hypercapnia, (3) Hypoxia/Air, (4) Normoxia/Hypoxia, (5) Normoxia/Hyperoxia, (6) Hypoxia/Hyperox-i a and (7) Hyperoxia/Air. Aquatic P0 2 (P w0 2) i n normoxia was 140<PwO2<160 Torr. P w°2 ^ n hypoxia/air was 54.6+6.5 Torr, and i n hypoxia/hyperoxia was 50.0+4.0 Torr. The PC0 2 of a e r i a l hypercapnia (5% C0 2 i n a i r ) and P0 2 of a e r i a l hypoxia (8% 0 2 b a l . N 2) was confirmed with Radiometer C0 2 and 0 2 electrodes, r e s p e c t i v e l y . The ele c t r o d e s were maintained at the same temperature as the f i s h (22 + 2 °C) and c a l i b r a t e d with pre-cisely-mixed gases from a Wosthoff gas mixing pump (C0 2 elec-28 trode) or a i r - s a t u r a t e d water and a zero P0 2 s o l u t i o n (0 2 electrode) or N 2~equilibrated water. 100% oxygen was used for aquatic or a e r i a l hyperoxia and the r e s u l t i n g P0 2 i n e i t h e r phase measured with the 0 2 electrode. The water was changed every 5-7 days with dechlorinated water p r e - e q u i l i b r a t e d to room temperature. After each water change, f i s h were returned to c o n t r o l c o n d i t i o n s (normoxic water and a i r ) . The f i s h showed no apparent i l l e f f e c t s from disturbance due to t h i s procedure and there was no c o r r e l a t i o n between ai r - b r e a t h i n g rates and water change of the aquarium. A i r Bladder D e f l a t i o n and I n f l a t i o n Four bowfin (mean mass=348 g) were brought i n t o the laboratory and acclimated to room temperature (22 + 2.0 C) as described above. Following acclimation, f i s h were anesthe-t i z e d i n i t i a l l y in 1:7,500 buffered t r i c a i n e methanesulfonate (MS 222; Sydell Laboratories) u n t i l g i l l v e n t i l a t o r y movements ceased, t r a n s f e r r e d to a s u r g i c a l table where the g i l l s were a r t i f i c i a l l y v e n t i l a t e d with oxygenated, dechlorinated water (22 ° C ) c o n t a i n i n g a lower c o n c e n t r a t i o n of a n e s t h e t i c (1:20,000). The g l o t t i s was located i n the dorsal gut wall and held open with forceps. A 50 cm length of PE 90 tubing with a f l a r e d end containing side holes to minimize blockage was placed approximately 3-5 cm i n t o the lumen of the a i r bladder. The cannula was secured to the p a l a t e with two sutures (0 s i l k ) and passed through a PE 200 grommet that had been placed through the f r o n t a l bone into the buccal c a v i t y . The outside diameter of PE 50 i s 0.965 mm, while the diameter 29 of the g l o t t a l aperture in Amia of the size used i n t h i s study was about 5 mm (pers. obs.); therefore, cannulating the a i r bladder by t h i s method would be expected to occlude the g l o t -t i s by less than 5%. Each f i s h was allowed to recover for 24 h i n an aquarium c o n t i n i n g aerated water. A f t e r recovery, d e f l a t i o n s and i n f l a t i o n s , of various volumes were done with the f i s h i n normoxic (P w0 2=155 Tor r ) , hypoxic (P w0 2=55 Torr) and hyperoxic (P w0 2=270) water over 2-3 days. A i r bladder volume manipulations were done with as l i t t l e disturbance as possible to the f i s h . Air-breathing behavior was observed for 10 min, with the experimenter out of view of the f i s h , follow-ing each d e f l a t i o n . At l e a s t 30 min was allowed between t r i a l s . During the 10 min observation period, the time taken fo r the f i s h to i n i t i a t e an a i r - b r e a t h following d e f l a t i o n , the number of air-breaths taken during the 10 min observation period and types of air-breaths were recorded. Maximum A i r Bladder Volume Measurements Six bowfin ranging i n mass from 217 g to 1201 g were over-anaethetized i n a concentrated solution (1.5 g/L) of MS-222. A f t e r 2-3 hours, the f i s h were weighed to the nearest 0.1 g with a t r i p l e beam balance and then opened v e n t r a l l y and the lung c a r e f u l l y dissected from the surrounding musculature and connective t i s s u e . A 1 ml syringe tube, cut i n half, was placed through the g l o t t i s into the lumen of the a i r bladder and t i e d i n place with a short length of 0 s i l k . The outside end of the tube was connected to an a i r source and the a i r bladder was slowly i n f l a t e d to maximum distension, taking care 30 not to rupture the lung wall. The i n f l a t e d a i r bladder was then grasped at the bottom with long forceps and immersed completely i n a 1500 ml Erlynmeyer f l a s k f i l l e d to the top with a 0.7% s a l i n e s o l u t i o n . The immersed a i r bladder d i s -placed the s a l i n e which flowed over the top of the f l a s k and was c o l l e c t e d i n a p l a s t i c container. The container was weighed to the nearest 0.1 g before and a f t e r c o l l e c t i n g the displaced s a l i n e . The mass of saline displaced by the forceps and d e f l a t e d a i r bladder were subtracted from mass displaced by the i n f l a t e d a i r bladder. The d i f f e r e n c e i n mass repre-sented the volume displaced by the i n f l a t e d a i r bladder. The procedure was repeated three times and the average mass of displaced saline taken as a i r bladder volume. Branchial Nerve Denervation A t o t a l of 16 bowfin (Mean Mass = 433 g) were brought i n t o the l a b o r a t o r y and anesthetized. These f i s h were not acclimated to room temperature before surgery. Following the experiment, the f i s h were over-anesthetized with MS-222 for post-mortem dissec t i o n (see below). The b r a n c h i a l ( g i l l ) branches of c r a n i a l nerves IX (glossopharyngeal) and X (vagus) were exposed by l i f t i n g the operculum and c u t t i n g the t h i n membrane l y i n g medial to the g i l l s . The vagus nerve innervates a l l four branchial arches, while the glossopharyngeal nerve innervates only the f i r s t g i l l arch and pseudobranch. In some f i s h , the pseudobranch i s also innervated by c r a n i a l nerve VII (f a c i a l ) (Nilsson 1984); however, i t i s not known whether the f a c i a l nerve innervates 31 the pseudobranch i n Amia. Attempts to locate a f a c i a l nerve innervation to the pseudobranch by dissec t i o n were unsuccess-f u l . The nerves were i d e n t i f i e d , c a r e f u l l y separated from surrounding f a s c i a , and cut with s c i s s o r s . Bleeding was con-t r o l l e d with cotton applicators. Following the operation, the membrane was closed with 5-0 s i l k sutures. The entire opera-t i o n l a s t e d between 30-60 min and the f i s h were allowed to recover i n oxygenated dechlorinated water u n t i l spontaneous g i l l v e n t i l a t o r y movements returned, and the f i s h had f u l l y righted i t s e l f . Sham-operated control animals were treated in the same manner, but the nerves were not cut. The operation and i n i t i a l recovery were performed at the temperature to which the f i s h were acclimated (4-6 °C); the f i s h were then put into holding tanks at the same temperature and allowed to recover for at least 3 weeks. A f t e r the minimum 3 week recovery p e r i o d , f i s h were brought back into the laboratory and allowed to acclimate to room temperature for f i v e days. Fish were placed i n d i v i d u a l l y i n t o aquaria with aerated, dechlorinated water at 22 + 2°C. A i r flowed through the b r e a t h i n g f u n n e l at 200 ml m i n - 1 . Videotaping began on day 2 with the f i s h i n normoxic water (P wO 2>140 T o r r ) . On day 3, the aquarium water was bubbled with a mixture of nitrogen and a i r to lower P w 0 2 to 49.0+1.0 Torr. When the desired P0 2 was reached, aft e r 2-3 hours, the videotape was started. On day 4, a second normoxic treatment was recorded on videotape and on day 5 the f i s h was over-anaesthetized with MS-222. After each recording session, a i r -breathing events and, whenever p o s s i b l e , g i l l v e n t i l a t o r y 32 rates, were counted d i r e c t l y from the videotape. The e f f i c a c y of each s u r g i c a l operation was c a r e f u l l y checked by post-mortem d i s s e c t i o n . B r a n c h i a l nerves were i d e n t i f i e d and compared with published anatomical descriptions ( A l l i s 1897). F i s h were placed i n t o one of three c a t e g o r i e s based on the r e s u l t s of the post-mortem d i s s e c t i o n : P a r t i a l l y - d e n e r v a t e d (PD) f i s h were those i n which nerve re-growth was indicated or i f a l l b r a n c h i a l branches of nerves IX and X were not sec-t i o n e d (5 f i s h ) ; sham-operated (SH) c o n t r o l s were those i n which no nerves were cut (4 f i s h ) ; t o t a l branchial denervates (TD) were those f i s h in which a l l branchial branches of c r a n i a l nerves IX and X were c l e a r l y sectioned (3 f i s h ) , or i f a l l branchial branches plus the branch of c r a n i a l nerve IX inner-vating the pseudobranch (psb) was also sectioned (4 f i s h ; for a t o t a l of 7 f i s h i n t h i s category). The study was b l i n d i n that the r e s u l t s of the videotaped breathing patterns could not be c o r r e l a t e d with any p a r t i c u l a r category u n t i l a f t e r post-mortem dissection. Intra-Cranial Perfusion A t o t a l of eight bowfin were used to examine the ef f e c t s of i n t r a - c r a n i a l perfusion with mock extradural f l u i d (EDF) on branchial and a e r i a l v e n t i l a t i o n . Fish were i n i t i a l l y anes-t h e t i z e d as described i n the preceding section. The dorsal aorta was cannulated with PE 50 tubing using a canine catheter placement unit as described by Smatresk and Cameron (1982). A PE 160 cannula was implanted in the buccal cavity through the nasal bone to monitor v e n t i l a t i o n . 33 Each f i s h was prepared for c r a n i a l perfusion by d r i l l i n g two small holes i n the mid-saggital plane of the cranium using a dental d r i l l . The f i r s t hole was d r i l l e d at an angle 10-15° from perpendicular, and 2-5 mm from the p o s t e r i o r margin of the p a r i e t a l bone. A second hole was d r i l l e d , at approximate-l y the same angle as the f i r s t hole, i n the f r o n t a l bone about 10 mm anterior to the f i r s t hole. F l u i d samples were taken at t h i s time with a syringe and placed into c a p i l l a r y tubes for l a t e r ion analysis. The f l u i d was extradural f l u i d (EDF), and not true cerebrospinal f l u i d (CSF), since i t was sampled from the meningeal space (Davson 1967). Erythrocytes were present i n some EDF samples and was subsequently removed by c e n t r i f u -gation. Perfusion cannulae made from the shanks of 18 gauge s t a i n l e s s - s t e e l hypodermic needles, approximately 15 mm to 20 mm long, were implanted i n the c r a n i a l h o l e s . The s t e e l tubing extended down into the meningeal space about 10 mm i n the posterior (inflow) port and 6 mm i n the anterior (outflow) p o r t . I t was found i n p r e l i m i n a r y experiments t h a t t h i s c o n f i g u r a t i o n s i t u a t e d the inflow tube about 2 mm above the roof of the fourth v e n t r i c l e and the outflow tube above the cerebellum. Patency of the perfusion arrangement was checked by perfusing mock EDF with a syringe connected to the inflow tube. Animals i n which flow could not be f r e e l y maintained, or i n which there was excessive bleeding, were not used i n the experiment. Once patency was ensured, the perusion tubing and meningeal space were f i l l e d with mock EDF and plugged. The f i s h were subsequently t r a n s f e r r e d to a darkened perspex box with continuously flowing, normoxic water. The box had a 34 forward a i r space to allow a i r - b r e a t h i n g . Surgery seldom required longer than 20 min to complete and a l l f i s h recovered quickly a f t e r withdrawal of the anesthetic. EDF concentrations of Na + and K + were determined by flame photometry (IL model 143); C l ~ concentration was determined with a Buchler d i g i t a l chloridometer. Experiments began a f t e r a 24 h recovery period and were conducted at water temperatures between 14 and 16 °C. The temperature throughout a s i n g l e experiment d i d not vary by more than 0.5 °C. Mock EDF was freshly made with the follow--i ing composition ( a l l concentrations in mmol 1 ) : NaCl (120) , KCI (4.0), MgS04 (1.0), C a C l 2 (1.0) and NaHC03 (10.0). This f l u i d was placed into 500 ml b e l l j a r s which were maintained at the same temperature as the f i s h throughout the experiment. The mock E D F was e q u i l i b r a t e d with one of four gas mixtures: (1) a i r (P0 2=156.4+1.0 T o r r , pH=7.77+0.02); (2) 100% N 2 (PO 2=4.5±0.7 Torr, pH=8.03±0.04); (3) 100% 0 2 (PO2=705+8.4 T o r r , pH=8.03+0.04); (4) 3% C 0 2 i n a i r [ t h r e e f i s h ; P02=145+3.8 Torr, PC02=21.1 Torr (c a l c u l a t e d ) , pH=7.02+0.03] or 5% C0 2 i n a i r mixture [ f i v e f i s h ; P0 2=151+2.5 T o r r , PC02=35.3 Torr (calculated), pH=6.74+0.01]. Oxygen and carbon dioxide p a r t i a l pressures and pH of blood and mock E D F were measured with a Radiometer PHM-71 acid-base a n a l y z e r and associated electrodes maintained at the same temperature as the f i s h . The pH el e c t r o d e was c a l i b r a t e d with Radiometer standard pH buffers; the oxygen electrode was cal i b r a t e d with a i r - s a t u r a t e d water and a Radiometer zero P0 2 s o l u t i o n ; the CO? electrode was c a l i b r a t e d with precise gas mixtures from 35 Wosthoff gas-mixing pumps. In order to measure PC0 2 of blood, the C0 2 electrode was adjusted to give f u l l - s c a l e readings for samples i n the expected range (0 to 2 0 Torr). This precluded accurate measurements of PC0 2 i n mock EDF equili b r a t e d with 3% or 5% C0 2 gas mixtures. These samples were, however, injected onto the C0 2 electrode and, i n both cases, r e s u l t e d i n read-ings that were o f f - s c a l e , i n d i c a t i n g that PC0 2 was at least 20 Torr. A blood sample was taken from the d o r s a l a o r t i c (DA) cannula before an experiment began i n normoxic or hypoxic water for measurement of P0 2, PC0 2 and pH. The DA and buccal c a v i t y cannulae were then attached to Statham P23Db and Hew-le t t - P a c k a r d 267BC pressure transducers, r e s p e c t i v e l y , to monitor blood and buccal pressures. The outputs of the trans-ducers were recorded on a Gulton Techni-rite (model 722) two-channel c h a r t r e c o r d e r . A 2 min r e c o r d of DA and buccal pressures was taken as the p r e - p e r f u s i o n b a s e l i n e of these v a r i a b l e s . P e r f u s i o n then began by switching a three-way stopcock which connected the perfusate to the inflow cannula attached to the f i s h ' s cranium. The mock EDF flowed through the meningeal space for 30 min while DA and buccal pressures were recorded from 0 to 6 min, then at 2 min i n t e r v a l s span-ning the 10, 15, 20, 25 and 30 min p e r f u s i o n time periods; therefore, t o t a l recording time was about 18 min. The perfu-sate flow through the meningeal space varied between 1 and 3 ml m i n - 1 and was maintained by a d j u s t i n g the pressure head between the inflow and outflow cannuale; t h i s pressure d i f f e r -e n t i a l ranged between 3.7 and 19 Torr and had no apparent 36 e f f e c t on r e s t i n g c a r d i o v a s c u l a r or v e n t i l a t o r y v a r i a b l e s . Normoxic water (PO2=156-160 Torr) flowed continuously through the perspex box at a rate of about 1000 ml m i n - 1 . Fish were allowed 30-60 min recovery time between perfusate treatments u n t i l a l l four treatments were completed. Each experiment began with the normoxic (air-equilibrated) perfusate, but the other three perfusates were given i n random order. The f i s h were allowed to recover overnight and the perfusate treatments repeated the next day with the f i s h i n hypoxic (PO 2 = 30-40 Torr) water. Experiments were also performed i n which sodium cyanide (NaCN) dissolved i n mock EDF at i n i t i a l concentrations ranging from 5 to 1000 ug m l - 1 , or h y d r o c h l o r i c a c i d (HCl) with pH ranging from 3.6 to 7.0, were added to the mock perfusate with f i s h i n normoxic water. Although the f i n a l outflow pH was greater than the i n i t i a l pH owing to the d i l u t i o n of H + in the c r a n i a l space, the f i n a l H + concentrations measured were one to two orders of magnitude higher than those used to stimulate c e n t r a l chemoreceptors i n mammals (Shams 1985) and t u r t l e s ( H i t z i g and Jackson 1978). NaCN i s a metabolic i n h i b i t o r of the mi t o c h o n d r i a l e l e c t r o n t r a n s p o r t chain and i s known to s t i m u l a t e a l l o x y g e n - s e n s i t i v e chemoreceptors, i n c l u d i n g c a r o t i d body oxygen chemoreceptors i n mammals (Mulligan and L a h i r i 1981) and oxygen-sensitive chemoreceptors i n f i s h g i l l s (Burleson and Milsom 1990). NaCN at the concentrations used here has also been used to e l i c i t v e n t i l a t o r y responses i n gar (Smatresk et a l . 1986). After an experiment was completed, the f i s h was k i l l e d by 37 over-anesthetization with concentrated MS-222. Sudan black dye d i s s o l v e d i n 95% ethanol, which has been used previously as a neural s t a i n ( F i l i p s k i and Wilson 1984), was perfused through the c r a n i a l space for 3 0 min at the same pressure as during the experiment. After dye perfusion, fresh ethanol was perfused through the cranium to r i n s e out excess dye. The b r a i n was then removed and examined fo r the presence of dye. The l o c a t i o n of s t a i n was taken to i n d i c a t e where the mock perfusate had come in contact with CNS structures during the experiment. Spectral Analysis of the Intermittent Air-Breathing Pattern The temporal air-breathing pattern of undisturbed bowfin was more c l o s e l y examined in the frequency domain using spec-t r a l a n a l y s i s as outlined in a commercially a v a i l a b l e time-s e r i e s a n a l y s i s software package (ITSM; Brockwell and Davis 1991). One 8 h recording session of air-breathing events for each of 6 f i s h i n (1) a q u a t i c / a e r i a l normoxia, (2) aquatic h y p o x i a / a e r i a l normoxia, and (3) a q u a t i c n o r m o x i a / a e r i a l hypoxia (8% 0 2) were used i n the a n a l y s i s . Air-breathing i n Amia occurs intermittently as single events with inter-breath i n t e r v a l s of varying lengths. A t y p i c a l series of type I and type II a i r breaths, expressed as inter-breath i n t e r v a l s , as a function of time i s shown in figure 4 (top panel; see Results for descriptions of a i r breaths). Air-breaths were analyzed as a series of discrete events i n time (see French and Holden 1971); d i s c r e t e events (or spikes) i n a time series have also been c a l l e d delta functions 38 Figure 4. Top panel: Plot of inter-breath i n t e r v a l (IBI) as a function of cumulative time (min) for a single f i s h i n normox-i c conditions. Open symbols denote type II a i r breaths and closed symbols denote type I breaths (see Results for descrip-t i o n s of both a i r b r e a t h s ) . Bottom panel: The same time series as the top panel, except the breaths have been replaced by p o s i t i v e and negative delta functions representing type II and type I a i r breaths, r e s p e c t i v e l y (see text f o r d e t a i l s ) . The delta functions were plotted a f t e r 5-point smoothing. 3 9 40 ffi M 30 20 IO O O IOO 200 300 400 500 Cumulative Time (min) TYPE II TYPE I 1 r IOO 200 300 400 500 O Cumulative Time (min) 40 (DeBoer et a l . 1984) . The spectrum of t h i s type of signal i s a l s o known as the spectrum of counts (DeBoer et a l . 1984). Air-breaths were treated as delta functions and given negative v a l u e s i f they corresponded with type I a i r breaths, and p o s i t i v e values i f they were type II a i r breathing events. The s i z e of the d e l t a f u n c t i o n s i s a r b i t r a r y and does not a f f e c t the analysis. Each time series of discrete a i r breaths (delta functions) was smoothed using a 5-point weighted moving average (Brockwell and Davis 1991) . This removes high f r e -quency components and i s analgous to passing the data through a low-pass f i l t e r . The inter-breath i n t e r v a l series ( f i g . 4, top) expressed as the time s e r i e s of d e l t a f u n c t i o n s a f t e r smoothing i s shown i n fi g u r e 4 (bottom panel). Each 500 min data set was subdivided into 4 equal data sets of 125 min for a t o t a l of 24 data sets (4 time series each for 6 f i s h ) . This was done to eliminate low-frequency trends in the 500 min data sets and thus maintain s t a t i o n a r i t y in the data, an assumption of Fourier a n a l y s i s ( C h a t f i e l d 1980) . The frequency compo-nents of the smoothed d e l t a f unctions were then c a l c u l a t e d using a discrete Fourier analysis (Brockwell and Davis 1991). The ITSM software c a l c u l a t e s the t o t a l power of the periodo-gram and div i d e s by the inverse square root of the number of data p o i n t s i n the a n a l y s i s ; t h e r e f o r e , the ordinates were expressed as normalized power. Following the c a l c u l a t i o n of the periodogram, the data were then t r a n s f e r r e d to a spread sheet (Quattro Pro) and spectrum averaged; that i s , normalized power at each frequency i n t e r v a l was averaged for a l l the data s e t s and p l o t t e d as a f u n c t i o n of frequency and p e r i o d . 41 Frequencies i n the spectrum contributing s i g n i f i c a n t amounts of power i n each 125 min data set were analyzed with F i s h e r ' s Exact t e s t (Brockwell and Davis 1991). Significance deviating from the n u l l hypothesis of random noise was accepted at the 5% l e v e l . Data A n a l y s i s and S t a t i s t i c s Most v a r i a b l e s are summarized as mean + 95% confidence i n t e r v a l s (95% C.I.)/ unless otherwise stated. A v a r i e t y of s t a t i s t i c a l t e s t s were used to e s t a b l i s h s i g n i f i c a n c e where appropriate: paired and unpaired t - t e s t s , one-way and two-way a n a l y s i s of variance, followed by the Student-Newman-Keuls (SNK) multiple range test or Tukey's multiple comparison test, and l e a s t squares l i n e a r regression analyses (Zar 1974). The p r o b a b i l i t y of committing a Type I error (rejection of a true n u l l hypothesis) was accepted at the 5% l e v e l . A l l s t a t i s t i -c a l t e s t s were done with commercially a v a i l a b l e s t a t i s t i c a l software (Systat or Statgraphics). Departures from the above methods of s t a t i s t i c a l analyses are indicated i n the text. 42 RESULTS A i r - B r e a t h i n g Rates and Behavior for Undisturbed Amia I n t i t i a l observations from Amia indicated that two d i f -ferent breathing behaviors occurred under normoxic conditions with two d i f f e r e n t patterns of a i r flow that were q u a l i t a t i v e -l y i n v a r i a n t ; that i s , every f i s h examined showed two, and only these two, a i r - b r e a t h i n g patterns. Since previous work on Amia has i d e n t i f i e d only one type of air-breathing behavior (Johansen et a l . 1970; Deyst and Liem 1985), the breaths described here have been named type I and type II a i r breaths. The a i r flow patterns f o r these breath types are shown i n figure 5. Type I a i r breaths were characterized by an exhala-t i o n phase, r e s u l t i n g i n increased a i r flow at the pneumotach-ograph, followed by an i n h a l a t i o n phase r e s u l t i n g i n a de-creased a i r flow ( f i g . 5, l e f t column) . The inhalation phase of the type I breath was u s u a l l y o f f s e t from the o r i g i n a l b a s e l i n e , owing to the c h a r a c t e r i s t i c s of the measurement system (see Methods). The f i s h projecting i t s snout above the water's surface also contributed to the change i n baseline; nevertheless, decreased a i r flow below the o f f s e t baseline was obviously correlated with a depression of the gular plate when viewed on videotape and, therefore, the i n h a l a t i o n phase of the type I breathing cycle. The small p o s i t i v e change i n flow that occurred consistently during type I breaths approximately 200-400 ms before exhalation was due to the f i s h t r a n s f e r r i n g lung gas to the buccal cavity upon approaching the surface and r a i s i n g the water l e v e l in the funnel (T=transfer phase, f i g . 43 Figure 5. Recordings of a i r flow (ml s - ± ) from two f i s h i n normoxic conditions. Type I a i r breaths ( l e f t column) and the corresponding type II a i r breath (right column) are from the same f i s h . Type I a i r breaths are characterized by a transfer phase (T), followed by exhalation and inhalation. The expira-i tory time i n t e r v a l f o r type I breaths ( T E ) , analogous to T E f o r the c a l i b r a t e d volumes ( f i g . 3), was used as the c r i t e -r i o n for integrating the area under the expiratory flow curve f o r type I breaths (shaded area) for measurement of expired t i d a l volume. Type II a i r breaths show only an i n h a l a t i o n phase, corresponding with a negative flow p a t t e r n . Shaded areas f o r type I and type II i n h a l a t i o n s correspond with i n h a l e d volumes, but were not q u a n t i f i e d owing to gas l o s t during i n h a l a t i o n and t r a n s f e r (see text for d e t a i l s ) . The time scale bar equals 100 ms and applies to a l l traces. 44 T Y P E I tn CC > tr O CC X HI 80 ni 40-1 0 - J T Y P E II °] 10 J INHALE i INHALE i INHALE I o -20 3 100 ms 5). This p o s i t i v e flow pulse was not due to the animal pro-j e c t i n g i t s snout above the water's surface since t h i s pulse was not apparent during type II breaths when the f i s h a l s o projected i t s snout s l i g h t l y above the surface. Type II a i r breaths were characterized by a decreased a i r flow, i n d i c a t i n g a s i n g l e inhalation ( f i g . 5, r i g h t column). Both types of breaths were e a s i l y distinguished from each other by d i r e c t observation. In type I breathing, there was a cl e a r depression of the gular plate twice during the breathing cycle. In type II breaths, the gular plate was depressed only once; furthermore, the amount of gular depression was notice-ably l e s s i n type II than i n type I breathing. The sequence of events during both types of breathing are shown schemati-c a l l y i n figures 6 and 7. During the compression phase, there was u s u a l l y , but not always, a loss of i n s p i r e d gas from the opercular c a v i t i e s as the f i s h descended below the surface (D i n f i g . 6, B i n f i g . 7) . Although the amounts l o s t could not be q u a n t i f i e d , a greater amount of gas appeared to be l o s t during type I breathing, and in cases where there appeared to be l a r g e amounts of i n s p i r e d gas l o s t , r e g a r d l e s s of the aquatic or a e r i a l conditions, type II breaths, often within one minute, occurred. Thus, owing to the l o s s of i n s p i r e d gas, i n s p i r e d volume could not be q u a n t i f i e d . Inspired v o l -umes are given only as estimates since measurement of these volumes would overestimate, by an unknown amount, the actual in h a l e d volume; breath volumes were, t h e r e f o r e , q u a n t i f i e d only for the expiratory phase of the type I breaths. There was a considerable range of i n t r a - i n d i v i d u a l and 46 Figure 6. Diagram of the sequence of events involved i n a type I a i r breath. There are four d i s t i n c t phases of the type I breath: (1) t r a n s f e r of a i r bladder gas to buccal cavity, (2) e x h a l a t i o n of the t r a n s f e r r e d gas, (3) i n h a l a t i o n of atmospheric gas and (4) compression ( i . e . transfer of inhaled gas to a i r bladder). During the compression phase there i s usually a loss of gas from the opercular c a v i t i e s . 47 TYPE 1. Transfer AIR-BREATH 4. Compression Figure 7. Diagram of the sequence of events in v o l v e d i n a type II a i r breath. There are only two phases: (1) inhalation of atmospheric gas and (2) compression (transfer of inhaled gas to the a i r bladder). 49 TYPE II 1. Inhalation A I R AIR-BREATH 2. Compression i n t e r - i n d i v i d u a l values for expired volumes (V E) i n the d i f -f e r e n t aquatic and a e r i a l c o n d i t i o n s . In 6 of the 8 f i s h examined i n undisturbed conditions, V E ranged from 6.3 ml k g - 1 to 36.4 ml k g - 1 . The average maximum V E values f o r these 6 f i s h was 25.1 + 6.2 ml k g - 1 . The responses of undisturbed Amia to changes i n aquatic and a e r i a l gas concentrations are summarized i n Table I and presented as frequency histograms figures 8-13. The frequency d i s t r i b u t i o n s under a l l c o n d i t i o n s were p o s i t i v e l y skewed ( i . e . the median value was less than the mean), due to breaths that occurred at time i n t e r v a l s greater than 90-100 min (not shown). The d i s t r i b u t i o n s ( f i g s . 8-13) are presented as stacked bars, rather than overlapping, to d i f f e r e n t i a t e the two breath types. In normoxic conditions, Amia took approximately 3 a i r -breaths h - 1 : inter-breath i n t e r v a l (IBI) was 19.8 + 0.9 min (Mean + 95% C.I.; n=1950 observations from 8 f i s h , f i g . 8), and there were s l i g h t l y more type I (60%) than type II a i r breaths (40%). Carbon dioxide (5% i n a i r ; f i g . 9) d e l i v e r e d through the a e r i a l phase had no s i g n i f i c a n t e f f e c t on a i r -breathing r a t e s compared with f i s h i n normoxic c o n d i t i o n s (Student-Newman-Keu1s (SNK) m u l t i p l e range t e s t , c[3744 2 = 1 - 9 8 ) - When bowfin were exposed to aquatic or a e r i a l hypoxia (8% 0 2 ) , a i r - b r e a t h i n g increased s i g n i f i c a n t l y to about 5 breaths h - 1 compared with normoxia: IBI decreased to 12.3+0.8 min (n=579) i n a q u a t i c hypoxia ( f i g . 10; SNK, q 3 7 4 4 4=13.4, P<0.001) and 12.9+2.0 min (n=328) i n a e r i a l hypoxia ( f i g . 11; SNK, ^2744 3 = 9- 7/ P<0.001); however, there 51 Table I. Mean Inter-breath Intervals (IBI; min) f o r undis-turbed Amia i n various combinations of aquatic and a e r i a l c o n d i t i o n s . Values are Mean + 95% C.I. and the number of observations (n). Type I and type II breaths are given as the percentage of t o t a l breaths for each condition. 52 CONDITION IBI BREATHS Aquatic A e r i a l Type I Type II Hyperoxia A i r 57 , .8 + 15 .1 (30) 0 100 Normoxia 5%C02* 21. . 6 + 2 . 2 (185) 55 45 Normoxia A i r 19. .8 + 0. 9 (1950) 60 40 Normoxia ioo%o 2 15. . 3 + 1. 6 (350) 0.6 99. 4 Normoxia 8%02 12 . 9+2 . 0 (328) 94 6 Hypoxia A i r 12 . 3+0. 8 (579) 80 20 Hypoxia ioo%o 2 11. . 1+1. 0 (329) 0.3 99. 7 N=7 f i s h , N=8 in other groups 53 F i g u r e 8. Histogram of the d i s t r i b u t i o n of i n t e r - b r e a t h i n t e r v a l s (IBI; min) f o r eight f i s h i n aquatic and a e r i a l normoxia at 22 + 2 °C. Closed bars represent numbers of type I a i r breaths, open bars denote type II a i r breaths. The bars are stacked rather than overlapped. IBI greater than 90 min are not shown. The mean (X) and median (m) values f o r the d i s t r i b u t i o n are indicated. 54 55 Figure 9. Histogram of the frequency d i s t r i b u t i o n of IBI (min) for eight f i s h i n aquatic normoxia and a e r i a l hypercap-nia (5% C0 ? i n a i r ) . 56 m X • Type I • Type H •JL I O 2 0 3 0 4 0 5 0 6 0 7 0 1 Inter - breath Interval (min.) Figure 10. Histogram of the frequency d i s t r i b u t i o n of IBI for eight f i s h i n aquatic hypoxia and a e r i a l normoxia ( a i r ) . 58 O H M ft ft >> >> H H O LO s I i 59 Figure 11. Histogram of the frequency d i s t r i b u t i o n of IBI (min) f o r eight f i s h i n aquatic normoxia and a e r i a l hypoxia (8% 0 2 ) . 6 0 m X Inter - breath Interval (min.) was no s i g n i f i c a n t d i f f e r e n c e i n IBI between the two hypoxic treatments (SNK, 2 = 0 ' 7 ) * There was also a c l e a r change i n the breathing p a t t e r n when Amia were placed i n hypoxic conditions: type I air-breaths were used predominantly, ac-counting f o r about 80% of t o t a l breaths i n aquatic hypoxia (Table I) ; of the remaining 2 0% of a i r breaths (type II) that occurred i n aquatic hypoxia, the large majority of these (ca. 75%) occurred at int e r v a l s of less than 10 min ( f i g . 10) . In a e r i a l hypoxia, 94% of a l l breaths were type I (Table I ) . When Amia were given 100% oxygen to breathe from the a e r i a l phase, and aquatic conditions remained normoxic, IBI s i g n i f i c a n t l y declined to 15.3+1.6 min (n=350) compared with a i r f l o w i n g i n t h e b r e a t h i n g f u n n e l ( f i g . 12; SNK, q 3 7 4 4 2 = 6 - 5 ' P<0.001); furthermore, with the exception of 2, every one of 350 breaths was type II ( f i g . 12) . When aquatic conditions were made hypoxic, and 100% oxygen remained i n the a e r i a l phase, type II breaths were s t i l l used almost exclu-s i v e l y ( f i g . 13), and IBI declined even further to 11.1+1.0 (n=329) min (SNK, q 3 7 4 4 4=4.6, P<0.01) as compared with f i s h i n normoxic water and breathing 100% 0 2 from the a e r i a l phase (Table I ) . Aquatic hyperoxia resulted i n a s i g n i f i c a n t reduc-t i o n of air-breathing frequency, to about 1 breath h - 1 (Table I ) . Of the few breaths that occurred i n t h i s condition, a l l were type II a i r breaths. A i r Bladder Deflation and I n f l a t i o n A i r bladder d e f l a t i o n i n i t i a t e d air-breathing responses. The time taken to i n i t i a t e an air - b r e a t h following d e f l a t i o n 62 F i g u r e 12. H i s t o g r a m of t h e f r e q u e n c y d i s t r i b u t i o n o f IBI (min) f o r e i g h t f i s h i n a q u a t i c normoxia and a e r i a l hyperoxia (100% 0 2 ) . 63 m X 30n | | 25 20 15 IO 5 O 11 1 1 • Type I • Type H TrThrmnlKIn n n n r m 1 1 1 'r 1 O IO 20 3 0 4 0 5 0 6 0 70 8 0 Inter - breath Interval (min.) Figure 13. Histogram of the frequency d i s t r i b u t i o n of IBI (min) f o r eight f i s h i n aquatic hypoxia and a e r i a l hyperoxia (100% 0 2 ) . 65 > 0 ) m X 40 30 20 IO O o • Type I • Type H rrfU n rlT I I I I T " ' "I" " ' " 1 " ' I XL IO 20 30 40 n 50 60 Inter - breath. Interval (min.) was s i g n i f i c a n t l y dependent upon the volume removed from the a i r bladder ( f i g . 14). A h y p e r b o l i c f u n c t i o n (Y=f(l/X)) between time to i n i t i a t e an a i r breath and the volume removed was i n i t i a l l y assumed s i n c e asymptotic values of X and Y should p h y s i c a l l y e x i s t . An asymptote w i l l e x i s t at some volume (X) where volume removal from the a i r bladder could not be detected by the f i s h . An asymptote should also e x i s t as a l i m i t to the speed (Y) at which the f i s h could make an excur-sion to the water's surface for an air-breath. Thus, a l i n e a r r e g r e s s i o n model of time (s) to i n i t i a t e an a i r breath was . — i . plotted as a function of (volume) removed to obtain a li n e a r equation f o r the r e l a t i o n s h i p between these v a r i a b l e s . The e q u a t i o n r e l a t i n g t h e s e v a r i a b l e s was : Time = 468.3*(l/vol.)+46.6 ; F 1^ 6 4=4.66, r 2=0.07, P<0.05. The curve i n figure 14 was calculated by converting the equation to i t s c u r v i l i n e a r form. The threshold (T, f i g . 14) was approximate-l y 3 ml k g - 1 ; d e f l a t i o n s below t h i s value produced no r e -sponses within 10 min following d e f l a t i o n . The fastest time for an a i r breath response to d e f l a t i o n was 5 s. A l l 66 a i r -b r e a t h s observed f o l l o w i n g lung d e f l a t i o n were type II breaths, regardless of the aquatic P0 2 ( f i g . 14). Most of the a i r breaths (82%) were i n i t i a t e d within 2 min a f t e r a i r blad-der d e f l a t i o n . There was a l s o a s i g n i f i c a n t r e l a t i o n s h i p between the number of breaths taken during the 10 min observation period, and the volume removed from the a i r bladder ( f i g . 15): No. Breaths = 0.089*Vol. + 0.81; F± ? 1=33.0; r 2=0.32; P<0.001. During the 10 min observation p e r i o d f o l l o w i n g d e f l a t i o n , 67 Figure 14. Relationship between the time taken to i n i t i a t e an a i r breath (seconds) as a function of volume of a i r removed (ml kg - 1) from the a i r bladder (N=4, n = 6 6 ) in aquatic normoxia ( • ) , aquatic hypoxia ( A ) and aquatic hyperoxia (O)* A H breaths observed following a i r bladder d e f l a t i o n were type II a i r breaths. The li n e a r equation r e l a t i n g these variables i s : Time = 468.3*(1/Vol.) + 4 6 . 6 ; r 2=0.07; P<0.05. The regression l i n e i n the f i g u r e was c a l c u l a t e d from t h i s equation and converted to a c u r v i l i n e a r form. -The threshold (T) for detec-t i o n of volume removed i s approximately 3 ml k g - 1 . 6 8 Air Bladder Inflation 300H I •H H 200 IOO O • • O ° o o A a n ftp O T T ACt] _ •o tr -QD-O T T IO 20 3 0 4 0 • N o r m o x i a A H y p o x i a O H y p e r o x i a • 50 2 m i n 1 m i n 60 Volume Removed (ml/kg) Figure 15. Relationship between the number of a i r - b r e a t h i n g events i n 10 min a f t e r a i r bladder d e f l a t i o n as a function of the volume removed (ml kg ). The l i n e a r equation r e l a t i n g these v a r i a b l e s i s : Breaths = 0.089*Vol. + 0.81; r 2=0.32; P<0.001. 70 there were occasionally type I breaths; however, they occurred only when the f i s h was i n hypoxic water and were u s u a l l y the t h i r d or f o u r t h breath i n the s e r i e s . The i n i t i a l breath following d e f l a t i o n was always a type I I , regardless of the aquatic oxygenation ( f i g . 14) . A i r bladder i n f l a t i o n s produced a v a r i e t y of responses from the f i s h . Small i n f l a t i o n s (3 ml kg - 1) that were d e l i v -ered as the f i s h approached the surface had no d i s c e r n a b l e e f f e c t ; that i s , ai r - b r e a t h i n g proceeded normally, with both type I and type II breaths evident. Larger i n f l a t i o n s some-times caused the f i s h to stop i t s ascent to the surface. A i r -breathing sometimes proceeded despite large i n f l a t i o n s , but the breaths were nearly always of type I when t h i s occurred. Maximum A i r Bladder Volume Maximum a i r bladder volume (ml), measured in v i t r o , i n c r e a s e d l i n e a r l y w i t h i n c r e a s i n g body mass (g) (Vol.=0.08*Mass+12.2; r 2=0.86, P<0.001; f i g . 16) over the range used i n t h i s study. The slope (0.08) indicated that a i r bladder volume increased at a rate of 8% per 100 g of body mass over t h i s range. The intercept (12.2 ml) was not s i g n i f -i c a n t l y d i f f e r e n t from zero (T5=1.05; P>0.05). B r a n c h i a l Nerve Denervation Denervation of b r a n c h i a l branches of c r a n i a l nerves IX and X were done to tes t the hypothesis that type I air-breaths are caused by a f f e r e n t stimulation of chemoreceptors located on the g i l l s innervated by c r a n i a l nerves IX and X. A l l three 72 Figure 16. Relationship between maximum a i r bladder volume (ml) and body mass (g) determined in v i t r o . Values represent the mean of three determinations for each a i r bladder. The l i n e a r equation r e l a t i n g these variables i s : Vol = 0.08*Mass + 12.2; r 2=0.86; P<0.001. 73 74 groups of s u r g i c a l l y t r e a t e d f i s h e x h i b i t e d both types of breaths i n aquatic normoxia and increased a i r breathing f r e -quency i n aquatic hypoxia ( f i g s . 17-19). Since air-breathing frequency f o r the two normoxic t r i a l s were not s i g n i f i c a n t l y d i f f e r e n t within a group, only the f i r s t normoxic t r i a l was presented i n f i g u r e s 17-19. When both breath types were con-sidered together, there was a s i g n i f i c a n t decrease i n IBI for t o t a l denervate (TD) f i s h compared with sham-operated (SH) or p a r t i a l denervate (PD) f i s h in normoxic conditions (Table I I ) . W i t h i n group comparisons r e v e a l e d t h a t IBI s i g n i f i c a n t l y decreased i n aquatic hypoxia compared with e i t h e r normoxic treatment. The increased air-breathing rate for the TD group was due e n t i r e l y to large numbers of type II breaths that occurred at IBI of 1 min or less i n aquatic normoxia ( f i g . 19A) and aquat-i c hypoxia ( f i g . 19B), thus s i g n i f i c a n t l y reducing the mean IBI (Table II) . D i r e c t observations from TD f i s h indicated q u a l i t a t i v e l y that larger amounts of inhaled gas were l o s t during the transfer phase of the air-breathing event i n these f i s h compared with SH and PD groups. Since the hypothesis of t h i s experiment was to test whether type I breaths were stimu-la t e d by chemoreceptors located on the g i l l arches, the re-s u l t s were re-analyzed after removing the type II breaths from the data set. A f t e r removing type II breaths, there were no s i g n i f i c a n t d i f f e r e n c e s i n the frequency of type I breaths between groups i n normoxia and hypoxia (Table III) . One exception was a s i g n i f i c a n t l y lower IBI in SH during the f i r s t normoxic treatment compared with the other groups (Table I I I ) . 75 Figure 17. Histogram of the frequency d i s t r i b u t i o n of IBI (min) of type I (closed bars) and type II (open bars) a i r breaths f o r sham-operated (SH) f i s h ( N = 4 ) . A. Aquatic and a e r i a l normoxia; B. Aquatic hypoxia/aerial normoxia. The mean (X) and median (m) values are indicated. 76 20i 16 12 8 m X I I 4 O • Type I • Type H • l A J L . HTI rBi. XL O IO 20 30 40 50 Inter-Breath Interval (min) 60 • Type I • Type H 20 30 40 50 GO Inter-Breath Interval (min) 77 Figure 18. Histogram of the frequency d i s t r i b u t i o n of IBI (min) of p a r t i a l b r a n c h i a l denervate (PD) f i s h (N=5). A. Aquatic and a e r i a l normoxia; B. Aquatic hypoxia/aerial normox-i a . 78 • Type I • Type H ,nrl,n n f l , ri, W-30 40 50 60 Inter-Breath Interval (min) 79 Figure 19. Histogram of the frequency d i s t r i b u t i o n of IBI (min) f o r t o t a l b r a n c h i a l denervate (TD) f i s h (N=7). A. Aquatic and a e r i a l normoxia; B. Aquatic hypoxia/aerial normox-i a . Note the d i s t i n c t mode of type II breaths at 1 min com-pared with SH or PD groups in aquatic normoxia ( f i g s . 16, 17). 80 A > fl> 0 • Type I • Type H 20 30 40 50 60 Inter-Breath Interval (min) 81 Table I I . Mean Inter-breath Intervals (min) + 95% C.I. and the number of o b s e r v a t i o n s (n) f o r b r a n c h i a l d e n e r v a t i o n groups. Groups: sham-operated (SH), partial-denervates (PD) and total-denervates (TD). Values are means for combined type I and type II a i r breaths. NI and N2 denote f i r s t and second aquatic normoxia treatments separated by an aquatic hypoxia (H) treatment. 82 NI H N2 SH 1 6 . 4 ± 5 . 1 (103) 6 . 3 ± 0 . 4 a (306) 17.3+3.0 (105) PD 2 0 . 5 ± 5 . 5 (132) 8 . 2 + 0 . 8 a (292) 16.7+3.8 (124) TD 7 . 5 + 1 . 0 b (451) 5 . 1 + 0 . 4 a ' c ( 6 8 0 ) 9 . 0 + 1 . 2 b (339) a S i g n i f i c a n t l y d i f f e r e n t NI and N2 w i t h i n same g r o u p . b S i g n i f i c a n t l y d i f f e r e n t from SH and PD f o r same t r e a t m e n t . c S i g n i f i c a n t l y d i f f e r e n t from PD-H f o r same t r e a t m e n t . 83 Table I I I . Mean + 95% C.I. of Inter-breath In t e r v a l s (min) f o r b r a n c h i a l denervation groups. Values are f o r type I a i r breaths only. Same abbreviations as Table II. 84 NI H N2 SH 23.7+4.0 (57) 8.6+0.73 (224) 22.6+5.9 (59) PD 35.5+12.5b (58) 12.9+0.9a (178) 25.2±9.1 (70) TD 31.3+5.5 (92) 12.6+0.8a (262) 23.7+4.3 (117) a S i g n i f i c a n t l y d i f f e r e n t from NI and N2 within same group, k S i g n i f i c a n t l y d i f f e r e n t from SH-N1. 85 A l l t h r e e groups i n c r e a s e d the frequency of type I a i r breaths to about 5 breaths h - 1 when exposed to aquatic hypoxia compared with 2-3 breaths h - 1 i n normoxic conditions (Table III) . Mean g i l l v e n t i l a t i o n r a t e s ranged between 4.5 to 10.5 breaths m i n - 1 and were s i g n i f i c a n t l y lower i n the TD group i n normoxic and hypoxic conditions (Table IV). Within groups, fg did not change s i g n i f i c a n t l y i n aquatic hypoxia (Table IV). I n t r a - C r a n i a l P e r f u s i o n G i l l v e n t i l a t i o n rates increased s i g n i f i c a n t l y from about 10 cycles min i n aquatic normoxia to about 17 cycles mm i n aquatic hypoxia before the s t a r t of perfusion (Table V). A i r - b r e a t h i n g r a t e s i n aquatic hypoxia were c a l c u l a t e d by d i v i d i n g the t o t a l number of a i r breaths during a perfusion treatment by the t o t a l minutes recorded f o r the treatment; — i these values were converted to breaths h . Air-breathing was not a f f e c t e d by any perfusate treatment (ANOVA, F 3 2 6 = 0 , 1 5 ) and the r a t e of a i r - b r e a t h i n g d u r i n g a q u a t i c hypoxia was 3.0+0.6 breaths h - 1 (n=8), and was s i g n i f i c a n t l y d i f f e r e n t from zero (T7=5.0, P<0.001; Table VI). The v e n t i l a t o r y responses to i n t r a c r a n i a l perfusion with NaCN were v a r i a b l e . NaCN at a concentration of 500 ug m l - 1 increased v e n t i l a t i o n i n one f i s h , 1000 ug m l - 1 increased v e n t i l a t i o n in another f i s h , but there was no e f f e c t at lower doses; however, at concentrations ranging from 5 to 500 ug ml" 1 , NaCN depressed v e n t i l a t i o n i n three other f i s h . V e n t i l a -t i o n was unaffected i n f i v e f i s h perfused i n t r a c r a n i a l l y for 86 Table IV. Mean + 95% C.I. values for g i l l v e n t i l a t i o n rate (breaths min - 1) f o r branchial denervation groups. Abbrevia-tions same as Tables II and I I I . 87 NI H N2 SH 9.6+1.8 (14) 10.5+2.0 (18) 7.0+1.5 (8) PD 9.5+2.5 (10) 8.9+1.6 (16) 6.6+0.8 (21) TD 6.1±2.4 a(14) 5.7±0.7 b (40) 4.5±0.7 (15) a S i g n i f i c a n t l y d i f f e r e n t from SH-N1 for same treatment. b S i g n i f i c a n t l y d i f f e r e n t from SH and PD for same treatment. 88 Table V. Mean + 95% C.I. f o r g i l l v e n t i l a t i o n r a t e ( f g ; breaths m i n - 1 ) , buccal pressure amplitude (Pb; mmHg) and a i r -breathing (AB; breaths h _ 1 ) during i n t r a - c r a n i a l p e r f u s i o n with mock EDF e q u i l i b r a t e d with four d i f f e r e n t gas mixtures under aquatic normoxic or hypoxic c o n d i t i o n s . Values are pr o v i d e d f o r p r e - p e r f u s i o n and 15 min a f t e r the s t a r t of perfusion. 89 Aquatic Normoxia 15 min Perfusate Variable Pre-perfusion perfusion AB Normoxia fg 9.8+3.0 9.8+2.3 (N=8) Pb 0.44±0.2 0.51±0.2 0 Hypoxia fg 11.8+2.5 11.1+2.1 (N=8) Pb 0.51±0.2 0.51+0.2 0 Hypercapnia fg 11.2+2.1 11.6+2.5 (N=8) Pb 0.44+0.2 0.51±0.2 0 Hyperoxia fg 10.3+2.8 9.7±2.5 (N=7) Pb 0.44+0.2 0.44+0.2 0 Aquatic Hypoxia 15 min Perfusate Variable Pre-perfusion perfusion Normoxia fg 16.4+3.5 16.1+4.1 (N=7) Pb 0.74±0.3 0.59+0.3 Hypoxia fg 17.6±4.1 16.7+3.7 (N=7) Pb 0.74+0.3 0.74+0.3 Hypercapnia fg 18.1±3.9 18.5+3.7 (N=8) Pb 0.81+0.3 0.74+0.3 Hyperoxia fg 18.8+3.7 18.3+3.9 (N=8) Pb 0.81±0.3 0.81+0.3 AB 3.2+2.1 3.2+2.1 2.4+2.1 3.1+2.3 Three f i s h perfused with 3% C0 2 perfusate and f i v e f i s h perfused with 5% C0 2 perfusate. 90 Table VI. Summary of dorsal aorta P0 2, pH and PC0 2 and a i r -breathing r a t e s (AB) during aquatic normoxia and hypoxia. Values are mean ± 95% C.I. and number of animals used i n each treatment (N). 91 Condition Aquatic Normoxia PO 2 (mmHg) 6 8 . 6 ± 2 7 . 6 (8) PH 7 . 74 + 0.05 (8) PCO 2 (mmHg) 3.0+0.5 (8) AB ( h " 1 ) 0 (8) Aquatic Hypoxia 19.3+4.4 (8) 7.81+0.07 (6) 1.4+0.2 (8) 0 + 1.4 (8) S i g n i f . (P) <0.005 <0. 02 <0.001 <0.001 92 30 min with HCl solutions ranging i n pH from 3.8 to 6.8. Sudan Black dye stained extensive areas of the brain and associated structures. Dye was found occasionally within the t h i r d v e n t r i c l e and usually i n the fourth v e n t r i c l e . Dye was always present on structures surrounding the brain, including the v e n t r a l surface of the medulla, cerebellum, o p t i c tectum and telencephalon. S p e c t r a l A n a l y s i s of the I n t e r m i t t e n t A i r - B r e a t h i n g P a t t e r n The temporal air-breathing pattern i n Amia was intermit-tent, and appeared i r r e g u l a r . The c o n t r a s t i n g patterns of a i r - b r e a t h i n g i n normoxia and aquatic hypoxia are shown i n f i g u r e 20. In normoxia, there was usually an a l t e r n a t i o n of type I and type II a i r breaths ( f i g . 20A), while i n hypoxia, a i r - b r e a t h i n g frequency was higher, and type I breaths were used predominantly ( f i g . 20B). Data from these f i s h , and from 4 others, were used i n the s p e c t r a l a n a l y s i s . Plots of i n t e r - b r e a t h i n t e r v a l over time f o r f i s h i n a e r i a l hypoxia were very s i m i l a r to those i n aquatic hypoxia. The averaged periodogram f o r the 6 f i s h i n normoxia revealed a s i g n i f i c a n t low frequency peak, corresponding to a period of about 25 min ( f i g . 21A). Two other peaks, with less power, occurred at about 12 min and 7 min i n normoxia. The dominant peak was halfway between the mean inter-breath i n t e r -vals for type I breaths (29.8+2.5 min) and type II a i r breaths (20.5+2.1 min) for these 6 f i s h . Before spectrum averaging, however, the dominant peak for the s i x f i s h was centered at 29.6 min i n normoxia ( f i g . 21B), indicating that the peri o d i c -93 Figure 20. Inter-breath I n t e r v a l (IBI; min) p l o t t e d as a function of cumulative time (min) for two f i s h , one i n aquatic and a e r i a l normoxia (A) , and the other i n aquatic hypoxia (P wO 2=50 Torr) and a e r i a l normoxia (B). Open symbols denote type II a i r breaths; closed symbols denote type I a i r breaths. 9 4 Figure 2 1 . A. Spectrum-averaged periodogram for 6 f i s h (n=24 data sets) in aquatic normoxia ( s o l i d line) and aquatic hypox-l a (dashed l i n e ) . Normalized power (mm) i s p l o t t e d as a function of frequency (cycles/min) and as an equivalent time p e r i o d (min). A s i g n i f i c a n t peak occurred i n normoxia at about 25 min; two smaller peaks are evident at 12 and 7 min. In aquatic hypoxia, the s i g n i f i c a n t peak has s h i f t e d to about 1 0 min, with a smaller peak at about 6 min. B. Periodogram f o r the same 6 f i s h i n 21A before spectrum averaging. The dominant peak i s centered at 29.6 min, i n d i c a t i n g t h i s peak contributes most of the power to the l a r g e s t peak i n A (see t e x t ) . 96 B Period (min) F r e q u e n c y ( c y c l e s / m i n ) 9 7 i t y i n figure 21 was associated with type I, rather than type II, breaths. Spectrum averaging smooths the periodogram even f u r t h e r and thus has the e f f e c t reducing and spreading out power among c l o s e l y related frequencies. At the low frequency end of the s c a l e , r e s o l u t i o n becomes f u r t h e r reduced, thus d i f f e r e n t i a t i o n between the 30 min and 20 min peaks was not possible a f t e r spectrum averaging. In aquatic hypoxia, the 30 min low frequency peak disap-peared and a s i g n i f i c a n t , higher frequency peak at 10 min was revealed ( f i g . 21A). A second peak, occurring at approximate-l y 7 min was al s o evident i n aquatic hypoxia. This 10 min peak was also c l o s e l y associated with the mean inter-breath i n t e r v a l f o r type I a i r breaths. The r e s u l t s from f i s h i n a e r i a l hypoxia were s i m i l a r to those i n aquatic hypoxia, but the f i s h f e l l into two c h a r a c t e r i s t i c groups ( f i g . 22). Half of the f i s h showed a s i g n i f i c a n t high frequency peak at ap-proximately 6 min, and a second broader peak with a period of about 9 min ( s o l i d l i n e , f i g . 22) . There was also a 9-10 min peak i n the 3 remaining f i s h ; at higher frequencies, however, several peaks with lower amounts of power were evident (dotted l i n e , f i g . 22). Overall, bowfin i n normoxic or hypoxic conditions exhib-i t e d some p e r i o d i c i t y around 10 min; i n normoxia there was a s i g n i f i c a n t low frequency peak occurring at about 30 min. In aquatic or a e r i a l hypoxia, dominant periods occurred i n the range of periods between 5 and 10 min. The dominant periods in normoxic or hypoxic conditions were most cl o s e l y associated with the mean int e r v a l s between type I a i r breaths. 98 Figure 22. Spectrum-averaged periodogram f o r 6 f i s h (n=24 data sets) i n aquatic normoxia and a e r i a l hypoxia (8% 0 2 ) . The f i s h were d i v i d e d i n t o two groups a c c o r d i n g to t h e i r periodograms. The s o l i d l i n e represents the average of 3 f i s h with a s i g n i f i c a n t peak centered at 6 min. A smaller peak occurs at about 9 min. Three other f i s h had a s i g n i f i c a n t peak around 10 min, and a s l i g h t l y smaller peak at 7 min. Note there was a common period of about 9-10 min f o r a l l 6 f i s h . 99 Period (min) 20 IO 5 2 . , , r - 1 o 0.1 0.2 0.3 0.4 0 .5 Frequency (cycles/min) DISCUSSION Although the n a t u r a l h i s t o r y of Amia i s not we l l known, the few observations made to date i n d i c a t e that i t does use a e r i a l r e s p i r a t i o n under natural conditions (Reighard 1903; Doan 1938). Air-breathing has been observed during breeding, and occurs at the same water temperatures used i n t h i s study (Reighard 1903). Thus, the p h y s i o l o g i c a l r e s u l t s presented here may have d i r e c t relevance to i t s n a t u r a l behavior and ecology, but these issues are not d i r e c t l y assessed i n t h i s study. A i r - B r e a t h i n g P a t t e r n s and the Responses to Changes i n A q u a t i c and A e r i a l Gas Composition Since the seminal study e s t a b l i s h i n g that Amia exhales and inhales atmospheric gas (Wilder 1877), subsequent studies have confirmed the double pulse breathing mechanism of a e r i a l v e n t i l a t i o n i n Amia (Johansen et a l . 1970; Randall et a l . 1981; Deyst and Liem 1985; Liem 1988, 1989): exhalation f o l -lowed by inhalation, which i s the breathing pattern described here as type I. This study used the technique of pneumota-chograph^ to d i r e c t l y measure a i r flow generated by the f i s h during a i r breathing events. The a p p l i c a t i o n of t h i s tech-nique has independently established the exhalation/inhalation breathing sequence i n Amia, in addition to finding a previous-l y undescribed b r e a t h i n g p a t t e r n . The type II b r e a t h i n g pattern has not been described i n Amia p r i o r to t h i s study, but makes up a large p r o p o r t i o n (ca. 40%) of the t o t a l a i r 101 breaths under normoxic conditions (Table I, f i g . 8). The most comprehensive study of the mechanism of a e r i a l v e n t i l a t i o n i n Amia, using X-ray cine f i l m , pressure record-ings and electromyographic (EMG) a n a l y s i s , was conducted by Deyst and Liem (1985). In a d d i t i o n to confirming previous studies showing that exhalation occurred before i n h a l a t i o n , X-ray cinematography revealed that a large residual volume i n i t s a i r bladder remained f o l l o w i n g e x h a l a t i o n . I t i s not cl e a r why t h e i r study f a i l e d to observe type II breaths. One reason may be that the experiments described by Deyst and Liem (1985) were done with f i s h i n hypoxic water, a condition that resulted i n a predominance of type I breaths i n t h i s study. The breathing pattern of Amia (Type I in t h i s study) has been described as a four-phase process (Liem 1989): (1) Trans-f e r phase; depression of the buccal f l o o r which creates a negative pressure in the buccal cavity, thus t r a n s f e r r i n g gas from the a i r bladder, (2) Expulsion phase; a c t i v e exhalation of buccal gas by e l e v a t i o n of the buccal f l o o r , (3) Intake phase; a second depression of the buccal f l o o r draws a i r from the atmosphere i n t o the buccal c a v i t y , and (4) Compression phase; t r a n s f e r of inhaled gas from the buccal c a v i t y to the a i r bladder by a second e l e v a t i o n of the buccal f l o o r . In t h i s study, phases 1, 2 and 3 were detectable by pneumotachog-raph^. The i n i t i a l t r a n s f e r phase (phase 1) appeared as a b r i e f p o s i t i v e flow (T i n f i g . 5), owing to the l o c a l change i n water l e v e l created as the f i s h expanded i t s buccal cavity when approaching the surface during gas t r a n s f e r . This flow was u s e f u l i n determining the s t a r t of the t r a n s f e r phase 102 during the type I breathing c y c l e . Exhalation (phase 2), as expected, produced a p o s i t i v e change in a i r flow at the pneu-motach as the f i s h a c t i v e l y e x p e lled buccal gas ( f i g . 5). I n h a l a t i o n (phase 3), produced a negative flow immediately a f t e r exhalation, but the baseline was raised as a function of the recording system (see Methods), and probably also due to the animal protruding i t s snout above the water surface. The compression phase (phase 4), although not apparent from the pneumotach recordings, was c l e a r l y v i s i b l e on videotape, and was u s u a l l y c h a r a c t e r i z e d by gas bubbles escaping from-the opercular c a v i t i e s as the f i s h descended below the surface. Deyst and Liem (1985) also reported inhaled gas loss during the compression phase i n Amia. Thus, observations from the video recordings and pneumotachographic analyses of type I a i r breaths are e n t i r e l y consistent with the mechanisms described by Deyst and Liem (1985). The a i r breath sequence i n c l u d i n g transfer, exhalation and inhalation for the f i s h in t h i s study occurred i n about 0.5 s, which i s s i m i l a r to reported values for Amia (Deyst and Liem 1985) , e l e c t r i c eel (Farber and Rahn 1970) and gar (Rahn et a l . 1971). Measurements of e x p i r e d t i d a l volume and a i r bladder volume were si m i l a r to measurements obtained for Lepisosteus. Maximal expired t i d a l volume f o r undisturbed Amia averaged 25.1 ml k g - 1 , which i s s i m i l a r to values of 24.9 and 31.7 ml k g - 1 measured for gar (Rahn et a l . 1971; Smatresk and Cameron 1982b). The expired t i d a l volumes for gar were obtained by d i r e c t c o l l e c t i o n of a i r bladder gas, therefore, the i n d i r e c t measurement technique used in t h i s study y i e l d s comparable re-103 s u i t s . Wilder (1877) measured expired t i d a l volumes by d i r e c t c o l l e c t i o n ranging from 18 to 105 ml k g - 1 , with a mean of 44 ml k g - 1 , f o r a s i n g l e Amia. Although the absolute volumes were lower i n t h i s study, compared with the s i n g l e f i s h used i n Wilder's o r i g i n a l observations, comparisons i n d i c a t e that expired t i d a l volume v a r i e s considerably w i t h i n i n d i v i d u a l animals. A i r bladder volumes f o r Amia ( f i g . 16) and Lepi-sosteus (Rahn et a l . 1971) are approximately 8% of body mass, which i s well within the range reported f o r freshwater f i s h (Jones 1957), and in the expected t h e o r e t i c a l range for fresh-water f i s h (Alexander 1966) . When expressed as a f r a c t i o n of a i r bladder volume, expired t i d a l volume for Amia i s about 31% of t o t a l a i r bladder volume, and about 40% in gar (Rahn et a l . 1971). These r e s u l t s support the d i r e c t observations of Deyst and Liem (1985) indicating that a substantial residual volume remains following expiration. In t h i s study, type II breaths were i d e n t i f i e d by nega-t i v e a i r flow at the pneumotachograph ( f i g . 5), with no i n i -t i a l transfer phase that was evident i n type I breaths, and no p o s i t i v e a i r flow preceding the c h a r a c t e r i s t i c negative flow pattern. Observations from the video recordings showed that the buccal f l o o r was depressed only once when the f i s h reached the surface (shown s c h e m a t i c a l l y i n f i g . 7) . There was no evidence of exhalation since a i r was never seen to escape as the f i s h approached the surface. During the compression phase of type II breaths, however, there were a l s o o c c a s i o n a l l y small losses of inhaled gas through the opercular c a v i t i e s , as in the type I breaths, but t h i s did not always occur. Since 104 type II breaths involved a b r i e f i n h a l a t i o n at the surface, the time sequence was faster (ca. 0.1 s) than type I breaths. I t i s not c l e a r why previous s t u d i e s have f a i l e d to observe single inhalation breaths (type II) in Amia. A l i k e l y reason i s that pneumotachography has not been previously used to examine a e r i a l v e n t i l a t i o n i n t h i s s p e c i e s . I n h a l a t i o n without e x h a l a t i o n would be d i f f i c u l t to detect without a d i r e c t measure of a i r flow. A recent re-examination of the air-breathing mechanism in Amia (Liem 1988, 1989) suggested that two types of breaths were used, but one type of b r e a t h i n g i n v o l v e d a p a s s i v e , rather than a c t i v e , t r a n s f e r of gas from the swim bladder to the buccal cavity. The analyses are based on EMG data showing that the sternohyoideus muscle, used p r i m a r i l y f o r lowering the buccal f l o o r during active t r a n s f e r , did not f i r e during "passive transfer'; therefore, Liem (1989) concluded that a i r bladder gas was t r a n s f e r r e d to the buccal c a v i t y passively, f a c i l i t a t e d by the hydrostatic pressure gradient. His obser-vations suggest, however, that exhalation was always a c t i v e . The r e s u l t s from t h i s study do not support a passive transfer hypothesis. There was no evidence from t h i s study to suggest t h a t "passive t r a n s f e r ' occurred. In t h i s study, t r a n s f e r (phase 1) and exhalation (phase 2) were e i t h e r a c t i v e (type I ) , i n which the buccal f l o o r was r a i s e d and lowered twice during a cy c l e and i s consistent with Liems's a n a l y s i s , or not present (type I I ) . There was no i n d i c a t i o n of a t h i r d breath type that could be interpreted as using "passive trans-f e r ' coupled with active exhalation. Upon closer examination 105 of Liem's (1989) re s u l t s , the EMG pattern and buccal pressure recordings corresponding with "passive transfer' could be re-i n t e r p r e t e d as a type II breathing pattern. The a c t i v i t y of the sternohyoidues muscle once only i n h i s d e s c r i p t i o n of "passive transfer' i s consistent with the observations i n t h i s study i n d i c a t i n g t h a t the buccal f l o o r i s depressed once during the type II breathing pattern. Since Liem (1989) did not measure expiratory or inspiratory a i r flow, i t would have been d i f f i c u l t to d i s t i n g u i s h , on the basis of EMG measure-ments, whether gas transfer had indeed occurred. The evidence from t h i s study, therefore, indicates that two separate mecha-nisms f o r generating a i r flow occur i n Amia. One mechanism, which has been previously described in several other studies, i n v o l v e s a c t i v e e x h a l a t i o n of a i r bladder gas, and a c t i v e i n h a l a t i o n of atmospheric a i r ; the use of the buccal force pump during these events i s evident. The second mechanism, not previously described in Amia, involves a single aspiratory inhalation by the action of the buccal cavity, with subsequent t r a n s f e r of i n h a l e d gas to the a i r bladder. There i s no evidence of exhalation during t h i s type of breathing sequence. Although the precise neuromuscular motor patterns responsible f o r generating these two a i r - b r e a t h i n g events are not known, type II breaths may be the l a t t e r h a l f of the type I a i r breath cycle. The responses of Amia to a l t e r a t i o n s i n aquatic or a e r i a l gas concentrations showed obvious changes in both the frequen-cy and pattern of air-breathing. In l i g h t of the finding that Amia use two breathing mechanisms, c l o s e r a t t e n t i o n must be 106 paid to the pattern of breathing i n order to more accurately interpret the r e s u l t s . Breathing frequency increased s i g n i f i -c a n t l y i n aquatic or a e r i a l hypoxia, a response t y p i c a l of Amia (Johansen et a l . 1970; Randall et a l . 1981; McKenzie 1990) and many other species of a i r - b r e a t h i n g f i s h e s (see Shelton et a l . 1986). In aquatic or a e r i a l hypoxia, Amia took about 5 breaths h - 1 , which i s considerably less than rates of 15-20 breaths h - 1 observed by Johansen et a l . (1970). There i s no apparent reason f o r t h i s large discrepancy, but the breathing frequencies i n t h i s study are closer to those found i n other studies (Horn and Riggs 1973; Randall et a l . 1981; McKenzie 1990). In addition to increased breathing frequency i n hypoxia, there was also a clear s h i f t to a predominance of type I air-breaths ( f i g s . 10,11) compared with f i s h i n normox-i a , suggesting that type I breaths are stimulated by 0 2 - s e n s i -t i v e chemoreceptors. There i s al s o an i n d i c a t i o n that 0 2~ s e n s i t i v e chemoreceptors monitor i n t r a v a s c u l a r 0 2 t e n s i o n . Amia exposed to a e r i a l hypoxia, with aquatic P0 2 held constant above 140 Torr, increased the o v e r a l l frequency of breathing and used predominantly type I a i r breaths. This i n d i c a t e s t h a t i n t e r n a l chemoreceptor s i t e s , at l e a s t , mediate the hypoxic v e n t i l a t o r y r e f l e x e s . The r e s u l t s with Amia exposed to either aquatic or a e r i a l hypoxia are, therefore, consistent w i t h s t u d i e s on gar (Smatresk et a l . 1986) and l u n g f i s h (Johansen and Lenfant 1968) showing that i n t e r n a l hypoxia stimulates a i r breathing. A e r i a l hyperoxia caused dramatic changes i n the a i r -breathing pattern. Amia exposed to 100% 0 P i n the gas phase 107 maintained, or increased, air-breathing rates (Table I, f i g s . 12,13), i n contrast with several other species of air-breath-ing f i s h where a e r i a l hyperoxia reduced air-breathing frequen-cy (Garey and Rahn 1970; Lomholt and Johansen 1974; Burggren 1979; P e t t i t and Beitinger 1985; Smatresk et a l . 1986). The high rate of type II breathing, e s p e c i a l l y i n aquatic hypoxia, suggests that type I breaths are in h i b i t e d by elevated i n t e r -nal PC>2, since type I breaths normally occur i n aquatic nor-moxia and are predominant i n aquatic hypoxia ( f i g 10). Gee and Graham (1978) reported that in the air-breathing c a t f i s h , Brochis splendens, a i r - b r e a t h i n g frequency was maintained i n a e r i a l hyperoxia at ra t e s s i m i l a r to when exposed to a i r . Their study indicated that buoyancy regulation, rather than blood PC>2 per se, was responsible for the maintenance of a i r -breathing frequency. This occurred since the increased 0 2 d i f f u s i o n gradient i n a e r i a l hyperoxia, compared with a i r i n the air-breathing organ (ABO), resulted i n a faster decline in volume. Thus, a c l e a r hydrostatic function, i n ad d i t i o n to the r e s p i r a t o r y f u n c t i o n , was i n d i c a t e d f o r the ABO i n B. splendens. This mechanism i s probably r e s p o n s i b l e f o r the maintained frequency of type II air - b r e a t h i n g during a e r i a l hyperoxia i n Amia. The o v e r a l l view of the control of air-breathing i n Amia, i n response to various combinations of gases i n the aquatic and a e r i a l environment, suggests that decreased intravascular P0 2 s t i m u l a t e s , and i n c r e a s e d P0 2 a b o l i s h e s , type I a i r breaths, i n d i c a t i n g an oxygen-related r o l e for type I breaths. Aquatic hyperoxia uniformly depressed air-breathing but, when 108 i t d i d occur, type II breaths were used e x c l u s i v e l y . Type I a i r breaths, therefore, appear to be modulated p r i m a r i l y by changes of i n t e r n a l P0 2 l e v e l s : low blood P0 2 s t i m u l a t e s i n t r a v a s c u l a r chemoreceptors which i n c r e a s e type I a i r -breathing, while high blood P0 2 i n h i b i t s type I a i r breaths. Previous work on other water-breathing and air-breathing fishes indicates the most l i k e l y s i t e s for the r e f l e x stimula-t i o n of v e n t i l a t i o n through 0 2 - s e n s i t i v e chemoreceptors are the central nervous system (Bamford 1974; Jones 1983) and the p e r i p h e r a l b r a n c h i a l v a s c u l a t u r e (Powers and C l a r k 1942; Saunders and S u t t e r l i n 1971; Milsom and B r i l l 1986; Burleson and Smatresk 1990; Burleson 1991). Since type I breaths were stimulated by aquatic and a e r i a l hypoxia, p o t e n t i a l c e n t r a l and peripheral s i t e s c o n t r o l l i n g type I breaths were i n v e s t i -gated. Increased r a t e s of type II a i r breathing i n a e r i a l hyperoxia suggested one of two hypotheses: (1) elevated blood P0 2 simultaneously i n h i b i t s type I breathing and causes a behavioral switch to type II air-breathing, or (2) elevated a i r bladder P0 2 r e s u l t s i n a faster reduction i n lung volume, owing to an increased 0 2 d i f f u s i o n gradient as suggested by Gee and Graham (1978), indicating type II breaths are mediated by s t r e t c h receptors located i n the a i r bladder w a l l . The second hypothesis does not exclude the p o s s i b i l i t y of high blood P0 2 i n h i b i t i n g type I breaths, but the r o l e of s t r e t c h receptors i n type II breathing could be tested independently of hypothesis 1. 109 A i r Bladder Mechanoreceptors i n the Control of Air-Breathing I t has been shown that Amia have slowly-adapting pulmo-nary stretch receptors (PSR), car r i e d i n the ramus intestina-lis branch of the vagus nerve, that respond to both dynamic and s t a t i c changes i n a i r bladder volume (Milsom and Jones 1985) . A r a p i d off-response i n the discharge frequency of vagal a f f e r e n t s was caused by rapid a i r bladder d e f l a t i o n s , i l l u s t r a t i n g the dynamic aspects of these r e c e p t o r s . In a d d i t i o n , PSR showed a discharge frequency p r o p o r t i o n a l to lung volume t y p i c a l of t o n i c s t r e t c h receptors i n the a i r bladders of gar (Smatresk and A z i z i 1987) and lungfish (DeLa-ney et a l . 1983). The i n i t i a l response to a i r bladder d e f l a -t i o n e l i c i t e d only type II a i r breaths, suggesting that a phys i o l o g i c a l and behavioral correlate for the neural observa-t i o n s e x i s t s : the rapid off-response of af f e r e n t nerve d i s -charge recorded from PSR aff e r e n t s (Milsom and Jones 1985) probably e l i c i t s the i n i t i a l type II breath r e f l e x since most def l a t i o n s i n t h i s experiment were done rapidly (<5 s ) . Also, about 60% of the responses to d e f l a t i o n occurred i n less than 1 min, suggesting a r a t e - s e n s i t i v e component of the r e f l e x . When Amia use type I breaths under normal conditions, expired volume i s occasionally low, or gas i s l o s t during the inhala-t i o n and transfer phase; in these instances, type II breaths often occur i n less than 1 min, suggesting the rate- s e n s i t i v e properties of PSR may be involved. This off-response behavior of PSR may be important i n maintaining a i r bladder volume by evoking a type II breath when the t i d a l volume of type I breaths alone are not s u f f i c i e n t to increase stretch receptor 110 f i r i n g above threshold. Since the off-response of PSR shown by Milsom and Jones (1985) was r a t e - s e n s i t i v e , r a t her than volume-sensitive, i t may account f o r the weak, although s i g -n i f i c a n t , c o r r e l a t i o n between the response time of i n i t i a l a i r breath and the volume removed from the a i r bladder ( f i g . 14) . Volume removed from the a i r bladder only accounted f o r 7% of the v a r i a t i o n i n the i n i t i a l response time a f t e r d e f l a t i o n ; the i n i t i a l a i r breath response was largely independent of the volume removed. Since t o n i c discharge of PSR are d i r e c t l y p r o p o r t i o n a l to a i r bladder volume (Milsom and Jones 1985), increased frequency of type II breaths would be expected to be r e l a t e d to the degree of a i r bladder d e f l a t i o n . Indeed, a s i g n i f i c a n t c o r r e l a t i o n was found between a i r - b r e a t h i n g f r e -quency and volume removed ( f i g . 15). Since type II breaths involve inhalation only, increased frequency of these breaths would have the e f f e c t of increasing a i r bladder volume over time. Type II breathing a f t e r d e f l a t i o n probably continues u n t i l PSR a f f e r e n t s cause feedback i n h i b i t i o n of breathing. The c o n s i s t e n t type II breathing behavior upon a i r bladder d e f l a t i o n , that i s also s i g n i f i c a n t l y related to the volume of d e f l a t i o n , supports the hypothesis that type II breaths are stimulated by mechanoreceptor afferents from the a i r bladder. Type II breaths appear to play a r o l e i n maintaining lung volume which d e c l i n e s during i n t e r - b r e a t h i n t e r v a l s by the d i f f u s i o n of 0 2 from the a i r bladder. The most l i k e l y reason f o r the maintenance of a i r bladder volume i s that type II breaths have a buoyancy, rather than gas exchange, function. The r e s u l t s from a i r bladder deflations, therefore, favor the 111 second hypothesis proposed i n the previous s e c t i o n : Amia breathing 100% 0 2 continue to take type II breaths i n response to a constant reduction i n PSR tonic discharge as 0 2 d i f f u s e s from a i r bladder to blood. I t i s a l s o l i k e l y t h at type I breaths are i n h i b i t e d by high blood P0 2 i n a e r i a l hyperoxia, since aquatic hypoxia normally stimulates type I a i r breaths. Given that type II a i r breaths involve i n h a l a t i o n , with no exhalation, t h i s type of breath would appear i d e a l l y suited for maintaining constant a i r bladder volume and neutral buoy-ancy i n Amia. During inter-breath i n t e r v a l s oxygen d i f f u s e s from the a i r bladder, without being replaced by equal amounts of C0 2 (Johansen 1970) ; the r e s u l t i s a decline i n a i r blad-der volume and a r e d u c t i o n i n buoyancy (Alexander 1966) . Unless the gas i s replaced, or other mechanisms are used, the f i s h w i l l become negatively buoyant and sink. Intermittent e x c u r s i o n s to the s u r f a c e to r e p l a c e reduced a i r bladder volume i s a necessary function of a e r i a l r e s p i r a t i o n i n f i s h where a c t i v e s e c r e t i o n of gas into the lung i s not an option (Jones and Marshall 1953). Bi s h a i (1961) demonstrated that the response of a number of physostomous t e l e o s t s to sudden compression, causing a reduction in a i r bladder volume, was to swim to the surface and gulp a i r . There i s evidence that some physostomous t e l e o s t s can secrete gas into the swim bladder (Alexander 1966), however, the rates of secretion are very low compared with physoclists; there i s no evidence that Amia can a c t i v e l y secrete gases i n t o the swim bladder. I t has been argued that physostomous f i s h are dependent upon a i r gulping for buoyancy regulation (Jones and Marshall 1953; Jones 1957). 'Amia, t h e r e f o r e , appears to f i t the general p a t t e r n f o r a physostomous f i s h : a i r - b r e a t h i n g i s requ i r e d f o r buoyancy regulation; Type II breaths probably provide t h i s function. P r evious s t u d i e s on Amia (Johansen et a l . 1970) and Lepisosteus (Smatresk and Cameron 1982b) have shown that a i r bladder d e f l a t i o n s r e s u l t i n a i r breathing events. Although the authors suggested that pulmonary s t r e t c h receptors may have been involved i n mediating the responses, they suggested that air-breathing was an in d i r e c t 0 2-chemoreceptor response, due to changes i n P0 2 i n the lung and blood r e s u l t i n g from the change i n lung volume. This explanation i s u n l i k e l y i n Amia for two reasons. F i r s t , the rapid response to a i r blad-der d e f l a t i o n i n d i c a t e s that s t r e t c h receptor pathways, not i n t r a v a s c u l a r 0 2 - s e n s i t i v e s t i m u l i , mediate the response. Although lung d e f l a t i o n would have the e f f e c t of decreasing t o t a l lung 0 2 stores, a i r bladder P0 2 would not change immedi-a t e l y a f t e r d e f l a t i o n , therefore, i t could not bring about a chemoreceptor-driven v e n t i l a t o r y response. Secondly, i f the change i n 0 2 stores were important i n determining the d e f l a -t i o n r e f l e x , i t would be expected to produce a greater number of type I breaths, as i n a e r i a l hypoxia i n t h i s study ( f i g . 9), r a t h e r than the 100% of type II breaths that were ob-served. Further evidence against a chemoreceptor hypothesis for lung d e f l a t i o n responses i s shown by c a l c u l a t i n g the gain i n lung 0 2 that would occur in a t y p i c a l type II breath. Assum-ing that a 500 g Amia has an a i r bladder volume of about 8% of body mass, or 4 0 ml, and the average P0 2 in the lung i s about 113 60-80 Torr (Johansen et a l . 1970; pers. obs.), an a i r bladder with a P0 2 of 80 Torr has an oxygen store of 4.32 ml 0 2. I f the apparent threshold for detection (3 ml k g - 1 , f i g . 13) of gas removal from the a i r bladder i s used, then 1.5 ml 0 2 d i f f u s e s from the lung to the blood, without being replaced by C0 2. The new a i r bladder 0 2 content and P0 2 become 2.82 ml 0 2 and 54.2 T o r r , r e s p e c t i v e l y . The f i s h r e p l a c e s the l o s t volume with a 1.5 ml type II breath, which contains about 0.32 ml 0 2. A f t e r the type II breath, the r e s u l t i n g P0 2 i s 58.1 Torr, a change of only 3.9 Torr. Thus, simple c a l c u l a t i o n s suggest i t i s un l i k e l y that type II breaths contribute s i g n i f -i c a n t l y to gas exchange. Furthermore, a small change i n P0 2 of 3-4 Torr i n the blood aft e r d i f f u s i o n from the a i r bladder i s u n l i k e l y to be undetected by in t r a v a s c u l a r chemoreceptors (Burleson 1991). Type I breaths, a l t e r n a t i v e l y , which are of lar g e r volume and occur predominantly i n hypoxic s i t u a t i o n s , would appear to serve mainly a gas exchange function by exhal-ing a lower P0 2 lung gas and r e p l a c i n g i t with ambient a i r containing a higher P0 2. A d d i t i o n a l support f o r t h i s view of Amia a i r bladder mechanoreceptors being involved i n buoyancy r e g u l a t i o n comes from experiments i n which these nerves were b i l a t e r a l l y sec-tioned. In t h i s condition, Amia continued type II air-breaths but the a i r bladder eventually became very distended u n t i l the f i s h floated at the surface, indicating that buoyancy control was severely compromised. In the aquatic anuran amphibian, Xenopus laevis, the lung also becomes distended and the animal loses buoyancy control with lung denervation (Evans and Shel-ton 1984), suggesting that v e n t i l a t i o n may also play an impor-tant r o l e i n buoyancy control in t h i s species. Bowfin were occasionally subjected to a i r bladder i n f l a -t i o n s upon approaching the surface. Although the responses are more d i f f i c u l t to i n t e r p r e t , a i r bladder i n f l a t i o n some-times caused Amia to abandon the approach to the surface but,, more importantly, a i r bladder i n f l a t i o n s f a i l e d to i n h i b i t type I a i r breaths when they did occur. Although t h i s i s not conclusive evidence, i t does further support the contention t h a t a f f e r e n t pathways c o n t r o l l i n g type I and type II a i r breaths are separate. Pulmonary mechanoreceptor inputs i n ectothermic v e r t e -brates appear to play a s i g n i f i c a n t r o l e i n modulating the pattern of breathing i n those animals that have been studied (see Milsom 1990, for review). For instance, i n the A f r i c a n lungfish, Protopterus annectens, lung i n f l a t i o n s i g n i f i c a n t l y increased inter-breath i n t e r v a l , and the e f f e c t was potentiat-ed by i n f l a t i n g the lung with high 0 2 concentrations (Pack et a l . 1990). I t i s not known whether equivalent v e n t i l a t o r y responses to chemoreceptor and mechanoreceptor stimulation are p r e s e n t i n l u n g f i s h ; the data from Amia suggest they are d i f f e r e n t . In the t u r t l e , Chrysemys picta, reductions i n lung volume reduced the length of the non-ventilatory period (NVP), but did not a f f e c t o v e r a l l minute v e n t i l a t i o n (Milsom and Chan 1986). The change i n NVP in the t u r t l e could not be accounted f o r on the b a s i s of changes i n lung 0 2 s t o r e s , as i n t h i s study. The general e f f e c t of increased pulmonary a f f e r e n t discharge i n Amia, and other ectothermic vertebrates, appears to i n c r e a s e the i n t e r - b r e a t h i n t e r v a l (IBI), or NVP, with i n f l a t i o n , and reduce IBI or NVP with d e f l a t i o n . The remain-ing question, for which there are no data, i s how these pulmo-nary afferents are integrated i n the central nervous system to produce the various breathing patterns i n ectothermic verte-brates . The general features of mechanoreceptor involvement i n c o n t r o l l i n g v e n t i l a t o r y patterns i n ectotherms are si m i l a r to those seen i n mammals. Lung i n f l a t i o n s in mammals character-i s t i c a l l y shorten i n s p i r a t i o n and prolong e x p i r a t i o n , while d e f l a t i o n s shorten e x p i r a t i o n and can increase r e s p i r a t o r y frequency (Knox 1973) . These reflexes in mammals are c o l l e c -t i v e l y known as the Breuer-Hering r e f l e x , which are mediated through s l o w l y - a d a p t i n g pulmonary s t r e t c h r e c e p t o r s with afferent f i b e r s contained in the vagus nerve (see Pack 1981). The s i m i l a r i t i e s of v e n t i l a t o r y responses to lung or a i r bladder i n f l a t i o n and d e f l a t i o n s across a broad phylogenetic spectrum i n d i c a t e s t h a t Breuer-Hering-1ike r e f l e x e s were selected early i n vertebrate evolution. Chemoreceptor Sites in the Control of Air-Breathing Peripheral Sites of Chemoreception Denervation of b r a n c h i a l branches IX and X to the g i l l arches and, i n a few f i s h , the pseudobranch, f a i l e d to abolish the v e n t i l a t o r y responses to aquatic hypoxia. The hypothesis tested was that peripheral 0 2-chemosensitive s i t e s on the g i l l arches, innervated by branches of the glossopharyngeal (crani-116 a l n. IX) and vagus ( c r a n i a l n. X), were r e s p o n s i b l e f o r s t i m u l a t i n g the type I a i r breaths during aquatic hypoxia. Mechanoreceptors innervating the a i r bladder are also inner-vated by a branch of the vagus nerve, but t h i s branch was l e f t i n t a c t ; only those branches of IX and X innervating the four branchial arches were sectioned. Since type II breaths appear to be influenced mainly by a i r bladder mechanoreceptors, there was no a priori assumption that a l l a i r - b r e a t h i n g responses c o u l d be a b o l i s h e d ; thus, any involvement of a i r bladder mechanoreceptors i n the a i r - b r e a t h i n g responses would have remained. I t was apparent that e l i m i n a t i n g the innervation to the four b r a n c h i a l arches and the pseudobranch d i d not a b o l i s h type I or type II air-breathing, nor did i t prevent or attenu-ate the increase i n air-breathing during aquatic hypoxia. The f a i l u r e to reduce type I a i r - b r e a t h i n g frequency with t o t a l b r anchial nerve section indicates there are probably extra-branchial s i t e s f o r detecting low P0 2 - Most experiments on water-breathing f i s h have also f a i l e d to abolish g i l l v e n t i l a -tory responses to hypoxia after branchial and/or pseudobranch denervation (Hughes and Shelton 1962; Saunders and S u t t e r l i n 1971; Randall and Jones 1973; Bamford 1974); however, a recent study i n d i c a t e d that hypoxic v e n t i l a t o r y r e f l e x e s are abol-ished by complete branchial nerve section in channel c a t f i s h , Ictalurus punctatus (Burleson and Smatresk 1990). In lung-f i s h , branchial nerve section attenuated, but did not abolish, the v e n t i l a t o r y responses to hypoxemia and cyanide i n j e c t i o n ( L a h i r i et a l . 1970). More recently, McKenzie (1990) reported 117 that air-breathing was completely abolished by t o t a l branchial nerve section combined with pseudobranch ablation i n Amia. In t h i s study, i t i s possible that f a i l u r e to abolish the hypoxic v e n t i l a t o r y r e f l e x e s were due to i n t a c t i n n e r v a t i o n of the f a c i a l nerve (n. VII) to the pseudobranch. The pseudobranch i n Amia i s a glandular structure l y i n g beneath the epithelium of the p a l a t e ( A l l i s 1897). I t re c e i v e s an a r t e r i a l blood supply v i a the inter n a l carotid artery, but probably does not sense immediate changes i n e x t e r n a l water P0 2 due to i t s s u b e p i t h e l i a l l o c a t i o n . I f the pseudobranch i n Amia func-tioned i n chemoreception, i t would be i n p o s i t i o n to detect changes i n a r t e r i a l P0 2. It i s not yet known, however, wheth-er the pseudobranch i n Amia has a chemoreceptive function or whether i t also receives neural innervation from the f a c i a l nerve. Cutting i t s glossopharyngeal innervation had no ef f e c t on the responses to aquatic hypoxia. The approach of McKenzie (1990), however, examined the ven t i l a t o r y resposes to aquatic hypoxia applied for only 15 min, which i s probably too short a time period to determine whether hypoxic r e f l e x e s were s t i l l i n t a c t . Amia i n t h i s study were exposed to aquatic hypoxia f o r several hours, which may have caused release of humoral factors, such as catecholamines, that could influence v e n t i l a -t i o n . Adrenaline and noradrenaline have been implicated i n the v e n t i l a t o r y responses to hypoxia i n water-breathing f i s h (Peyraud-Waitzenegger 19,79; Aota et a l . 1990). However, i n t r a - v a s c u l a r i n j e c t i o n s of catecholamines i n Amia do not stimulate a i r - b r e a t h i n g (McKenzie 1990). Despite the demon-s t r a t i o n of 0 2 - s e n s i t i v e chemoreceptors with a f f e r e n t s con-118 t a i n e d i n the glossopharyngeal (Burleson and Milsom 1990; Burleson 1991) and vagal (Milsom and B r i l l 1986) branches of the g i l l arches, as well as the pseudobranch (n. IX) nerves (Laurent and Rouzeau 1972), the f a i l u r e to a b o l i s h hypoxic v e n t i l a t o r y responses by denervating these nerves suggests that extra-branchial chemoreceptive s i t e s are probably present in f i s h . C l e a r l y - d e f i n e d oxygen receptor s i t e s i n a i r - b r e a t h i n g f i s h have not been demonstrated, but i n d i r e c t evidence i n d i -cates that vascular chemoreceptors stimulate air-breathing in lungfish ( L a h i r i et a l . 1970) and gar (Smatresk 1986; Smatresk et a l . 1986). Injections of hypoxic blood or NaCN into the a n t e r i o r arches of l u n g f i s h produces a greater v e n t i l a t o r y stimulus than injections into more posterior arches ( L a h i r i et a l . 1970). In gar, denervation of vagal b r a n c h i a l nerves abolished the v e n t i l a t o r y depression of aquatic hypoxia and NaCN on g i l l v e n t i l a t i o n , but air-breathing reflexes were only s l i g h t l y attenuated (Smatresk 1987). Although the evidence i s limited, the air-breathing f i s h that have been studied to date may have e x t r a - b r a n c h i a l s i t e s f o r 0,2 chemoreception that modulate air-breathing reflexes. The w a l l s of the p o s t e r i o r c a r d i n a l v e i n i n Amia have been shown to contain chromaffin c e l l s , associated with nerve t e r m i n a l s , which resemble c e l l s of the adrenal medulla i n higher vertebrates (Youson 1976). The u l t r a s t r u c t u r a l charac-t e r i s t i c s of these c e l l s indicate they may release catechola-mines since they c l o s e l y resemble c e l l s derived from neural crest t i s s u e belonging to the APUD (Amine Precursor Uptake and 119 Decarboxylation) series (Youson 1976). Glomus c e l l s i n mamma-l i a n and avian carot i d bodies, with a known 0 2-chemoreceptive function, are also considered part of the APUD series (Pearse 1969). Thus, i f the granulated c e l l s i n the venous vascula-t u r e i n Amia are indeed s i m i l a r to the APUD c e l l s of other vertebrates (see Jones and Milsom 1982), they may be involved in chemoreception. Their location i n the walls of the poste-r i o r c a r d i n a l vein would place them i n an i d e a l p o s i t i o n to monitor blood P0 2 d i s t a l to the a i r bladder, before the blood enters the g i l l v a s c u l a t u r e . This would suggest a venous l o c a t i o n f o r 0 2 - s e n s i t i v e chemoreception. There i s some experimental evidence for a venous location for chemoreception i n water-breathing f i s h ( B a rrett and Taylor 1984). I t i s s t i l l not known, however, i f such c e l l s are involved i n 0 2 chemoreception in f i s h . Amia normally increase g i l l v e n t i l a t i o n in aquatic hypox-i a i n ad d i t i o n to increasing a e r i a l v e n t i l a t i o n (Johansen et a l . 1970; McKenzie 1990). In t h i s study, g i l l v e n t i l a t i o n in normoxic and hypoxic conditions did not appear to be affected to as great an extent as air- b r e a t h i n g (Tables I I I , IV). In contrast with other studies, Amia did not increase g i l l v e n t i -l a t i o n i n aquatic hypoxia i n t h i s experiment, although g i l l v e n t i l a t o r y increases in aquatic hypoxia were noted with Amia using a d i f f e r e n t experimental approach (next s e c t i o n ) . In previous studies, g i l l v e n t i l a t i o n rates, at si m i l a r tempera-tures and le v e l s of oxygenation, were 10 to 100% greater than reported here (Johansen et a l . 1970; McKenzie 1990). In those studies, g i l l v e n t i l a t i o n rates i n normoxia were higher and 120 increased s i g n i f i c a n t l y i n aquatic hypoxia. The d i f f e r e n c e s in g i l l v e n t i l a t i o n rates between studies suggests the type of experimental approach used to monitor v e n t i l a t i o n may have an e f f e c t on the normal g i l l v e n t i l a t o r y p a t t e r n and the r e -sponses to hypoxia. G i l l v e n t i l a t i o n rates have been usually determined using invasive techniques that involve implanting cannulae i n t o the buccal or opercular c a v i t i e s to monitor v e n t i l a t i o n . G i l l v e n t i l a t i o n rates reported here were deter-mined by d i r e c t o b s e r v a t i o n from videotape. The g r e a t e r l a b i l i t y of air-breathing over branchial responses to aquatic hypoxia was a l s o noted by d i r e c t o b s e r v a t i o n i n r e e d f i s h ( P e t t i t and Beitinger 1985). The large number of type II breaths that occurred follow-ing t o t a l b r a n c h i a l denervation i n Amia was an unexpected r e s u l t ( f i g . 19). It was apparent from the videotape record-ings these f i s h had considerable d i f f i c u l t y with gas capture and transfer (phases 3 and 4 of the type I cy c l e ) , p a r t i c u l a r -l y during type I breaths; i t appeared that abnormally large amounts of gas escaped from the opercular c a v i t i e s . In these cases, l a r g e numbers of type II breaths occurred, probably owing to an i n a b i l i t y of the f i s h to maintain a constant volume i n the swim bladder. This i s i n d i r e c t evidence to support the hypothesis that type II breaths are mediated by a i r bladder mechanoreceptors. The observation that t o t a l branchial denervation adverse-l y a f f e c t s gas t r a n s f e r during a i r - b r e a t h i n g i n Amia r a i s e s questions concerning the e f f i c a c y of t h i s experimental ap-proach i n d e l i m i t i n g chemoreceptive s i t e s i n a i r - b r e a t h i n g 121 f i s h . The underlying assumption of b r a n c h i a l denervation experiments i s that e f f e r e n t and af f e r e n t information, other than chemoreceptor a f f e r e n t s , c a r r i e d i n c r a n i a l nerves IX and X i s unimportant i n determining the motor output f o r the branchial pumps, which are mainly innervated by the trigeminal (c r a n i a l n. V) and f a c i a l (cranial n. VII) nerves (see Nilsson 1984) . Powers and Clark (1942) did observe that c u t t i n g the g i l l innervation of c r a n i a l nerve IX (glossopharyngeal) pro-duced gasping movements i n tro u t that were not present when nerve X (vagus) alone was eliminated. Nerve sections i n Amia, although a f f e c t i n g the f i s h ' s a b i l i t y to breathe a i r normally, d i d not produce any e f f e c t s that resembled gasping. Nerve f i b e r s eliminated by c u t t i n g c r a n i a l nerves IX and X to the g i l l s provide motor con t r o l for p o s i t i o n i n g g i l l arch rakers and vasomotor c o n t r o l f o r b r a n c h i a l v a s c u l a t u r e ( N i l s s o n 1984). Afferent information, other than 0 2 chemosensory, from baroreceptors (Mott 1951), nociceptors (Poole and S a t c h e l l 1979) and g i l l mechanoreceptors (De Graaf.et a l . 1987), would also be eliminated by denervation. The i n a b i l i t y of Amia to e f f e c t i v e l y capture and t r a n s f e r i n h a l e d a i r a f t e r t o t a l b r a n c h i a l denervation suggests that i n t a c t b r a n c h i a l motor and/or sensory information i s required for normal air-breath-ing f unction. G i l l v e n t i l a t i o n , although depressed with r e-spect to rate, appeared normal. It i s d i f f i c u l t to determine, at present, which nervous pathways and structures innervated by nerves IX and X are important i n f a c i l i t a t i n g normal a i r -breathing i n Amia. The most l i k e l y candidates are g i l l arch motor nerves and mechanoreceptors that c o n t r o l and transduce 122 information about g i l l arch p o s t i o n . These f u n c t i o n s are probably important f o r p o s i t i o n i n g and sensing the changes that occur during a i r - b r e a t h i n g when the buccal c a v i t y i s b r i e f l y f i l l e d with a i r . These r e s u l t s would appear to v i o -l a t e the assumptions of t h i s experimental approach and r a i s e more questions about the r o l e of b r a n c h i a l i n n e r v a t i o n i n determining v e n t i l a t o r y neuromuscular motor output, p a r t i c u -l a r l y i n air-breathing f i s h . Central Sites of Chemoreception The hypothesis that c e n t r a l chemoreceptors may be i n -volved i n the v e n t i l a t o r y responses to hypoxia and/or hyper-capnia was not supported by the EDF p e r f u s i o n experiments (Table V). The only s i g n i f i c a n t e f f e c t on branchial or a e r i a l v e n t i l a t i o n occurred with aquatic hypoxia; changing EDF gas concentrations had no ef f e c t on v e n t i l a t i o n . The manipulation of EDF f o r the experimental p r o t o c o l necessitated the use of a d i f f e r e n t type of arrangement than other experiments in t h i s study. Fish had to be confined to a small box i n order to change EDF concentrations. The type of holding box described here, which has been used previously for v e n t i l a t i o n s t u d i e s i n Amia (Randall et a l . 1981; McKenzie 1990) made d i f f e r e n t i a t i n g between type I and type II breaths impossible, since air-breathing was distinguished on the basis of changes i n opercular pressure during these events. The use of t h i s type of apparatus, which i s more c o n f i n i n g to the animal, might be expected to produce d i f f e r e n t branchial or a e r i a l v e n t i l a t i o n responses from f i s h than freely-swimming 123 animals i n larger aquaria. C o n t r o l g i l l v e n t i l a t i o n r a t e s i n t h i s experiment were considerably higher than values obtained by d i r e c t observation from f i s h i n aquaria (Table IV), despite the lower temperature (15 °C) i n the EDF experiment. A i r - b r e a t h i n g i n Amia i s highly temperature dependent (Johansen et a l . 1970; Horn and Riggs 1973; pers. obs.), which probably accounts for the lack of air-breathing i n aquatic normoxia (Table IV). In the same type of experimental apparatus used here, Amia have been shown to use a e r i a l r e s p i r a t i o n i n normoxia, and increase the f r e -quency of a i r - b r e a t h i n g i n aquatic hypoxia (Randall et a l . 1981; McKenzie 1990), although the reported frequencies were s l i g h t l y lower than f i s h i n aquaria i n t h i s study. Despite the caveats regarding experimental s i t u a t i o n s , the r e l a t i v e responses to aquatic hypoxia remain i n t a c t regardless of the approach; therefore, the conclusion that EDF manipulations do not a f f e c t a e r i a l v e n t i l a t i o n would appear to remain v a l i d . In s e l e c t e d s p e c i e s from a l l c l a s s e s of t e r r e s t r i a l vertebrates studied thus far, central chemoreceptors have been shown to stimulate v e n t i l a t i o n (amphibians: Smatresk and Smits 1991; r e p t i l e s : H i t z i g and Jackson 1978; birds:- Milsom et a l . 1981; mammals: M i t c h e l l et a l . 1963; M i l l h o r n and El d r i d g e 1986). In these t e r r e s t r i a l groups, areas near the ventrolat-e r a l medulla oblongata, and acce s s i b l e from the v e n t r i c u l a r system (Pappenheimer et a l . 1965; H i t z i g and Jackson 1978), a f f e c t breathing i n response to changes in CC>2 and/or pH. In t h i s experiment, cerebrospinal f l u i d (CSF) was not manipulated d i r e c t l y but, instead, mock EDF was perfused throughout the 124 meningeal space. In f i s h e s , the meningeal space l i e s between t h e p e r i o s t e u m o f t h e cra n i u m and t h e meninx o v e r l y i n g t h e b r a i n s u r f a c e ; t h i s communicates through b l o o d v e s s e l s w i t h t h e v e n t r i c u l a r system and CSF. These pathways have been confirmed w i t h r a d i o a c t i v e t r a c e r s and v i t a l dyes (see Davson 1967). The dye p e r f u s i o n r e s u l t s i n t h i s experiment i n d i c a t e d t h a t t h e c r a n i a l p e r f u s i o n method a l l o w e d mock EDF t o come i n t o c o n t a c t w i t h areas of the b r a i n c o r r e s p o n d i n g w i t h cen-t r a l r e f l e x o g e n i c areas of t e r r e s t r i a l v e r t e b r a t e s . The l a c k of s i g n i f i c a n t v e n t i l a t o r y e f f e c t s i n response t o hypercapnia and low pH s o l u t i o n s i n d i c a t e s t h a t Amia do not pos s e s s cen-t r a l c hemoreceptive s i t e s analogous t o those of t e r r e s t r i a l v e r t e b r a t e s . A r e c e n t study i n the skate (Raja ocellata) a l s o showed t h a t EDF pH was not c o r r e l a t e d with changes i n v e n t i l a -t i o n (Graham e t a l . 1990). The l a c k of v e n t i l a t o r y responses t o changes i n 0 2 i n EDF a l s o argues a g a i n s t the hyp o t h e s i s of a c e n t r a l oxygen chemo-r e c e p t o r i n f i s h . Experiments i n which i s o l a t e d , spontaneous-l y b r e a t h i n g carp heads are made hypoxemic, show t h a t v e n t i l a -t o r y movements e v e n t u a l l y stop, i n d i c a t i n g t h a t c e n t r a l hypox-i a has a d e p r e s s a n t e f f e c t on v e n t i l a t i o n (Kawasaki 1980); however, t h e r e was no e v i d e n c e f o r a d e p r e s s a n t e f f e c t o f c e n t r a l hypoxia on v e n t i l a t i o n i n Amia. The l i m i t e d e v i d e n c e from Amia and Raja would i n d i c a t e t h a t c h o n d r i c h t h y e a n and a c t i n o p t e r y g i a n f i s h e s p r o b a b l y do n o t h a ve c e n t r a l c h e m o r e c e p t o r s ; i t i s l i k e l y t h e y have e v o l v e d w i t h the t r a n s i t i o n t b t e r r e s t r i a l i t y i n S a r c o p t e r y -g i a n v e r t e b r a t e s . 125 Intermittent Air-Breathing i n Amia The intermittent breathing pattern of Amia usually showed an a l t e r n a t i o n between type I and type II a i r breaths i n normoxic conditions ( f i g . 2OA). Although there did not appear to be an obvious p a t t e r n , s p e c t r a l a n a l y s i s of long-term recordings c l e a r l y i n d i c a t e d , however, that a i r - b r e a t h i n g occurred with a period of 30 min ( f i g . 21B) i n normoxic condi-t i o n s , and 10 min i n aq u a t i c hypoxia ( f i g . 21A). I f the breathing p a t t e r n of Amia were i r r e g u l a r or random, there would be no s i g n i f i c a n t peaks from s p e c t r a l a n a l y s i s ; the spectrum would resemble, i n s t e a d , broad-band n o i s e . The re s u l t s here indicate the pattern i s not ir r e g u l a r or random. Although i t has been acknowledged that i n t e r m i t t e n t l y -breathing f i s h , when undisturbed, air-breathe somewhat regu-l a r l y (see Milsom 1991), the use of spectral analysis c l e a r l y demonstrates the u n d e r l y i n g r h y t h m i c i t y i n the b r e a t h i n g pattern of Amia. In previous studies, the rhythmicity of the breathing patterns has been assessed subjectively, or has been examined using measurements of variance, which cannot y i e l d much information about rhythmicity in the data (see van den Aardweg and Karemaker 1991). The f i n d i n g that Amia breathe r h y t h m i c a l l y has important i m p l i c a t i o n s f o r the c o n t r o l of air-breathing i n Amia and perhaps other intermittently-breath-ing species. The reasons that rhythmic a i r - b r e a t h i n g have not been previously demonstrated i n air-breathing f i s h are at least two-f o l d . F i r s t , long-term observations of breathing patterns are usually not recorded i n the laboratory. Most of the recording 126 s e s s i o n s i n t h i s study were 8 h i n l e n g t h , w i t h a minimum of o u t s i d e d i s t u r b a n c e t o the f i s h , which a l l o w e d time enough f o r u n d i s t u r b e d b r e a t h i n g p a t t e r n s t o become e s t a b l i s h e d . I t i s w e l l known t h a t b e h a v i o r a l e f f e c t s , such as the s i m u l a t i o n o f a p r e d a t o r , i n c r e a s e s t h e i n t e r - b r e a t h i n t e r v a l i n a i r -b r e a t h i n g f i s h (Gee 1980; Smith and Kramer 1986). Second, a l t h o u g h s p e c t r a l a n a l y s i s has been e x t e n s i v e l y a p p l i e d i n a n a l y z i n g the fr e q u e n c y components of v e n t i l a t i o n i n humans (Goodman 1964; H l a s t a l a e t a l . 1973; Waggener e t a l . 1982; Pack e t a l . 1988), t h i s a n a l y s i s does not appear t o have been used t o examine b r e a t h i n g p a t t e r n s i n ectotherms. T h e r e i s ample e v i d e n c e t h a t v e n t i l a t o r y p a t t e r n s o f humans e x h i b i t a wide r a n g e o f p e r i o d i c i t e s (see van den Aardweg and Karemaker 1991, f o r a r e c e n t r e v i e w ) . For i n -stance, Goodman (1964) f i r s t demonstrated s i g n i f i c a n t v e n t i l a -t o r y o s c i l l a t i o n s o c c u r r i n g a t approximately 50-80 s, 2-3 min, 6-8 min, and 2.5-3.5 h. H l a s t a l a e t a l . (1973) i d e n t i f i e d s e v e r a l f r e q u e n c i e s i n humans, with p e r i o d s up t o 28 min, t h a t were a s s o c i a t e d with end t i d a l P0 2 and PC0 2, t i d a l volume and f u n c t i o n a l r e s i d u a l c a p a c i t y . H l a s t a l a e t a l . (1973) suggest-ed f u r t h e r t h a t any system w i t h feedback l o o p s , such as the r e s p i r a t o r y system, would be e x p e c t e d t o o s c i l l a t e . Thus, s p e c t r a l a n a l y s i s has been shown t o be a p o w e r f u l t o o l f o r r e v e a l i n g p e r i o d i c i t y i n time s e r i e s data i n humans t h a t would o t h e r w i s e be o v e r l o o k e d . T h i s t y p e o f a n a l y s i s has wide a p p l i c a t i o n , and should be a p p l i e d i n s t u d i e s i n i n t e r m i t t e n t b r e a t h e r s b e f o r e d e c i d i n g whether rhy t h m i c o s c i l l a t i o n s are pr e s e n t . 127 The major question that a r i s e s from t h i s a n a l y s i s i s : what i s r e s p o n s i b l e f o r the underlying p e r i o d i c i t y i n the breathing pattern i n Amia? One possib l e explanation i s that an i n t r i n s i c " c l o c k " l o c a t e d i n the CNS generates a c t i v i t y every 3 0 min i n aquatic normoxia, which decreases to 1 0 min in aquatic hypoxia. This concept would be a n t i t h e t i c a l to the view that a i r - b r e a t h i n g i n f i s h i s an on-demand phenomenon (see Smatresk 1 9 9 0 ) , generated by p e r i p h e r a l feedback. I f ai r - b r e a t h i n g i s indeed on-demand, a rhythmic breathing pat-tern would imply periodic stimulation of peripheral chemo- or mechanoreceptors. Since type I and type II breaths appear rela t e d to oxygen a v a i l a b i l i t y and buoyancy, respectively, i t seems reasonable that either of these factors could contribute s i g n i f i c a n t l y to the rhythmic breathing pattern. The evidence from spectral analysis that shows a dominant p e r i o d i c i t y i d e n t i c a l to the average inter-breath i n t e r v a l for type I breaths i s strong evidence that p e r i o d i c breathing i n Amia i s dependent on feedback from 0 2 _ s e n s i t i v e chemorecep-t o r s ; however, t h e i r locations could not be determined from t h i s study. Although centrally-generated periodic rhythmogen-e s i s cannot be r u l e d out as a mechanism, a more a t t r a c t i v e hypothesis f o r the generation of rhythmic a i r - b r e a t h i n g i n Amia i s that p e r i o d i c stimulation of per i p h e r a l 0 2 - s e n s i t i v e chemoreceptors are r e s p o n s i b l e . In a q u a t i c hypoxia, the dominant p e r i o d i s reduced to about 1 0 min, which i s al s o associated with the mean of inter-breath i n t e r v a l f o r type I a i r breaths. Since type I a i r breaths are sensi t i v e to exter-n a l and/or i n t e r n a l P 0 ? , t h i s suggests the rhythmic a i r -1 2 8 breathing behavior i s driven by p e r i o d i c stimulation of 0 2 ~ s e n s i t i v e chemoreceptors. The p e r i o d i c i t i e s occur i n condi-tions of constant aquatic P0 2, which suggests that i n t e r n a l l y f l u c t u a t i n g oxygen levels are probably responsible for estab-l i s h i n g the v e n t i l a t o r y r h y t h m i c i t y . Whatever the p r e c i s e mechanism for generating the breathing pattern, i t i s apparent that Amia breathe rhythmically in undisturbed conditions, and the dominant frequency of the rhythm changes from normoxic to hypoxic conditions. The next section tests these p o s s i b i l i t e s using a computer model to simulate the intermittent breathing pattern i n Amia. A Computer Model of Intermittent Air-Breathing i n Amia A computer model, using empirically-derived observations from t h i s study and elsewhere, was created to simulate the breathing pattern of Amia. The main purposes of t h i s model were to explain the v e n t i l a t o r y responses of Amia to changes in r e spiratory gases, and tes t some of the underlying hypothe-ses generated from the empirical data. The basis for the model i s the assumption that a i r blad-der oxygen mixes with systemic venous return before entering the g i l l vasculature. The vascular anatomy of Amia i n d i c a t e s t h i s i s a v a l i d assumption (Randall et a l . 1981). G i l l v e n t i -l a t i o n and i t s c o n t r i b u t i o n to the oxygenation of the a i r bladder v a s c u l a t u r e was not considered i n t h i s model. In r e a l i t y , a r t e r i a l blood from a branch of the fourth g i l l arch i n Amia serves as the af f e r e n t blood supply to the a i r blad-der. Oxygen obtained by air-breathing, however, u l t i m a t e l y 129 d i f f u e s into the venous, rather than a r t e r i a l , vasculature; i t i s t h i s step i n a e r i a l v e n t i l a t i o n that has been modeled. At present, there appears to be only one comparable model that has simulated the v e n t i l a t o r y pattern of i n t e r m i t t e n t l y breathing vertebrates (Shelton and Croghan 1988). Their model used data from an a i r - b r e a t h i n g t e l e o s t , Electrophorus elec-tricus, and an aquatic anuran amphibian, Xenopus laevis, to simulate intermittent breathing patterns i n reponse to changes i n lung, blood and tissue oxygen stores. Both models use the concept of a blood P0 2 threshold to t r i g g e r a i r - b r e a t h i n g events, but there are some differences between them. Simulat-ed type I breaths i n the Amia model are triggered i n response to reductions i n blood P0 2 to a set-point threshold, which i s s i m i l a r to air-gulping simulations in the Shelton and Croghan (1988) model. In both models v e n t i l a t i o n i s dependent on a feedback response to a decline i n blood P0 2 rather than any e x p l i c i t change i n chemoreceptor d i s c h a r g e . T h i s i s more convenient than hypothesizing chemoreceptor discharge per se s i n c e the l o c a t i o n s and d i s c h a r g e c h a r a c t e r i s t i c s of the chemoreceptors are not known. A major d i s t i n c t i o n of the Amia model i s the a d d i t i o n of a i r b l a d d e r v o l u m e - r e l a t e d a i r breaths suggested by the empirical data; consequently, there are two independent variables for t r i g g e r i n g a i r breaths. The Amia model also incorporates a method to simulate the pattern of breathing that i s c h a r a c t e r i s t i c of intermittent breathers. The Shelton and Croghan (1988) model predicts that v e n t i l a t o r y bouts are always evenly spaced. Both models, which use a c t i v -i t y of v a r i a b l e s d e c r e a s i n g (or i n c r e a s i n g ) to t h r e s h o l d 1 3 0 values and then r e s e t t i n g , are of a general type known as integrate and f i r e (Glass and Mackey 1988). A schematic diagram of the e s s e n t i a l f e a t u r e s of the model i s i n f i g u r e 23. A more p r e c i s e d e s c r i p t i o n of the model, i n c l u d i n g the parameters used, t h e i r assigned values and the equations f o r numerical c a l c u l a t i o n s are given i n Appendix 1. The source code, written i n Turbo Pascal, used to solve the equations i t e r a t i v e l y i s l i s t e d i n Appendix 2. The values used i n the model (Appendix 1) are based on empirical data from t h i s study, or estimates from other studies on Amia or Lepisosteus, for a 500 g animal. A i r breaths, whether type I or type I I , were modeled as di s c r e t e events triggered by independent, thresholds ( f i g . 23). I t was assumed that following either a type I or type II a i r -breath, 0 2 diffused into the blood flowing past the a i r blad-der. Oxygen f l u x a f t e r an a i r breath was modeled i n two d i s c r e t e steps: (1) 0 2 d i f f u s i o n from a i r bladder to blood, and (2) convective f l u x of oxygen i n blood flowing past the a i r bladder. Oxygen d i f f u s i n g from the a i r bladder was assumed not to be r e p l a c e d by C0 2, so any d e c l i n e i n a i r bladder volume was e n t i r e l y dependent on 0 2 removal. Respira-tory exchange r a t i o (R) values i n the a i r breathing organs of a i r - b r e a t h i n g f i s h are low owing to the p r e f e r e n t i a l loss of C0 2 to the environment through d i f f u s i o n across g i l l surfaces (see Shelton et a l . 1986), and so t h i s seems a reasonable assumption. Oxygen d i f f u s i o n from the a i r bladder to blood, according to F i c k ' s Law, i s d i r e c t l y p r o p o r t i o n a l to the d i f f u s i o n 131 Figure 23. A schematic diagram of the e s s e n t i a l features of the model used to simulate ai r - b r e a t h i n g i n Amia. See text for d e t a i l s . 132 A i r - Breathing Model Volume SenSOr ( S t r e t c h Receptor Feedback) I T Y P E II — • r AIR B R E A T H AIR B L A D D E R O z DIFFUSION T Y P E I _ P O 2 _ Sensor (O2 Chemoreceptor) EFFERENT A F F E R E N T Q A B 1 coeffiencent for 0 2 in tissue (D0 2), surface area (SA) for gas exchange and the P0 2 gradient (AP0 2) between a i r bladder and blood, and inversely proportional to d i f f u s i o n distance (Ax) . D0 2, SA and Ax are often d i f f i c u l t to determine e m p i r i c a l l y , so i t i s convenient to combine them into a s i n g l e parameter, pulmonary d i f f u s i o n c a p a c i t y ( D L 0 2 ) . Oxygen f l u x from a i r bladder to blood i s then the product of D L 0 2 and the P0 2 gradient. There are no d i r e c t measurements of D L0 2 for Amia, so a value was estimated based on empirical data and estimates from amphibians and gar (Lepisosteus). Measured and estimated values of D L0 2 for amphibians are in the range of 0.03 ml min - 1 T o r r - 1 at 25 °C (Glass et a l . 1981; Withers and Hillman 1988). Values for air-breathing f i s h should be less than t h i s , owing to lower a l v e o a r i z a t i o n of the a i r bladder. Indeed, empirical measurements of a i r bladder surface area i n Lepi-sosteus are approximately 3 to 4 times lower than amphibians of comparable body mass (Rahn et a l . 1971), so D L 0 2 was scaled proportionately lower for Amia. A value of 0.0085 ml m i n - 1 T o r r - 1 f o r D L0 2 was used i n the model c a l c u l a t i o n s . D L0 2 was c a l l e d a i r bladder d i f f u s i o n capacity (D A B0 2) i n t h i s model. The c o n v e c t i v e step was s i m p l i f i e d and modeled as a s i n g l e vessel containing blood flowing past the a i r bladder ( f i g . 23) . 0 2 f l u x from the a i r bladder was added to blood flow i t e r a t i v e l y in discrete steps in the program. The i t e r a -t i o n process r e s u l t e d i n the i n t e g r a t i o n of the 0 2 f l u x or blood flow r a t e s over each d i s c r e t e time i n t e r v a l . Since volume (ml 0 2 or ml blood) i s the int e g r a l of 0, flux or blood 134 flow r a t e , volumes of 0 2 or blood were c a l c u l a t e d at each i t e r a t i o n . Therefore, the convective step was modeled as the summation of blood 0 2 content u n t i l the threshold was reached. Blood was thus treated as a pool, rather than a flow rate i n the s t r i c t sense. Changing blood flow would have the e f f e c t of changing the size of the blood pool. Blood 0 2 content at any l e v e l of blood P0 2 was calculated by i n t e r p o l a t i n g p r e v i o u s l y published oxygen d i s s o c i a t i o n curves of Amia (Johansen et a l . 1970). Blood flowing toward the a i r bladder containing low oxygen was designated as a f f e r -ent (aff) blood. Blood d i s t a l to the a i r bladder, a f t e r the accumulation of 0 2 from the a i r bladder, was termed e f f e r e n t (e f f ) blood. A f f e r e n t blood i n the model i s analogous to systemic venous blood returning from the body. This blood i s mixed with 0 2 d i f f u s i n g from the a i r bladder. The P0 2 and 0 2 content of afferent blood was kept constant i n the model. The amount of oxygenation of systemic venous blood i n the animal would r e f l e c t the degree of 0 2 e x t r a c t i o n from r e s p i r i n g t i s s u e s . In other words, the constant afferent input term i s analogous to a constant t i s s u e metabolic rate. The issue of metabolism cannot be addressed here, since only the a i r blad-der and i t s immediate vascular environment i s modeled. E f f e r -ent blood i n the model i s analogous to blood that i s d i s t a l to the a i r bladder and i s a mixture of systemic venous blood and 0 2 ~ s a t u r a t e d blood from the a i r bladder. These two blood pools i n Amia are mixed before entering the heart, p r i o r to entering the v e n t r a l aorta and g i l l vasculature (Randall et a l . 1981). The PCU sensor placed i n t h i s l o c a t i o n i n the 135 model i s thus analogous to a chemoreceptor sensing venous P0 2 between the a i r bladder and the g i l l vasculature i n the a n i -mal. There i s experimental evidence f o r a venous oxygen receptor i n water-breathing f i s h (Barrett and Taylor 1984), and a venous receptor has also been used i n a model of v e n t i -l a t o r y c o n t r o l i n water-breathing f i s h (Taylor et a l . 1968). A t h r e s h o l d value of 20 Torr was used f o r the e f f e r e n t P0 2 sensor, which represents an 0 2 saturation of 60% (Johansen et a l . 1970) measured for ve n t r a l a o r t i c blood i n Amia (Randall et a l . 1981). A i r bladder volume was set i n i t i a l l y at 42 ml, which i s approximately 8% of body mass f o r a 500 g f i s h , based upon measurements from t h i s study. A threshold of 1.5 ml ( i . e . 3 ml k g - 1 ; c f . f i g . 14) l e s s than i n i t i a l a i r bladder volume t r i g g e r e d a simulated 1.5 ml type II i n h a l a t i o n of a i r to replace the l o s t volume. A key f e a t u r e of the model i s the i n c o r p o r a t i o n of a v a r i a b l e error term, associated with both breath types. The error term was added to take into account the observation that Amia usually lose some proportion of the inhaled volume during the transfer of inhaled gas from the buccal cavity to the a i r bladder i n both type I and type II breaths (Deyst and Liem 1985; t h i s study); therefore, gas capture and transfer i s not 100% e f f i c i e n t . The error term was varied over a range from 0 to a maximum of +15% of the mean values for both breath types to examine the e f f e c t s of i n e f f i c i e n t gas exchange on the simulated p a t t e r n of breathing. A random walk si m u l a t i o n (Treloar 1975) was used to approximate a Gaussian d i s t r i b u t i o n 136 f o r breath volumes over the desired error range. For a par-t i c u l a r percentage e r r o r , a pseudo-random number generator picked a value within the range of the approximated d i s t r i b u -t i o n . For instance, i f an error of + 10% of the mean value for a type II breath (1.5 ml) i s considered, then the computer picked a value at random from the approximated Gaussian d i s -t r i b u t i o n ranging from 1.5 ml + (1.5*0.1), or 1.35 ml to 1.65 ml. Results and Comparison with Data from Undisturbed Amia In the absence of any error associated with the simulated breaths ( i . e . 100% e f f i c i e n c y ) , there was a regular a l t e r n a -t i o n of type I and type II a i r breaths ( f i g . 24A) . Type I breaths occurred every 22 min after a type II breath, and type II breaths followed 8 min after each type I breath. Efferent blood P0 2 ( p eff°2^ increased to a maximum of 32 Torr a f t e r each type I breath, then declined to the threshold value of 20 Torr every 3 0 min which triggered another type I breath ( f i g . 24B). I t i s apparent that type II breaths i n the model had very l i t t l e a f f e c t on PeffC>2 o v e r time except s l i g h t l y length-ening the time to reach the P e f f 0 2 t h r e s h o l d . This i s not sur p r i s i n g since each simulated type II breath, 1.5 ml of a i r , contains only about 0.3 ml 0 2 and, therefore, did not appre-c i a b l y a f f e c t the a i r bladder to blood P0 2 gradient. With f i x e d breath volumes, the model produced a constant i n t e r -breath i n t e r v a l f or both types of breaths. In t h i s respect, the Amia model produces a constant simulated breath i n t e r v a l s i m i l a r to the Shelton and Croghan (1988) model for simulated 137 Figure 24. A. Simulated IBI (min) as a function of Cumulative Time (min) f o r the model with 0 e r r o r i n both breaths (see t e x t ) . Open symbols i n d i c a t e type II breaths i n the model output, closed symbols denote type I breaths. B. Simulated changes i n e f f e r e n t blood P 0 2 (Torr) associated with IBI (A, above) as a function of Cumulative Time (min). 138 25n 139 air-gulping events in Electrophorus. The ad d i t i o n of small amounts of random error (+ 10% of the mean breath volume) i n both breath types had a marked e f f e c t on the simulated breathing p a t t e r n ( f i g . 25A). The timing between type I and type II breaths with v a r i a b l e v o l -umes was no longer f i x e d , which gave the o v e r a l l breathing pattern an i r r e g u l a r appearance. The int r o d u c t i o n of error i n t o breath volumes produced a simulated breathing p a t t e r n that q u a l i t a t i v e l y resembled the breathing pattern of Amia i n normoxic conditions ( f i g . 20A). In comparison, the model and data from bowfin, showed an a l t e r n a t i o n between type I and type II a i r breaths, with type II breaths often occurring at short time i n t e r v a l s after type I breaths. In the model, the sh o r t i n t e r v a l type II breaths can be e x p l a i n e d by t h e i r occurrance a f t e r type I breaths that were much less than 100% e f f i c i e n t (<12 ml). In these cases, a i r bladder volume thresh-old was not met and, therefore, a type II breath was t r i g -gered to r a i s e a i r bladder volume above i t s t h r e s h o l d . I f type I breaths were at the high end of the d i s t r i b u t i o n , then a i r bladder volume was a l s o high and the P0 2 t h r e s h o l d was reached before the volume t h r e s h o l d . When t h i s occurred, t h e r e were no type II br e a t h s between s u c c e s s i v e type I breaths. The simulated pattern produced by the model that resem-bles data from bowfin can be p a r t i a l l y explained by examining the simulated P e f f 0 2 values over time ( f i g . 25B). Variable amounts of inhaled a i r contained d i f f e r e n t amounts of 0 2, which changed the a i r bladder to blood P0 ? gradient with each 140 Figure 25. A. Simulated IBI (min) p l o t t e d as a f u n c t i o n of Cumulative Time (min) i n the model with +10% e r r o r i n both breaths. Symbols same as f i g . 26. B. Simulated changes i n efferent blood P0 2 (Torr) associated with IBI (A, above) as a function of Cumulative Time (min). 141 Cumulative Time (min) 36-i 18 1 1 1 1 1 1 1 1 O IOO 200 300 400 Cumulative Time (min) breath. This created variable decay times for P eff02 -*-n w n i ° n one t h r e s h o l d or the other was reached at d i f f e r e n t times depending on the value of blood P0 2 immediately following an a i r breath. Even small v a r i a t i o n s i n breath volume, there-f o r e , had profound e f f e c t s on the q u a l i t a t i v e p a t t e r n of breathing i n the model. The e r r o r i n breath volumes that changed the simulated v e n t i l a t o r y p a t t e r n from constant to variable breath i n t e r v a l s i s analogous to incorporating feed-back delays i n models of human respiratory control (Glass and Mackey 1988). Despite the i r r e g u l a r appearance of simulated breathing from model simulations with the a d d i t i o n of an e r r o r term, s p e c t r a l a n a l y s i s showed that the underlying rhythmicity i s maintained ( f i g . 26) . Ten data sets taken at random from the model, a l l with + 12.5% error i n both breaths, were analyzed by spectral analysis as previously described. The largest peak from the s p e c t r a l analysis of model data was associated with the mean i n t e r v a l of 30 min f o r between successive type I breaths. This r e f l e c t e d the decay i n P e f f 0 2 t n a t occurred a f t e r each type I breath which reached the PeffC>2 threshold, on average, at the same i n t e r v a l i n which there was a f i x e d breath s i z e ( f i g . 24A) . Thus, the d e t e r m i n i s t i c , r e g u l a r breathing pattern evident with fixed breath volumes was quan-t i t a t i v e l y revealed by spectral analysis, despite the incorpo-r a t i o n of an e r r o r term that produced i r r e g u l a r - a p p e a r i n g v e n t i l a t o r y patterns. In undisturbed Amia, a periodic breath-in g p a t t e r n was a l s o a s s o c i a t e d with type I breaths t h a t occurred at 30 min i n t e r v a l s (see Results). The occurrance of 143 Figure 26. Averaged periodogram generated from 10 random data sets from the model with +12.5% error i n both breaths. 144 1 4 5 s i g n i f i c a n t p e r i o d i c i t i e s i n the natural breathing pattern i n undisturbed Amia i s good evidence of agreement between model assumptions and r e a l data. Since breathing frequencies i n the model and from Amia can be attributed to the i n t e r v a l between type I breaths, there i s the strong suggestion that the under-l y i n g p e r i o d i c i t y during a i r - b r e a t h i n g behavior i s dependent on p e r i o d i c feedback from 0 2 chemoreceptors i n contact with the blood. A sensor placed i n the g i l l a r t e r i a l vasculature would probably face a nearly constant environmental P0 2 rather than l a r g e r f l u c t u a t i o n s i n 0 2 t e n s i o n . Therefore, a g i l l a r t e r i a l receptor would not be expected to produce p e r i o d i c large fluctuations i n output since the receptor i n t h i s loca-t i o n probably e q u i l i b r a t e s with the environmental P0 2. The periodic stimulation of type I breathing i n f i s h with a near-constant aquatic P0 2 would p r e d i c t , from model simulations, that a chemoreceptor sensor i n the blood should be located between the a i r bladder and g i l l vasculature before e q u i l i b r a -t i o n with aquatic P0 2 occurs. This does not rule out a bran-c h i a l receptor located in the afferent g i l l vasculature. A i r bladder volume-related breaths are an important element of the breathing pattern in Amia, since they contribute to the appearance of ir r e g u l a r breathing and tend to mask the normal p e r i o d i c i t y of type I breaths. Type II breaths appear to be responsible for making moment by moment adjustments of lung volume on a f i n e r temporal scale than 0 2 ~ r e l a t e d breaths. The model a l s o a c c u r a t e l y p r e d i c t e d the q u a l i t a t i v e changes i n breathing pattern when a e r i a l hyperoxia (100% 0 2) or hypoxia (8% 0 2) was used in place of a i r (21% 0 2) in simu-146 l a t e d breaths. In a e r i a l hyperoxia, PeffC>2 remained above threshold because type II breaths replaced l o s t volume with 100% 0 2, not a i r (21% 0 2 ) , so type I breaths were not stimu-lated. The larger P0 2 gradient resulted i n a greater 0 2 flux from the a i r bladder, causing continuous reductions i n a i r bladder volume that were corrected by type II breaths. The r e s u l t was a simulated breathing pattern dominated exclusively by type II breaths ( f i g . 27A). This i s the pattern displayed by Amia breathing pure 0 2 from the a e r i a l phase ( f i g . 27B) . In simulated a e r i a l hypoxia, the reverse occurred: the blood P0 2 t h r e s h o l d was reached more often than the a i r bladder volume threshold, and the model predicted a greater frequency of type I breaths ( f i g . 28A), as occurs when Amia are made to breathe hypoxic gas ( f i g . 28B). The s i m i l a r i t y between model res u l t s and data from undis-turbed Amia, suggest the basic assumptions incorporated in the model are probably r e a l i s t i c . Model predictions with simulat-ed a e r i a l normoxia, hypoxia and hyperoxia, which c l o s e l y resemble, both q u a l i t a t i v e l y and q u a n t i t a t i v e l y , data from Amia indicate the breathing pattern i s largely governed by two d i s t i n c t inputs. One input monitors intravascular P0 2, while the other monitors changes i n a i r bladder volume, which sug-gests that buoyancy regulation i s probably a major determinant of the v e n t i l a t o r y pattern. Thus, a f u l l understanding of the v e n t i l a t o r y p a t t e r n i n Amia should incorporate a i r bladder volume and blood oxygen information. 147 Figure 27. A. Plot of IBI (min) vs. Cumulative Time (min) for model data with simulated 100% 0 2 in inspired gas and + 15% e r r o r i n both breaths. A l l simulated breaths were type II (open symbols). In t h i s example, Q A B was lowered to 10 ml . — i — i — i min x and D A B 0 2 was reduced to 0.002 ml mm x Torr B. Plot of IBI vs. Cumulative Time for a single data set from one Amia breathing 100% 0 2 from a e r i a l phase in aquatic normoxia shows only type II breaths. 148 251 20-e 1 0 5-1 100 200 300 400 Cumulative Time (min) 149 Figure 28. A. Plot of IBI (min) vs. Cumulative Time (min) for model data with simulated 8% 0 2 i n i n s p i r e d gas and + 10% er r o r i n both types of breaths. Closed symbols denote simu-l a t e d type I breaths, open symbols i n d i c a t e type II breaths. B. IBI vs Cumulative Time for a single data set from one Amia breathing hypoxic gas (8% 0 2) from the a e r i a l phase. Closed symbols i n d i c a t e type I breaths, open symbols denote type II breaths. 150 a 25i 20 15 IO O O IOO 200 300 Cumulative Time (min) 400 25 20 e S 15 IO 5 O O IOO 200 300 400 Cumulative Time (min) 500 151 E v o l u t i o n a r y I m p l i c a t i o n s The current view of the e v o l u t i o n of a i r - b r e a t h i n g i s that lungs o r i g i n a l l y evolved as gas exchange organs and were retained f o r that function i n extant a i r - b r e a t h i n g f i s h such as Amia (see Romer 1972; Randall et a l . 1981). A p r i m i t i v e lung i n f i s h , or any g a s - f i l l e d structure, however, would have a u t o m a t i c a l l y assumed a buoyancy f u n c t i o n . I t i s l i k e l y , then, that s t r a t e g i e s to overcome the problems of buoyancy would have evolved with the development of p r i m i t i v e lungs. The r e s u l t s from t h i s study are important i n suggesting that Amia have evolved two d i f f e r e n t r e s p i r a t o r y s t r a t e g i e s f o r coping with the problems imposed by having a single organ for gas exchange and buoyancy regulation. This raises a question concerning the primary selection pressure for the development of air-breathing i n Amia: did air-breathing evolve f i r s t for gas exchange or to cont r o l buoyancy in the aquatic environ-ment? I t has been suggested t h a t a i r - b r e a t h i n g mechanisms evolved from r e - o r g a n i z a t i o n of p r e - e x i s t i n g neuromuscular patterns for aquatic v e n t i l a t i o n , feeding or coughing (McMahon 1969; Smatresk 1990). ' Presumably, p r i m i t i v e f i s h e s evolved feeding mechanisms much e a r l i e r than a i r - b r e a t h i n g , so the neuromuscular motor patterns would have been i n place before the evolution of air- b r e a t h i n g . Lauder (1980) has suggested that the neuromuscular motor pattern for the feeding mechanism in Amia represents the primitive condition for the Teleostomi. The feeding pattern i n Amia involves a suction mechanism i n which negative pressures are generated by lowering the buccal 152 f l o o r (Lauder 1980). This pattern resembles s u p e r f i c i a l l y a type II breath in which a single aspiratory mechanism creates a negative pressure i n the buccal c a v i t y to inhale a i r . The resemblance between the two patterns suggests that aspiration of a i r or prey into the buccal cavity may have evolved f i r s t . T h i s would argue that a i r bladder volume r e g u l a t i o n , hence buoyancy, rather than gas exchange, may have been the primary s e l e c t i o n pressure f o r a i r - b r e a t h i n g i n Amia. A d d i t i o n a l mechanisms for exhalation could have evolved l a t e r since Amia would have been pre-adapted for inhalation and swallowing a i r . This s c e n a r i o i s , of course, s p e c u l a t i v e , and a l t e r n a t i v e hypotheses are indeed p o s s i b l e . I t does suggest, however, that mechanisms to incorporate the c o n f l i c t i n g demands of gas exchange and buoyancy may have sim u l t a n e o u s l y evolved i n pri m i t i v e f i s h e s . 153 LITERATURE CITED A l l i s , E.P. (1897). The c r a n i a l muscles and c r a n i a l and f i r s t s p i n a l nerves i n Amia calva. J. Morphol. 12:487-772. Aota, S . , K.D. Holmgren, P. G a l l a u g h e r and D.J. R a n d a l l (1990). A possible r o l e for catecholamines i n the v e n t i l a t o r y responses associated with inte r n a l acidosis or external hypox-i a i n rainbow trout Oncorhynchus mykiss. J. Exp. Biol. 151:57-70. Alexander, R.M. (1966). Physical aspects of swimbladder func-t i o n . Biol. Rev. 41:141-176. Babiker, M.M. (1979). Respiratory behaviour, oxygen consump-t i o n and r e l a t i v e dependence on a e r i a l r e s p i r a t i o n i n the A f r i c a n l u n g f i s h (Protopterus annectens, Owen) and an a i r -breathing t e l e o s t (Clarias lazera, C ) . Hydrobiologia 65:177-187. B a l l i n t i j n , CM. (1982). Neural c o n t r o l of r e s p i r a t i o n i n fishes and mammals. In: Exogenous and Endogenous Influences on Metabolic Control, edited by A.D.F. Addink and N. Spronk. pp. 127-140. Oxford:Pergamon Press. Bamford, O.S. (1974). Oxygen reception i n the rainbow trout (Salmo gairdneri). Comp. Biochem. Physiol. 48A:69-76. Bar r e t t , D.J. and E.W. Taylor (1984). Changes i n heart rate during progressive hypoxia in the dogfish, Scyliorhinus cani-cula L.: evidence for a venous oxygen receptor. Comp. Biochem. Physiol. 78A:697-703. Bevelander, G. (1934). The g i l l s of Amia calva s p e c i a l i z e d for r e s p i r a t i o n i n an oxygen d e f i c i e n t habitat. 1934(3):123-127. B i s h a i , H.M. (1961). The e f f e c t of pressure on the d i s t r i b u -t i o n of some Nil e f i s h . J. Exp. Zool. 147:113-124. Bishop, I.R. and G.E.H. Foxon (1968). The mechanism of breath-ing i n the South American lu n g f i s h , Lepidosiren paradoxa; a r a d i o l o g i c a l study. J. Zool. (Lond.). 154:263-271. Boereske, J.R. (1974). A review of the North American f o s s i l amiid fi s h e s . Bull. Mus. Comp. Zool.146:1-87. B o u t i l i e r , R.G. (1984). Chara c t e r i z a t i o n of the inte r m i t t e n t breathing pattern i n Xenopus laevis. J. Exp. Biol. 110:291-309. Brainerd, E.L., K.F. Liem and C.T. Samper (1989). A i r v e n t i l a -t i o n by r e c o i l a s p i r a t i o n i n P o l y p t e r i d f i s h e s . Science. 246:1593-1595. 154 Brockwell, P.J. and R.A. Davis (1991). ITSM: An Interactive Time Series Modelling Package for the PC. New York:Springer-Verlag. Burggren, W.W. (1979). Biomodal [ s i c ] gas exchange during v a r i a t i o n i n environmental oxygen and carbon dioxide i n the a i r breathing f i s h Trichogaster trichopterus. J. Exp. Biol. 82:197-213. Burggren, W.W. and K. Johansen (1986). C i r c u l a t i o n and r e s p i -r a t i o n i n lungfishes (Dipnoi). J. Morph. Suppl. 1:217-236. Burleson, M.L. (1991). Oxygen-sensitive Chemoreceptors and Cardiovascular and Ventilatory Control i n Rainbow Trout. Ph.D. Thesis. Univ. of B r i t i s h Columbia. 153 pp. Burleson, M.L. and W.K. Milsom (1990) . Propranolol i n h i b i t s oxygen-sensitive chemoreceptor a c t i v i t y in trout g i l l s . Am. J. Physiol. (Regulatory Integrative Comp. Physiol. 27):R1089-R1091. Burleson, M.L. and N.J. Smatresk (1990a). Evidence f o r two oxygen-sensitive chemoreceptor l o c i i n channel c a t f i s h , Icta-lurus punctatus. Physiol. Zool. 63:208-221. Burleson, M.L. and N.J. Smatresk (1990b). Effects of section-ing c r a n i a l nerves IX and X on cardiovascular and ve n t i l a t o r y r e f l e x responses to hypoxia and NaCN i n channel c a t f i s h . J. Exp. Biol. 154:407-420. C a r r o l l , R.L. (1988) . Vertebrate Paleontology and Evolution. New York:W.H. Freeman and Co. 698 pp. Carter, G.S. (1957). A i r breathing. In: The Physiology of Fishes, edited by, M.E. Brown. New York: Academic, v o l . 1, p. 65-79. C h a t f i e l d , C. (1980). The Analysis of Time Series: An Intro-duction. London:Chapman and H a l l . 2 68 pp. Davson, H. (1967). Physiology of the Cerebrospinal Fluid. Boston, MA:Little, Brown. Daxboeck, C. and G.F. Holeton (1977). Oxygen receptors i n the rainbow trout, Salmo gairdneri. Can. J. Zool. 56:1254-1259. Daxboeck, C., D.K. Barnard and D.J. Randall (1981). Functional s i g n i f i c a n c e of the g i l l s of the bowfin, Amia calva, with s p e c i a l reference to t h e i r s i g n i f i c a n c e durin air-exposure. Respir. Physiol. 43:349-364. Dean, B. (1898) . On the dogfish (Amia calva) , i t s habits and breeding. State of N.Y., Fish., Game and Forest Comm.4:246-256. 155 DeBoer, R.W., J.M. Karemaker and J. Strackee (1984). Comparing spectr a of a s e r i e s of point events p a r t i c u l a r l y f o r heart rate v a r i a b i l i t y data. IEEE Trans. Biomed. Eng. 31:384-387. DeGraaf. P.J.F., CM. B a l l a n t i j n and F.W. Maes (1987). Mecha-noreceptor a c t i v i t y i n the g i l l s of the carp. I. G i l l filament and g i l l raker mechanoreceptors. Respir. Physiol. 69:173-182. Dejours, P. (1975). Principles of Comparative Respiratory Physiology.Amsterdam: Elsevier/North-Holland. 253 pp. DeLaney, R.G., P. Laurent, R.Galante, A.I. Pack and A.P. Fishman (1983). Pulmonary mechanoreceptors i n the d i p n o i l u n g f i s h Protopterus and Lepidosiren. Am. J. Physiol. 244(Regulatory Integrative Comp. Physiol. 13): R418-R428. Dence, W.A. (1933). Notes on a l a r g e bowfin (Amia calva) l i v i n g i n a mud puddle. Copeia. 1933(1):35. Deyst, K.A. and K.F. Liem (1985). The muscular basis of a e r i a l v e n t i l a t i o n of the primitive lung of Amia calva. Resp. Physi-ol. 59: 213-223. Doan, K.H. (1938). Observations on dogfish (Amia calva) and t h e i r young. Copeia 1938(4):204. Douse, M.A. and G.S. M i t c h e l l (1990). E p i s o d i c r e s p i r a t o r y r e l a t e d discharge i n t u r t l e c r a n i a l motoneurons: in vivo and in vitro studies. Brain Res. 536:297-300. Eclancher, B. and P. Dejours (1975). Controle de l a r e s p i r a -t i o n chez l e s poissons teleosteens: existence de chemorecep-teur s physiologiquement analogues aux chemorecepteurs des vertebres superieurs. C. R. Acad. Sci. Ser. D 280:451-453. Endler, J.A. (1986). Natural Selection in the Wild. Princeton Univ. Press. 336 pp. Evans, B.K. and G. Shelton (1984). V e n t i l a t i o n i n Xenopus laevis a f t e r lung or c a r o t i d body l a b y r i n t h denervation. 1st Int. Congr.: Comp. Physiol. Biochem. Liege, Belgium. Abstract A75. Fange, R. (1976) . Gas exchange i n the swimbladder. In: Respi-ration of Amphibious Vertebrates, edited by G.M. Hughes, pp. 189-211. London:Academic Press. Farber, J. and H. Rahn (1970). Gas exchange between a i r and water and the v e n t i l a t i o n pattern in the e l e c t r i c e e l . Respir. Physiol. 9:151-161. Feldman, J.L., J.C. Smith, H.H. Ellenberger, CA. Connelly, G. Liu, J . J . Greer, A.D. Lindsay and M.R. Otto (1990). Neurogene-s i s of respiratory rhythm and pattern: emerging concepts. Am. J. Physiol. 259(Regulatory Integrative Comp. Physiol. 28): R879-R886. 156 F i l i p s k i , G.T. and M.V.H. Wilson (1984). Sudan black B as a nerve s t a i n for whole cleared fishes. Copeia 1984(1):204-208. French, A.S. and A.V. Holden (1971). A l i a s - f r e e sampling of neuronal spike t r a i n s . Kybernetik 8:165-171. Funk, G.D., C.L. Webb and W.K. Milsom (1986). Non-invasive measurement of r e s p i r a t o r y t i d a l volume i n a q u a t i c , a i r -breathing animals. J. Exp. Biol. 126:519-523. Gans, C. (1970). Strategy and sequence i n the evolution of the external gas exchangers of ectothermal vertebrates. Forma et Functio 3:61-104. Garey, W.F. and H. Rahn (1970). Normal a r t e r i a l gas tensions and pH and the breathing frequency. Respir. Physiol. 9:141-150. Gee, J.H. (1980). Respiratory patterns and antipredator re-sponses i n the c e n t r a l mudminnow, Umbra limi, a continuous, fa c u l a t i v e , air-breathing f i s h . Can. J. Zool. 58:819-827. Gee, J.H. (1981). Coordination of respiratory and hydrostatic functions of the swimbladder i n the c e n t r a l mudminnow, Umbra limi. J. Exp. Biol. 92:37-52. Gee, J.H. and J.B. Graham (1978). Respiratory and hydrostatic f u n c t i o n s of the i n t e s t i n e of the c a t f i s h e s Hoplosternum thoracatum and Brochis splendens ( C a l l i c h t h y i d a e ) . J. Exp. Biol. 74:1-16. G l a s s , L. and M.C. Mackey (1988). From Clocks to Chaos. Princeton, NJ: Princeton Univ. Press. 248 pp. Glass, M.L., W.W. Burggren and K. Johansen (1981). Pulmonary d i f f u s i n g capacity of the b u l l f r o g (Rana catesbeiana). Acta Physiol. Scand. 113:485-490. Glass, M.L., R.G. B o u t i l i e r and N. Heisler (1983). Ventilatory control of a r t e r i a l P0 2 in the t u r t l e Chrysemys picta b e l l i i : e f f e c t s of temperature and hypoxia. J . Comp. Physiol. 151:145-153 . Goodman, L. (1964). O s c i l l a t o r y behavior of v e n t i l a t i o n i n re s t i n g man. IEEE Trans. Biomed. Eng. 11:82-93. Graham, M.S., J.D. Turner and CM. Wood (1990). Control of v e n t i l a t i o n i n the hypercapnic skate Raja ocellata: I. Blood and extradural f l u i d . Respir. Physiol. 80:259-277. Greenwood, P.H. and K.F. Liem (1984). Aspiratory r e s p i r a t i o n i n Arapaima gigas (Teleostei, Osteoglossomorpha): A reapprais-a l . J. Zool. (Lond.). 203:411-425. 157 G r i g g , G.C. (1965a). Studies on the Queensland l u n g f i s h , Neoceratodus forsteri ( K r e f f t ) . I. Anatomy, h i s t o l o g y , and functioning of the lung. Aust. J. Zool. 13:243-253. Gri g g , G.C. (1965b). Studies on the Queensland l u n g f i s h , Neoceratodus forsteri ( K r e f f t ) . I I I . A e r i a l r e s p i r a t i o n i n r e l a t i o n to habits. Aust. J. Zool. 13:413-421. H i t z i g , B.M. and D.C. Jackson (1978). Central chemical control of v e n t i l a t i o n i n the unanesthetized t u r t l e . Am. J. Physiol. 235(Regulatory Integrative Comp. Physiol.): R257-R264. H l a s t a l a , M.P., B.Wranne and C.J. Lenfant (1973). C y c l i c a l v a r i a t i o n s i n FRC and other r e s p i r a t o r y v a r i a b l e s i n r e s t i n g man. J. Appl. Physiol. 34:670-676. Horn, M.H. and CD. Riggs (1973). E f f e c t s of temperature and l i g h t on the rate of a i r breathing of the bowfin Amia calva. Copeia 1973 (4): 653-657. Hughes, G.M. and G. Shelton (1962). Respiratory mechanisms and t h e i r nervous c o n t r o l i n f i s h . Adv. Comp. Physiol. Biochem. 1:275-364. Ishimatsu, A. and Y. Itazawa (1981). V e n t i l a t i o n of the a i r -breathing organ i n the snake-head Channa argus. Jpn. J. Ich-thyol. 28:276-282. Janssen, R.G. and D.J. Randall (1975). The ef f e c t s of changes i n pH and PC0 2 i n blood and water on breathing i n rainbow trout Salmo gairdneri. Respir. Physiol. 25:235-245. Johansen, K. (1970). A i r breathing in fishes. In: Fish Physi-ology IV, edited by W.S. Hoar and D.J. Randall, pp. 361-411. London:Academic. Johansen, K. and C. Lenfant (1967). Respiratory function i n the South American lungfish. J. Exp. Biol. 46:205-218. Johansen, K., C. Lenfant and G. Grigg (1967). Respiratory control i n the lungfish Neoceratodus forsteri ( K r e f f t ) . Comp. Biochem. Physiol. 20:835-854. Johansen, K. and C. Lenfant (1968). Respiration i n the African lungfish Protopterus aethiopicus. II. Control of breathing. J. Exp. Biol. 49:453-468. Johansen, K., C. Lenfant, K. Schmidt-Nielsen and J.A. Peterson (1968). Gas exchange and control of breathing in the e l e c t r i c e e l , Electrophorus electricus. Z. Vergl. Physiol. 61:137-163. Johansen, K., D. Hanson and C. Lenfant. (1970). Respiration i n a pri m i t i v e a i r breather Amia calva. Respir. Physiol. 9: 162-174. 158 Jones, D.R. (1983). Ontogeny and phylogeny of the oxygen response. Proc. Physiol. Soc.NZ 3:79-81. Jones, D.R. and W.K. Milsom (1982). P e r i p h e r a l r e c e p t o r s a f f e c t i n g b r e a t h i n g and c a r d i o v a s c u l a r f u n c t i o n i n non-mammalian vertebrates. J. Exp. Biol. 100:59-91. Jones, F.R.H. (1957) . The swimbladder. In:The Physiology of Fishes, V o l . 2, e d i t e d by M.E. Brown, pp. 305-322. New York:Academic Press. Jones, F.R.H. and N.B. Marshall (1953). The s t r u c t u r e and functions of the teleostean swimbladder. Biol. Rev. 28:16-83. Kawasaki, R. (1980). Maintenance of r e s p i r a t o r y rhythm-generation by vascular perfusion with physiological saline i n the i s o l a t e d head of the carp. Jpn. J. Physiol. 30:575-589. Knox, C.K. (1973). Characteristics of i n f l a t i o n and d e f l a t i o n r e f l e x e s d u r i n g e x p i r a t i o n i n the c a t . J. Neurophysiol. 36:284-295. Kramer, D.L. (1978). V e n t i l a t i o n of the respiratory gas blad-der i n Hoplerythrinus unitaeniatus ( P i s c e s , C h a r a c o i d e i , Erythrinidae). Can. J. Zool. 56:931-938. Kramer, D.L. (1987). D i s s o l v e d oxygen and f i s h behavior. Environ. Biol. Fish. 18:81-92. L a h i r i , S., J.P. Szidon and A.P. Fishman (1970). P o t e n t i a l r e s p i r a t o r y and c i r c u l a t o r y adjustments to hypoxia i n the Afr i c a n lungfish. Fed. Proc. 29:1141-1148. Lauder, G.V. (1980) . Evolution of the feeding mechanism i n p r i m i t i v e A c t i n o p t e r y g i a n f i s h e s : a f u n c t i o n a l anatomical a n a l y s i s of Polypterus, Lepisosteus, and Amia. J. Morph.163:283-317. Lauder, G.V. and K.F. Liem (1983). The evolution and i n t e r r e -l a t i o n s h i p s of the Actinopterygian f i s h e s . Bull. Mus. Comp. Zool.150:95-197 . Laurent, P. and J.D. Rouzeau (1972) . Afferent neural a c t i v i t y from the pseudobranch of t e l e o s t s . Respir. Physiol. 14:307-331. Liem, K.F. (1980). A i r v e n t i l a t i o n i n advanced t e l e o s t s : biomechanical and e v o l u t i o n a r y aspects. In: Environmental Physiology of Fishes, e d i t e d by M.A. A l i . pp. 57-91. New York:Plenum Press. Liem, K.F. (1984). The muscular basis of aquatic and a e r i a l v e n t i l a t i o n i n the air-breathing teleost f i s h Channa. J. Exp. Biol. 113:1-18. 159 Liem, K.F. (1987). Functional design of the a i r v e n t i l a t i o n apparatus and overland excursions by t e l e o s t s . Fieldiana (NS) 37:1-29. Liem, K.F. (1988). Form and function of lungs: the evolution of a i r breathing mechanisms. Am. Zool. 28:739-759. Liem, K.F. (1989). R e s p i r a t o r y gas bladders i n t e l e o s t s : Functional conservatism and morphological d i v e r s i t y . Am. Zool. 29: 333-352. Lomholt, J.P. and K. Johansen (1974). Control of breathing i n Amphipnous cuchia, an amphibious f i s h . Respir. Physiol. 21:325-340. McKenzie, D.J. (1990). V e n t i l a t i o n in Amia calva: A Comparison with Water-Breathing F i s h . Ph.D. Thesis. Univ. of B r i t i s h Columbia. 191 pp. McMahon, B.R. (1969). A functional analysis of the aquatic and a e r i a l respiratory movements of an African lungfish, Protopt-erus aethiopicus, with reference to the evolution of the lung-v e n t i l a t i o n mechanism i n vertebrates. J. Exp. Biol. 51: 407-430. M c N i f f , J.M., A.I. Pack, R.J. Galante and A.P. Fishman. (1987). Innervation of the g l o t t i c musculature in the African lungfish (Protopterus). Fed. Proc. 46(3): 344. Millhorn, D.E. and F.L. Eldridge (1986). Role of vent r o l a t e r a l medulla i n r e g u l a t i o n of r e s p i r a t o r y and cardiovascular sys-tems. J. Appl. Physiol. 61:1249-1263. Milsom, W.K. (1990). Mechanoreceptor modulation of endogenous r e s p i r a t o r y rhythms i n v e r t e b r a t e s . Am. J. Physiol. 259 (Regulatory Integrative Comp. Physiol. 28):R898-R910. Milsom, W.K. (1991). Intermittent breathing i n vertebrates. Ann. Rev. Physiol. 53:87-105. Milsom, W.K., D.R. Jones and G.R.J. Gabbott (1981). On chemo-receptor c o n t r o l of v e n t i l a t o r y responses to C0 2 i n unanaes-thetized ducks. J. Appl. Physiol. 50:1121-1128. Milsom, W.K. and D.R. Jones. (1985). Ch a r a c t e r i s t i c s of mecha-noreceptors i n the a i r - b r e a t h i n g f i s h , Amia calva. J. Exp. Biol. 117: 389-399. Milsom, W.K. and R.W. B r i l l (1986). Oxygen se n s i t i v e afferent information a r i s i n g from the f i r s t g i l l arch of y e l l o w f i n tuna. Respir. Physiol. 66:193-203. Milsom, W.K. and P. Chan (1986). The re l a t i o n s h i p between lung volume, respiratory drive and breathing pattern in the t u r t l e , Chrysemys picta. J. Exp. Biol. 120:233-247. 160 M i t c h e l l , R.A., H.H. Loeschcke, J.W. Severinghaus, B.W. Rich-ardson and W.H. Massion (1963). Regions of respiratory chemo-s e n s i t i v i t y on the surface of the medulla. Ann. N.Y. Acad. Sci. 109:661-681. Mott, J.C. (1951). Some facto r s a f f e c t i n g the blood c i r c u l a -t i o n i n the common e e l (Anguilla anguilla). J. Physiol. (Lond.). 114:387-398. Mull i g a n , E. and L a h i r i , S. (1981). Dependence of c a r o t i d chemoreceptor s t i m u l a t i o n by metabolic agents on Pa0 2 and PaC0 2. J. Appl. Physiol. 50:884-891. N e i l l , W.T. (1950). An est i v a t i n g bowfin. Copeia. 1950:240. Nilsson, S. (1984). Innervation and pharmacology of the g i l l s . I n : F i s h Physiology, V o l . XA, edi t e d by W.S. Hoar and D.J. Randall, pp. 185-227. New York:Academic Press. Ohya, N., J . Huang and T. Fukunaga (1988). A new method for measuring the pneumotachometer c h a r a c t e r i s t i c s using a sy-ringe. Jpn. J. Physiol. 38:577-584. Pack, A.I. (1981). Sensory inputs to the medulla. Ann. Rev. Physiol. 43:73-90. Pack, A.I., R.J. Galante and A.P. Fishman (1984). Breuer-Hering r e f l e x e s i n the A f r i c a n l u n g f i s h (Protopterus annec-tens). Fed. Proc. 43:433. Pack, A.I., D.A. Silage, R.P. Millman, H. Knight, E.T. Shore and D.-C. C. Chung (1988). Spectral analysis of v e n t i l a t i o n in el d e r l y subjects awake and asleep. J. Appl. Physiol. 64:1257-1267 . Pack, A.I., R.J. Galante and A.P. Fishman (1990). Control of interbreath i n t e r v a l in the African lungfish. Am. J. Physiol. 259(Regulatory Integrative Comp. Physiol. 29): R139-R146. Packard, G.C. (1974). The evolution of air-breathing i n paleo-zoic gnathostome fishes. Evolution 28:320-325. Pappenheimer, J.R., V. Fencl, S.R. Heisey and D. Held (1965). Role of cerebral f l u i d s i n con t r o l of r e s p i r a t i o n as studied i n unanesthetized goats. Am. J. Physiol. 208:436-440. Pearse, A.G.E. (1969). The cytochemistry and ult r a s t r u c t u r e of polypeptide hormone-producing c e l l s of the APUD series and the embryologic, ph y s i o l o g i c and pathologic implications of the concept. J. Histochem. Cytochem. 17:303-313. P e t t i t , M.J. and T.L. Beitinger (1985). Oxygen ac q u i s i t i o n of the r e e d f i s h , Erpetoichthys calabaricus. J. Exp. Biol. 114:289-306. 1 6 1 Peyraud-Waitzenegger, M. (1979). Simulataneous modifications of v e n t i l a t i o n and a r t e r i a l P0 2 by catecholamines i n the eel, Anguilla anguilla L.: p a r t i c i p a t i o n of alpha and beta e f f e c t s . J. Comp. Physiol. 129:343-354. Poole, CA. and G.H. Satchell (1979). Nociceptors i n the g i l l s of the dogfish Squalus acanthias. J. Comp. Physiol. A 13 0:1-7. Powers, E.B. and R.T. Clark (1942). Control of normal breath-ing i n fishes by receptors located in the regions of the g i l l s and innervated by the IXth and Xth c r a n i a l nerves. Am. J. Physiol. 138:104-107. Rahn, H., K.B. Rahn, B.J. Howell, C.Gans and S.M. Tenney (1971). A i r breathing of the g a r f i s h (Lepisosteus osseus). Respir. Physiol. 11:285-307. Randall, D.J. and D.R. Jones (1973). The e f f e c t of deafferen-t a t i o n of the pseudobranch on the r e s p i r a t o r y response to hypoxia and hyperoxia in the trout (Salmo gairdneri). Respir. Physiol. 17:291-3 01. Randall, D.J., W.W. Burggren, A.P. F a r r e l l and M.S. Haswell (1981). The Evolution of Air Breathing in Vertebrates. Cam-bridge, UK:Cambridge Univ. Press. 133 pp. Ra n d a l l , D.J., J.N. Cameron, C. Daxboeck and N. Smatresk (1981). Aspects of bimodal gas exchange i n the bowfin Amia calva L. ( A c t i n o p t e r y g i i : Amiiformes) . Respir. Physiol. 43: 339-348. Reighard, J. (1903). The natural history of Amia calva Linnae-us. Mark Anniversary Volume: 59-109. Richter, D.W., D. Ballantyne and J.E. Remitters (1986). How i n the r e s p i r a t o r y rhythm generated? A model. News in Physiol. Sci. 1(June):109-112. Romer, A.S. (1972). Skin breathing - primary or secondary? Respir. Physiol. 14:183-192. Romer, A.S. and T.S. Parsons (1977). The Vertebrate Body. F i f t h E d i tion. Philadelphia: W.B. Saunders. Rovainen, CM. (1974). Respiratory motoneurons i n lampreys. J. Comp. Physiol. 94:57-68. Rovainen, CM. (1977) . Neural co n t r o l of v e n t i l a t i o n i n the lamprey. Fed. Proc. 36:2386-2389. Saunders, R.L. and A.M. S u t t e r l i n (1971). Cardiac and r e s p i r a -t o r y responses to hypoxia i n the sea raven, Hemitripterus americanus, and an i n v e s t i g a t i o n of possible c o n t r o l mecha-nisms. J. Fish. Res. Bd. Can. 28:491-503. 162 Shams, H. (1985). D i f f e r e n t i a l e f f e c t s of C0 2 and H + as cen-t r a l s t i m u l i of r e s p i r a t i o n i n the cat. J. Appl. Physiol. 58:357-364. Shelton, G., D.R. Jones and W.K. Milsom (1986). Cont r o l of breathing i n ectothermic vertebrates. In: Handbook of Physiol-ogy, section 3, The Respiratory System, vol. II, Control of Breathing, part 2, edited by A.P. Fishman, N.S. Cherniak, J.G. Widdicombe and S.R. Geiger. pp. 857-909. MD: American Physio-l o g i c a l Society. Shelton, G. and P.C. Croghan (1988). Gas exchange and i t s c o n t r o l i n non-steady s t a t e systems: the consequences of evolution from water to a i r breathing i n the vertebrates. Can. J. Zool. 66:109-123. Smatresk, N.J. (1986). V e n t i l a t o r y and c a r d i a c r e f l e x r e -sponses to hypoxia and NaCN i n Lepisosteus osseus, an a i r -breathing f i s h . Physiol. Zool. 59:385-397. Smatresk, N.J. (1987). Vagal a f f e r e n t c o n t r o l over water and a i r - b r e a t h i n g p a t t e r n s i n Lepisosteus oculatus, an a i r -breathing f i s h . Am. Zool. 27(4):111A. Smatresk, N.J. (1990). Chemoreceptor modulation of endogenous respiratory rhythms in vertebrates. Am. J. Physiol. (Regulato-ry Integrative Comp. Physiol. 28): R887-R897. Smatresk, N.J. and J.N. Cameron (1982a). Respiration and acid-base physiology of the spotted gar, a bimodal breather. I. Normal values, and the response to severe hypoxia. J. Exp. Biol. 96:263-280. Smatresk, N.J. and J.N. Cameron (1982b). Respiration and acid-base physiology of the spotted gar, a bimodal breather. I I I . Response to a transfer from fresh water to 50% sea water, and control of v e n t i l a t i o n . J. Exp. Biol. 96:295-306. Smatresk, N.J., M.L. Burleson and S.Q. A z i z i (1986). Chemore-f l e x i v e responses to hypoxia and NaCN i n longnose gar: e v i -dence for two chemoreceptor l o c i . Am. J. Physiol. (Regulatory Integrative Comp. Physiol. 21): R116-R125. Smatresk, N.J. and S.Q. A z i z i (1987). Charact e r i s t i c s of lung mechanoreceptors in spotted gar, Lepisosteus oculatus. Am. J. Physiol. 252 (Regulatory Integrative Comp. Physiol. 21): R1066-R1072. Smatresk, N.J. and A.W. Smits (1991). E f f e c t s of c e n t r a l and p e r i p h e r a l chemoreceptor s t i m u l a t i o n on v e n t i l a t i o n i n the marine toad, Bufo marinus. Respir. Physiol. 83:223-238. Smith, R.S. and D.L. Kramer (1986). The e f f e c t of apparent predation r i s k on the respiratory behavior of the F l o r i d a gar (Lepisosteus platyrhincus). Can. J. Zool. 64:2133-2136. 163 Smith, F.M. and D.R. Jones (1982) . The e f f e c t of changes i n blood oxygen-carrying capacity on v e n t i l a t i o n volume i n the rainbow trout (Salmo gairdneri). J. Exp. Biol. 97:325-334. Taylor, W., A.H. Houston and J.D. Horgan (1968). Development of a computer model simulating some aspects of the cardiovas-cular - r e s p i r a t o r y dynamics of the salmonid f i s h . J. Exp. Biol. 49:477-493. T r e l o a r , L.R.G. (1975). The Physics of Rubber Elasticity. Oxford Univ. Press 310 pp. van den Aardweg, J.G. and J.M. Karemaker (1991). Respiratory v a r i a b i l i t y and associated cardiovascular changes in adults at r e s t . Clin. Physiol. 11:95-118. Waggener, T.B., I.D. Frantz III, A.R. Stark and R.E. Kronauer (1982). O s c i l l a t o r y b r e a t h i n g p a t t e r n s l e a d i n g to apneic s p e l l s i n infants. J. Appl. Physiol.:Respirat. Environ. Exer-cise Physiol. 52:1288-1295. Walker, R., R.J. Galante, A.P. Fishman and A.I. Pack (1990). E f f e c t of GABA on g i l l and lung v e n t i l a t i o n i n an in vitro i s o l a t e d brainstem preparation i n the tadpole. The Physiolo-gist. 33:A-35. Wilder, B.G. (1877). On the r e s p i r a t i o n of Amia. Proc. Am. Acad. Adv. Sci. 26: 306-313. Withers, P.C. and S.S. Hillman (1988). A steady-state model of maximal oxygen and carbon dioxide transport in anuran amphib-ians. J. Appl. Physiol. 64:860-868. Wood, CM., J.D. Turner, R.S. Munger and M.S. Graham (1990). Control of v e n t i l a t i o n i n the hypercapnic skate Raja ocellata: I I . Cerebrospinal f l u i d and i n t r a c e l l u l a r pH i n the brain and other ti s s u e s . Respir. Physiol. 80:271-290. Youson, J.H. (1976). Fine structure of granulated c e l l s i n the p o s t e r i o r c a r d i n a l and renal veins of Amia calva L. Can. J. Zool. 54:843-851. Zar, J.H. (1974). Biostatistical Analysis. Englewood C l i f f s , NJ:Prentice-Hall, Inc. 164 APPENDIX 1 Table A l . Summary of i n i t i a l parameters used i n the model of intermittent breathing. Values based on a 500g bowfin. Parameter A i r Bladder Volume Abbreviation A i r Bladder 0 2 Content A i r Bladder P0 o A i r Bladder Blood Flow A i r Bladder D i f f u s i o n Capacity Afferent 0 2 Content Efferent 0 2 Content Afferent P0 2 Efferent P0 2 H i l l n C o e f f i c i e n t H i l l K C o e f f i c i e n t VAB VAB°2 PAB°2 A^B DAB°2 Caff°2 Ceff°2 Paff°2 Peff°2 Value 42 ml 4.32 ml 0-80 Torr 25 ml min -1 0.0085 ml min" 1 T o r r " 1 0.0443 ml 0 2/ml blood Variable (= c a f f ° 2 ' at threshold) 20 Torr Variable (= P a f f 0 2 , at threshold) 1.58 0.0116 Outline of the Model The model s t a r t e d with i n i t i a l values f o r a i r bladder and blood parameters given i n Table A l . Oxygen fl u x (A0 2/At) was c a l c u l a t e d i n two steps, one d i f f u s i v e and the other convec-t i v e (see Discussion). The d i f f u s i v e step, from a i r bladder to blood, r e q u i r e d a P0 2 g r a d i e n t to support an 0 ? f l u x . In 165 o r d e r t o e s t a b l i s h t h e a i r b l a d d e r t o b l o o d P 0 2 g r a d i e n t , a v a l u e f o r mean bloo d P0 2 ( pb°2^ w a s c a l c u l a t e d . T h i s was a c c o m p l i s h e d i n E q u a t i o n s 1 t o 3. An e f f e r e n t -a f f e r e n t 0 2 content g r a d i e n t ( c e f f _ a f f ° 2 ) : ceff-aff°2 = ceff°2 " caff°2 I t i s apparent t h a t a t i n i t i a l v a l u e s c e f f - a f f ° 2 e c 3 u a l s zero. An average b l o o d 0 2 c o n t e n t ( C b 0 2 ) i s then c a l c u l a t e d from ceff-aff°2 a s ' C b ° 2 " n C e f f „ a f f 0 2 ) / 2 ] + C a f f 0 2 [2], which equals C a^^0 2 when c e f f - a f f ° 2 i S z e r o * c b ° 2 a S u s e c * t o c a l c u l a t e an a v e r a g e b l o o d P 0 2 ( P b 0 2 ) . The r e l a t i o n s h i p between b l o o d oxygen s a t u r a t i o n (S) and b l o o d P 0 2 i s n o t l i n e a r , but can be approximated as a s i g m o i d a l f u n c t i o n de-s c r i b e d by the H i l l e q u ation: K P n S/100 = [3] . 1 + KP n V a l u e s f o r K and n (Table A l ) were c a l c u l a t e d from p u b l i s h e d oxygen d i s s o c i a t i o n c u r v e s f o r Amia (Johansen e t a l . 1970). E q u a t i o n 3 was l i n e a r i z e d by l o g - t r a n s f o r m a t i o n and rearranged t o c a l c u l a t e P^C^ from C b 0 2 • E q u a t i o n 3 was used f o r C b 0 2 v a l u e s r a n g i n g from 0 t o 0.0702 ml 0 2/ml b l o o d , which i s 90% s a t u r a t e d w i t h oxygen a t 0.0702 ml Op/ml b l o o d , and c o r r e -16 6 sponds to a blood P0 2 of 67 Torr. A simple l i n e a r regression was used t o c a l u l a t e c o n t e n t and P 0 2 between 0 .0702 and 0.07988 ml 0 2/ml blood (100% s a t u r a t i o n ) . An upper l i m i t of 110 Torr was set for the blood P0 2 corresponding with 100% 0 2 saturation. The determination of P^C^ thus allowed c a l c u l a t i o n of oxygen f l u x ( A 0 2/At), which was dependent upon a i r blad-der d i f f u s i o n capacity ( D A B0 2) and the a i r bladder to blood P0 2 gradient, ]]>[ A0 2/At = D A B 0 2 AP0 2/At [4], where AP0 2/At was the d i s c r e t e change i n the a i r bladder to blood P0 2 gradient. At each step, AP0 2/At was the difference between P A B 0 2 and Pj~,02 . Since the change i n a i r bladder volume (AV A B) and a i r bladder 0 2 content ( a vAB^2^ w e r e depend-ent only upon the oxygen flux , AV A B/At = V A B - ^ A 0 2 / A t [5a], t and A V a b 0 2 = V A B 0 2 - ^ A 0 2 / A t [5b]. t A f t e r 0 2 d i f f u s e d from the a i r bladder, the new P A B 0 2 was calculated from the changes i n V A B and v A B 0 2 as: P A B ° 2 = ( V AB°2/ V AB> * ( P B " W V P ) ^ ' where P B i s barometric pressure at sea l e v e l (760 Torr) and wvp i s the water vapor pressure at 22 °C (20 T o r r ) . 167 The oxygen fl u x calculated from Eq. 4 was used to deter-mine the amount of 0 2 added to e f f e r e n t blood at each i t e r a -t i o n (see Discussion). The c a l c u l a t i o n of A0 2/At and Q A B at each step by the i t e r a t i o n procedure integrated each variable so volumes of 0 2 and blood were produced, from which C e f f 0 2 was calculated, £ A0 2/At Ceff°2 = ™ + Caff°2 ^'3 • t The new C e^^0 2 value was used as the input variable for equa-t i o n 1. Equations 1-7 were solved numerically i n d i s c r e t e steps u n t i l e i t h e r a i r bladder volume threshold (40.5 ml) or a r t e r i a l blood P0 2 threshold (20 Torr) was reached. I f the volume threshold was met, the program added 1.5 ml of a i r to the a i r bladder and new parameters were calculated and used i n the next i t e r a t i o n . I f blood P0 2 threshold was reached, the program decreased a i r bladder volume to an end e x h a l a t i o n volume of 30 ml, and then added 12 ml of a i r . A f t e r each simulated breath, Equations 1-7 were again i t e r a t i v e l y solved to calculate new blood and a i r bladder parameters. 168 APPENDIX 2 PROGRAM AmexII3; { A p r i l 27, 1991 } { This program was written i n Pascal using Borland's Turbo Pascal compiler (Version 5.0) with Borland's Turbo Pascal Graphix Toolbox (Version 4.0). ) USES Crt, Dos, Gdriver, P r i n t e r , Gkernel, GWindow, Gshell; CONST { Set i n i t i a l constants. ) VenousPo2 =20; { Torr } VenousContent = 0.0443; { Volumes % } BloodFlow =25; { Mis/ Minute } VAR ( Set global variables. ) DataFileName : STRING; TypeOne, TypeTwo : REAL; One, Two : REAL; PROCEDURE GetFileName; { Get f i l e name to store generated values of blood Po2 and lung volume. } VAR Name : STRING; BEGIN WRITELN('This i s the Normoxic Version.'); WRITELNJ'What i s the F i l e Name?'); READLN(Name); DataFileName := 'B:'+ Name + '.DAT'; WRITE('What i s the Noise Level i n the Type One Breaths? (%)'); READLN(One); WRITE('What i s the Noise Level i n the Type Two Breaths? (%)'); READLN(Two) ; TypeOne := (One/100); TypeTwo := (Two/100); END; {GetFileName} PROCEDURE GetArterialPressure(Var A, B, M, N, S, T : PlotArray); { Generates the values for blood Po2 and lung volume for ) { p l o t t i n g . ) 169 VAR { Variables l o c a l to procedure "GetArterialPressure". } ArterialContent, ArterialPressure : REAL; LungPo2, LungVolume, Lung02Volume : REAL; Lung02Content, LungBloodGradient : REAL; I : INTEGER; DataFile : TEXT; FUNCTION Log(X : REAL) : REAL; BEGIN Log := Ln(X)/Ln(10); END; FUNCTION PressureFromContent(ArterialContent : REAL) : REAL; { Calculates blood Po2 from blood content using data from Johansen et a l . (1970). The sigmoid curve i s solved from a content of 0.00 to 0.0702 ml 02/ ml blood. I t i s l i n e a r l y approximated from 0.0702 to 0.07988 ml 02/ ml blood. } VAR b, c, Saturation : REAL; BEGIN IF ArterialContent > 0.07988 THEN PressureFromContent := 110.0 ELSE IF ArterialContent > 0.0702 THEN PressureFromContent := (ArterialContent * 4438.016529) - 244.5087 ELSE BEGIN Saturation := (ArterialContent/7.7988) * 100; b := Log(Saturation/(1 - Saturation)); c := (b + 1.9355)/1.582264; PressureFromContent := EXP(c * Ln(10)); END E N D ; PROCEDURE Iterate; { Calculate new values f o r a l l physiological v a r i a b l e s . ) VAR MeanArterioVenousContent, LungBloodGradient : REAL; Hold, 02Flux, ArterioVenousGradient : REAL; 170 BEGIN ArterioVenousGradient := (ArterialContent - VenousContent); MeanArterioVenousContent := (ArterioVenousGradient/2) + VenousContent; Hold := PressureFromContent(MeanArterioVenousContent); LungBloodGradient := LungPo2 - Hold; 02Flux := 0.0085 * LungBloodGradient; LungVolume := LungVolume - 02Flux; Lung02Volume := Lung02Volume - 02Flux; LungPo2 := (Lung02Volume/LungVolume) * 740; { PB(760) - wvp(20) = 740 } ArterialContent := VenousContent + (02Flux / BloodFlow); ArterialPressure := (PressureFromContent(ArterialContent)); END; PROCEDURE TypelE; { Exhale for Type I breath. } BEGIN LungVolume := 30; { ml } Lung02Volume := (LungPo2/740)*LungVolume; { ml } END; PROCEDURE Typell; { Inhale for Type I breath. ) VAR DeltaLungVolume, Test : REAL; Al, A2, SinAl, SinA2 : REAL; DeltaA : REAL; BEGIN Al := Random * 360; { Two-step random walk simulation. ) SinAl := Sin(Al * (PI/180)); A2 := Random * 360; SinA2 := Sin(A2 * (PI/180)); DeltaA := (SinAl + SinA2)/2; Test := (12 * TypeOne) * DeltaA; DeltaLungVolume := 12 - Test; LungVolume := LungVolume + DeltaLungVolume; Lung02Volume := Lung02Volume + (DeltaLungVolume * 0.21); LungPo2 := (Lung02Volume/LungVolume) * 740; END; PROCEDURE Type2; £ Inhale for Type II breath. } VAR DeltaLungVolume, Hold : REAL; Bl, B2, SinBl, SinB2, DeltaB : REAL; 171 BEGIN Bl := Random * 360; SinBl := Sin(Bl*(PI/180)); B2 := Random * 360; SinB2 := Sin(B2*(PI/180)); DeltaB := (SinBl + SinB2)/2; Hold := (1.5 * TypeTwo) * DeltaB; DeltaLungVolume := 1.5 - Hold; LungVolume := LungVolume + DeltaLungVolume; { ml ) Lung02Volume := Lung02Volume + (DeltaLungVolume * 0.21); LungPo2 := (Lung02Volume/LungVolume) * 740; END; BEGIN { GetArterialPressure } Assign (DataFile, DataFileName); Rewrite (DataFile); RANDOMIZE; I := 1; { Set i n i t i a l conditions. } ArterialPressure := 20; { Torr } ArterialContent := 0.044 3; { mis } LungVolume := 40; { ml } Lung02Volume := 5.945911; { ml ) LungPo2 := 80; { Torr } LungBloodGradient := 60.0; = I; = ArterialPressure; = I; = LungVolume; := 1 to 400 DO = I, A [ l , l ] A[l,2] M [ l , l ] M[l,2] FOR I BEGIN A[I,1] A[I,2] M[I,1] M[I,2] S[I,1] S[I,2] Iterate; IF LungVolume < 40.5 THEN BEGIN Type2; S[I,1] S[I,2] END ELSE IF ArterialPressure < 20.1 THEN ArterialPressure; I; LungVolume; I; = 2; I; 4 172 BEGIN TypeIE; S[I,1] := I; S[I,2] := 0; Ty p e l l ; END; WRITELN(DataFile, A[I,1], A[I,2], M[I,2], S[I, END; Close (dataFile); END; { GetArterialPressure } PROCEDURE ComputeAndDisplay; VAR A, B : PlotArray; M, N : PlotArray; S, T : PlotArray; Ch : Char; XI, X2 : Integer; OneStr, TwoStr : STRING; BEGIN ClearScreen; GetArterialPressure(A, B, M, N, S, T); DefineWindow(l, 0, 5,' XMaxGlb, YMaxGlb-220); DefineHeader(l, 'Blood Po2 Through Time. ' ) ; Defineworld(1, 0, 15, 405, 50); SelectWorld(1); SelectWindow(1); SetBackground(0); SetForeGroundColor(12); SetHeaderOn; DrawBorder; DrawAxis(7, -8, 0, 0, 0, 0, 0, 0, False); DrawPolygon(A, 10, 400, 0, 0, 0); DrawLine(20, 41.6, 200, 41.6); DefineWindow(2, 0, YMaxGlb-215, XmaxGlb, YMaxGlb-80); DefineHeader(2, 'Lung Volume Through Time.'); DefineWorld(2, 0, 34.5, 405, 45); SelectWorld ( 2 ) ; SelectWindow(2); SetForeGroundColor(12); SetHeaderOn; DrawBorder; DrawAxis(8, -5, 0, 0, 0, 0, 0, 0, Fa l s e ) ; 173 DrawPolygon(M, 10, 400, 0, 0, 0); DrawLine(20, 41.7, 200, 41.7); STR(TRUNC(One), OneStr); STR(TRUNC(Two), TwoStr); DrawTextW(25, 38, 1, 'Noise Level: Type I:' + OneStr + ' % , ' ) ; DrawTextW(130, 38, 1, 'Type I I : ' + TwoStr + ' % ' ) ; DefineWindow(3, 0, YMaxGlb-75, XMaxGlb, YMaxGlb); DefineHeader(3, 'Breath Occurance'); DefineWorld(3, 0, -1, 405, 5); SelectWorld(3); SelectWindow(3); SetForeGroundColor(12); SetHeaderOn; DrawBorder; DrawAxis(8, 1, 0, 0, 0, 0, 0, 0, False); DrawPolygon(S, 10, 400, 0, 0, 0); DrawTextW(3, 2.2, 1, 'Type I ' ) ; DrawTextW(3, 0.5, 1, 'Type I I ' ) ; Repeat U n t i l Keypressed; END; { ComputeAndDisplay } BEGIN { main program. ) GetFileName; InitGraphic; ComputeAndDisplay; Repeat U n t i l KeyPressed; LeaveGraphic; END. { main program. } 174 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

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"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0100451/manifest

Comment

Related Items