UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Locomotor-respiratory synchrony in the Canada goose Funk, Gregory Douglas 1990

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

Item Metadata

Download

Media
831-UBC_1990 A1 F96.pdf [ 10.11MB ]
Metadata
JSON: 831-1.0106749.json
JSON-LD: 831-1.0106749-ld.json
RDF/XML (Pretty): 831-1.0106749-rdf.xml
RDF/JSON: 831-1.0106749-rdf.json
Turtle: 831-1.0106749-turtle.txt
N-Triples: 831-1.0106749-rdf-ntriples.txt
Original Record: 831-1.0106749-source.json
Full Text
831-1.0106749-fulltext.txt
Citation
831-1.0106749.ris

Full Text

L O C O M O T O R - R E S P I R A T O R Y S Y N C H R O N Y IN T H E C A N A D A G O O S E By G R E G O R Y D O U G L A S F U N K B . S c , T h e University of British C o l u m b i a , 1985 A T H E S I S S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Department of Zoology) W e accept this t h e s i s a s conforming to the required s t a n d a r d T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A April 1990 ©Gregory Douglas Funk, 1990 ln 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. (Signature) Department of Zoology The University of British Columbia Vancouver, Canada Date May 22, I990 DE-6 (2/88) The one to one synchronization between wing motion and respiration during "Active" flight in the absence of phasic peripheral feedback, however, was different from the predominant three to one relationship seen during free-flight and the range of relationships seen during electrically induced wing flapping in the presence of feedback (one wing beat per breath to four wing beats per breath). Proprioceptive feedback modification of the centrally derived coordination could account for these differences. In addition, hypoxia was associated with decreases in entrainment between hindlimb and respiratory rhythms, thus chemoreceptor drive also appears to be involved in modulating locomotor-respiratory coordination. Results indicate there is considerable redundancy in the mechanisms involved in the production and control of locomotor-respiratory entrainment. The type of locomotor-respiratory coordination expressed in free-ranging animals is likely due to an interaction of these mechanisms, where proprioceptive as well as chemoreceptive feedback modify a central coordinating mechanism. iii ABSTRACT U s i n g a variety of preparations, (intact birds during treadmill and overground walking/running a n d free-flight; d e c e r e b r a t e birds during elec t r i c a l l y i nduced walking/running, p a s s i v e wing flapping, and elect r i c a l l y i n d u c e d wing flapping and "Active" flight), this t h e s i s e x a m i n e s s o m e of the m e c h a n i s m s involved in the production a n d control of locomotor-respiratory coordination (entrainment) during alternating hindlimb locomotion (walking/running) a n d s y n c h r o n o u s forelimb locomotion ("flight") in the C a n a d a g o o s e . Significant coordination of locomotor a n d respiratory rhythms w a s o b s e r v e d during both modes of locomotion in intact a s well a s dec e r e b r a t e birds. While coordination of forelimb motion a n d respiration w a s virtually complete, coordination of hindlimb motion a n d respiration w a s sp o r a d i c . T h e level of entrainment during hindlimb locomotion, however, i n c r e a s e d with i n c r e a s e d stride frequency, rather than i n c r e a s e d work rate, s u g g e s t i n g that proprioceptive f e e d b a c k from the limbs is involved in the production of locomotor-respiratory entrainment. Further e v i d e n c e for a role of proprioceptive f e e d b a c k in the production of entrainment was provided by the entrainment of respiration by p a s s i v e wing motion in decerebrate g e e s e . Although feedback from the periphery w as sufficient for the entrainment of wing motion and respiration, motor nerve outputs to the wing a n d respiratory musculature were also s y n c h r o n i z e d in p a r a l y z e d , decerebrate birds during electrically induced activity, in the complete a b s e n c e of ph a s i c afferent feedback. Thus, both feedback a n d feedforward m echanisms a p p e a r sufficient for the coordination of wing beat a n d respiration. ii TABLE OF CONTENTS Title P a g e i Abstract ii T a b l e of contents iv List of T a b l e s vi List of F i g u r e s vii List of A b b r e v i a t i o n s x A c k n o wledgement xi G e n e r a l Introduction 1 C h a p t e r 1 - R o l e of the t e l e n c e p h a l o n in the 11 s y n c h r o n i z a t i o n of locomotor a n d respiratory f r e q u e n c i e s during walking in the C a n a d a g o o s e Introduction 12 Materials a n d methods 15 R e s u l t s 2 5 D i s c u s s i o n 37 C o n c l u s i o n 43 C h a p t e r 2 - Effects of c h a n g e s in locomotor intensity, 4 5 hypoxia and h y p e r c a p n i a on locomotor-respiratory s y n c h r o n y during walking/running in the C a n a d a g o o s e Introduction 46 Materials a n d methods 47 R e s u l t s 54 D i s c u s s i o n 67 C o n c l u s i o n 74 C h a p t e r 3 - Coordination of wing beat an d respiration in the 76 C a n a d a goo s e . I. Free-flight Introduction 77 Materials a n d methods 80 R e s u l t s 85 D i s c u s s i o n 98 C o n c l u s i o n 107 C h a p t e r 4 - Coordination of wing beat an d respiration in the 108 C a n a d a goo s e . II. P a s s i v e wing flapping Introduction 109 Materials a n d methods 111 R e s u l t s 123 D i s c u s s i o n 153 C o n c l u s i o n - 162 C h a p t e r 5 - Coordination of wing beat a n d respiration in the 163 C a n a d a goo s e . III. "Fictive" flight. Introduction 164 iv Materials a n d methods R e s u l t s D i s c u s s i o n C o n c l u s i o n G e n e r a l D i s c u s s i o n R e f e r e n c e s LIST OF TABLES Table 1-1. Respiratory values in decerebrate and intact Canada geese 28 recorded prior to and over the last 4 minutes of a 10 minute (decerebrate) and 8 minute (intact) walking period. Table 3-1. The relationship between wing beat and respiration in 101 different species. Table 3-2. Duty cycle of the locomotor and respiratory patterns, and 106 the portion of inspiration occuring during upstroke. Table 5-1. Different wing beat-respiratory coordination schemes in 180 decerebrate birds. Table 5-2. Relationship between stimulation site and iJ1v. 183 Table 5-3. Magnitude of the increase in several variables during 221 free-flight and electrically induced "flight". Table 5-4a. Respiratory responses of several avian species to flight. 224 Table 5-4b. Metabolic and cardiovascular responses of several avian 225 species to flight. vi LIST OF FIGURES Chapter 1 Figure 1-1. Figure 1-2. Figure 1-3. Figure 1-4. Chapter 2 Figure 2-1. Figure 2-2. Figure 2-3. Figure 2-4. Figure 2-5. Figure 2-6. Chapter 3 Figure 3-1. Figure 3-2. Figure 3-3. Figure 3-4. Figure 3-5. Figure 3-6. Chapter 4 Figure 4-1. Figure 4-2. Figure 4-3. Figure 4-4. Experimental a p p a r a t u s u s e d for running intact g e e s e on 18 the treadmill. Ventilatory r e s p o n s e s of intact a n d dec e r e b r a t e g e e s e to 30 treadmill w a l k i n g . R e l a t i o n s h i p between minute ventilation a n d metabolic 33 rate during walking/running in intact a n d dec e r e b r a t e g e e s e . Time c o u r s e of the c h a n g e s in blood g a s status during 35 treadmill walking in intact a n d decerebrate g e e s e . Experimental apparatus u s e d during overground running. 51 Ventilatory r e s p o n s e s of intact g e e s e to treadmill 56 running at three different v e l o c i t i e s . Effects of treadmill velocity on respiration a n d 58 metabolic rate. Effects of velocity on respiratory a n d locomotor 60 fre q u e n c i e s a n d entrainment during treadmill a n d overground running. 0 2 a n d C 0 2 ventilatory r e s p o n s e s during rest and 64 ex e r c i s e . Effects of inspired g a s composition on entrainment. 66 Effects of vari o u s diets on respiratory frequency. 87 Free-flight v e l o c i t i e s , a n d wing beat a n d 89 respiratory f r e q u e n c i e s . Free-flight entrainment ratios. 91 Co m p u t e r generated trace of free-flight wing e x c u r s i o n 93 and respiratory movement. Com p u t e r generated trace of free-flight wing e x c u r s i o n 95 and respiratory movement. P h a s e relationship between respiration and wing beat. 97 Method for generation of phase response cu r v e s . 117 A n a l y s i s of the p h a s e relationship between wing 121 beat and respiration. Relationship between p a s s i v e limb movement 126 frequency a n d respiration. Effects of pa s s i v e wing flapping on respiration. 128 vii Figure 4-5. Figure 4-6. Figure 4-7. Figure 4-8. Figure 4-9. Figure 4-10. Figure 4-11. Figure 4-12. Chapter 5 Figure 5-1. Figure 5-2. Figure 5-3. Figure 5-4. Figure 5-5. Figure 5-6. Figure 5-7. Figure 5-8. Figure 5-9. Figure 5-10. Figure 5-11. Figure 5-12. Figure 5-13. Figure 5-14. T r a c e s s h o w i n g the relationship between p a s s i v e 130 wing e x c u r s i o n and air flow. Re l a t i o n s h i p between p a s s i v e wing flapping f r e q u e n c y 132 a n d breathing frequency. S u m m a r y of the effects of p a s s s i v e wing flapping on 136 breathing frequency. T r a c e s s h owing effects of wing s t a l l s on respiration. 140 Effects of wing st a l l s on the p h a s e relationship 142 between wing beat a n d respiration. Time taken to restore the p h a s e relationship between 144 p a s s i v e wing beat a n d respiration following perturbation of the wing c y c l e . P h a s e r e s p o n s e c u r v e for entire respiratory period. 148 P h a s e r e s p o n s e c u r v e s for inspiratory a n d expiratory 150 periods . Sa g g i t a l s e c t i o n through bird brain s h o w ing level of 169 C N S tr a n s e c t i o n . Bilateral external intercostal nerve activity during 174 active respiration. Activity of internal a n d external intercostal 176 nerves showing "fictive" respiration. "Fictive" respiration s h o w n by activity in external 178 intercostal nerve a n d cranial nerve IX. Stimulation s i t e s s h o w n on c r o s s s e c t i o n s through 186 the a v ian mid- a n d hind brain. C h a n g e s in ventilation, heart rate, b l ood p r e s s u r e 188 and pectoralis E M G activity during electrically i n d u c e d wing flapping in a decerebrate duck. C o m p a r i s o n of respiratory and c a r d i o v a s c u l a r v a r i a b l e s 190 at rest a n d during electrical activation of wing activity, before a n d after p a r a l y s i s . Kinetics of the ventilatory r e sponse to electrically 192 induced wing flapping. 1:1 coordination between pectoralis E M G activity a n d 194 respiration in a decerebrate duck. 1:1 coordination between pectoralis E M G , intercostal 196 E M G a n d respiration in a decerebrate duck. 1:1 coordination between pectoralis E M G a n d respiration 198 in a decerebrate g o o s e . 2:1 coordination between pectoralis E M G and respiration 204 in a decerebrate duck. 4:1 coordination between pectoralis E M G a n d respiration 206 in a decerebrate goose. 1:1 coordination between pectoralis and external 208 intercostal E N G activity during "fictive" flight in a decerebrate, p a r a l y z e d duck. viii Figure 5-15. 1:1 coordination between pectoralis a n d cr a n i a l nerve 210 IX E N G activity during "Active" flight in a d e c e r e b r a t e , p a r a l y z e d duck. Figure 5-16. 1:1 coordination between pecto r a l i s a n d internal 2 1 2 intercostal E N G activity during "fictive" flight in a d e c e r e b r a t e , p a r a l y z e d duck. Figure 5-17. 1:1 coordination between pectoralis a n d external 2 1 4 intercostal E N G activity in a d e c e r e b r a t e , p a r a l y z e d g o o s e . Figure 5-18. 2:1 co o r d i n a t i o n between pectoralis a n d whole 216 intercostal E N G activity during s p o n t a n e o u s "fictive" flight in a d e c e r e b r a t e , p a r a l y z e d duck. Figure 5-19. Effects of i n c r e a s i n g in s p i r e d C 0 2 on the relationship 218 between pectoralis a n d external intercostal E N G activity in a decerebrate, p a r a l y z e d g o o s e . ix LIST OF ABBREVIATIONS V E - minute ventilation V T - tidal volume f v - respiratory frequency V D - d e a d s p a c e ventilation V A C - air capillary ventilation (similar to alveolar ventilation in mammals) voz ' oxygen consumption v co2 - carbon dioxide production f w - wing beat frequency f 6 - stride frequency y f y - cou p l i n g ratio; number of strides per breath Vf v - coupling ratio; number of wing beats per breath Paco2 - partial p r e s s u r e of arterial c a r b o n dioxide Pao 2 - partial pressure of arterial oxygen p H a - arterial pH B P - b l o o d p r e s s u r e H R - heart rate T B - body temperature T A - ambient temperature U D V - unidirectional ventilation E M G - electro myogram ITC - iliotibialis c r a n i a l i s muscle (similar to the sartorius of mammals) E N G - electroneurogram C n d - nucleus reticularis medullaris, pars d o r s a l i s C n v - nucleus reticularis medullaris, pars ventralis T T D - nucleus and tract of the d e s c e n d i n g trigeminal nerve R P - nucleus reticularis pontis c a u d a l i s , pars gigantocellularis M R F - m e s e n c e p h a l i c reticular formation ICo - nucleus intercollicularis |iA - micro a mps x ACKNOWLEDGEMENT S e v e r a l people have b e e n i n d i s p e n s a b l e in helping me complete this t h e s i s . Dr. G e r r y Sholomenko, not only taught me most of the t e c h n i q u e s d e s c r i b e d in the following c hapters, but a s s i s t e d with many of the experiments a s w e l l . He baby sat g o s l i n g s , performed admirably a s a motor c y c l e j o c k e y a n d e v e n m a n a g e d to k e e p smiling at 4:00 A.M., when, after being up all night with A a r o n , he would find himself at the bottom of the W r e c k B e a c h hill for the third time in an hour, trying to recapture renegade birds. I can't thank you enough Gerry. S p e c i a l thanks a l s o go to Capt. Zot (Dr. Bill Miisom) an d Dr. J o h n S t e e v e s for the friendship, encouragement and direction they have provided through the c o u r s e of this investigation (not to mention all the trips they g e n e r o u s l y funded). The high level of interaction brought about by their "open-door" policy was a major contributor to the early s u c c e s s of these experiments. I would also like to thank Dr. Dave J o n e s a n d e s p e c i a l l y Bill for the careful reading of this document. Jo r g e Ignacio V a l l e n S a n Martin E r r e r a R a m i r e z P e n a Y Lillo (he is the Ig Man, he is the walrus) w a s my engineering i d e a man, a n d a s s u c h , invaluable. He was a l s o involved in many of the experiments a n d ran up a n d down Wreck B e a c h hill f aster than anyone. In addition, Ignacio's latin ancestry bestowed upon him an inspiring film p r e s ence, thus his V H S version of 'La C u c a r a c h a ' will thrill a udiences at meetings for years to come. Thank you Ig. Finally, Ignacio a n d S o h a i l H a s a n , were instrumental in reviving Data G o d worship. Although I initially found the s c r e a m s (officially referred to as the D a t a G o d Cry) and gyrations (the Data G o d Dance) u s e d by S o h a i l a n d J o r g e Ignacio V a l l e n S a n Martin E r r e r a R a m i r e z P e n a xi Y Lillo to a p p e a s e the D a t a G o d s a little disruptive, my c o n v e r s i o n w a s a s s o c i a t e d with immediate s u c c e s s in the lab. T hen, upon wit n e s s i n g t h e s e s a m e rituals performed high on a s a n d s t o n e pinacle d e e p in the V a l l e y of the D a t a G o d s (just north of Sedona) s e v e r a l months later, I knew I had made the correct c h o i c e . T h i s th e s i s is therefore, in part, a tribute to t h e s e d e i t i e s . I would also like to thank S a r a h Bradley, Arthur V a n der Horst, L a u r a S a b o r i o a n d J e a n Antoine for their a s s i s t a n c e with the care a n d training of birds as well as their a s s i s t a n c e with s o m e of the experiments. I am a l s o indebted to Mark B u r l e s o n for his a s s i s t a n c e with everything having to do with computers. T h e gratitude is limited, however, b e c a u s e if it wasn't for N Y E T , I may have had time to learn something about about computers myself (doubtful?). A s i d e from simply providing te c h n i c a l a n d a c a d e m i c support, t h e s e people are the main reason I have enjoyed myself over the last four a n d a half y e a r s . There are a number of others who, although not directly involved in my r e s e a r c h , have al s o greatly e n r i c h e d my life a s a G r a d student. First, a s Mark reminded me, I would like to thank the board of directors at C a r l i n g O'Keefe Breweries, a n d the w omen of S.I.; they always managed to raise my spirits. I would also like to thank Marek an d C a t h y Hudon, Marianne Morgan, G r a h a m Dodd, C h e r y l W e b b and Heather M c L e a n for the support and friendship they have provided over the years. I would e s p e c i a l l y like to thank D o n n a M a c k e n z i e for the s m i l e s , the encouragement and most of a l l , for putting up with me through the completion of two degrees. I would also like to thank my parents for all the s a c r i f i c e s , financial a n d otherwise, they have made over the years, and, most importantly, for providing an environment where I was free to pursue any goal, no matter how c r a z y (or financially xii irresponsible). Finally, I would like to acknowledge N S E R C of C a n a d a a n d the Killam F oundation w h o s e support greatly expedited the completion of this work. xiii GENERAL INTRODUCTION It i s of interest to note that an early g e n e r a l s c h e m e , pu b l i s h e d by Zuntz a n d Geppert in 1886 (cited in Eldridge et al., 1985), of the m e c h a n i s m s that might participate in the control of the e x e r c i s e h y p e r p n e a has not c h a n g e d substantially o v e r the intervening century. The m e c h a n i s m s p r o p o s e d to be involved fall into f e e d b a c k or feedforward cat e g o r i e s . The former holds that the stimulation of breathing is a c h i e v e d through s e n s o r y f e e d b a c k following the onset of e x e r c i s e . N e urogenic f e e d b a c k from mechanical or c h e m i c a l receptors in the working m u s c l e s was p r o p o s e d to account for the rapid i n c r e a s e in ventilation s e e n at the onset of e x e r c i s e . Alternatively, humoral f e e d b a c k from unknown s u b s t a n c e s r e l e a s e d into the b lood from the working m u s c l e s w a s pr o p o s e d to c a u s e the slow, subsequent i n c r e a s e in ventilation (Geppert & Zuntz, 1888; ci t e d in Eldridge et a l . , 1985). Alternatively, feedforward theory holds that c o m m a n d s i g n a l s originating in the suprapontine a r e a of the brain or in the motor cortex c a n drive the locomotor a n d respiratory s y s t e m s proportionately without the intervention of f e e d b a c k mechanisms. Although many a d v a n c e s have been made sin c e the early study of Zuntz & Geppert (1886), attempts to explain the control of ventilation during e x e r c i s e b a s e d on a single m e c h a n i s m have proven inadequate. There is evidence that e a c h of the pro p o s e d f e e d b a c k m e c h a n i s m s (see reviews by Dejours, 1964; D e m p s e y et a]., 1985; W a s s e r m a n et al., 1986) as well a s feedforward m e c h a n i s m s (Eldridge et a]., 1985) exert s o me influence on respiration under appropriate experimental conditions. It is most likely that they all contribute to its control during normal e x e r c i s e . T h e relative importance of e a c h to the overall control of ventilation during e x e r c i s e , however, remains unclear. 2 O n e factor often d i s r e g a r d e d in studies examining the control of ventilation during e x e r c i s e is the interaction between the locomotor s y s t e m and the respiratory s y s t e m . A potential conflict exists b e c a u s e in many s p e c i e s , the s a m e m u s c l e s s u b s e r v e both respiration and locomotion (Carrier, 1987; C e l l i et a l . , 1988; Ainsworth et al., 1989). During activity, oxygen requirements i n c r e a s e a n d ventilation must i n c r e a s e accordingly. However, if the respiratory m u s c l e s are se r v i n g locomotion, how is their respiratory function a c h i e v e d ? T h i s conflict is most pro n o u n c e d in s o m e reptiles (Carrier, 1987). At moderate e x e r c i s e intensity (< 0.42 m/s), lizards a p p e a r able to i n c r e a s e ventilation in proportion to metabolic rate (Mitchell et a l . , 1981). C a r r i e r (1987) found, however, that a bove 0.40 m/s, ventilation a p p e a r e d to d e c r e a s e or c e a s e as velocity i n c r e a s e d (Carrier, 1987). T h u s the respiratory musculature a p p e a r e d to be totally c o n s t r a i n e d by locomotion at high work rates. Although mammals have d e v e l o p e d a separate inspiratory muscle, the diaphragm, a potential conflict still e x i s t s . Many of the a c c e s s o r y inspiratory m u s c l e s (external intercostals, sternocleidomastoid) a n d expiratory m u s c l e s (internal intercostals, abdominals, triangularis sterni) are also likely to s e r v e a locomotor function, a s has been shown for the transverse abdominus muscle in the dog (Ainsworth et al., 1989). Respiratory m u s c l e s also s e r v e postural and stabilizing functions (Adams et a]., 1989). Thus, during arm e x e r c i s e in humans, some of the inspiratory m u s c l e s of the rib cage take part in nonventilatory functions a n d e x e r c i s e capacity d e c r e a s e s (Celli et al., 1988). E v e n if the conflict between respiratory v e r s u s locomotor muscle function is minimal, the mechanical deformations of the abdomen and/or thorax a s s o c i a t e d with 3 locomotion in a variety of animals (horse, dog, rabbit; Bramble & C a r r i e r , 1983; wallaby; Baudinette et al., 1987; bird; J e n k i n s et al., 1988) c o u l d s e v e r e l y impair the ability of the respiratory s y s t e m to move air if the two s y s t e m s were not coordinated. R e cent observations have s hown that locomotion a n d respiration are tightly c o ordinated in many vertebrate groups in a number of different ways. Two very different t y p e s of coordination o c c u r in f i s h . During cruise swimming, many pelagic f i s h e s (see review by R o b e r t s & R o w e l l , 1988) employ ram ventilation. U p o n reaching a certain velocity, respiratory movements c e a s e , the mouth a n d operculi o p e n , and water is f o r c e d over the gills due to the forward movement of the fish a n d the inertia of the medium. The work of breathing is transferred from the respiratory to the swimming musculature. In contrast, in C y m a t o g a s t e r a g g r e g a t a (Webb, 1975) and other cic h l i d fish ( S a t c h e l l , 1968) that s w i m with their pectoral fins, respiratory movements are s y n c h r o n i z e d 1:1 with pectoral fin movements s o that the mouth open s a s the fish is propelled forward. Locomotor-respiratory coordination has also b e e n d o c u m e n t e d during flight in birds (see chapter 3 for summary) a n d bats (Suthers et al., 1972; Thomas, 1981) a n d during q uadrupedal locomotion in rabbits, dogs, an d horses (Bramble & Carr i e r , 1983) and bipedal locomotion in wallabies (Baudinette et al., 1987) and man (Bramble & Carrier, 1983). This contrasts with the involvement of respiratory muscles in locomotion in s o m e lizards, mentioned earlier, in which ventilation appears to virtually c e a s e during e x e r c i s e at high work rates (Carrier, 1987). The variety of different coordination s c h e m e s o b s e r v e d between locomotion and respiration suggests that the interaction between locomotor a n d respiratory s y stems during exercise has 4 p l a y e d a n important role in sh a p i n g the ventilatory r e s p o n s e s of many vertebrates to e x e r c i s e . T h e s e previous s t u d i e s , however, were primarily descriptive a n d revealed little about the m e c h a n i s m s res p o n s i b l e for the production of locomotor-respiratory coordination. T h e first s t u d i e s that b e g a n to ex a m i n e locomotor-respiratory entrainment from a control standpoint were performed o n humans during either treadmill walking/running (Kay et a]., 1975; B e c h b a c h e & Duffin, 1977; P a t e r s o n et al., 1987), free running (Bramble, 1983; P a t e r s o n et a]., 1987) or cy c l e e x e r c i s e (Kay et al., 1975; B e c h b a c h e & Duffin, 1977; J a s i n s k a s et a[., 1980; Kohl et al., 1981; Y o n g e & P e t e r s e n , 1983; Pa t e r s o n et al., 1986). T h e s e s t u d i e s , which a d d r e s s e d q u e s t i o n s pertaining to the effects of limb movement rate on breathing pattern produced inconsistent results. During low level e x e r c i s e in humans, entrainment is minimal. K a y et a]. (1975) a n d K e l m a n & W a t s o n (1973) failed to find entrainment during cycling or cycling a n d treadmill walking respectively. B e c h b a c h e & Duffin (1977) and J a s i n s k a s et a l . (1980), using c r o s s correlation a n a l y s i s , later found significant entrainment during running/cycling a n d c y c l i n g respectively. Although c r o s s correlation a n a l y s i s , a s u s e d in t h e s e two studies, c o u l d l e a d to error in positive detection of entrainment (Yonge & P e t e r s e n , 1983), more appropriate a n a l y s i s techniques still s u g g e s t e d that entrainment d i d oc c u r during c y c l i n g (Kohl et a]., 1981; Y o n g e & P e t e r s e n , 1983; Pa t e r s o n et al., 1986). With i n c r e a s e s in the level of e x e r c i s e , entrainment levels a l s o i n c r e a s e d , but it was unclear whether the s e effects were due to i n c r e a s e d work rate or i n c r e a s e d limb movement frequency. S o m e studies c l a i m e d there was an increase in entrainment with increasing frequency of limb movement (Bechbache 5 & Duffin, 1977). Other s t u d i e s attributed this to i n c r e a s e s in work rate ( J a s i n s k a s et al. , 1980; S c h w a r z , 1973; referenced in Kohl et a l . , 1981). Still other s t u d i e s found there w a s no effect of in c r e a s i n g limb movement frequency or work rate on entrainment (Kohl et a l . , 1981). R e c e n t work examining the coordination between locomotor a n d respiratory patterns during treadmill ( P a t e r s o n et al., 1987) or overground running (Bramble, 1983; P a t e r s o n et a l . , 1987) indicated that entrainment c o u l d o c c u r to a high degree in some individuals. Part of this earlier c onfusion may have been due to a failure to examine se p a r a t e l y the effects of limb movement rate a n d metabolic rate on entrainment (Bramble & Ca r r i e r , 1983). In addition, there was little agreement o n how entrainment s h o u l d be a s s e s s e d . C o r t i c a l influences may also have been a confounding factor. T h i s last possibility is supported by the observation that entrainment r e a c h e d 8 0 % in trained runners but was not detectable in untrained individuals (Bramble, 1983). Entrainment w a s also greater during cy c l i n g in a group of trained v e r s u s untrained s u b j e c t s (Kohl et al., 1981). T h e i n c r e a s e d coordination in t r ained subjects c o u l d relate to a motor pattern that has b e e n laid d own that c a n be activated without cortical involvement. However, cortical involvement is further supported by the finding that entrainment i n c r e a s e s when locomotion is timed with audible pacing c u e s ( B e c h b a c he & Duffin, 1977; J a s i n s k a s et al, 1981). T h e s e studies e m p h a s i z e the intermittent nature of locomotor-respiratory entrainment in humans, which confounds all attempts to delineate the me c h a n i s m s involved in the production a n d control of locomotor-respiratory coordination. Thus, although it is clear that entrainment d o e s occur during bipedal locomotion in humans, it remains unclear whether it is due to voluntary respiratory control, or 6 dependent on limb related afferent feedback, or work rate. S t u d i e s on cats a n d rabbits, however, indicate that, as in the control of ventilation during e x e r c i s e , both f e e d b a c k an d feedforward m e c h a n i s m s may be involved in the production of entrainment. Iscoe and P o l o s a (1976) and K a w a h a r a et a l . (1988) have s hown that respiration c a n be entrained over s m a l l ranges by periodic stimulation of s p e c i f i c s o m a t i c afferent nerves in c a t s . It is difficult to interpret the importance of this type of f e e d b a c k in entraining respiration during actual locomotion, however, s i n c e the afferent traffic p roduced in the two situations may be markedly different. V i a l a et a l . (1987b) and P e r s e g o l et a l . (1988), using p a r a l y z e d , decerebrate rabbits, a n d K a w a h a r a et a l . (1989a) using p a r a l y z e d , decerebrate c a t s , have shown that the locomotor and respiratory neural outputs may be intermittently c o u p l e d centrally in the a b s e n c e of p h a s i c afferent feedback. The physiological s i g n i f i c a n c e of these events, however, remains unclear. The major limitation of t h e s e latter s t udies a r i s e s from the fact that the animals e xamined demonstrate gait dependent entrainment. T hus, in rabbits, d o g s and horses (Bramble & Carrier, 1983), entrainment of respiration a n d locomotion is minimal during walking, but virtually 1 0 0 % upon the transition to trot or gallop. E v e n in humans, entrainment a p p e a r s to i n c r e a s e from 2 8 % during walking (Hill et a]., 1988) to > 8 0 % during free running (Bramble, 1983). The decerebrate, p a r a l y z e d preparations d e s c r i b e d above must be induced to walk, either through electrical or c h e m i c a l activation of s p e c i f i c neural pathways. The locomotor pattern or gait produced va r i e s in an intensity dependent fashion, i.e. higher stimulation intensities or greater chemical concentrations are required to 7 induce running than walking (Shik et al., 1966; S t e e v e s et al., 1987). T h o s e gaits during which entrainment is o b s e r v e d in active animals are the most difficult to initiate in t h e s e p a r a l y z e d preparations. In addition, the gait p r o d u c e d is not always constant, shifting from one to another between a n d within trials ( Sholomenko et al., 1990 a & b; G a r c i a - R i l l et al., 1985). The locomotor pattern, a s indicated by activity in the nerves leading to the locomotor m u s c l e s , was not recorded bilaterally. T h u s the intermittent entrainment d e s c r i b e d above may simply be a function of changing gait. C o n v e r s e l y , entrainment at the central level may only be an intermittent event. C l o s e l y monitoring the gait of the animal under investigation or using an animal where gait d e p e n d e n c e of entrainment is not a factor would help resolve this i s s u e . C l e a r l y this r e p r e s e n t s a fundamental question to the control of entrainment. S p o r a d i c central coordination would imply that p h a s i c afferent f e e d b a c k p l a y s a major role in the production of the continuous locomotor-respiratory coordination o b s e r v e d in a variety of freely moving vertebrates. C o n v e r s e l y , a tight central coordination would imply that feedback plays only a modulatory role. While it is generally felt that locomotor rhythm influences respiratory rhythm to produce entrainment (Iscoe & P o l o s a , 1976; B e c h b a c h e & Duffin, 1977; Y o n g e & P e t e r s e n , 1983; Bramble & Carrier, 1983), the effects of chemoreceptor drive on entrainment must also be c o n s i d e r e d . Entrainment d e c r e a s e s in humans during progressive hypoxia (Paterson et al., 1987), and may d e c r e a s e with increasing e n d tidal C 0 2 in cat s ( K a w a h a r a et al., 1989a). Entrainment also shifts from 1:1 to 1:2 fin beats per breath in the f i s h , C y m a t o g a s t e r aqqreqata during hypoxia (Webb, 1975). Thus, to a s s u m e that the relationship between locomotion and respiration is solely dependent on locomotor drive is simplistic. A complete understanding of the 8 p r o c e s s e s involved in producing entrainment is unlikely until the influence of non-e x e r c i s e related respiratory drives o n the relationship b etween locomotor a n d respiratory s y s t e m s is c o n s i d e r e d . This t h e s i s a d d r e s s e s s e v e r a l q uestions c o n c e r n i n g the control of entrainment through experiments performed o n the C a n a d a goo s e . T h e r e are numerous ad v a n t a g e s to using this animal, most of which will be d i s c u s s e d in individual c h a pters. T h e main advantage, however, is that the bird has two s e p a r a t e modes of locomotion. T h e alternating bipedal hindlimb form of locomotion of the g o o s e i s s i m i l a r both in pattern a n d in its d e s c e n d i n g control to that in humans ( M c C l e l l a n , 1986; S h o l o m e n k o & S t e e v e s , 1987; S t e e v e s et al., 1987). T h u s this model c a n be u s e d to help unravel the controversy surrounding the initial work on the control of entrainment in humans. S p e c i f i c a l l y , I will examine; 1) the importance of the t e l e n c e p h a l o n (cortical influences) to the production of entrainment during bipedal locomotion; 2) the effects of c h a n g i n g work rate, p r oduced independently of c h a n g e s in locomotor frequency, on entrainment; 3) the effects of i n c r e a s e d locomotor frequency on locomotor-respiratory synchrony; a n d 4) the effects of i n c r e a s e d non-exercise related c h e m i c a l respiratory drives on entrainment. G e e s e a l s o demonstrate bilaterally s y nchronous forelimb locomotor movements during free-flight. Unlike bipedal locomotion in man where intermittent entrainment greatly c o m p l i c a t e s the examination of entrainment control, wing a n d respiratory movements are highly coordinated during flight in many s p e c i e s (see chapter 3 for summary). E v e n in decerebrate g e e s e during electrically i nduced locomotion, the wings always beat synchronously o nce activated (Sholomenko et al., 1990 a,b & c). In addition, the problem of gait dependent entrainment that may have confused the 9 results of V i a l a et a l . (1987b), P e r s e g o l et a[. (1988) a n d K a w a h a r a et a l . (1989a), is removed during "flight". T h i s s e c o n d mode of locomotion in birds, therefore, facillitates the examination of the potential f e e d b a c k a n d feedforward mechanisms involved in the production a n d control of locomotor-respiratory entrainment. Sp e c i f i c a l l y , the effects of afferent f e e d b a c k on respiratory pattern will be examined during p a s s i v e wing flapping. T h e importance of feedback to the sync h r o n i z a t i o n of wing beat a n d respiration will a l s o be a s s e s s e d by examining the relationship between motor outputs to locomotor a n d respiratory m u s c l e s in the complete a b s e n c e of afferent feedback. T h e s e experiments will determine the extent to which feedforward a n d f e e d b a c k m e c h a n i s m s alone, c a n produce entrainment. 10 CHAPTER 1 ROLE OF THE TELENCEPHALON IN THE SYNCHRONIZATION OF LOCOMOTOR AND RESPIRATORY FREQUENCIES DURING WALKING IN THE CANADA GOOSE. 11 INTRODUCTION T h e locomotor a n d respiratory s y s t e m s are so m e t i m e s very tightly c o u p l e d during locomotion in a variety of a n i m a l s , e s p e c i a l l y t h o s e which display s y n c h r o n o u s (in phase) movements of the limbs (birds in flight; Hart & Roy, 1966; rabbits hopping; V i a l a et al., 1987; wa l l a b i e s bounding; Baudinette et al . , 1987; h o r s e s galloping; B r a m b l e & Carrier, 1983). T h e relationship between breathing pattern a n d bipedal locomotion employing alternating (out of phase) movements of the limbs is much l e s s o b vious. S e v e r a l studies using humans performing bicycle (Kay et a]., 1975; B e c h b a c h e & Duffin, 1977; J a s i n s k a s et al., 1980; Kohl et al., 1981; Y o n g e & P e t e r s e n , 1983; P a t e r s o n et al., 1986) a n d running e x e r c i s e (Kay et al., 1975; B e c h b a c h e & Duffin, 1977; Bramble, 1983; P a t e r s o n et al., 1987) have yield e d conflicting results. Thus, although Kay et a l . (1975) failed to s h o w a relationship between respiratory and locomotor c y c l e s during treadmill walking, more recent work s u g g e s t s that respiratory and locomotor c y c l e s may be c o u p l e d up to 8 0 % of the time during unrestrained, overground human running (Bramble, 1983; P a t e r s o n et aj., 1987). Furthermore, s o m e studies have s h o w n i n c r e a s e s in entrainment with i n c r e a s e d work rate ( J a s i n s k a s et a[., 1980) or limb movement rate ( Bechbache & Duffin, 1977) while others have shown neither (Kohl et a]., 1981). Part of this confusion may be the result of the failure to independently examine the effects of metabolic rate and limb movement rate on entrainment (Bramble & Carrier, 1983). It has been s u g g e s t e d that mechanical interactions between the locomotor musculature a n d the respiratory s y s t e m force a coupling of the two systems 12 (Bramble & Carrier, 1983). Alternatively, recent experiments on paralyzed rabbits suggest that locomotor-respiratory coupling is produced via a central neural circuit (Viala et al., 1987b). Somatic afferent feedback also appears to be involved in the production of this entrainment (Iscoe & Polosa, 1976). In addition to these factors, the greater levels of entrainment observed in trained athletes relative to non-athletes (Bramble, 1983; Kohl et al., 1981) and the increase in entrainment associated with audible pacing cues (Bechbache & Duffin, 1977; Yonge & Petersen, 1983) suggests that telencephalic influences may be partly responsible for the coupling of locomotor and respiratory patterns. It has been shown in a wide variety of species that, following decerebration, the vertebrate brainstem is capable of producing normal locomotor (McClellan, 1986) and respiratory behaviours (Feldman et al., 1988). Thus the first purpose of this study was to determine whether locomotor-respiratory interactions at the level of the brainstem are sufficient to produce entrainment of the two motor patterns or are contributions from the telencephalon required for the production of entrainment? To this end, I compared the degree of entrainment found in a group of intact trained, walking geese with that seen in a group of decerebrate, brainstem stimulated walking geese. Since general anaesthesia greatly depresses locomotor as well as respiratory function, these experiments could not be performed on anaesthetized animals. Surgical removal of all brain regions rostral to the brainstem (the telencephalon in birds) produces a decerebrate animal that is incapable of any perception of pain and therefore does not require the subsequent use of general anaesthetics (Loeser & Black, 1975; Wall, 1975; Adams, 1980). Unanaesthetized decerebrate animals are capable of a wide range of spontaneous and stimulated 1 3 motor behaviours (eg. breathing and walking; Sherrington, 1910; 1915; McClellan, 1986; Feldman et al, 1988). This type of experimental preparation is the only way to humanely examine the relationship between locomotor and respiratory patterns in the absence of telencephalic influences and without the confounding affects of general anaesthetics. The second purpose of this study was to examine the effects of changes in metabolic rate on entrainment independent of changes in locomotor frequency. By supporting the decerebrate birds in a sling overlying the treadmill and using low intensity electrical stimulation to induce walking movements, it was possible to produce and sustain locomotion without the associated large increases in metabolic rate normally seen during walking in intact birds. Thus, by matching the walking speeds (i.e. stride frequencies) of the intact geese with those of the decerebrate birds, we were able to examine the relationship between locomotor and respiratory patterns where similar stride frequencies were produced at markedly different metabolic work rates. Geese were the animal of choice because the neural control of bipedal walking in geese is very similar to that of man (Sholomenko & Steeves, 1987; Steeves et al., 1987), yet they are "true" bipeds and thus have complete separation of forelimb and hindlimb function. Humans on the other hand, still possess quadrupedal motor programs and arm swing has been shown to entrain ventilation in humans (Paterson et al., 1986). Thus in walking birds such as geese, forelimb-respiratory interactions are minimized and a more direct examination of the role hindlimb-respiratory interactions play in the development of entrainment is possible. 14 MATERIALS & METHODS Intact Geese (i) Training & Equipment. F o u r C a n a d a g e e s e (Branta c a n a d e n s i s ; 4.2 + 0.2 kg) were r a i s e d from hatching a n d imprinted upon the author. T h e g o s l i n g s spent at least 4 hours per d a y with their "foster parent" for their first three w eeks, one hour of which w a s spent wearing a f a c e m a s k that formed an airtight s e a l c a u d a l to the nares and b e a k opening. F rom 3 w e e k s to 3 months of age, they spent at least 2 hours per day with their "foster parent" in daily w a l k s with the s a m e m a s k s in p l a c e . T h u s a s the birds i n c r e a s e d in s i z e , the contribution of the mask to overall d e a d s p a c e volume d e c r e a s e d both in absolute a n d relative terms. In addition, e a c h gosling was run on the treadmill at least o n ce a week to become a c c u s t o m e d to the lab environment. At three months of age more intensive treadmill training w a s started. First, the birds were fitted with a slightly h e avier (60 g), form-fitting mask. A port (10 mm inside diameter) on the dorsal s u rface of the mask, overlying the nares, allowed e a s y attachment a n d removal of a pneumotachograph (Fle i s c h Model #00) to monitor ventilation. The mask and peumotachograph had a c o m b i n e d d e a d s p a c e volume of 4.0 mL a n d therefore i n c r e a s e d d e a d s p a c e by approximately 1 0 % (Fedde et aj,, 1986). The birds became ac c u s t o m e d to the new mask after l e s s than 1 w e ek of daily training runs (10-15 minutes duration; 0.30-0.72 m/s). However, an additional 8 w eeks of training was required before the birds would run calmly with the pneumotachograph an d a s s o c i a t e d equipment attached to the mask. 15 T h e pneumotachograph w a s attached to a differential p r essure t r a n s d u c e r (Validyne D P 103-18) a n d an integrating amplifier (Gould) in order to obtain tidal v o l u m e (V T), breathing frequency (fv) a n d minute ventilation ( V E ) . T h e p neumotachograph w a s heated to prevent c o n d e n s a t i o n . Ca l i b r a t i o n w a s a c h i e v e d by injection a n d withdrawal of known g a s v o l u m e s through the mask before a n d after e a c h run. E x p i r e d g a s fractions were monitored using a paramagnetic 0 2 a n a l y z e r ( B e c k m a n OM-11) and infra red C 0 2 a n a l y z e r (Beckman LB-2), c o n n e c t e d in s e r i e s , with s a m p l i n g from the distal end of the pneumotachograph at 600 mL/min (fig 1-1A). O x y g e n uptake (v" 0 2),and C 0 2 output ( V ^ ) were m e a s u r e d by p a s s i n g 10 L of air/minute past the e n d of the pneumotachograph through lightweight anaesthetic tubes and collecting the effluent g a s in a 2 L Douglas bag (fig 1-1B). The Douglas bag was then emptied and the fractional concentrations of 0 2 a n d C 0 2 in the mixed g a s was determined u s i n g the 0 2 a n d C 0 2 a n a l y z e r s d e s c r i b e d above. V 0 2 a n d were then c a l c u l a t e d using the F i c k equation. E a c h measurement took approximately 30 s e c o n d s and c o u l d not be performed while mixed expired gas measurements were being taken. T h u s 2 runs were carried out for e a c h bird, one during which mixed expired g a s concentrations were measured and one during which V 0 2 a n d v ^ were measured (except for bird #1 who c o u l d not be trained to wear the g a s delivery s y s t e m while walking). The birds were run in a 37 cm wide x 80 cm long x 60 c m high plywood box overlying a variable s p e e d , motor driven treadmill. A window was cut in the front of the box, s i n c e this facilitated their w i l l i n g n e s s to walk. Th e treadmill belt lay over a 16 Figure 1-1: S c h e m a t i c diagram illustrating the experimental arrangement for monitoring (A) mixed expired g as concentrations a n d (B) metabolic rate in the intact g e e s e . S e e text for further d e t a i l s . P E C 0 2 , mixed expired C 0 2 ; P E 0 2 , mixed expired 0 2 ; V E, minute ventilation; V T, tidal volume; f v, breathing frequency; f s, stride frequency; VCOJ, C 0 2 output; V 0 2, 0 2 uptake. Arrows indicate direction of air flow. 17 A P n e u m o t a c h V t / xV) J f l o w m e t r e s mask 18 force plate constructed from a 7 x 15 c m piece of piezo electric film s a n d w i c h e d between two insulating p i e c e s of paper c e m e n t e d to the back of a 31 x 60 c m piece of 1/8 inch mild s t e e l . T h e signal from the force plate, indicating footfall, w a s amplified a n d recorded on an electrostatic recorder (Gould E S 1 0 0 0 B). (ll)Protocol. E a c h bird w a s fitted with the ported mask a n d p l a c e d on the treadmill. They were allowed at least 5 minutes before the pneumotachograph and either the intake line to the g a s a n a l y z e r s (fig. 1-1 A) or the air delivery s y s t e m w a s attached to the mask (fig. 1-1B). Pre-locomotion ("resting") d a t a including V E, V T, f ¥, and mixed expired g a s concentrations ( P E 0 2 & P E C 0 2) or V o z and were recorded for 1 minute before the treadmill w a s turned on. Treadmill velocity w as i n c r e a s e d gradually from 0 to 0.40 m/s over 2 s e c o n d s to prevent startling the bird. T h i s particular velocity w a s c h o s e n to match the step frequency of intact birds to that of the decerebrate birds (which in turn was a function of stimulation intensity; s e e next se c t i o n on "Decerebrate Animals"). E a c h run lasted 8 minutes. R e c o v e r y was followed for 5 minutes only, s i n c e the birds had a tendency to struggle if this period w as extended. A n i m a l s were not run more than once per day during d a t a collection a n d only trials free of disturbance were included in the d a t a a n a l y s i s . (iii) Measurements. Resting measurements of V E, V T, f v, V 0 2, and R (tyxj/O represent means (+ S.E.) of 21 measurements on 4 animals with the exception of V 0 2, and R which represent means (+ S.E.) of 9 measurements on 3 animals. Walking v a l u e s 19 of VE, VT a n d f ¥ represent m e ans ( ± S.E.) of 7 trials on 4 an i m a l s while v a l u e s of V02, VcoZ a n d R represent means (+ S.E.) of 3 trials on 3 ge e s e . Following completion of t h e s e measurements, c l o a c a l temperature w as monitored throughout walking and recovery in 1 animal on 2 separate d a y s . Arterial blood gas tens i o n s a n d p H were measured in 1 animal using cannulation, sampling a n d an a l y s i s p r o cedures d e s c r i b e d below for decerebrate g e e s e . S i m i l a r measurements were attempted on the three remaining intact g e e s e , however, the procedures greatly altered the ventilatory r e s p o n s e s of these birds and thus these d a t a were rejected. Decerebrate Animals (i) Materials and Methods. All s u r g i c a l p r o c edures were performed under general a n a e s t h e s i a (induction level = 3 % h a l o t h a n e / 2 0 % nitrous oxide; maintenance level = 1-2% h a l o t h a n e / 2 0 % nitrous oxide; both on a background of 9 5 % OJ 5 % C 0 2 ) . The right brachial vein w a s cannulated for infusion of fluids when n e c e s s a r y and the right brachial artery w a s can n u l a t e d for monitoring blood pressure a n d arterial blood s a m p l i n g . B l o o d s a m p l e s of 0.6 mL were t a k e n a n d the catheter w a s flushed with s a l i n e containing 100 I.U./mL of heparin. B l o o d s a m p l e s were then stored on ice until they could be an a l y z e d for arterial C 0 2 tension (Paco 2), arterial oxygen tension ( P a ^ ) and arterial p H (pHa) using electrodes (Radiometer BMS-Systems, C o p e n h a g e n , Denmark) thermostated to body temperature (41 °C) . The blood gas electrodes were calibrated with humidified g a s mixtures delivered by a mixing pump (Radiometer G M A 2 g a s 20 mixing pump) and the p H electrode w a s calibrated using precision buffer solutions (Radiometer). A tracheostomy w a s then performed to allow continued administration of general anaesthetic a n d s ubsequent monitoring of ventilation v i a pneumotachography as d e s c r i b e d above. V 0 2 a n d were m e a s u r e d a s d e s c r i b e d previously, except a 10 L Douglas b a g w a s u s e d for air col l e c t i o n . Thirty mL air s a m p l e s were taken from the collection bag using g l a s s , leuer-lock s y r i n g e s . T h e s e were later injected into the B e c k m a n gas a n a l y z e r s to determine the fractional concentrations of 0 2 and C 0 2 in the effluent gas. A temperature probe was p l a c e d approximately 25 cm d own the e s o p h a g u s to monitor body temperature. The animal was then s u s p e n d e d in a sling overlying the treadmill belt. The bird's h e a d was then p l a c e d in a stereotaxic headholder and it w a s d e c e r e b r a t e d as previously d e s c r i b e d by S t e e v e s et a]. (1987). A l l incision sites and stereotaxic headholder pressure points were generously infiltrated with l o c a l anaesthetic ( 2 % xylocaine) throughout the c o u r s e of surgery a n d during all subsequent experimental tri a l s . Following decerebration, E M G electrodes were implanted percutaneously in the iliotibialis c r a n i a l i s (ITC) m u s c l e s of the leg (homologous with mammalian sartorius m u s c l e s ) . G e n e r a l a n a e s t h e s i a was then discontinued and the birds were allowed a minimum of 1.5 hours for recovery to remove all effects of the an a e s t h e t i c before electrical brainstem stimulation was initiated (complete washout of halothane takes between 20-30 min. following up to 6 hours of surgery in man; C o m p e n d i u m of P h a r m a c e u t i c a l s and S p e c i a l t i e s , 1982). Stride frequency (f s), defined a s the number of complete step c y c l e s performed by the right or left leg per minute, was obtained from E M G s i g n a l s which were amplified a n d filtered prior 21 to storage on an electrostatic recorder. Brainstem stimulation (pulse duration, 2 msec; pulse frequency, 60 Hz; stimulation intensity range, 30-120 uA) w a s carried out on 9 C a n a d a g e e s e (3.6 + 0.2 kg) using procedures d e s c r i b e d by S t e e v e s et a l . (1987). High intensity stimulation (100 \iA) w a s u s e d to l o c a l i z e the stimulation s i t e . Stimulation intensity w a s then d e c r e a s e d to z e r o and gradually i n c r e a s e d to est a b l i s h the current t h r e s h o l d intensity n e c e s s a r y to evoke walking. Typ i c a l l y it took between 30 a n d 60 minutes to e s t a b l i s h optimal electrode position a n d current threshold for e v o k e d locomotion ( S t e e v e s et a]., 1987). Therefore anaesthetic had been removed for a minimum of two hours before any data were c o l l e c t e d . During higher intensity stimulation, the animals generated sufficient force during walking to be completely self supporting. However, at the lower stimulation intensities u s e d here during d a t a co l l e c t i o n , the sling supported a large portion of the animals' weight, enabling the decerebrate animal to generate locomotor patterns of si m i l a r frequency a n d timing to intact birds at much lower metabolic work rates. T e n minute locomotion trials were then initiated at current stimulation intensities 2 0 % greater than threshold. T h i s level of stimulation produced stepping of a s p e c i f i c frequency immediately with the onset of brainstem stimulation and avoided the delay often a s s o c i a t e d with stimulation at threshold current strengths. T h e use of low stimulation intensities made it possible to produce walking for greater than 10 minute periods. By i n c r e a s i n g brainstem stimulation intensity, f s, stride force a n d work rate could all be i n c r e a s e d . S i n c e another purpose of this study was to examine the effects of work rate on the relationship between locomotor a n d breathing patterns, however, the lower stimulation intensities were u s e d to maximize 22 the metabolic work rate difference between decerebrate and intact groups walking at simi l a r stride f r e q u e n c i e s . Stimulation sites were identified histologically a s d e s c r i b e d by S t e e v e s et a l . (1987). V a l u e s of V E, V T a n d f v represent m e a n s (±S.E.) of 13 runs on 9 a n i m a l s while v a l u e s of V 0 2, V ^ and R represent means (+S.E.) of 9 trials on 9 animals. (ii) Protocol. V E, V T a n d f v were recorded for 2 minutes prior to stimulation, during 10 minutes of brainstem e v o k e d walking at a s p e e d of 0.40 m/s and for 15 minutes of post-stimulation recovery. Arterial blood s a m p l e s a n d body temperature m e a s urements were taken at 30 s e c o n d s prestimulation, 1, 2, 4, 6 and 8 minutes of e v o k e d walking a n d 5, 10 a n d 15 minutes of recovery. Effluent g a s s a m p l e s for determining V 0 2 a n d V ^ were taken at 30 s e c o n d s prestimulation, 1, 6 and 10 minutes of e v o k e d walking a n d 5, 10 a n d 15 minutes of recovery. Stride frequency w a s recorded throughout the brainstem stimulated walking period. T o avoid any transient i n c r e a s e in V E a s s o c i a t e d with switching the treadmill on and off, all measurements, from pre-stimulation through post-stimulation, were made with the treadmill on. In order to ensure that all effects of the general anaesthetic had worn off and that the animals had sta b i l i z e d following surgery, 4 animals were run twice (hence 13 runs on 9 birds). The first run, a s described above, took place a minimum of 2 hours following the removal of the general anaesthetic. T h e s e c o n d run was performed 2 hours following the first; 4 hours after general anaesthetic removal. There were no significant differences between the res p o n s e s s e e n at 2 and 4 23 hours. T h u s it was felt that the birds had st a b i l i z e d by two hours a n d that sufficient time had be e n allowed for complete removal of the effects of general anaesthetic. Data Analysis Entrainment has be e n defined mathematically a s a c l o s e integer or 1/2 integer relationship between stride frequency (fs) and breathing frequency (fv) ( P a v l i d i s , 1973). A s s u m i n g random a s s o c i a t i o n of fs and fv a n d a s s i g n i n g confidence limits of ± 0.05 for entrained v a l u e s , c h a n c e coupling of locomotor a n d breathing patterns c o u l d reach 2 0 % (Paterson et al., 1987). T o determine the degree of entrainment, c o u p l i n g ratios (Vfv) were c a l c u l a t e d for e a c h 10 s e c o n d interval throughout the locomotor periods of both intact and decerebrate birds. In order to determine the importance of the sampling interval duration on ca l c u l a t e d levels of entrainment, coupling ratios were also c a l c u l a t e d on the b a s i s of 2 s e c o n d intervals in one animal. O n the few o c c a s i o n s when breathing rates e x c e e d e d 3 0 per minute (i.e., e a c h breathing c y c l e l a s t e d < 2 se c ) in this animal, coupling ratios were c a l c u l a t e d on a breath by breath basis. L e v e l s of entrainment c a l c u l a t e d using 2 s e c o n d intervals were 2 . 5 % greater than when calc u l a t e d using 10 s e c o n d intervals. Thus, although a n a l y s i s b a s e d on 10 s e c o n d periods will l ead to slightly c onservative v a l u e s of entrainment relative to shorter sampling intervals (i.e. < 1 s e c o n d , P a t e r s o n et al., 1987), it e n s u r e d that to be counted as entrainment, any coupling of fs to fv, had to be sust a i n e d for at least 10 s e c o n d s (approximately 6 strides or 12 consecutive steps). After normalizing the percent entrainment d a t a (arcsin transformation), a t-test was u s e d to ascertain whether the percent of entrainment was significantly greater than c h a n c e (i.e. 2 0 % ) . Although it 24 w a s difficult to determine exactly how long it took for the birds to come to a new stead y state following the onset of e x e r c i s e , from f i g . 1-2 a n d from previous work with birds (Brackenbury et al-- 1981b), it a p p e a r e d that new "steady states" were re a c h e d within 4 minutes. T h u s ventilatory d a t a were a v e r a g e d over the final 4 minutes of e a c h walking period. A N O V A a n d Tukey's multiple c o m p a r i s o n test were u s e d to test the difference between means of all respiratory a n d entrainment data. R e g r e s s i o n a n a l y s i s a n d A N C O V A were us e d to test for differences in the relationship between V E and V 0 2 and V E and V ^ in both intact and decerebrate groups. V a l u e s of p<0.05 were a s s u m e d significant. RESULTS A l l effective locomotor electrical stimulation s i t e s fell within the a r e a s d e s c r i b e d by S t e e v e s et a]. (1987). T h e y were histologically verified within the mid-medulla, between the nucleus d o r s a l i s pars centralis (Cnd), ventromedially, and the nucleus and tract of the Trigeminal nucleus (TTD), d o r s o l a t e r a l ^ . The stimulation sites e xtended ca u d a l l y from 300jj.m c a u d a l to the caud a l extent of the 12th cranial nerve nucleus, rostrally to the rostral extent of the 12lh nerve nucleus. It sho u l d be noted that with the onset of brainstem stimulation in the decerebrate animals, three different transient ventilatory r e s p o n s e s , lasting l e s s than 20 s e c , were ob s e r v e d . Three animals stimulated within the TTD/Cnd region, 150-300 u.m caudal to the cauda l extent of the 12th cranial nerve nucleus, s h o w e d immediate d e c r e a s e s in V E (>400 mL/min) at the onset of stimulation. In contrast, the three animals stimulated within the TTD/Cnd region, 300-600 |im rostral to the caudal margin of the 1 2m 25 nerve nucleus s h o w e d immediate i n c r e a s e s in V E (>800 mL/min). Th e final three animals stimulated in TTD/Cnd, between the two stimulation s i t e s d i s c u s s e d above, s h o w e d s m a l l c h a n g e s in V E (< 200 mL/min); 1 animal exhibited a n i n c r e a s e , 1 s h o w e d a d e c r e a s e and another s h o w e d no change. A s i d e from t h e s e transient differences in the ventilatory r e s p o n s e s a s s o c i a t e d with the onset of electrical stimulation, the overall r e s p o n s e s of the decerebrate animals were very similar. T h e pre-walking a n d steady-state walking v a l u e s of the v a r i o u s respiratory v a r i a b l e s monitored are s h o w n in table 1-1 for both intact and decerebrate g e e s e . Pre-walking tidal v o l u m e s (V T) were very similar between the two groups, but a slightly larger f v in the intact a n imals led to a larger "resting" V E in this group. Similarly, V 0 2 w a s slightly elevated in the intact animals relative to the decerebrate birds at rest. w a s significantly greater in the intact animals. T h e s e differences were most likely a c o n s e q u e n c e of a greater reliance on fat metabolism (R=0.7) by the decerebrate group due to a 12 hour pre-surgery starvation period v e r s u s carbohydrate utilization (R=1.0) by the non-starved intact birds (Schmidt-Nielsen, 1979). With the onset of walking, V E i n c r e a s e d significantly in both intact a n d decerebrate groups (fig. 1-2); however, there was a s m a l l overshoot in V E in the intact a n imals at the start of walking that was not normally s e e n in the decerebrate animals. T h e overall magnitude of the ventilatory r e sponses, as well as the c h a n g e s in breathing pattern, were quite different in the two groups. Both V T and f v i n c r e a s e d marginally during e v o k e d walking in the decerebrate birds to produce a small but significant 4 0 % i n c r e a s e in V E. In the intact birds however, the much larger increase 26 T a b l e 1-1: R e s p i r a t o r y v a l u e s in decerebrate an d intact C a n a d a g e e s e recorded prior to a n d over the last 4 minutes of a 10 minute (decerebrate) an d 8 minute (intact) walking period, (mean + S.E., n = number of experiments) % C h a n g e ; percent c h a n g e from rest to e x e r c i s e . R; Pearson's Cor r e l a t i o n Coefficient #; s i g . difference between rest a n d e x e r c i s e within a group *; s i g . difference between the two groups during rest or e x e r c i s e . 27 TABLE 1 V, (ml/min/Kg) VT (ml/Kg) £, (min -') f. (min "'I V w (ml/min/Kg) V c (ml/min/Kg) R T k (*C) Ent % Slope V,/V„ Slope V./V^ Blood Gases Pa„ (mmHg) PaTO, (mmHg) pHa N> Oo DECEREBRATE Rest n 314 + 28 13 29.9 + 1.5 13 10.7 + 1.0 13 11.3 + 0.8 8 7.6 + 0.7 8 0.67 + 0.03 8 40.4+0.7 5 Exercise % Change 446 + 27 # 42.2 34.1+2.3 14.6 13.4 + 0.8 ' 25.2 57.5 + 5.0 15.9 + 1.1 # 46.2 12.6 + 0.6 I 68.3 0.80 + 0.03 40.6 + 0.6 28.9+3.3 24.43 R - 0.787 24.74 R - 0.988 93.4+7.1 8 94.9+7.3 24.9 + 1.2 12 23.9 + 0.9 7.50 + 0.02 12 7.54 + 0.02 INTACT n Rest n Exercise % Change n 13 417 + 30 21 887 + 65 #* 119 7 13 28.2 + 1.3 21 26.5 + 2.4 -4.6 7 13 14.8 + 0.80 21 34.9 + 3.0 #* 137 7 13 ~ — 55.2 + 2.3 — 7 9 13.5 + 1.5 9 32.7 + 3.8 I* 142 3 9 13.1 + 1.3 * 9 29.9 + 3.8 i * 128 3 9 0.98 + 0.04 * 9 0.91 + 0.03 — 3 5 42.3 2 42.7 — 2 9 — — 28.3 + 4.0 — 7 18. 48 R - 0.985 18. 67 R - 0.982 8 89.0 1 103.0 1 12 27.5 1 24.5 — 1 12 7.52 1 7.56 — 1 Figure 1-2: Effect of e x e r c i s e on minute ventilation (V E), tidal v olume (V T), breathing frequency (fv) and stride f requency (f 6). Walking was initiated at t=0 and terminated at t=10 min for the decerebrate g e e s e (filled circles) a n d at t=8 min for the intact g e e s e (filled s q u a r e s ) , (mean + S.E.) 29 (ml/tnin/Kg) (ml/Kg) (1/min) 1000 -800 -600 400 200 30 Y 10 40 20 H ^i-H-i-i-H-H-H-H (1/rnin) 60 40 • • • • • J I I L . I I I I I I I l l _ i u -I I I I 1 I I I ' • ' 4 8 12 16 Time (min) 20 24 30 in V E ( 1 2 0 % ) w a s entirely due to a 1 3 7 % inc r e a s e in f ¥; V T, following an initial i n c r e a s e , d e c r e a s e d marginally over the co u r s e of the 8 min walking trial for the intact birds. T h e i n c r e a s e in V 0 2 between "rest" a n d walking w as a l s o 3 times greater in the intact than in the decerebrate g e e s e . V 0 2 i n c r e a s e d 1 4 2 % from 13.5 to 32.7 ml/min/kg in the intact g e e s e , but it only i n c r e a s e d 4 6 . 2 % from 11.3 to 15.9 ml/min/kg in the decerebrate g e e s e . Despite the large differences in the overall magnitude of the V E a n d V C 2 r e s p o n s e s of the two groups to walking a n d the resultant reduced range over which the V E-V Q 2 regression w as ca l c u l a t e d for decerebrate birds, the s l o p e s a n d y-intercepts of the relationship between V E a n d V C 2 were not different between the two groups (table 1-1, fi g . 1-3). Similarly, the relationship between V E a n d w a s not significantly different between intact a n d decerebrate birds (table 1-1, f i g . 1-3). Despite a transient i ncrease in Pao2 of 5.0 mmHg, a d e c r e a s e in Ps^^ of 3.0 mmHg, and an in c r e a s e in p H a of 0.05 U over the first minute of evoked locomotion in the decerebrate g e e s e (fig. 1-4), blood g a s e s , p H a a n d body temperature ( T J did not c h a n g e significantly between rest and the final 4 minutes of walking in the decerebrate birds (table 1-1). Pao2 only i n c r e a s e d 1.5 mmHg, P ^ d e c r e a s e d 1 mmHg, p H a i n c r e a s e d 0.04 U and T b also i n c r e a s e d marginally (0.2°C) from rest to ex e r c i s e . B l o o d g a s and T b measurements were only possible in 1 intact animal, making c o m p a r i s o n s between intact a n d decerebrate g e e s e difficult. Arterial p H al s o i n c r e a s e d by 0.04 U in the intact animals, however, P a ^ , P a ^ and T b showed much larger c h a n g e s in the intact bird than in the decerebrate geese. P a ^ inc r e a s e d 14 mmHg, d e c r e a s e d 3 mmHg while T b i n c r e a s e d 0.4°C. 31 Figure 1-3: R e l a t i o n s h i p between minute ventilation (V E) and oxygen uptake (V 0 2) a n d V E a n d C 0 2 output ( V ^ ) in decerebrate (filled circles) and intact (filled squares) g e e s e . E a c h point represents the mean of 9 and 7 trials in decerebrate and intact g e e s e respectively. V a l u e s from resting and walking birds are included. 3 2 1200 (ml/min/Kg) 8 0 0 *oo 0.0 1 1 10.0 20.0 30.0 40.0 50.0 VO, (ml/min/Kg) 1200 (ml/min/Kg) 8 0 0 j LOO 0' 0.0 • 1 • • • 1 1 10JO 20.0 30.0 LOQ 50.0 V Q Q {ml/min/Kg) 33 Figure 1-4: Effect of e x e r c i s e on arterial P a ^ , arterial P a ^ a n d arterial p Ha. Arrows indicate duration of the walking period in decerebrate (filled c i r c l e s , n=8) a n d intact (filled s q u a r e s , n=1) g e e s e , (mean + S.E.) 34 Time (min) Locomotor and respiratory rhythms were coupl e d significantly more than the maximum predicted by c h a n c e ( 2 0 % ; P a t e r s o n et al., 1987) in both groups. There w a s no significant variation in the degree of entrainment over the co u r s e of an 8 minute walking trial a n d the coupling w a s always subharmonic. S i n c e the stride frequency was the s a m e in both groups but the level of f v was higher in the intact a nimals, the coupling ratios (fs/fv) in the intact group were much lower than in decerebrate birds. R a t i o s of 3 a n d 4 steps per breath (ie: ^=1.5 & 2.0) were most commonly found in the intact g e e s e , ranging from 2 to 6 steps per breath. T h e coupling ratios were consistently larger at the onset of walking, declining a s f v i n c r e a s e d over time. In decerebrate birds, ratios ranged from 6 to 18 st e p s per breath, with the most commonly o b s e r v e d ratios v a rying between 6 a n d 10 s t e p s per breath (i.e.: y f v =3 to 5) - The highest ratios o c c u r e d at the very beginning of walking in the three a n i m a l s that s h o w e d large, transient d e c r e a s e s in V E a n d f v at the onset of stimulation. S i n c e f v quickly recovered in these birds, the coupling ratios a l s o rapidly returned to between 6 and 10 steps per breath. C o u p l i n g ratios in the remaining 6 decerebrate g e e s e , as s e e n in intact g e e s e , were also highest at the onset of locomotion a n d de c l i n e d very gradually a s f v i n c r e a s e d slightly over the co u r s e of the walking period. In spite of the differences in breathing pattern r e s p o n s e s of the two groups and the resultant differences in v a l u e s , the degree of locomotor-respiratory coupling in intact (28.3 ± 4.0%) and decerebrate g e e s e (28.9 + 3.3%) was virtually identical. 36 DISCUSSION a) Ventilation during treadmill walking in intact birds. The ventilatory r e s p o n s e s of birds to treadmill running appear to be dependent on both work rate (as reflected in this and other studies by c h a n g e s in treadmill velocity: Kiley et al., 1979; Brackenbury et al., 1981a; 1982a; Nomoto et a]., 1983; Brackenbury, 1986) and c h a n g e s in body temperature (T b) (Kiley et al., 1979; Brackenbury et al., 1981 a & b; 1982a; Nomoto et al., 1983; Brackenbury & G l e e s o n , 1983). S i m i l a r to the present results, during submaximal ex e r c i s e at ambient temperatures near 2 0°C, V E typically i n c r e a s e s between 2 and 4 fold, d e pending on velocity, in response to treadmill running in both the c h i c k e n (Brackenbury et al., 1982a; Brackenbury & G l e e s o n , 1983) a n d the duck (Kiley et al., 1979; Kiley et al., 1982). A s shown by others, i n c r e a s e s in f v are primarily responsible for the initial rise in V E at the onset of exercise (Brackenbury et al., 1982a; Kiley et a l . , 1982; Brackenbury & G l e e s o n , 1983; Brackenbury, 1986), although V T has a l s o been shown to i n c r e a s e significantly at the onset of e x e r c i s e in s o m e studies (Brackenbury et al., 1982a; Brackenbury, 1986). R e g a r d l e s s of the initial r e s ponse to e x e r c i s e , it is cl e a r that as treadmill walking pro g r e s s e s and T b r i s e s , V T gradually d e c r e a s e s and f v rises, contributing more and more to the i n c r e a s e in V E in the g e e s e , as has also been shown for c h i c k e n s (Brackenbury et a]., 1982a) an d d u c k s (Kiley et a]., 1979). If T b i n c r e a s e s sufficiently, thermal panting is commonly observed (Kiley et al., 1979; Brackenbury et al., 1981b) and f ¥ c a n reach levels greater than 150 breaths per minute (Kiley et 37 a l . , 1979). W h e n body temperature fluctuations are minimized (i.e.: maintained below 0.5°C; Brackenbury, 1986), either by using lower treadmill s p e e d s a n d reducing metabolic heat production (fig. 1-2; Brackenbury, 1986), d e c r e a s i n g ambient temperature (Kiley et al., 1982; Brackenbury, 1986), or artificially c o o l i n g the a n imals (Brackenbury & G l e e s o n , 1983), fv s e l d o m i n c r e a s e s beyond 55 breaths/min. b) Effects of decerebration on locomotion and ventilation. Although there are extensive neurological differences between intact a n d decerebrate g e e s e , it has been shown for a large number of s p e c i e s that locomotion in decerebrate animals, whether produced spontaneously or by electrical/chemical stimulation, is similar to locomotion in intact animals (Grillner, 1975; M c C l e l l a n , 1986; S t e e v e s et al., 1987). The step c y c l e of the individual limb r e s e m b l e s that of the intact animal both with regard to the duration and the amplitudes of the movements at different s p e e d s as well a s E M G pattern (Grillner, 1975; M c C l e l l a n , 1986; S t e e v e s et a]., 1987). In the c a s e of electrical stimulation, with increasing intensity, the force generated by the limbs increase a s d o e s the s p e e d of walking. T h e s e findings hold true for the bird a s well (Sholomenko & S t e e v e s , 1987). Similarly, with respect to respiration, the essential neural networks responsible for the production of normal respiratory pattern have long been known to be contained within the brainstem (Lumsden, 1923). The question remains, however, as to whether the ventilatory r e s p o n s e s to ex e r c i s e are the s a m e in decerebrate g e e s e 38 during electrically i n d u c e d walking a s in intact birds. Although the decerebrate g e e s e were only made to perform at very low work rates in the present study, our d a t a suggest that the ventilatory r e s p o n s e s to electrically i nduced locomotion in decerebrate g e e s e are very similar to those s e e n in intact birds. V E, V Q 2 a n d i n c r e a s e d significantly in response to walking in both the intact a n d decerebrate g e e s e , a n d although the i n c r e a s e s were significantly greater in the intact animals, the sl o p e of the relationships between V E and V 0 2 a n d V E a n d were s i m i l a r for both groups (fig. 1-3). Thus, for a given i n c r e a s e in work rate, the decerebrate and intact birds r e sponded with a similar increase in V E. It would have been useful to extend this c o m p a r i s o n to higher work rates. Although, higher work rates c o u l d be p r o d u c e d with greater stimulation intensities, high intensity stimulation frequently led to a d e c r e a s e in the duration of e a c h run or a d e c r e a s e in f s a s stimulation p r o g r e s s e d , probably due to in c r e a s e d electrolytic d amage to the surrounding t i s s u e . That this d e c l i n e in activity w as not due to fatigue w as indicated by the observation that stimulation at new brainstem sites c ould still elicit high levels of activity. F o r the purpose of our experiments it was more important to produce longer periods of steady locomotion at low work rates repeatedly a n d thus the decerebrate birds were not run at the higher work rates. In addition, although the sample s i z e for the blood gas and p H a va l u e s for the intact g e e s e limited our ability to make a firm comparison between intact and decerebrate g e e s e , it appears from the literature that the blood gas and p H a res p o n s e s of intact c h i c k e n s to exercise (Brackenbury & G l e e s o n , 1983) are similar to those of decerebrate g e ese. While P a C 0 2 has been frequently observed to de c r e a s e a n d P a ^ to increase during exercise (intact geese, table 1-1; Kiley et al., 39 1979), when e x e r c i s e is performed isothermically, c h a n g e s in blood g a s e s a n d p H a are greatly reduced or absent (decerebrate g e e s e , table 1-1; B rackenbury & G l e e s o n , 1983). Therefore the larger c h a n g e s in arterial blood g a s tensions o b s e r v e d in intact birds in other s t u d i e s are probably due to the greater rise in T b that o c c u r e d relative to the decerebrate birds d e s c r i b e d here. It s e e m s probable that overall ventilatory control during locomotion is not greatly affected by decerebration. T h e effects of d e c e r e b r a t i o n on the control of respiratory pattern were l e s s clear. At rest, breathing patterns were very similar in both groups of birds. With the onset of walking, however, i n c r e a s e s in both V T and fv were responsible for the steady state i n c r e a s e in V E in the decerebrate birds (25.2 & 1 4 . 6 % respectively), while i n c r e a s e s in f ¥ were s o l e l y r esponsible for the i n c r e a s e in V E in the intact birds (table 1-1, f i g . 1-2). It has been documented that both V T and fv increase to produce the i n c r e a s e in V E during isothermic e x e r c i s e in birds (Brackenbury et al., 1982a; Br a c k e n b u r y & G l e e s o n , 1983). However, when T b is not controlled a n d rises during locomotion, breathing frequency continues to rise while V T d e c r e a s e s to control levels or below (Brackenbury et al., 1982a). In the present study, the decerebrate g e e s e e v i d e n c e d a maximum increase in T b of only 0.2°C during walking while the one intact animal in which T b was monitored underwent a 0.4°C increase. T h u s the differences in breathing pattern responses of the two groups to e x e r c i s e may have been the direct c o n s e q u e n c e of different thermoregulatory influences o n respiration. Similarly, aside from developing an hypocapnia, the ventilatory r e s p o n s e s of decerebrate cats to electrically or chemically induced locomotion are very similar to those of normal intact animals (Eldridge et al., 1985). In addition, it now appears that e x e r c i s e hyperpnea is not i s o c a p n i c in most 40 s p e c i e s . Dogs, ponies, rats, goats, lizards, (Dempsey et a l , 1985) and birds (Kiley et al., 1979) d e v e l o p hyperventilation a n d h y p o capnia during e x e r c i s e . c) Effects of work rate and decerebration on entrainment. D e s p i t e the attention given to the control of ventilation during locomotion in birds (Kiley et al., 1979; 1982; Brackenbury et al., 1981a,b; 1982a; B e c h & Nomoto, 1982; Brackenbury & G l e e s o n , 1983; Brackenbury, 1986), no one has directly e x a mined the relationship between ventilatory pattern and hindlimb locomotor rhythms in t h e s e animals. Substantial work has been performed on humans in this regard, and, although it now a p p e a r s that under appropriate conditions there c a n be a high degree of coupling between locomotor and respiratory s y stems during bipedal running (Bramble, 1983; P a t e r s o n et al-. 1987), the m echanisms responsible for this entrainment remain unclear. In particular, b a s e d on existing data, it is difficult to isolate the effects of i n c r e a s e d limb movement during walking/running from the separate effects of a secondarily produced increase in work rate ( B e c h b a c h e & Duffin, 1977). W h e n limb movement rates were controlled in intact and decerebrate g e e s e , levels of entrainment were virtually identical in spite of three-fold differences in V 0 2 (table 1-1). A s s o c i a t e d with the three-fold difference in V 0 2 between the two groups during walking, the breathing frequencies of the intact group were also 3 times greater than decerebrate birds. In turn, since both groups walked with similar step frequencies, the number of steps taken per breath by the intact geese was approximately 1/3 that of the decerebrate birds. Thus, although the number of steps 41 taken per breath is related to metabolic rate, it a p p e a r s that the degree to which locomotor an d respiratory patterns are entrained is completely independent of metabolic work rate during bipedal locomotion in g e e s e . Increased limb movement rate may therefore prove to be a major factor a s s o c i a t e d with i n c r e a s e d entrainment in t h e s e t y p e s of experiments. Afferent information from e x e r c i s i n g limbs has been s h own to play a role in the entrainment of respiration rate with limb movements (Iscoe & P o l o s a , 1976; K a w a h a r a et a l . , 1988) a s have central nervous s y s t e m feedforward effects ( V i a l a et al., 1987b). Although the results of this study do not help elucidate the relative contributions of afferent v e r s u s central efferent interactions in developing entrainment, they do serve to isolate the level at which th e s e interactions take; place. S ome investigators have s u g g e s t e d that locomotor-respiratory coupling in man is under complete or at least partial t e l e n c e p h a l i c control (Yonge & P e t e r s e n , 1983). T h i s suggestion is supported by the tight coupling of locomotor and breathing rhythms in trained athletes v e r s u s the complete lack of entrainment in sedentary subjects (Bramble, 1983). Th e i n c r e a s e in entrainment a s s o c i a t e d with audible pacing c u e s in humans during c y c l e e x e r c i s e ( B e c h b a c h e & Duffin, 1977; Y o n g e & P e t e r s e n , 1983) also supports this hypothesis. O u r data, however, suggest that, although coupling ratios in intact a n d decerebrate birds are different, removal of the telencephalon and thalamus has no effect on the degree of entrainment between locomotor a n d respiratory patterns in g e e s e (table 1-1). Similarly, entrainment has been observed during galloping in decerebrate c a t s ( Kawahara et a l . , 1989a) Humans ( 2 9 % ; Hill et al., 1988) and geese walking on treadmills show virtually identical d e g r e e s of entrainment. During treadmill running, humans show 42 entrainment around 4 5 % of the time (Paterson et al., 1987). During free-running in man, le v e l s of entrainment a s high as 8 0 % have been o b s e r v e d (Paterson et a l . , 1987). T h e s e obser v a t i o n s suggest that entrainment may be related to locomotor frequency or gait in the human. Gait dependent entrainment has been o b s e r v e d in s e v e r a l q u a d r u p e d s including decerebrate cats ( Kawahara et al., 1989b), rabbits, h o r s e s a n d d o gs (Bramble & Carrier, 1983). T h e y also suggest that a moving treadmill f o r c e s an unnatural locomotor rhythm, thereby d e c r e a s i n g the level of entrainment o b s e r v e d during treadmill running (Paterson et al., 1987). It is a l s o p o s s i b l e that g e e s e never demonstrate entrainment much more than 3 0 % of the time a n d that the greater level of entrainment s e e n in trained humans during free-running is l e a r n e d . If this is the c a s e , it is difficult to envision the physiological s i g n i f i c a n c e of t h e s e relatively low coupling rates in g e e s e . However, it is also p o s s i b l e that g e e s e s h o w i n c r e a s e s in entrainment during treadmill running at higher locomotor f r e q u e n c i e s and/or during free-running. CONCLUSION It has been argued that the entrainment of locomotor and ventilatory patterns during human bipedal locomotion may be strongly affected by telencephalic factors ( Bechbache & Duffin, 1975; Y o n g e & P e t e r s e n , 1983). The findings of the present study using another biped, the C a n a d a goose, however, suggest that the telencephalon is not ess e n t i a l for the low degree of coupling s e e n between the locomotor a n d respiratory patterns during treadmill walking in this s p e c i e s . Locomotor-respiratory interactions within the brainstem and/or spinal cord are 43 sufficient to produce significant coupling of locomotor and breathing rhythms during bipedal locomotion. In addition, having controlled for the po s s i b l e effects of limb movement rate on entrainment, this study s u g g e s t s that, although the number of st e p s taken per breath is related to work rate, the degree of locomotor-respiratory coordination is completely independent of metabolic work rate. 44 CHAPTER 2 EFFECTS OF CHANGES IN LOCOMOTOR INTENSITY, HYPOXIA AND HYPERCAPNIA ON LOCOMOTOR-RESPIRATORY SYNCHRONY DURING WALKING/RUNNING IN THE CANADA GOOSE 45 INTRODUCTION E x p e r i m e n t s examining the effects of limb proprioception on entrainment in humans during cy c l i n g a n d running activity have yiel d e d inconsistent results. S o m e indicate an effect of limb movement frequency on entrainment (Bechbache & Duffin, 1977), others indicate a n effect of metabolic work rate ( J a s i n s k a s et al., 1980), while still others s h o w neither (Kohl et al., 1981; P a t e r s o n et a!., 1986). At the centre of this problem has been the failure to examine the effects of i n c r e a s e d limb movement rate on entrainment rate while maintaining a constant metabolic work rate, a n d v i c e v e r s a (Bramble & Carrier, 1983). Furthermore, in the s t e a d y state of moderate e x e r c i s e , ventilation has been shown to be tightly c o u p l e d to metabolic C 0 2 production (see review by W a s s e r m a n et ah, 1986). Thus, when limb movement rate a n d metabolic rate are allowed to change concomitantly, it is not po s s i b l e to distinguish between locomotor drive, metabolic rate and ventilatory drive a s the factor affecting the degree of locomotor-respiratory coupling. Experiments with g e e s e d e s c r i b e d in C h a p t e r 1, however, have s hown that when stride f requency is constant, entrainment rates are independent of the c h a n g e s in oxygen uptake ( V 0 2 ) , C 0 2 output ( V ^ ) a n d ventilation produced through i n c r e a s e d work rate. A s a c o n s e q u e n c e , g e e s e offer an opportunity to examine the independent effects of locomotor drive on entrainment. Entrainment c o u l d result from either the reciprocal interaction between two oscillators, e a c h oscillator affecting the output of the other, or from unidirectional interactions where the driving oscillator forces the s e c o n d oscillator to the frequency of the first ( V i a l a , 1986). During exercise, it is generally a c c e p t e d that locomotor 46 rhythm influences respiratory rhythm (Iscoe & P o l o s a , 1976; B e c h b a c h e & Duffin, 1977; Y o n g e & P e t e r s e n , 1983; Bramble & Carrier, 1983). This s h o u l d not be taken to imply, however, that other non-exercise drives to ventilation have no effect on entrainment. Entrainment d e c r e a s e s during hypoxia in humans (Paterson et a l . 1987) and h y p e r c a p n i a in c a t s ( K a w a h a r a et al., 1989a), and hypoxia l e a d s to a shift in the ratio of fin beat to respiration in C v m a t o a a s t e r agare a t a (Webb, 1975). T h us, c h a n g e s in ventilatory drive also affect the coupling of locomotor and respiratory s y s t e m s . The g o a l s of this study were therefore: (1) to determine whether incr e a s i n g locomotor movements enhance entrainment and (2) to examine the effects of i n c r e a s e d c h e m i c a l drives to respiration on entrainment in C a n a d a g e e s e . To this end we e x a m i n e d the relationship between locomotor and respiratory patterns: (1) while birds walked/ran over a range of velocities on a treadmill and overground and (2) while the birds respired normal air, and hypoxic and hypercapnic gas mixtures during treadmill walking. MATERIALS & METHODS Treadmill Experiments (i) Training and equipment. The four intact birds d e s c r i b e d in chapter 1 (body weight, 4.2 + 0.2 kg) were also u s e d in t h e s e st u d i e s . T hus the training procedures a n d equipment u s e d to measure minute ventilation (V E), tidal volume (V T), breathing frequency (f v), oxygen uptake ( V 0 2 ) , C 0 2 output ( V ^ ) , mixed expired gas concentrations ( P E 0 2 and P E C 0 2 ) , 47 a n d stride f requency (f.) during treadmill running have been d e s c r i b e d , in de t a i l , previously (Chapter 1). H y p o x i c an d hypercapnic gas mixtures were produced using precision gas flow meters. L e v e l s of 0 2 a n d C 0 2 were monitored using a n a l y z e r s and d e l i v e r e d at 10 L/min v i a the s y s t e m previously d e s c r i b e d for the measurement of V 0 2 a n d V ^ . (ii) Protocol T h e birds were run on the treadmill at 0.40, 0.52 and 0.72 m/s. V E, V T, f v, V 0 2, Vco2- PEO2> PECC* A N D h w e r e measured continuously for 1 min pre-activity, 8 min of walking a n d 5 minutes of recovery. V a l u e s of V E, V T, f v and f 6 represent means (+ S.E.) of s e v e n runs on four animals and V 0 2 and V ^ v a l u e s represent means (+ S.E.) of three runs on three animals. In another s e r i e s of experiments, the birds were run at 0.52 m/s while breathing one of three different gas mixtures: (1) air, (2) 5 % C 0 2 and (3) 1 2 % 0 2 . The 5 % C 0 2 w a s administered on a hyperoxic background of 3 0 % Cv/balance N 2. Brackenbury (1986) e x p o s e d running ducks to 1 0 % 0 2 and found a 4.0 torr d e c r e a s e in Paco2. Addition of 3 % C 0 2 to the mixture during e x e r c i s e reversed this and resulted in a 1.0 torr i ncrease in Paco;, relative to resting l e v e l s . Thus, in this study, the 1 2 % 0 2 mixture was delivered in a background of 1 % C 0 2 to minimize the large d e c r e a s e in P a ^ j a s s o c i a t e d with hypoxic e x e r c i s e but not c a u s e an a c i d o s i s . U s i n g the s a m e apparatus d e s c r i b e d above, preliminary gas response studies were performed on resting animals (n=4), sitting quietly in a small holding pen, for later c o m p a r i s o n with the gas responses observed during exercise. T h e s e studies indicated that the level of V E attained following 2 min of exposure to a new 48 g a s w a s not different from that s e e n after 10 min of exposure. J o n e s & P u r v e s (1970) a l s o found that 9 0 % of the increase in V E a s s o c i a t e d with inspiration of 4 % C 0 2 o c c u r s within the first 48 s of exposure. Thus, following 2 min of exposure to a new g a s mixture, V E, V T, f v > V 0 2, and were monitored for 1 min pre-activity, 8 min walking a n d 5 min recovery during the administration of the 5 % C 0 2 a n d 1 2 % 0 2 gas mixtures. A l l v a l u e s represent means (+ S.E.) of three runs on three animals. Overground, Free-Running Experiments (i) Training and equipment. T o examine the diffe r e n c e s in locomotor-respiratory coupling between treadmill running a n d overground free-running, a s well as examine the effects of f s on entrainment over a larger velocity range than was possible on the treadmill, a s e c o n d group of g e e s e was raised from hatching. T h i s group was treated virtually the s a m e a s the previous group. For their first 3 months they wore a cylindrical f ace mask for 1 h per day. T h e s e birds were also fitted with lightweight nylon h a r n e s s e s . A c a n v a s strap was attached to the mesh harness and p a s s e d just c a u d a l to the wings onto the middle of the back. The harness did not interfere with wing or leg movements in any way. At 3 months of age, the birds were introduced to a walled circular walkway (fig. 2-1). In the centre of the circular walkway was a double aluminum pole. The outer pole was fixed while the center pole rotated within the outer pole. The center 49 Figure 2-1: S c h e m a t i c diagram of the apparatus u s e d in the free-running experiments as s e e n in top view (A). T h e arrow in A indicates the angle from which the s i d e v i e w (B) w a s t a k e n . In the s i d e view, the outer wall is not s h o w n a n d the inner wall h as been cut away to show the central support. f s, stride frequency; f v, breathing frequency. 50 Outer and inner walls (100 cm high) Tether • Grass preamplif ier -Wheatstone bridge C e m r a l „ support •Harness v \ ,„ n I Cross b e a m (540 cm) Mask/thermistor | ^ | | ^Counter-weight Inner wall (100 cm high) Wheatstone bridge Grass preamplif ier Tether L-bar (150 cm) Centra l |— support (165 cm) Chart recorder J I Strain gauge Harness Commuta to r 51 pole supported a cross-beam which had a counterweight at one end a n d an L-s h a p e d b a r at the other. T h i s additional b ar (L-bar) supported a d e c o y g o o s e at its leading e nd. T h e g e e s e were tethered to the trailing e n d of the L-bar v i a a 200 c m tether that c l i p p e d to the c a n v a s strap on the harness. The c h a i n c o n t a i n e d two se t s of recording l e a d s . O n e set carried breathing frequency information from a thermistor (Fenwall 112-202EAJ-B01) p l a c e d in the mask, overlying a nostril, to a Wheatstone bridge that w a s attached to the trailing end of the L-bar. The mask had a d e a d s p a c e volume of l e s s than 2.0 mL, a n d thus i n c r e a s e d total d e a d s p a c e by l e s s than 5 % (F e d d e et al., 1986). T h e s e c o n d set of leads carried the sig n a l from a strain gauge to a preamplifier ( G r a s s model P15D) and indicated stride frequency. T h e birds were not bothered by the chai n and leads a n d eagerly c h a s e d the decoy g o o s e around the cir c u l a r walkway. A major objective of this part of the study was to ensure that we did not force (i.e. pace) the locomotor rhythm a s occurs on a moving treadmill. T hus the " w a l k e r d e v i c e was not motor driven. T h e experimenters manually rotated the sy s t e m within a given veloc i t y range, but e a c h bird ultimately s e l e c t e d its own precise velocity and locomotor rhythm. Forward tension was never pla c e d on the chain . A n event marker indicated the completion of e a c h lap and allowed calculation of average locomotor velocity p er revolution. (Ii) Protocol. Stride frequency a n d fv were recorded during running at three velocities. S i n c e we did not want to force locomotor rhythm a n d the birds d i d not always run at a constant velocity, the three s p e e d levels e n c o m p a s s e d fairly wide ranges. The first 52 range e n c o m p a s s e d the s p e e d s produced on the treadmill (0.40 - 0.72 m/s). T h e s e c o n d range v a r i e d between 1.0 a n d 1.5 m/s, while the third range v a r i e d from 1.5 to 2.0 m/s. T h e birds were run for a minimum of 8 min at the lowest s p e e d , 5 min over the middle range, a n d 4 min at the highest velocity. O n e replicate w as performed, s o that v a l u e s of f s, f v, a n d percent entrainment obtained during free running represent means (+ S.E.) of 10 trials on five animals. DATA ANALYSIS The degree of locomotor-respiratory entrainment during treadmill and over-ground running w a s determined a s d e s c r i b e d in Chapter 1. Res p i r a t o r y d a t a c o l l e c t e d during treadmill walking/running at different v e l o c i t i e s were a v e r a g e d over the last 4 min of each run before comparison between s p e e d s . V a l u e s of f s, f v a n d percent entrainment from the free-running experiments represent a v e r a g e v a l u e s r e c o r d e d over the co u r s e of e a c h trial. S i n c e the lowest velocity range o b s e r v e d during overground running e n c o m p a s s e d the entire range of s p e e d s produced on the treadmill, all f 8, f v and entrainment data from the treadmill experiments were p o o l e d (i.e. 21 measurements from four animals) before c o m p a r i s o n with the free-running v a l u e s of f s, f v a n d percent entrainment. Unless otherwise stated, A N O V A and Tukey's multiple comparison test were u s e d to test the differences between means of all entrainment a n d respiratory d a t a. V a l u e s of p<0.05 were a s s u m e d significant. 53 RESULTS Effects of Treadmill Speed on Entrainment V E i n c r e a s e d approximately 2.5 - 3.0 times immediately upon the onset of treadmill walking (fig. 2-2). V T remained unchanged from resting levels and thus the c h ange in f v w a s s o l e l y responsible for the initial i ncrease in V E from rest to walking. Increases similar in magnitude to those s e e n in V E were also o bserved in V D 2 and with the onset of treadmill walking. The magnitude of these initial i n c r e a s e s in V E, f ¥ > V 0 2 a n d a p p e a r e d to be independent of treadmill velocity. C o n v e r s e l y , "steady-state" levels of these variables (i.e. v a l u e s taken over the last 4 min of the trial) were affected by treadmill velocity. T h e increase in velocity from 0.40 m/s to 0.72 m/s produced s m a l l , further i n c r e a s e s in and V 0 2 a n d significant i n c r e a s e s in V E, f v and f s (fig. 2-2, 2-3). A g a i n f v was solely responsible for the i n c r e a s e in V E a s s o c i a t e d with the i n c r e a s e s in velocity. Breathing frequency continued to rise throughout the last 4 min of treadmill walking at all s p e e d s , the degree d e p e n d i n g on velocity (20.5 ± 12.6% at 0.40 m/s; 26.9 ± 10. 8 % at 0.52 m/s; 30.1 + 7 . 1 % at 0.72 m/s). V T fell marginally over the last 4 min of e ach run, at all treadmill s p e e d s , s u c h that V E remained relatively constant at 0.40 and 0.52 m/s (fig. 2-2). At 0.72 m/s however, V E i n c r e a s e d continuously during the last 4 minutes as f v continued to i n c r e a s e more than V T d e c r e a s e d (fig. 2-2). Locomotor and respiratory patterns displayed significant coupling at all velocities. All entrainment w a s subharmonic, and coupling ratios of 1.5 and 2 strides/breath were the most common; however, coupling ratio v a l u e s ranged from 2:1 to 6:1, the higher ratios occurring at the onset of treadmill locomotion. Entrained 54 Figure 2-2: Effect of treadmill walking/running at 0.40 m/s (open ci r c l e s ) , 0.52 m/s (filled s q u a r e s ) and 0.72 m/s (filled circles) on minute ventilation (V E), tidal volume (V T) a n d breathing frequency (fv) in C a n a d a g e e s e , (four birds, n=7; mean + S.E.) 55 Time (min) 56 Figure 2-3: Effect of treadmill velocity on mean levels of minute ventilation (VE), breathing frequency (fv), stride frequency (fs) (four birds, n=7), 02 uptake (V02 (open circles)) and C0 2 output ( V ^ (filled circles)) (three birds, n=3) averaged over the last 4 min of 8-min walking/running trials, (mean + S.E.) 57 —I L. 1 0.00 0.20 0.40 0.60 Velocity (m/s) 58 Figure 2-4: Effect of locomotor velocity during free running (open circles) (five birds, n=10) a n d treadmill running (filled circles) (four birds, n=7) on stride frequency (f 6), breathing frequency (fv) and the degree of locomotor-respiratory coupling (% Entrainment; — = 2 0 % ; the maximum entrainment predicted by chance) (data points = mean + S.E.). S i n c e the lowest velocity range employed during free-running e n c o m p a s s e d all three s p e e d s u s e d during treadmill running, all data from the treadmill experiments were pooled prior to comparison with free-running v a l u e s . 59 n i t I I I I 1 1 1 « — 0 0.40 0.80 1-20 1.60 Velocity (m/s) 60 periods v a r i e d in duration from 10 s (experimental minimum) to 2 min, but were most frequently in the 30-60 s range. Se v e r a l s u c h periods were s e e n during e a c h tria l . T he distribution of entrained periods did not appear to be b i a s e d towards any particular portion of the tri a l s . Entrainment was as likely to occur at the onset of a trial a s at the end. Ther e were no statistically significant effects of treadmill velocity on the percentage of entrainment or the coupling ratio. However, a slight i ncrease in coupling from 28.3 to 3 4 . 8 % ac c o m p a n i e d the increase in treadmill velocity from 0.40 t o . 0.52 m/s. A s velocity w as increased further to 0.72 m/s, however, entrainment returned to 27.1 % (fig. 2-4, bottom panel). Effects of treadmill and overground locomotion on entrainment Stride f requency i n c r e a s e d similarly during both treadmill walking a n d overground walking (fig. 2-4). However, the f v recorded with the form-fitting mask and pneumotachograph on the treadmill were significantly greater than those measured with the mask a n d thermistor u s e d during the overground s e r i e s of experiments under conditions of both rest a n d ex e r c i s e . This difference may have been related to the slightly greater d e a d s p a c e or resistance of the form-fitting mask, although inc r e a s e s in d e a d s p a c e (Fedde et al., 1986) a n d resistance (Hof et a]., 1986) have not been a s s o c i a t e d with i n c r e a s e s in f v. Alternatively, the form-fitting mask may have reduced the potential for evaporative heat loss which, depending on ambient temperature, c a n play a significant role in thermoregulation (Hudson & Bernstein, 1981). A s i d e from the shift in baseline f ¥, however, it c a n be s e e n from fig. 2-4 that f v rises similarly with velocity during both treadmill and overground running, regardless of the method u s e d to measure it. 61 A s s e e n during treadmill running, locomotor and respiratory rhythms d i s p l a y e d significant coupling at all vel o c i t i e s during overground locomotion. All entrainment was s ubharmonic and, a s s e e n at s p e e d s similar to those observed during treadmill running, the coupling ratio of 2 strides/breath w as most common. In the middle velocity range (1.0-1.5 m/s), the most common ratio w as 2.5 strides/breath, while at the highest s p e e d s (>1.5 m/s), 3.0 strides/breath were most commonly observed. Co u p l i n g ratio v a l u e s ranged from 1.5:1 to 4:1 (i.e. 3 steps/breath to 8 steps breath). Entrained periods v a r i e d in duration from 10 s to complete entrainment from start to finish (4 min total, s e e n in one animal on one occ a s i o n ) . The d e g r e e of coupling between locomotor and respiratory rhythms was significantly greater during overground running than during treadmill running. In addition, there w a s a significant i n c r e a s e in the degree of entrainment when running velocity w a s i n c r e a s e d from 0.51 to 1.58 m/s during overground locomotion (one-tailed t-test, p<0.026). Effects of Inspired Gas Composition on Entrainment F i g . 2-5 illustrates the steady-state ventilatory responses of the birds to hypoxia and hypercapnia under rest and exercise conditions. S i n c e the resting values of V E, V T and f v shown in fig. 2-5 were taken from a separate s e r i e s of experiments (see Materials and methods), they do not correspond exactly to the val u e s s e e n at rest in fig. 2-2. Under resting conditions, exposure to 5 % C C y 3 0 % 0 2 resulted in a 2 0 % increase in f v and a significant 8 0 % increase in V T, resulting in a significant 1 0 0 % rise in V E (fig. 2-5). During treadmill walking, hypercapnia was again a s s o c i a t e d with increases in f v and V E. The absolute change in v*E, however, 62 Figure 2-5: Ventilatory r e s p o n s e s to changes in inspired C 0 2 and 0 2 during rest a n d e x e r c i s e . V a l u e s of V E, V T a n d f ¥ at rest (n=4) a n d e x e r c i s e (n=3) are s h own during air breathing (shaded columns) and during inspiration of 5 % C 0 2 (top three panels, o p e n columns) or 1 2 % 0 2 (bottom three panels, open columns), (mean + S.E.). S e e text for further de t a i l s . 63 C 0 2 BREATHING (ml/min/Kg) V j (rri/Kg) fy (1/min) 1500 Y 40 1000 0 2 BREATHING 1500 Y 1000 Y 20 Rest Walk Rest Walk Rest Walk 64 Figure 2-6: Percent of total time entrained while running at 0.52 m/s breathing air (four birds, n=7), 5 % C C y 3 0 % 0 2 and 1 2 % 0 / l % C 0 2 (three birds, n=3). Individual data points are shown (filled circles) a s well as mean levels (columns, ± S.E . ) (— = 2 0 % ; the maximum entrainment predicted by chance). 65 60 %Ent 50 AO 30 20 10 ¥ T 0" Air 5%C02/30% 0 2 12%02/1%C02 Inspired Gas Composition was only half that s e e n in response to 5 % C 0 2 at rest, a s V T did not inc r e a s e in response to C 0 2 during e x e r c i s e . E x p o s u r e to 1 2 % 0 / 1 % C 0 2 under resting conditions resulted in much sm a l l e r i n c r e a s e s in f v a n d V T than s e e n in response to C 0 2 (fig. 2-5) a n d only a 4 0 % increase in V E. In contrast, a significant 1 0 0 % increase in VE a c c o m p a n i e d hypoxia during e x e r c i s e . Breathing frequency rose continuously during hypoxic e x e r c i s e ( 3 5 . 3 % between 2 a n d 8 min) a n d was primarily responsible for the increase in VE . C o u p l i n g of locomotor a n d respiratory s ystems was only significantly greater than the maximum which c o u l d occur due to chance during air-breathing (fig. 2-6). T h e rate of entrainment o b s e r v e d during the hypoxic trials (19.0 ± 6.9%) was significantly l e s s than that s e e n during air-breathing (34.8 ± 3.9%) a n d the former could p o s s i b l y be ac c o u n t e d for by chance alone. In response to C 0 2 , two animals s howed approximately 2 0 % entrainment while one animal demonstrated entrainment 5 9 % of the time. B e c a u s e of this large variability, the mean level of coupling during hypercapnia (32.8 ± 13.0%) was not significantly different from either c h a n c e ( 2 0 % ) or air-breathing v a l u e s . DISCUSSION 0 2 and C0 2 Sensitivity During Rest and Exercise A s s h o wn by fig. 2-5, there were marked differences in the ventilatory responses to hypoxic and hypercapnic g a s e s between rest and treadmill e xercise conditions. Unfortunately, in the a b s e n c e of blood g a s data, it is very difficult to 67 account for t h e s e c h a n g e s . The magnitude of the ventilatory response to hypoxia a p p e a r e d to be greatly e n h a n c e d during treadmill walking in geese. Increases in hypoxic sensitivity have b e e n documented previously in the c h i c k e n (Brackenbury et a]. 1982b). A l t h ough t h e s e results may represent a state-dependent difference in hypoxic sensitivity, inspiration of 1 2 % 0 2 during exercise probably resulted in lower Pao 2 levels than t h o s e produced by inspiration of 1 2 % 0 2 at rest. We attempted to measure le v e l s of arterial blood g a s e s in the present study but this experimental procedure p r o d u c e d substantial c h a n g e s in the respiratory r e s p o n s e s of the g e e s e to simple treadmill walking and the d a t a were di s c a r d e d . In contrast to the apparent increase in hypoxic sensitivity, there appeared to be a d e c r e a s e in C 0 2 sensitivity of these birds during treadmill exercise (fig. 2-5). Breathing frequency i n c r e a s e d consistently during rest a n d e x e r c i s e . However, the large i n c r e a s e in V T o b s e r v e d during rest w a s not s e e n during treadmill walking. C onversely, Brackenbury et a l . (1982b) found no change in the ventilatory response to C 0 2 during e x e r c i s e in the c h i c k e n , although he did show a d e c r e a s e in the response of V T to C 0 2 from rest to exe r c i s e . Without arterial partial p ressures of C 0 2 , however, it is difficult to a s s e s s sensitivity properly. T h e s m a l l s a m p l e s i z e a n d the higher degree of variablity of the ventilatory responses to hypoxia and hypercapnia during e x e r c i s e relative to rest may also account for some of the differences. 68 Locomotor-Respiratory Entrainment C o n s i d e r a b l e attention has b e e n given to the relationship between locomotor a n d ventilatory patterns during bipedal activity (walking, running a n d cycling) in humans. However, there are few d a t a describing the relationship between stepping frequency and breathing frequency in birds. Evidence of air s a c pressure pulses related to footfall during treadmill running in d u c k s (Brackenbury, 1986) has been presented anecdotally, but the exact relationship between locomotor and respiratory rhythms has not b e e n previously e x a m i n e d in g e e s e . O n e advantage of using birds to examine the interaction of hindlimb locomotor activity a n d the respiratory system, is that birds, unlike humans, are true bipeds, and s h o w complete separation of hindlimb and forelimb function. C o n v e r s e l y , humans still p o s s e s s quadrapedal motor programs, so that the effects of arm swing on respiratory pattern must also be c o n s i d e r e d . That this is important is shown by the observation that human entrainment rates are consistently higher during running (Bechbache & Duffin, 1977; Bramble, 1983) than, during cycling (Bechbache & Duffin, 1977; Y o n g e & P e t e r s e n , 1983). T h i s difference has previously been attributed to d e c r e a s e d v i s c e r a l movement during cy c l i n g c o m p a r e d with running (Yonge & P e t e r s e n , 1983). In addition, entrainment of human ventilation by arm motion alone has been documented (Paterson et a l . , 1986). Decerebrate geese, however, have b e e n shown to have entrainment rates identical to intact animals, in spite of the fact that the decerebrate geese were s u s p e n d e d in a sling that drastically r educed v i s c e r a l movements during locomotion (Chapter 1). This, combined with the complete separation of hindlimb and forelimb function, may make birds the animal of c h o i c e for separating hindlimb-respiratory interactions from the 69 forelimb-respiratory interactions. T h e rates of co u p l i n g found in g e e s e during treadmill walking/running were significantly greater than c h a n c e (a maximum of 3 4 . 8 % ) , but substantially l e s s than the maximum s e e n in humans during s o m e studies of treadmill running. Although K a y et a l . (1975) found no evidence of entrainment between locomotor and respiratory patterns during human treadmill running, Hill et a l . (1988) found 2 9 % entrainment during treadmill walking and Pat e r s o n et a l . (1987) more recently found the two patterns to be c o u p l e d up to 4 5 % of the time during treadmill running. There are s e v e r a l factors that c o u l d account for this reduced degree of synchrony in g e e s e . First, the use of 10 s intervals c o m p a r e d with l e s s than 1-s intervals (Paterson et al., 1987) to calculate coupling ratios may account for s ome of the differences (Chapter 1). S e c o n d , as d i s c u s s e d above, forelimb interactions may c a u s e a n i n c r e a s e in entrainment in humans. Finally, in birds, respiration, through evaporative heat l o s s , plays a significant role in body temperature control. S m a l l i n c r e a s e s in body temperature c a n le a d to significant c h a n g e s in breathing pattern, primarily through i n c r e a s e s in breathing frequency (Brackenbury et a]., 1981). Thus, i n c r e a s e s in f v a s a result of c h a n g e s in body temperature (T B), c o u l d disrupt the relationship between locomotion and respiration. Collectively, these factors c o u l d account for the 1 0 % difference in entrainment rates s e e n between humans and g e e s e during treadmill running. 70 Effects of Increasing Locomotor Frequency on Entrainment (i) Treadmill J u s t a s the effects of limb movement rate on entrainment are not cl e a r in humans, the present results with g e e s e during treadmill running were not definitive. It s h o u l d be noted that, owing to the difficulty of increasing stride frequency independently of work rate a n d V E, no attempt was made to control work rate as the rate of limb movement was i n c r e a s e d . However, this was not felt to be crucial s i n c e the i n c r e a s e s in metabolic rate and V E a s s o c i a t e d with i n c r e a s e s in treadmill velocity o v er this range were s m a l l . Furthermore, previous work has shown entrainment rate to be independent of the c h a n g e s in metabolic rate and V E a s s o c i a t e d with i n c r e a s e s in work rate in t h e s e animals (Chapter 1). Like the i n c r e a s e in entrainment o b s e r v e d with increasing locomotor frequency in humans (B e c h b a c h e & Duffin, 1977), i n c r e a s e s in entrainment in g e e s e , although not significant, were a s s o c i a t e d with i n c r e a s e s in stride frequency over the 0.40 -0.52 m/s range. With further i n c r e a s e s in velocity to 0.72 m/s, however, f v i n c r e a s e d by 3 0 . 1 % o v e r the 8 min run and entrainment fell to 2 7 . 1 % . Panting was o c c a s i o n a l l y o b s e r v e d at the e n d of the 0.72 m/s runs. R e c t a l temperature measurements replicated twice on one animal at all three velocities s h o w e d in c r e a s e s of 0.85°C at 0.72 m/s. Thermal hyperventilation has been a s s o c i a t e d with i n c r e a s e s in body temperature of this magnitude during treadmill running in d u c k s (Kiley et al., 1979). Smaller temperature c h a n g e s of 0.50°C, similar to those s e e n in g e e s e running at 0.40 and 0.52 m/s, did not result in thermal panting in fowl (Brackenbury, 1986). Thus, although not conclusive, the d a t a from the treadmill 71 study s u g g e s t that i n c r e a s e d locomotor output, whether acting v i a afferent (Iscoe & P o l o s a , 1976) or efferent interactions (Viala et al., 1987b), i n c r e a s e s locomotor-respiratory c o u p l i n g . However, there appears to be a limit to this coupling, where further i n c r e a s e s in velocity are a s s o c i a t e d with sufficient i n c r e a s e s in T B that entrainment is s u p p r e s s e d in favor of thermoregulation. (ii) Free running. A s s e e n with humans (Paterson et al., 1987), entrainment during overground running w a s significantly greater than during treadmill running. Similarly, entrainment d e c r e a s e s when preferred c y c l e rhythm, c h o s e n by the subject, is replaced with an i m p o s e d pedalling rate ( J a s i n s k a s et al., 1980). It is possible that the treadmill, by forcing locomotor frequency, upsets any internal rhythm that the locomotor s y s t e m may prefer (Paterson et a]., 1987), an d thereby d e c r e a s e s entrainment. In contrast to treadmill running, the extent to which locomotor and respiratory s y s t e m s were c o u p l e d during overground running i n c r e a s e d significantly with increasing velocity. T h i s difference may have been due to the larger velocity range u s e d during free running (0.40-0.72 m/s on the treadmill v s 0.51-1.58 m/s outside). However, if this were the c a s e then it would be e x p e c t e d that entrainment would have s hown a continuous increase a s treadmill velocity i n c r e a s e d from 0.52 to 0.72 m/s, rather than the o b s e r v e d d e c r e a s e . Thus, a s already s u g g e s t e d , increased thermoregulatory d e m a nds on ventilation may have been responsible for this d e c r e a s e in entrainment during treadmill locomotion. During overground locomotion, although body temperature was not measu/ed, the birds s h o w e d no s i g n s of thermal s t r e s s at the termination of even the fastest trials. Thermal panting was 72 never o b s e r v e d . T h i s difference in heat s t r e s s between treadmill and overground locomotion is probably due to the impairment of the two major heat los s m e c h a n i s m s during treadmill walking. Evaporative heat l o s s , v i a the upper airway, would be reduced by the form fitting mask. In addition, convective heat l o s s , which is dependent on air flow o ver the body, would be greatly reduced during treadmill locomotion. During overground locomotion, the mask was not form fitting and air flowed over the body. Thus, from the results presented here a n d the previous demonstration that entrainment is not affected by work rate in g e e s e (Chapter 1), it a p p e a r s that if thermal s t r e s s e s are reduced, i n c r e a s e s in limb movement frequency alone e nhance locomotor-respiratory coupling. Effects of Hypoxia and Hypercapnia on Entrainment The prevailing ventilatory drive appears to affect the strength of locomotor-respiratory c o u p l i n g . Like previous results in humans (Paterson et a]. 1987), an in c r e a s e d hypoxic drive to breathe led to a reduction of entrainment to levels which c o u l d simply be due to c h a n c e in all three g e e se tested. Breathing frequency rose continuously over the c o u r s e of the 8 min run, c a u s i n g a two fold increase in V E relative to air-breathing l e v e l s . The reasons for the continuous rise in f v with acute exposure to hypox i a remain unclear. However, the s e data suggest that the effects of i n c r e a s e d peripheral chemoreceptor drive on breathing pattern were incompatible with the effects of the locomotor system on ventilatory pattern. Consequently entrainment w a s s u p p r e s s e d . In the teleost, Cvmatoqaster aqqreqata, although hypoxia d o e s not s u p p r e s s entrainment, the relationship between locomotor and respiratory movements is altered to accomodate increased ventilation (Webb, 1975). 73 The effects of hyper c a p n i a on the relationship between respiration a n d locomotion were l e s s c l e a r . Two animals s h owed a fall in coupling to c h a n c e levels while one s h o w e d an inc r e a s e to approximately 6 0 % entrainment. The wide variability of the entrainment d a t a makes it very difficult to determine the effects of a hypercapnic ventilatory drive on the relationship between respiration a n d locomotion in g e e s e . Recent experiments during "fictive" locomotion in cats suggest that entrainment d e c r e a s e s with increasing end tidal ( K awahara et al., 1989a). T h e low e n d tidal P ^ periods a s s o c i a t e d with high levels of entrainment o c c u r e d when the frequency of pump ventilation w as high (i.e. similar to that of locomotion). P u m p frequency w a s lower at higher end-tidal P ^ ' s . S i n c e respiration c a n be entrained by a number of rhythmic inputs (Eldridge, 1972a & b; Remmers & Marttila, 1975), it is not c l e a r from these studies whether locomotion and respiration were entrained. If they were entrained, it is not cle a r whether i n c r e a s e d C 0 2 or reduced pump freq u e n c y w as responsible for the decline in entrainment. H y p e r c a p n i a in horses, like hypoxia in Cvmatoqaster, leads to a shift in the relationship between locomotion and respiration without interupting entrainment (Dempsey, personal communication). CONCLUSION Although there is conflicting e v idence a s to the effects of velocity on locomotor-respiratory coupling, an increase in coupling with velocity during overground running in g e e s e has been clearly demonstrated. It has been suggested that these increases in entrainment are not due to incr e a s e s in metabolic rate, but to 74 i n c r e a s e s in limb movement frequency alone. Variations in experimental d e s i g n may account for much of the current confusion surrounding the effects of i n c r e a s e d velocity on entrainment, a s there are significant differences in the degree of entrainment o b s e r v e d during treadmill and overground running. Ventilatory drive has a l s o been shown to affect the strength of coupling between locomotor a n d respiratory s y s t e m s . Ventilatory stimuli s u c h as hypoxia and elevated body temperature a p p e a r to reduce entrainment. The various inputs affecting ventilatory pattern a p p e a r to be arranged in a hierarchy, with locomotor entrainment n e a r the bottom. U n d e r certain conditions the influence of locomotor pattern on respiration is e x p r e s s e d , but any additional ventilatory stimuli, e s p e c i a l l y those that mediate their effects through an increase in breathing frequency, s u c h as i n c r e a s e d body temperature or hypoxia, a p p e a r to override the entraining effects of locomotor pattern on respiratory rhythm and entrainment is either reduced or s u p p r e s s e d . The examination of locomotor-respiratory coordination during bipedal locomotion in the goose has been useful in sorting out several of the i s s u e s regarding the control of entrainment that a r o s e from work on humans. However, it i s still apparent that entrainment during this type of locomotion, at least in the laboratory, is intermittent. A s y s t e m which demonstrates tight coordination between locomotion a n d respiration would greatly facilitate initial s t udies of entrainment control. During free-flight, locomotion and respiration are tightly coordinated in a variety of birds (see C h a p t e r 3). T h u s the remainder of this thesis e xamines the mechanisms involved in the production of entrainment during flight. The problem has been approached from a broad perspective, examining the involvement of feedback a n d feedforward m e c h a n i s m s in the production of locomotor-respiratory coordination. 75 CHAPTER 3 COORDINATION OF WINGBEAT AND RESPIRATION IN THE CANADA GOOSE I. FREE-FLIGHT 76 INTRODUCTION V e r y little information is available on the physiology of flight in birds, primarily due to the tech n i c a l difficulties in obtaining measurements from flying animals. In spite of this, there were reports as early as 1890 (Marey, cited in Hart & Roy, 1966) that the respiratory and wingbeat frequencies of pigeons are coupled in a 1 to 1 f a s h i o n . G r o e b b e l s (1932) and Zimmer (1935; both cited in Hart & Roy, 1966) observed a similar s ynchronization. More recently, Tomlinson (1957), using high s p e e d cinematography of a pigeon in flight with a balloon p l a c e d over its beak, also found a 1:1 relationship between respiratory frequency and wingbeat frequency. In addition, he s h o w e d that inspiration occured during the upstroke and expiration during the downstroke. Later studies using radiotelemetry during short duration (10 s) tethered flight (Hart & Roy, 1966), sustained tethered flight (10 min) in a wind tunnel, and telemetry during sustained free-flight (Butler et a]., 1977), all corroborated these earlier findings. Studies on other, larger s p e c i e s of birds also indicated that respiration and wingbeat were coordinated. However, there was some variation from the 1:1 relationship between wingbeat and breathing frequency (1Jiv) s e e n in pigeons. Lord et al- (1962), again using radiotelemetry, found a 2:1 relationship between wingbeat frequency a n d breathing frequency during free-flight in the wild mallard (Anas platvrhvnchos). Berger et a l . (1970a), using telemetry during short term tethered flight, found that f v/f v varied from 1:1 to 5:1 depending on s p e c i e s , several s p e c i e s showing more than one ratio. More recently, Butler and W o a k e s (1980) recorded that the ratio of wingbeats per breath during sustained flight in trained (imp'rinted) 77 barnacle g e e s e (Branta leucopsis) was predominantly 3:1, although some individual breaths o c c u r e d with ratios of 2:1 and 4:1 (refer to table 3-1 for summary). Not all studies have recorded s u c h high degrees of entrainment. Tomlinson's (1963) observ a t i o n s on the western gull, lesser s c a u p , pintail and mallard during short duration tethered flight, s u g g e s t e d that wingbeat and respiration were not always c o u p l e d . T o m l i n s o n (1963) simply averaged the number of breaths taken over a unit time by the number of wingbeats s e e n over the s ame interval. A s the coupling ratio ( y f v ) appears to vary in many s p e c i e s (Berger et al., 1970a), this method of a n a l y s i s will not adequately a s s e s s entrainment. Thus, it a p p e a r s that many, if not a l l , s p e c i e s examined do coordinate their breathing movements with wing motion. In spite of these observations, the m echanisms responsible for the coupling of the two motor systems have not been examined. Before any of the questions pertaining to the mechanisms responsible for entrainment of wingbeat and respiration could be a d r e s s e d , however, it was first n e c e s s a r y to d e s c r i b e the relationship between these two variables in our animal of choice, the C a n a d a goose (Branta canadensis). C a n a d a g e e s e were well suited for these studies. First, their greater s i z e facillitated the use of externally mounted radiotransmitter. d e v i c e s , and the relative increase in drag a s s o c i a t e d with the externally mounted transmitters was less than it would be for a small bird. S e c o n d , the e a s e with which geese could be imprinted and trained made it possible to carry out the measurements during sustained free-flight, with minimal risk of equipment loss. Butler & W o a k e s (1980) first used the technique of imprinting goslings on an experimenter to obtain flight recordings for extended periods. I felt it was n e cessary 78 to repeat t h e s e experiments using C a n a d a geese for two reasons. First, Butler & W o a k e s were only s u c c e s s f u l in making recordings from 2 animals. T h e fact that they o b s e r v e d a range of coupling ratios from 2:1 to 4:1 with a predominance of 3:1, rather than a fixed ratio, left open the possibility that other birds would preferentially couple the two rhythms at some multiple other than 3:1. S e c o n d , C a n a d a g e e s e (average weight u s e d in our study; 3.8 + 0.2 kg) are considerably larger than the barnacle g e e s e u s e d by Butler & W o a k e s (1.5 and 1.7 kg). The s i z e of the bird may influence the nature of the locomotor-respiratory synchronization (Tomlinson, 1963). In addition, telemetry d e v i c e s have been shown to c a u s e significant i n c r e a s e s in drag under some conditions (Obrecht et al., 1988). This may also affect the relationship between wingbeat frequency and respiratory frequency. Although it is largely believed that the locomotor system entrains the ventilatory system (Iscoe & P o l o s a , 1976; B e c h b a c h e & Duffin, 1977; Y o n g e & P e t e r s e n , 1983), there is i n c r e a s i n g evidence that prevailing non-exercise ventilatory drives such a s hypoxia and/or hypercapnia, also affect the strength of the coupling. Hypoxia d e c r e a s e s the amount of locomotor-respiratory coupling in g e e s e (Chapter 2) and in humans (Paterson et al., 1987) during treadmill locomotion. In addition, there is s o m e ev i d e n c e that synchronicity of leg and phrenic nerve discharge d e c r e a s e s during fictive locomotion in paralyzed cats as end-tidal C 0 2 increases (Kawahara et a[., 1989). Thus the goals of this study were; 1) using telemetry to record f v, and v i d e o a n a l y s i s to record f w , to describe the relationship between wingbeat and respiration in the C a n a d a goose during free-flight; 2) to establish the relationship between respiration and wingbeat during flight in completely uninstrumented birds, using only video analysis of wing motion to establish f w and of 79 mouth movements (shown by Butler & W o a k e s (1980) to reliably indicate respiration) to measure respiration; and 3) to determine if alterations in (non-exercise) respiratory drives affect the way in which wingbeat and respiration are coupled. MATERIALS & METHODS The birds u s e d in this study were the same birds used in the free-running experiments d e s c r i b e d in chapter 2. The training these animals (N=5, Wt = 3.8 + 0.2 kg) r e c e i v e d for their first 4 months is described more fully in chapter 2. At approximately 4 months of age, the g e e se had fully d eveloped flight feathers and the flight training was initiated. Every morning, the mesh harnesses (described in chapter 2) that would later facilitate the attachment of a radiotransmitter, were placed on the birds. In addition, a small cylindrical mask was placed on each bird. The mask contained a thermistor bead (Fenwall 112-202EAJ-B01), positioned to overlie a nostril. The thermistor, sensitive to the changes in temperature of inspired versus expired air, provided a measure of breathing frequency. Once the equipment was in place, individual animals were taken outside and released. I would then get on my bicycle and ride down the road. Invariably the geese would try to run after me. When they were unable to keep up, they would attempt to fly. Initial attempts at flight met with varying results. Most birds achieved lift off with little difficulty, but very few mastered cornering or landing on first attempts. T h e road over which these training flights were performed was approximately 1.0 km long and each bird travelled up and down it twice per training 80 s e s s i o n . The birds were taken through this procedure daily for approximately three weeks, at which time their flight capabilities far e x c e e d e d my cycling ability. Experimental trials were then begun. Prior to the onset of the experimental trials, e a c h bird was fitted with a nylon jacket and face mask. A radiotransmitter (Wt = 19.5g; length=5.5cm; diameter=2.0cm; d e s i g n e d by Dr. F. Smith) was attached, v i a a plastic carrier, onto the back of the bird. T h e antenna (30.0cm of s i z e 8 piano wire) ran caudally along the dorsal surface of the animal and extended approximately 5-10 cm behind the bird. Two l e a d s from the thermistor were attached to the radiotransmitter. R e c o r d i n g s were made during flight by having the birds fly behind a motorcycle. Dr. G.N. Sh o l o m e n k o drove the motorcycle while I, facing backwards, recorded the birds' flight, a nd therefore wingbeat frequency, using a high s p e e d v i d e o c a m e r a ( J V C Model # GS-CD1U) and recorder ( J V C BR-1600U). A n internal digital c l ock recorded time to the 0.1 s e c o n d on the videotape. The cyclical c h a n g e s in the resistance of the thermistor, representing breathing frequency, were transmitted v i a the radiotransmitter on the animal's back a n d received using an F M radio. The output from the radio was, in turn, recorded on the audio channel of the video recorder (both F M radio and the video recorder were mounted on the motorcycle). The transmission range of the transmitter was less than 3.5 meters. Thus the need for having well trained animals was obvious. The birds generally flew within a corridor of 0.5 to 2 meters from either side of the cycl e , 0 to 3 meters behind the cycle and 1 to 3 meters above the ground. All experiments were carried out between .4:30 and 10:00 A.M. on an isolated 2 km section of road. S i n c e it was possible that flight s p e e d would affect the way in 81 which wingbeat an d respiration were coupled, the velocity, as read from the s p e e d o m e t e r of the motorcycle, was varied during each run. The range of velocities u sed w a s then recorded upon completion of each flight. Experimental trials c o n s i s t e d of approximately 4 consecutive, 60 s flights (although some were > 5min) from one e n d of the road to the other. The birds had between 10 and 60 s e c o n d s on the ground between consecutive flights. Flight velocity was varied randomly between 50 and 80 km/h. E a c h bird was only taken through this procedure once per day. Due to the difficulty in obtaining good transmissions from these animals, the frequency at which they flew away, the difficulty of recapturing them, and the a s s o c i a t e d risk of losing our only transmitter, we were satisfied with one good recording (minimum duration of 30 s) from each animal. Experiments were repeated on the same birds during uninstrumented flights. In these instances, f w was obtained as described above. Breathing frequency was also taken from the video signal since the birds' mouths open (inspiration) and c l o s e (expiration) with e a c h breath (present study; Butler & Woakes, 1980). In these trials, since there w a s no risk of equipment loss, a minimum of three recordings was made from e a c h animal. Alteration of Non-exercise Respiratory Drive During free-flight, it is not possible to alter respiratory drive by changing the composition of inspired gas. Thus, I attempted to induce a respiratory compensation for a metabolic a c i d o s i s or alkalosis (Barnas & Burger, 1984). To induce the acidosis, the birds were fed a diet for two days of poultry feed that had been 82 s o a k e d in 6 % NH 4CI (1.0L/100g feed). A l k a l o s i s was induced using a two day exposure to a diet of 6 % N a H C 0 3 (1.0L/100g feed). I also attempted to induce a more s e v e r e a l k a l o s i s with a two and then five day exposure to 1 0 % N a H C 0 3 (1.0L/1 OOg feed). Resting ventilation was measured after e a c h treatment (using the system d e s c r i b e d to monitor breathing frequency during overground running in Chapt e r 2) to determine the effects of the diet on resting breathing pattern. Flight trials were then performed to establish the relationship between wingbeat and respiration. Telemetry was not us e d for these experiments. Wingbeat and breathing frequencies were c a l c u l a t e d from visual analysis of the video tape. DATA ANALYSIS A n a l y s i s of the video tape was achieved using a video editing machine which allowed the tape to be replayed frame by frame. G i v e n the resolution of the videotape, when v i e w e d in this fashion, and that V H S is recorded at 30 frames per second, the timing of events during flight could be a s s e s s e d to within + 0.03 seconds. Wingbeat frequency was obtained from this a nalysis. Breathing frequency information was obtained from the output of the V H S audio channel. T h e signal w as demodulated and then synchronized, by means of the internal clock, with the visual signal indicating wingbeat frequency. From these records, the number of wingbeats per breath was calculated. The main purpose of these experiments was to establish, as previously demonstrated by Butler and Woakes (1980), that v i s u a l records of mouth movements indicated respiration. Due to the possible effects of drag as s o c i a t e d with the mask, harness and transmitter on 83 y f v (although none was noted over the short duration of the telemetered flights), uninstrumented flights were preferred. In addition, due to the time lag of the thermistor b e a d which varied between runs due to thermistor placement and air temperature, it w a s not possible to determine the exact timing of the respiratory c y c l e relative to the wingbeat cycle from these data. The coupling ratios of wing beat to respiration c a l c u l a t e d from the telemetered f v were compared with those c a l c u l a t e d from the v i s u a l a n a l y s i s and found to be identical. Thus, the data presented are taken from the uninstrumented flights. In addition to the phase relationship between f w and f v was c a l c u l a t e d . This was done by standardizing the respiratory period (T T 0 T) to equal one, then determining the portion of T T 0 T when either peak upstroke or downstroke occured. Inspiratory a n d expiratory intervals were established by following mouth movements. A s indicated by simultaneous records of mouth movements and tracheal temperature (indicating fv) v i a telemetry in barnacle geese (Butler & Woakes, 1980), inspiration o c c u r s when the mouth is open and expiration occurs through the nostrils after the mouth has c l o s e d . U n l e s s the d a t a for individual birds are given, values represent means + S.E.. Note from fig. 3-1 that there was no change in resting respiratory frequency of geese following administration of the different diets. The same was found for their flight behaviour, therefore all data, control and those from the diet experiments, were pooled. 84 RESULTS The range of wingbeat frequencies (f w), respiratory frequencies (fv) and the range of vel o c i t i e s over which the s e data were collected are presented for e ach bird in fig. 3-2. A s veloc i t y i n c r e a s e d 6 0 % from 50 (13.8 m/s) to 80 km/h (22.2 m/s), wingbeat frequency v a r i e d little, i n creasing only 2 5 % from 224 ± 13 to 281 + 1 8 beats per minute on average. Similarly, f v only increased 19.0 + 3.7 breaths per minute, from 75 + 4 to 94 + 6. In concert with these findings, although it was not possible to quantify amplitude of wing stroke from the video analysis, it was noted that the birds c o m p e n s a t e d for c h a n g e s in velocity more with c h a n g e s in wingstroke amplitude than with c h a n g e s in wingbeat frequency. A very tight relationship between f w and f v was also observed, so that, for 4 of the 5 birds, every breath a n a l y z e d was synchronized with wingbeat at three wingbeats/breath (fig. 3-3 & 3-4). The last bird also showed a predominant coupling ratio of three wingbeats per breath. However, this bird would sporadically switch from flying with three wingbeats per breath, to flying with two wingbeats per breath for a varying period of time, a n d then switch back to 3:1 (fig. 3-5). Approximately 2 0 % of the breaths a n a l y z e d from this bird were coupled at 2:1. A more thorough a n a l y s i s of the wing cycle and the respiratory cycle is shown in fig. 3-4. W ing excu r s i o n and respiratory movevents are shown for a typical 10 s econd period of flight. Inspiration occurs over the region of positive slope with the mouth o p e n (trough to peak). Expiration occurs over the region of negative slope when the mouth is c l o s e d . Superimposed on the respiratory record are symbols denoting peak of upstroke (circles) and bottom of downstroke (triangles). 85 Figure 3-1: Effects of the va r i o u s diets on resting breathing frequency (f v). The diets are listed on the x-axis, and the duration of each treatment is indicated in parentheses. 86 If) II c ro O O X D O r o O o ^ 2 CO ON v v CO o X z CO to i n m > Figure 3-2: R a n g e s of wingbeat frequency (f w), respiratory frequency (fv) and the velocity range (ground velocity, km/h) over which these data were collected for each bird, 1 through 5. 88 f w (min1) 280 240 200 160 120 80 40I f v (min-1) 120 90 60 30 • • i i — i— 1 2 3 4 5 100 80 60 40 20 Velocity (km/h) • • i i 1 2 3 4 5 BIRD No. J u 1 2 3 4 5 Figure 3-3: P e r c e n t a g e of breaths analyzed for each bird that were coupled with wingbeat at two, three or four wingbeats per breath (ie. = 2, 3 or 4). A n average of 80 breaths were a n a l y z e d per bird. 90 fw/fy CD Figure 3-4: C o m p u t e r simulated trace of a 10 s e c o n d flight s e q u e n c e showing wing excursion a n d respiratory movements. Inspiration occurs when the mouth is open. In the trace inspiration o c c u r s from trough to peak over the region of positive s l o p e . Expiration o c c u r s from peak to trough over the region of negative slope. S u p e r i m p o s e d on the respiratory trace are c l o s e d circl e s indicating the peak of upstroke and c l o s e d triangles indicating the bottom of downstroke. 92 MOUTH WING POSITION o POSITION o 5 ° o LT! m \% S= Figure 3-5: C o m p u t e r simulated trace of a 12 s e c o n d flight s e q u e n c e from bird #5 showing the frequent transition between a 3:1 and a 2:1 locomotor-respiratory coupling ratio. Inspiration o c c u r s while the mouth i s open and expiration while the mouth is c l o s e d . T h e c i r c l e s on the respiratory trace indicate the peak of upstroke. 94 1 Figure 3-6: P h a s e of the respiratory cycle when the wing as at the peak of upstroke. P o o l e d d a t a were derived from the analysis of 20 breaths (60 wingbeats) from e a c h of the 4 birds that s h o w e d exclusive 3:1 synchronization. Inspiration starts at 0.0. T h e end of expiration and the start of the next inspiration were normalized to o c c u r at 1.0. 96 4 0 co c CD > LU O 6 3 0 + 2 0 + 10 CO -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Phase of the Respiratory cycle when wing is at peak of u n=4; 20 breaths (60 wingbeats) from each bird From this it c a n be s e e n that the wings are at peak upstroke just after the start of inspiration, approximately 2/3 of the way through inspiration and again approximately 1/3 through expiration. Furthermore, it is c l e a r that this relationship is very consistent from one breath to the next. The relationship between wing excursion and respiration w as also very consistent in the bird that did not always follow 3:1 (fig. 3-5). T o determine whether the phase relationship between f w a n d f v observed in fig. 3-4 w a s maintained throughout flight at a variety of velocities or simply characteristic of this 10 s e c o n d flight in one animal, the phase relationship between peak upstroke and respiration was calculated for the four birds showing 3:1 coordination, b a s e d on the analysis of 20 breaths/bird. From the pooled data (fig. 3-6), it is c l e a r that the phase relationship between f w and f v is very tightly coordinated a n d that it s h ows little variation, both within and between birds. P e a k upstroke o c c u r s at approximately 0.05, 0.40 and 0.70 of the respiratory cycle (0.00 represents start of inspiration). DISCUSSION Metabolic Acid/Base Loading Experiments It is c l e a r from fig. 3-1 that the different diets were unsuccessful in invoking a respiratory c o m p e n s a t i o n under resting conditions. It was possible, however, that the different diets did represent a small stress to the birds which, when compounded with the additional s t r e s s of flight, would affect respiratory frequency, and therefore, the relationship between wingbeat and f v. This was not the c a s e . The relationship between f w and f v was the same in control, acid and base loaded birds, suggesting 98 that the different diets d id not affect the birds' acid-base status. T hus ail flight data were pooled. It is likely that, in addition to the altered diet, replacing their fresh water with a c i d i c and/or ba s i c water would have facilitated a significant shift in acid-base status. However, the physiological c o n sequences of s u c h aggressive dietary treatment were unclear. The birds were considered far too valuable to risk losing, thus this alternative was abandoned. General Discussion The flight velocitites recorded in these experiments (50-80 km/h) were somewhat higher than those recorded previously for free flying, wild C a n a d a g e e s e (13.9 m/s = 50.0 km/h; T u c k e r & Schmidt-Koenig, 1971). Although the velocities recorded here are ground ve l o c i t i e s rather than air velocitites (Tucker & Schmidt-Koenig, 1971), this is not likely to account for much of the difference. The birds were flown in both directions during each experimental trial and the range of velocities observed for the birds was similar regardless of direction, indicating that wind velocity was minimal. It is more likely that the birds' position relative to the motorcycle allowed them to attain higher flight v e l o c i t i e s due to a drafting effect. This possibility is supported by the similar observation in barnacle geese where greater flight velocities were obtained during flight behind a truck than during free- flight (Butler & Woakes, 1980). In addition, Butler and W oakes (1980) noted that below a velocity of approximatley 15 m/s (54 km/h) the g e e se would land. The C a n a d a geese in this study also landed when velocity was decreased below 50 km/h, which according to Tucker & Schmidt-Koenig (1971) is their normal flight speed. 99 During flapping flight, the wingbeat frequency of the C a n a d a geese remained fairly constant over a substantial range of flight velocities (fig. 3-2). This has been observed in a number of other s p e c i e s as well (barnacle goose; Butler & Woakes, 1980; fish crow; Bernstein et al., 1973; starling; Torre-Bueno & Larochelle, 1978). S i n c e f w v a r i e s little, and f w and f v remained coordinated at 3:1 in 4 of the 5 birds, any change in V E with changing velocity must have been achieved primarily through inc r e a s e s in respiratory amplitude. The observation that wingbeat amplitude is varied to achieve c h a n g e s in flight velocity may also indicate that changes in respiratory amplitude are used to increase ventilation in proportion to the increased 0 2 requirements of faster flight. However, it is not necessarily the c a s e that 0 2 requirements i n c r e a s e with flight velocity. The C a n a d a goose may be like the fish crow (Bernstein et al., 1973) a n d the starling (Torre-Bueno & Larochelle, 1978), but unlike the budgerigar (Tucker, 1968) and have a relatively constant rate of V 0 2 for a large range of flight v e l o c i t i e s . No data are currently available concerning the relationship between metabolic rate and flight s p e e d in the goose. It is very difficult to fly a bird of this s i z e in a wind tunnel. However, the recent development of a radiotransmitter system that will transmit values of oxygen consumption (Ikegami et al., 1988) will greatly facilitate the solution of this question. Although the relationship between wingbeat amplitude and respiratory amplitude could not be a s s e s s e d from the d a t a collected, a remarkable coordination between wingbeat and breathing frequency was very apparent. A s seen in barnacle g e e s e (Butler & W o a k e s , 1980), wingbeat and respiratory frequencies were coordinated predominantly at three wingbeats per breath in the C a n a d a goose. Coordination between wingbeat and respiration has been documented in a number of s p e c i e s . 100 Table 3-1. The relationship between wingbeat and respiration In different species. S p e c i e s Wt.(Kg) Wingbeats/breath 1 2 3 4 Study Pigeon Mallard G rosbeak Waxbill* Chickadee* Quail Gull Crow Wood Duck Black Duck Pheasant Barnacle goose C a n a d a goose 0.38 0.44 0.063 0.0069 0.0104 0.118 0.425 0.455 0.565 1.02 1.52 1.60 4.50 x+ x+ x x+ X X+ X X X X x+ X x+ x X X X X X T o m l i n s o n , 1957 Hart & Roy, 1966 Butler et al., 1977 Lord et al., 1962 Berger et al., 1970 Butler et al., 1980 Present study 101 TABLE 3-1 CONT'D. * denotes s p e c i e s that a l s o s h o w periods of flight where wingbeat and respiration are not c o u p l e d . + denotes that wingbeat and respiratory frequencies are also related at half integers. Thus, "+" following an "x" in the 1 column indicates that the particular s p e c i e s also c oordinates f w and f v at 1.5:1. 1 0 2 Ratios of 1:1 have been observed in the pigeon (Tomlinson, 1957; Hart & Roy, 1966; Butler et al., 1977) and crow (Bernstein et al., 1973). Ratios of 2:1 have been recorded in the mallard (Lord et al., 1962) and occasionally in the C a n a d a (fig. 3-3 & 3-5) and barnacle (Butler & Woakes, 1980) goose. Ratios as high as 5:1 have been recorded for the black duck, quail and pheasant (Berger et al., 1970a; s e e table 3-1 for complete list). T h e factors determining which s p e c i e s demonstrates which coupling ratio are unclear. T u c k e r (1968) s u g g e s t e d that there is a lower weight limit below which birds would not be able to show entrainment. He proposed that at the high wingbeat fr e q u e n c i e s s e e n in small birds, the birds would not have sufficient time to move an adequate volume of air per breath to meet metabolic demands. However, he was only c o n s i d e r i n g 1:1 coupling. The apparent flexibilty of the coupling ratios ( d iscussed below) undermines this argument. Tomlinson (1963), from his observations on the western gull and three s p e c i e s of ducks, suggested that a s weight i n c r e a s e s , the number of wingbeats per breath also increases. Further examination of similar s i z e d birds with different wing aspect ratios (wing length/wing breadth), however, indicated that body weight alone would not account for all the differences s e e n (Bernstein et a]., 1973). A s Berger et a]. (1970a) point out, the same type of coordination is observed in sev e n gram waxbills and 1440 g pheasants. It appears that the best method for finding a correlation between entrainment ratio and morphological features is by using body weight and wing a r e a (Berger et al., 1970a). For most of the sp e c i e s examined by Berger et a l . (1970a), the most notable exception being the pigeon, as the wing loading index (WM/A) increased, f w i n c r e a s e d and a high y f v ratio occurred. 103 Not only is the number of wingbeats per breath coordinated, the phase relationship between wingbeat and respiration is very consistent. In the pigeon, inspiration is always o b s e r v e d to occur during the upstroke and expiration during the downstroke (Tomlinson, 1957; Hart & Roy, 1966; Butler et al., 1977). In birds that do not s h o w 1:1 synchronization, the phase relationships are more complicated, but equally consistent. In general, the wings are fully elevated at the start of inspiration a n d fully d e p r e s s e d at the start of expiration. This is true of the C a n a d a goose (fig. 3-4), the barnacle goose (Butler & Woakes, 1980), a n d se v e r a l s p e c i e s examined by Berger et a l . (1970a). Although many s p e c i e s appear to have a preferred relationship between f w and f v, this relationship is not always fixed. The number of wingbeats taken per breath has been s h o w n to vary in C a n a d a g e e s e (present study), barnacle g e e s e (Butler & Woakes, 1980) and black ducks (Berger et al., 1970b). In addition, some s p e c i e s have been shown to switch from periods of coordinated to periods of non coordinated flight (Evening Grosbeak; Chickadee; Berger et a]., 1970a). This flexibility presumably en a b l e s the system to adjust to environmental perturbations. If the coordination was fixed, situations would inevitabely arise where the advantages of entrainment, whatever they are, would be outweighed by the constraints of obligatory coordination. If f v was rigidly fixed by locomotor rhythm, then V E could increase only through i n c r e a s e s in V T. This, in turn, would severely limit the ventilatory reserve upon which the animal could draw. Although the factors leading to a shift in the coupling ratio are unclear, this adaptability is essential if the coordination itself is not to become limiting to performance. 104 In spite of the c l e a r documentation that wingbeat and respiration are coordinated during flight, there is no empirical evidence indicating a function for this coordination. In the c a s e of the pigeon and fish crow where wingbeat and respiration are c o u p l e d at 1:1, the function of the coordination s e e m s obvious; support of the respiratory muscles by the locomotor muscles. For the C a n a d a goose, coordinated at 3:1, it is still possible that the wings assist respiration. Most of inspiration may occur during the upstroke (ie. two upstrokes and one downstroke per inspiration) and most of expiration during downstroke (ie. two downstrokes and one upstroke per expiration). To examine this possibility, the portion of inspiration during which the wings were moving up versus down was calculated for three animals (table 3-2). The remaining two birds were not examined because the transition between inspiration and expiration (mouth open and mouth closed) could not be precisely defined in these animals. In all three birds, upstroke occupied less than half of the inspiratory period. Thus there is no indication that the upstroke is coordinated to predominantly occur during inspiration. The final possibility is that, rather than the wing m u s c l e s supporting the entire respiratory cycle as is possible in the pigeon, there is an advantage to the coupling only during part of the respiratory c y c l e . For example, the wing action may assist in flow reversal; switching from inspiration to expiration and vice v ersa. This is supported by the observation in a number of s p e c i e s that the beginning of inspiration is predominantly found at the peak of upstroke and the beginning of expiration at the bottom of downstroke. At present, however, there are no data indicating that coordination of f w and f v during flight reduces the cost of locomotion and/or respiration. 105 Table 3-2. Duty cycle of the locomotor and respiratory pattern, and the portion of inspiration occuring during upstroke. N n X S.D. Wingbeat duty c y c l e : upstroke duration/cycle duration 4 100 0.47 0.03 downstroke duration/cycle duration 4 100 0.53 0.04 Respiratory duty cycle: Flight: T / T T 0 T : 3 20 0.58 0.02 TE/TTOT- 3 20 0.42 0.02 Rest: T / T T 0 T : 5 20 0.39 0.04 5 20 0.61 0.04 Portion of inspiration Bird #: occuring during upstroke: 1 20 0.50 0.006 2 20 0.44 0.006 3 20 0.36 0.006 N = number of animals n = number of events a n a l y z e d 106 CONCLUSION Although there is no evidence suggesting an energy saving function for entrainment in birds, the synchronization of wing beat and respiration has been documented in a large number of bird s p e c i e s and now also in the C a n a d a goose. From the s e obse r v a t i o n s it is becoming clear that the relationship between f w and f v is not fixed. The ratio of clearly varies between s p e c i e s , but it also varies within individuals of the s a m e s p e c i e s on a breath by breath basis. The flexibility of this relationship and the potential respiratory limitations a s s o c i a t e d with obligatory mechanical c o u p l i n g , s u g g e s t s that the relationship between f w and fv is the result of more than a mechanical interaction between the locomotor and respiratory systems. 107 CHAPTER 4 COORDINATION OF WINGBEAT AND RESPIRATION IN THE CANADA GOOSE II. PASSIVE WING FLAPPING 108 INTRODUCTION T h e d a t a in chapter 3 indicates that wing beat a n d respiratory rhythms are very tightly c o u p l e d during free-flight. T h e s e studies, however, are descriptive a n d do not address the qu e s t i o n of how these two motor patterns become coordinated. A s mentioned in the previous chapter, it d o e s not appear that the coordination is entirely due to a mechanical interaction between the wings and the respiratory system. A s with the control of ventilation during exercise, it is possible that feedback mechanisms relaying information about the locomotor rhythm play a role in establishing this entrainment. A large number of studies involving c r o s s perfusion techniques (Kao, 1963), electrical stimulation of muscle (Cross et a]., 1980) and ventral roots of s p i n a l nerves leading to limb muscles ( M c C l o s k e y & Mitchell, 1972; Mitchell, 1985; W a l d r o p et al., 1986a & b) as well a s stimulation of afferent fibres originating in the limbs ( K a l i a et al., 1972; Senapati, 1966) have indicated a significant role of afferent feedback from the limbs in generating the increase in ventilation during e x e r c i s e . From the few studies that have examined the effect of this afferent activity on respiratory pattern, it does appear that repetitive, electrical somatic afferent stimulation will entrain the respiratory rhythm to some degree in cats (Iscoe & P o l o s a , 1976; Kawahara et al., 1988) and dogs (Howard et al., 1969). It i s very difficult, however, to reproduce the afferent neural flow s e e n during actual locomotion through electrical stimulation of discrete nerves. Electrical stimulation activates the largest diameter fibres first, not necessarily the ones that are activated first during a given locomotor sequence. In-addition, it does not allow 109 s e l e c t i v e activation of speci f i c functional c l a s s e s of receptors s i n c e afferent information from e a c h c l a s s of receptors is transmitted along a range of fibre types (Swett & B o u r a s s a , 1981). Furthermore, the afferent traffic set up through electrical stimulation of muscle or cutaneous nerves is very speci f i c and not likely to represent the c o m plex afferent signal associated with actual wing flapping. During free-flight, there are likely to be cycl i c changes, not only in limb related feedback, but in th o r a c i c pressure and/or volume related information from the respiratory s y s t e m a n d chest wall a s s o c i a t e d with the up and downstrokes of the wings (Jenkins et al., 1988). There is abundant evidence in mammals of entrainment of respiratory rhythm by respiratory related afferent feedback from intercostal nerves (Remmers & Marttila, 1975), carotid s i n u s nerves (Eldridge, 1972a & b) and v a g a l l y mediated lung volume information in mammals (Petrillo et a]., 1983). Similarly, respiratory rhythm c a n be entrained by cycl i c oscillations in pulmonary mechanoreceptor information (Ballam et al., 1982; Ballam et al., 1985) and intrapulmonary chemoreceptor discharge in birds (Kunz & Miller, 1974; Kunz, 1987). Experiments that look only at the effects of afferent activity from the limbs on respiration are overlooking a potentially important entraining s i g n a l . T h u s the first goal of t h e s e studies w as to determine if afferent activity a s s o c i a t e d with a more physiological stimulus s u c h as passive wing flapping could entrain respiration. S i n c e both locomotor and respiratory related afferent information could be involved in s u c h entrainment, the sec o n d purpose of this study was to determine which of the afferent pathways activated by wing flapping were actually involved in the production of the coordination (mechanoreceptors from the thorax (lungs and chest wall) and/or the wings). This was approached from two directions. First, 110 sectioning of the sp i n a l nerves emanating from the brachial plexus determined whether afferent activity from the moving wings was essential for the entrainment of passive wing flapping a n d respiration or whether afferent activity from the lungs and/or chest wall w a s sufficient. The s e c o n d a p p r o a c h was b a s e d on a number of studies that have examined the effects of s i n g l e stimulations of specific locomotor and respiratory afferent pathways on respiratory rhythm (Howard et al., 1969; Eldridge, 1972a & b; Remmers & Marttila, 1975; Iscoe & P o l o s a , 1976; Y o u n e s & Polachek, 1981; K a w a h a r a et a l . , 1988). T h e respiratory effects of s u c h stimuli are dependent on when the stimulus is presented relative to the respiratory c y c l e , and the type of afferent pathway stimulated. The respiratory responses of the birds to single wing flaps during different p h a s e s of the respiratory cycle and the reported r e s p o n s e s of a number of s p e c i e s to activation of spec i f i c afferent pathways were compared in an attempt to gain s o m e insight into which afferent pathways activated by single passive wing flaps affected respiration. Pathways mediating the ventilatory responses to s i n g l e wing flaps may also be involved in the entrainment of wing beat and respiration during p a s s i v e wing flapping as well as free-flight. MATERIALS AND METHODS Experiments were carried out on 16 C a n a d a geese (body weight; 4.6 ± 0.1 kg). Care was taken to select animals of the same si z e to facilitate the use of comparable wing e x c u r s i o n s in each animal. The surgical procedures used in these experiments were all performed under halothane/nitrous oxide anaesthesia, and are 111 very similar to tho s e d e s c r i b e d in chapter 1 for decerebrate g e ese. A carotid artery and jugular v e i n were canulated for monitoring blood pressure and for the administration of I.V. fluids a s required. The tr a c h e a was cannulated to permit monitoring of ventilation v i a pneumotachography. The birds were then supported on a sl i n g , their h e a d s were placed in a stereotaxic head holder and a suction decerebration w a s performed along a plane extending dorsally from the caudal margin of the ha b e n u l a r n u c l e u s to the caudal edge of the optic c h i a s m ventrally. Since the effects of fe e d b a c k activity alone on respiratory rhythm were being examined, it w a s very important that the animal not resist the movements of the passive wing flapper through mechanisms which were central or feedforward in origin. T h i s low level of decerebration d e c r e a s e d the amount of spontaneous activity s e en in the birds a n d removed all signs of resistance to the action of the passive wing flapping d e v i c e . E M G electrodes were implanted percutaneously into the pectoralis major m u s c l e s to ensure that there was no activity in the pectoralis muscles. The body w a s supported from the dorsal surface v i a two bone clamps. One was attached to the pelvic girdle, the other to the dorsal vertebral s p i n e s just caudal to the brachial p l e x u s . D o r s a l s u s p e n s i o n was required in these experiments. Without it, the movement of the wing flapper compressed the thorax onto the sling during the downstroke a n d lifted it during upstroke, causing increases in the magnitude of non-respiratory c h a n g e s in air flow. E v e n when supported dorsally, the pa s s i v e motion of the wings generated small volume changes in the thorax. E M G electrodes were therefore implanted into an intercostal muscle group (5 th or 6 t h costal space; external intercostal E M G indicated inspiration and internal intercostal indicated 112 expiration) to en s u r e that the c h a n g e s in respiration indicated by the air flow trace were truly respiratory in origin a n d not movement artifacts. E a c h wing w a s unfolded a n d fastened to a bar that held it in the approximate position s e e n during free-flight (as recorded during free-flight experiments, chapter 3). In order to permit s h o u l d e r rotation, the bars extended from approximately 2 cm distal to the s h o u l d e r joint to the wing tip. E a c h wing bar was attached, v i a a hinged joint, to a ve r t i c a l b a r that w as hinged at its upper joint with a common horizontal bar. From the centre of this common horizontal bar, an additional vertical bar connected to the free e n d of the main support rod that lay over an eccentric, adjustable c am. T h e an c h o r e d end of the main support rod acted a s a pivot. A 1/4 HP D.C. motor rotated the c a m a n d c a u s e d the free end of the support rod to move up a n d down. T h e vertical motion of the main support rod w a s transferred to the wings v i a the multi-hinged apparatus described above. In addition, the hinge system w a s d e s i g n e d s o that the upstroke and downstroke c y c l e added a slight antero-posterior component to the wing cyc l e , as s e e n during free-flight. Wing flapping frequency ( Q w a s adjusted v i a a rheostat attached to the D.C. motor. Wing beat amplitude w a s adjusted to approximate the amplitude observed during free-flight at 50 - 60 km/h by adjusting the diameter of the eccentric cam. Wing excursion a n d f w were recorded using a movement potentiometer attached to one of the horizontal bars. Body temperature (T B) was monitored v i a a thermistor placed approximately 25 cm down the t r a c h e a and maintained at 41 °C using a heat lamp and a 15 cm, small diameter U-shaped copper tube that was inserted through the anus into the intestine. Hot or c o l d water was circulated through the tube a s required to maintain 113 T B . All equipment u s e d to monitor ventilation and E M G activity is the s a m e as that described in c h a p t e r 1. A n a e s t h e t i c w a s removed following completion of the surgical procedures and the birds w ere a l l o w e d a minimum of one hour to recover. All pressure points and incision s i t e s were generously infiltrated with local anaesthetic ( 2 % xylocaine). For denervation of the brachial plexus, extending from C 1 7 to T3 (5 segments), the a n i m a l s were rea n a e s t h e t i z e d . This denervation removed all afferent activity from the moving wings. A dorsal midline incision was made from T4 to C18. Several strap m u s c l e s were retracted to permit a c c e s s to the cerv i c a l air s a c s on either s i d e of the s p i n e . It w a s possible to section the C 1 7 and C18 spinal nerves through the air s a c s a s they exited the spinal column. The three caudal nerves ( T I -TS) b r a n c h e d extensively before entering the air s a c s . Thus, it was n e c essary to perform a partial laminectomy on these segments. Removal of the lateral trabeculae revealed the mixed nerves prior to their branching point and allowed complete nerve section. Prior to the actual sectioning of the nerves, e a c h root was liberally infused with xylocaine. F a i l u r e to do s o frequently resulted in immediate c a r d i a c arrest. Following nerve s e c t i o n , the air s a c s , the muscle layers and skin were sutured closed. Protocols and data analysis I. The first protocol involved the presentation of 30 single stimuli, each stimulus being one wing c y c l e (upstroke-downstroke), at various periods throughout inspiration (15 wing cycles) and expiration (15 cycles). P h a s e response curves were generated from t h e s e data. To begin with, phase response curves for the entire respiratory c y c l e were generated. The average duration of the two breaths 114 proceeding a stimulated breath were measured and represented control v a l u e s of respiratory period (T c). The period between onset of the stimulated breath a n d stimulus onset (TB) w a s a l s o measured, a s was the duration of the stimulated breath (T s + T f)(see f i g . 4 -1 A for illustration). The p h a s e of the respiratory c y c l e in which the stimulus w a s presented (6) was therefore given by, 9 = TJTC a n d the ph a s e c h a n g e of the respiratory rhythm was given by f(9) = 1 - (T s + T,)/Tc. From t h e s e equations, it is c l e a r that a positive p hase change indicates a d e c r e a s e in respiratory period (an inc r e a s e in f v), and a negative phase change indicates a n inc r e a s e in the respiratory period (a d e c r e a s e in f v ) . P h a s e response c u r v e s were then generated by plotting p hase change (f(9)) v e r s u s phase of stimulation (9). A n y c h a n g e s in the respiratory period produced through p a s s i v e wing flapping c o u l d be due to c h a n g e s in either inspiration, expiration, or both. Thus, additional p h a s e r e s p o n s e c u r v e s were generated to determine if the c h a n g e s in respiratory period were due to c h a n g e s in inspiratory or expiratory duration or both (see fi g . 4-1B) . F o r t hese, the av e r a g e inspiratory duration (TIC) a n d expiratory duration (T E C) of the two breaths preceeding the stimulated breath represented control v a l u e s . Inspiratory (T,) and expiratory duration (T E) of the stimulated breath were measured. T h e interval between onset of the stimulated inspiration (TSI) or expiration (T S E) a n d stimulus presentation w a s also measured to determine the phase of inspiration (0, = TSI/T,) or expiration (0 E = T S E/T E) when the stimulus was presented. The phase c h a n g e s of the inspiratory (f(9), = 1 - T/T1C) and expiratory intervals (f(9) E = 1 -Tg/Tec) c o u l d thus be ca l c u l a t e d separately. In addition, by measuring the length of the expiration immediately following an inspiratory stimulation (TE') a n d 115 Figure 4-1: Method for generation of phase response c u r v e s for the total respiratory c y c l e (A) a n d for the inspiratory a n d expiratory intervals (B). The s e q u e n c e of tra c e s is the s a m e in B a s in A. Definitions: A. T c is the duration of the control respiratory periods; T s is the time from the onset of inspiration to stimulus onset; a n d T, is the time from stimulus onset to e n d of the breath. From the s e v a l u e s , the following parameters were ca l c u l a t e d : 0 = Tj/T c; phase of respiratory c y c l e when stimulation occured; f(9) = 1 - (Ts+T,)/Tc; respiratory p hase shift. B. T I C a n d T E C represent the duration of the control inspirations a n d expirations; TSI a n d T S E represent the time from the onset of inspiration or expiration to stimulus onset; T, = TS1 + T, a n d corr e s p o n d s to the duration of the stimulated inspiration; T E' is the duration of the expiration following the stimulated inspiration; T E is the duration of the stimulated expiration; T,' is the duration of the inspiration following the stimulated expiration. From the s e v a l u e s , a similar set of parameters were c a l c u l a t e d for the inspiratory a n d expiratory stimulations. Inspiratory stimulations: 6, = TSI/TIC; p h a s e of inspiratory c y c l e when stimulation occured; f(8), = 1.- T/TIC; p h a s e shift of inspiratory c y c l e ; f(G) E' = 1 - T E7T E C; phase shift of the expiration following the stimulated inspiration. Expiratory stimulations: 9 E = T S E/T E C; phase of expiratory cycle when stimulation occured; f(6) E = 1 - TJTzc, phase shift of expiratory c y c l e ; f(9),' = 1 - T.VT.c; p h a s e shift of the inspiration following the stimulated expiration. 116 I A i • « f • i I I I i • _ i i i i a f ri ! ! n ••• M ! M - T - J * I * t I I I I I I I I — T„-?-T,-»r-T I ' - » j ; 117 the duration of the inspiration immediately following a n expiratory stimulation (Ty), the p h a s e c h a n g e s of t h e s e two va r i a b l e s relative to control v a l u e s (f(8) E' = 1 -TE'/TE a n d f(9),' = 1 - T.'/T, respectively) c o u l d be obtained. From this further a n a l y s i s , 4 additional p hase response c u r v e s were generated (1, f(9), v s 9,; 2, f(9) E' v s 9,; 3, f(8) E v s 9 E; 4, f(9),' v s 9E) to d e s c r i b e , in det a i l , how single wing flaps alter respiratory timing. II. T h e effects of continuous wing flapping on respiratory rhythm were examined in protocol II. Following the minimum one hour recovery period, resting ventilatory pattern w a s recorded for two, one minute periods, 10 minutes apart. Resting ventilatory frequency (fv) w a s ca l c u l a t e d and if respiratory frequency was stable (< 5% variation), the protocol w a s started. Having cal c u l a t e d f v, a wing beat frequency ( f j w as d i a l e d into the wing flapper that w as either 1.5, 2.5, or 3.5 times f v. T h e s e f w's were c h o s e n a s they bracketed the relationships between f „ a n d f v s e e n during free-flight. f w w a s then slowly i n c r e a s e d through a range of frequencies that e n c o m p a s s e d ratios of two wing beats per breath (2:1), three wing beats per breath (3:1) and/or four wing beats per breath (4:1). Typically, f w a n d f v would eventually b e come s y n c h r o n i z e d . f w a n d f ¥ were a n a l y z e d to determine the ratio of a n d the range of f w a n d f v over which the synchronization w a s maintained. Tidal volume (V T) an d minute ventilation (V E) were also measured in 6 animals at e a c h f w to determine if p a s s i v e wing flapping had any effect on ventilation and, if there was an effect, if it w a s frequency dependent. 118 III. S i n c e wing flap f requency was gradually i n c r e a s e d over a range that would e n c o m p a s s predetermined multiples of f v, there was the possibility that f w and f v would become s y n c h r o n i z e d fortuitously. The apparent synchronization in s u c h in s t a n c e s would not be due to any interaction between the two s y s t e m s . In order to e n sure that random a s s o c i a t i o n of the two rhythms was not responsible for any o b s e r v e d sy n c h r o n i z a t i o n , the effects of momentary stalls (delays) in the p a s s i v e wing flapping c y c l e on respiratory rhythm were examined during periods of synchronization (see fig. 4-2). If the synchronization w a s fortuitous, then the ratio between f w a n d f v would be unchanged, but the p h a s e relationship (S/F) between the two rhythms w o u l d , be shifted by the duration of the d e l a y (d). T h e s e experiments were c a r r i e d out on 4 non-denervated and 4 denervated birds. Duration of the d e l a y s (d), e x p r e s s e d a s a fraction of the wing period (d/F), ranged from 0.0 to 1.0 complete wing c y c l e . The phase relationship between f v a n d f w (the phase of the wing c y c l e where inspiration started, S/F) before and after the stall was c o m p a r e d . In addition, the number of respiratory c y c l e s e l a p s e d prior to reestablishment of prestall S/F was a n a l y z e d a s a function of stall duration. The relationship between the timing of the delay relative to the respiratory c y c l e ( 8 / 1 ^ ) and the number of breaths required to reestablish the phase relationship (S/F) between f v a n d f w following a perturbation was also examined. A s c a n be s e e n from f i g . 4-5 and 4-8, the motion of the wings generated by the wing flapper w a s not always s i n u s o i d a l . The upstroke took longer than the gravity a s s i s t e d downstroke. This mechanical problem was corrected part way through the experiments. A s a result, the phase of the .wing c y c l e corresponding to peak upstroke w a s 0.68 in the early experiments and 0.50 in the later experiments. In 119 Figure 4-2: Method u s e d to de s c r i b e the effects of wing c y c l e perturbations on the p h a s e relationship between wing beat a n d respiration. Definitions: F = wing c y c l e duration; T T O T = respiratory c y c l e duration; S, = time from start of upsroke to beginning of inspiration; S w = time from beginning of inspiration to onset of delay; d = duration of the delay. F r o m the s e v a l u e s , the following parameters were calc u l a t e d ; S/F = p h a s e relationship between wing beat and respiration; SJTJOT = p h a s e of respiratory c y c l e when stall w a s introduced; d/F = duration of stall e x p r e s s e d as a fraction of wing c y c l e period. 120 order to compare the different experiments and facilitate determination of wing position from wing phase data, all measurements from these early experiments were adjusted s o that upstroke and downstroke, rather than occupying 0.68 a n d 0.32 of the c y c l e , both o c c u p i e d half of the wing c y c l e . T h u s 0.50 of the wing c y c l e represented the peak of upstroke in all experiments. IV, V & VI. Protocols I, II a n d III were then repeated following denervation of the brachial plexus a n d c o r r e s p o n d to protocols IV, V and VI respectively. Only 2 of 13 birds s u c c e s s f u l l y c o m p l e t e d protocols I through VI. Most animals d id not tolerate the denervation procedure when it was performed following protocols I, II and III. A s a result, in order to complete protocols IV, V a n d VI on enough animals, it w a s n e c e s s a r y to modify the procedure s u c h that the denervation procedure was carried out during the initial surgery, immediatley following the decerebration. For logistical r e asons, not all protocols could be carried out on all animals. Thus, the number of individuals u s e d in protocols I through VI was 8, 8, 4, 2, 5 a n d 4 respectively. It s h o u l d be noted that no attempt was made in these experiments to generate wing beat f r e q u e n c i e s similar to those observed during free-flight. There were obvious mechanical difficulties in trying to pass i v e l y drive the wings at 4 to 5 c y c l e s per s e c o n d . T h e s e problems were not insurmoutable. However, if free-flight levels of f w were produced, free-flight levels of f v also had to be generated. Otherwise it would not have been possible to examine the effects of wing a s s o c i a t e d afferent feedback on respiration over frequency ranges .where the ratio of f w to f v was similar to that s e e n during free-flight (2:1 - 4:1). It was not possible to elevate respiratory 122 frequency to levels similar to those s e e n during free-flight except by generating extreme elevations in T B. Si n c e elevations in T B of the required magnitude would have independent effects on respiratory pattern and most likely mask the effects of wing a s s o c i a t e d afferent feedback on respiratory rhythm, this a p p r o ach was not tried. Instead, wing beat frequencies ranging from 1.5 to 4.5 times resting ventilatory frequency were s e l e c t e d . Minute ventilation (V E), tidal volume (V T), breathing frequency (f v), wing beat frequency ( f j , heart rate and blood pressure were monitored throughout e a c h protocol. U n l e s s otherwise stated, all v a l u e s represent means + S.E. A N O V A was u s e d to test the difference between means of all respiratory data. R e g r e s s i o n a n a l y s e s and A N C O V A were u s e d to test the differences in the relationships d e s c r i b e d in fi g . 4-9 & 4-10. V a l u e s of P<0.05 were a s s u m e d to be significant. RESULTS Continuous wing flapping The r e s p o n s e s of the geese to passive wing flapping were virtually identical before a n d after denervation of the brachial plexus. Thus, unless otherwise stated, all r e s p o n s e s d e s c r i b e d below apply to both groups. There was no consistent or significant relationship between minute ventilation, tidal volume or breathing frequency and the rate of passive wing movement in the 6 C a n a d a g e e s e examined (fig. 4-3). Two birds s h owed a small increase in V E, two showed small d e c r e a s e s and 2 did not change a s f w increased. S l o p e s of the regression lines of V E, V T and f v versus f w plotted for each animal were not 123 significantly different from z e r o . S i n c e there w as no relationship between any ventilatory v a r i a b l e m e asured a n d f w , average v a l u e s of V E, V T and f v during flapping were c a l c u l a t e d for the 6 birds and compared with preflapping v a l u e s (fig. 4-4). There were no significant effects of passive wing flapping on average levels of V E, V T or f v. O c c a s s i o n a l l y a small transient increase in V E was observed at the onset of wing flapping, but this returned to resting levels within 10 s e c o n d s . However, when the breathing pattern r e s p o n s e s of individual animals to pas s i v e wing flapping were a n a l y z e d relative to wing flapping pattern on a breath by breath b a s i s , it w a s c l e a r that the frequency of limb motion affected respiratory rhythm. In fig . 4-5, there are three s e t s of paired traces showing air flow a n d wing excursion at three different wing flap frequencies for one bird; top, f w=0, middle, fw=25.7 and bottom, f w=35.4. The resting breathing pattern (fv = 10.3 breaths/min) is shown in the top trace prior to the onset of pas s i v e wing flapping. Wing flap frequency was then i n c r e a s e d in small s t e p s , starting at approximately 24.0 wing beats per minute (2.5 times f v ) . E a c h step w a s held for 2 - 3 minutes. When fw reached 25.7 (fig. 4-5, middle), the respiratory period i n c r e a s e d (fy decreased) s o that f w a n d f v became entrained at 3 wing flaps per breath. f w and f v remained precisely coordinated in this f a s hion for the entire 3 minutes that f w was set at 25.7. O v e r four subsequent step c h a n g e s in f w , from 25.7 to 35.4 beats/min, the pattern of coordination was not altered. f v originally d e c r e a s e d 1 6 . 5 % from a resting f v of 10.3 breaths/min to 8.6 breaths/min in order to achieve initial entrainment, and eventually increased by 1 5 . 5 % to 11.8 breaths/min a s f w reached 35.4. When f w was increased past 35.4 flaps per minute, the 3:1 relationship between f w and f v broke down, and f v returned toward preflap levels (10.3 breaths/min). Coordination of f w 124 Figure 4-3: Relationship between passive wing flap frequency (fj and minute ventilation (VE), tidal volume (VT) and breathing frequency (f¥) for five geese. Regression lines are drawn through the data for each individual. 125 Figure 4-4: Effects of p a s s i v e wing flapping on overall levels of minute ventilation (V E), tidal v o l u me (V T) a n d breathing frequency (fv; v a l u e s represent means + S.E.; n = 5) in animals at rest and during periods of p a s s i v e wing flapping. 127 Figure 4-5: Thre e pairs of t r a c e s showing the relationship between air flow a n d wing e x c u r s i o n in o n e bird at three different wing beat frequencies (f w). T o p pair of tra c e s represents resting breathing frequency (fv) when the wing were not moving. T h e middle a n d bottom traces were recorded at f w's of 25.7 a n d 35.4 beats/min. T h e large horizontal arrows indicate the increase (middle) a n d d e c r e a s e (bottom) in duration of the respiratory period relative to the respiratory period w hen there w a s no wing flapping. 129 A I R now • ' I N I N O KXCtntSXOM >—-—\-GO o I 5 sac Figure 4-6. Relat i o n s h i p between wing beat frequency ( f j and respiratory frequency (fv) in bird #7 a s f w w a s i n c r e a s e d from zero to 47.0 beats/min, through 10 step i n c r e a s e s in f w , a n d then returned back to z e r o . 131 a n d f v o c c u r r e d over a limited range. To achieve this entrainment, the respiratory period both i n c r e a s e d a n d d e c r e a s e d relative to resting l e v e l s . In spite of t h e s e c h a n g e s in f v, the overall level of minute ventilation did not c h a n g e (fig. 4-3) indicating that c h a n g e s in f v were compensated for by opposing c h a n g e s in V T. Th i s c a n be s e e n by c l o s e inspection of fig. 4-5. V T i n c r e a s e d when f v fell from 10.3 to 8.6 breaths/min, and d e c r e a s e d relative to resting levels when f v i n c r e a s e d to 11.8 breaths/min. T h e effect of p a s s i v e wing flapping on breathing pattern is demonstrated for another bird (bird #7) in f i g . 4-6. Resting f v in this bird, represented by the point la b e l l e d 'pre', was 13.5 breaths/min. With the onset of passive wing flapping at 47.0 beats/min, f v i n c r e a s e d from 13.5 to 23.5 breaths/min s u c h that f w and f v were immediatley coordinated at 2 beats per breath. A s f w was in c r e a s e d through the 7 ste p s indicated in f i g . 4-6 from 47 beats/min to 67 beats/min, in c r e a s e s in fv from 23.5 to 33.5 matched the in c r e a s e s in f w s u c h that f w a n d f v remained coordinated at 2 beats per breath. Only when f w was increased to 72.5 beats/min (step 8) did the 2:1 coordination disintegrate. f w began to d e c r e a s e toward pre flap l e v e l s . W h en f w w a s returned to zer o , f v immediately returned to resting levels. T he response of this bird was atypical in two respects. First, whereas the range of f v over which entrainment occurs usually bracketed resting f v, in this bird (as in #5) f ¥ only i n c r e a s e d . S e c o n d , when entrainment was lost, f v usually returned abruptly to near resting l e v e l s . In bird #7, f v d e c r e a s e d once the entrainment was lost, but still remained considerably elevated relative to rest. It only returned to resting levels following the termination of wing flapping. 133 A summary of the results obtained during continuous p a s s i v e wing flapping is presented for non-denervated an d denervated birds in fig. 4-7. The so l i d c i r c l e s in this figure represent the resting breathing frequency for e a c h animal. Breathing frequency was very stable in these preparations. With the exception of bird # 12 that s h o w e d substantial variation in resting f¥ (maximum range was + 4 4 % ) , breath duration v a r i e d < + 5 % . The standard error of the resting breathing frequencies in fi g . 4-7 do not extend past the limits of the c l o s e d c i r c l e s , and thus are not shown. The vertical bars indicate the range of breathing frequencies over which f v a n d f w were s y n c h r o n i z e d during p a s s i v e wing flapping. The ratios above e a c h vertical bar, in turn, indicate the relationship between f w and f v produced during the period of s y n c h r o n i z a t i o n . There is no significance to the fact that different animals s h o w e d different t y p e s of entrainment. T h e observed coordination s c h e m e was more a function of the f w d i a l e d into the wing flapping machine. E a c h animal could show more than one type of synchronization if the wing flapping machine was adjusted appropriately. Time constraints did not permit the stepwise increase of f w at s e v e r a l multiples of f v. T h u s the relationship between f w a n d f v was only examined in detail for one multiple of f v (ie. 2:1, 3:1 or 4:1) for e a c h animal. All of the coordination s c h e m e s observed during free-flight c ould be reproduced through p a s s i v e wing flapping in non-denervated an d denervated animals. Of the 8 animals e x a m i n e d prior to denervation of the brachial plexus, 2:1, 3:1 and 4:1 synchronization w a s produced in three, three and two birds respectively. Entrainment between fw a n d f v appeared to occur over the largest range of f v when synchronization w a s 2:1. However, due to the limited sample s i z e , it is difficult to compare entrainment ranges a s a function of entrainment ratio (ijiv). Five birds 134 Figure 4-7. Summary of results showing the range of breathing frequencies (fv) over which fw and f¥ were sychronized for each animal. Solid circles represent resting fv. The vertical bars respresent the range of fv over which synchronization was maintained. The ratio above each bar indicates the ratio of y f v observed during the synchronization period. 1 3 5 eg 32 28 (min1) 24 20 12 8 CM CM I I • i 1 — J 1 «-CM » T7 • 1 CM j i i i BIRD* 5 6 7 B 9 H 1 2 1 3 intact 8 12 U 15 16 denervated 136 were e x a m i n e d following denervation of the brachial plexus (fig. 4-7). Two to one entrainment w a s p r o d u c e d in two animals, 3:1 in two more and 4:1 in the remaining bird. It is a l s o important to note that entrainment of f w and f v was not produced through a unidirectional response of f v to the imposed wing flapping. Entrainment was a c h i e v e d through both i n c r e a s e s and d e c r e a s e s in f v relative to rest in non-denervated a n d denervated groups. Thus it appears that continuous wing flapping c a n both shorten a n d extend the respiratory period. Continuous wing flapping: phase delays Description of the relationship between f w and f v solely in terms of ratio v a l u e s (L/f v) is incomplete. S i n c e approximate multiples of ventilatory frequency were s e l e c t e d as wing beat frequencies, entrainment of wing beat a n d respiration could occur fortuitously a n d not indicate an interaction between the two systems. There was, however, a ver y precise phase relationship between the imposed wing motion and the respiratory c y c l e . T h i s can be s e e n in fig. 4-8 where, during a period of coordination between f w a n d f ¥ in a denervated bird, the wing oscillation was briefly interupted. Prior to the delay, inspiration consistently started at the peak of wing upstroke. Introduction of a brief delay into the wing oscillation a d v a n c e d the onset of the next'inspiration relative to the wing cycle by a period equal to that of the delay (i.e. inspiration started 2/3 of the way through upstroke rather than at the peak). However, one breath later, the original phase relationship between f w and f v was reestablished, and inspiration again started at the peak of upstroke, if the entrainment was coincidental, although y f v would not have changed, all the 137 inspirations following the delay would have been shifted by the period of the delay, o c c u r i n g approximately 2/3 of the way through upstroke. Instead, the respiratory rhythm was reset and the original phase relationship between f w and f v was r e e s t a b l i s h e d . T h i s c a n be s e e n for a slightly longer delay in the bottom pair of t r a c e s (fig. 4-8). The reestablishment of the phase relationship between f w and fv (S/F) following a perturbation of the wing c y c l e is shown in fig. 4-9 for both intact (A) a n d denervated birds (B) o v e r a range of wing flapping frequencies. The s t a l l s v a r i e d in duration from 0.0 to 1.0 complete wing c y c l e s . The phase relationship (S/F) on the y-axis ranges from 0.0 to 1.0. Z e ro represents an inspiration starting at the onset of upstroke, 0.50 c o r r e s p o n d s to inspiration at the peak of upstroke a n d 1.0, inspiration starting at the termination of downstroke. R e g r e s s i o n lines were drawn through the control (prestall) and post stall data. Four major c o n c l u s i o n s c a n be drawn from this graph. First, identical v a l u e s of S/F for the control a n d the post stall periods indicate that the phase relationship between f w and f v was not disrupted by d e l a y s in the wing rhythm. The respiratory c y c l e was reset. S e c o n d , the s l o p e s of the r e g r e s s i o n lines were not significantly different from z e r o indicating that the p hase of the wing c y c l e at which inspiration was initiated did not vary with wing flap frequency. T h i r d , that most points occur slightly above the line of S/F = 0.50, indicates that inspiration started shortly after the the peak of upstroke. Finally, denervation of the brachial plexus did not affect these relationships. The stalls shown in f i g . 4-8 were of relatively short duration and S/F was reestablished in the first breath following the stimulus. The possibility of a relationship between magnitude of the perturbation (duration of the delay) and the 138 Figure 4-8. Two pairs of traces from bird # 8 (denervated) showing the effects of d e l a y s in the wing c y c l e on the phase relationship between respiratory (fv) and wing beat frequency ( f w ) . f w , f v a n d the duration of the delay are all i n c r e a s e d in the bottom pair of t r a c e s . 139 Figure 4-9. P h a s e relationship between wing beat a n d respiratory c y c l e s (Si/F) before (indicated by the open c i r c l e s , s o l i d regression line) a n d after (indicated by so l i d c i r c l e s , d a s h e d regre s s i o n line) a phase delay a s a function of wing flap frequency ( f j in non-denervated (A; n = 4) and denervated (B; n = 4) geese. T h e dotted line at Si/F = 0.50 represents the peak of upstroke in the wing cycle (see text for further de t a i l s ) . Note the different ranges of f w for the non-denervated v e r s u s denervated groups. 141 1.00 0.80 h 0.60 r Sl/F 0.40 h 0.20 f-0.00 25.0 26.0 27.0 28.0 1.00 20.0 29.0 30.0 f W (min 1) f, w -O control 30.0 35.0 (min""1) - • following stair 142 Figure 4-10. Time taken to reestablish the phase relationship between wing beat a n d respiration (standardized as number of breaths following the breath with the delay) a s a function of stall duration (A) a n d timing of the stall relative to the respiratory c y c l e (S/TTOT) (B) in non-denervated (open c i r c l e s , s o l i d regression line; n = 4) a n d denervated birds (open square, d a s h e d regression line; n = 4). Stall duration (d) is e x p r e s s e d a s a fraction of the wing c y c l e period (d/F). 143 c CD CO CD CD CT CD L-CO _ C -M D CD =*fc • O 8.0 7.0 -6.0 -5.0 -4.0 3.0 • 2.0 -1.0 0.0 -1.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 -1.0 O O D ODD __L"PB-BD o~o no o qp DU DD LTD rrnrE o o o D8> LTOO O • 0.00 0.20 0.40 0.60 0.80 Stall Duration ( d / F ) 1.00 • • B - • D - o o - O D u • • <gp _ •fj o CSD DD rm> OGD • C E P DD D D D O • A • 1 i - I . I . • 0.00 0.20 0.40 0.60 0.80 sw /TT 0 T O intact • denervated 1.00 144 time required for resetting w a s also examined. Increases in the duration of the stalls (d; e x p r e s s e d a s a fraction of the wing cycle period = d/F), d id not affect the time (standardized a s the number of breaths following the stimulus breath) required for the p h a s e relationship between f w and f ¥ to become reestablished in intact or denervated birds (fig. 4-1 OA). The data were a n a l y z e d s u c h that when S/F was reestablished on the breath immediately following the stimulated breath, the number of breaths required to reestablish rhythm was recorded as 0.0. R e g a r d l e s s of the duration of the s t a l l , the phase relationship between f w a n d f v was reestablished in the two breaths immediately following the dela y in > 8 0 % of the tri a l s . M any stu d i e s have s h own that the effects of an afferent stimulus on respiration are strongly p hase dependent ((Howard et a]., 1969; Eldridge, 1972a & b; Remmers & Marttila, 1975; Iscoe & P o l o s a , 1976; Y o u n e s & Polachek, 1981; K a w a h a r a et a]., 1988). Thus, it w as also important to examine the relationship between the timing of the stimulus relative to the respiratory period and the system's ability to reestablish S/F. There d id not, however, appear to be any relationship between the timing of the stimulus relative to respiration ( S / l " . ^ ) and the time required to reestablish S/F (fig. 4-1 OB). A g a i n , the relationship between f w and fv was reestablished by the first or s e c o n d breath following the delay in > 8 0 % of the trials, regardles of when during the respiratory c y c l e the delay was introduced. This was true for both intact and denervated birds. The only difference between intact and denervated birds was that, in a few inst a n c e s , the denervated birds took slightly longer to reestablish S/F than the intact birds. 145 Single wing flaps P h a s e response cu r v e s were generated for the entire respiratory period (f(8) v s 6) in 9 animals. In general, the respiratory period was shortened by single wing flap s (indicated a s a positive phase shift, f(9)). There was, however, a phase d e p e n d e n c y of the response. F i g . 4-11 shows the phase response curve obtained for one animal. There are two distinct regions on the curve. The first c orresponds to inspiratory stimulation a n d extends from 6 = 0.0 to approximately 0.45. Over this range, f(0) w a s positive a n d the slope of f(8) versus 8 was not different from z e r o . T h e respiratory period w a s shortened (f w increased) a similar amount regardless of when the stimulus w a s presented. O ver the remainder of the respiratory cycle (8 = 0.45 - 1.00) which roughly corresponded to expiratory stimulation, f(8) was positive a n d a negative s l o p e w a s obtained. The average maximum value of f(8) for all birds (0.19 + 0.03) was usually obtained shortly following the beginning of this s e c o n d region (8 = 0.47 + 0.02) and this d e clined toward zero by the end of the breath (8 = 1.00). Thus, stimuli presented in the latter half of the breath (expiration) produced a phase dependent p h a s e advance of the respiratory period. The magnitude of the p h a s e a d v a n c e (f(8)) d e c r e a s e d as the stimuli were presented later in the breath (i.e. a s 8 i n c r e a s e d ) . P h a s e response cu r v e s qualitatively similar to that shown in f i g . 4-11 were obtained in 6 of 9 animals. The responses of the three remaining birds to stimuli presented in the latter half of the respiratory cycle were also the s ame as shown in fi g . 4-11. However, the phase shifts recorded when stimuli were presented between 8 = 0.00 - 0.45 v a r i e d . In two birds, the magnitude of the increase in f(8) s h owed a phase dependent d e c r e a s e as 8 increased from 0.0 to 0.45. The final animal ' 146 Figure 4-11. Ty p i c a l p hase response curve for the entire respiratory period as ca l c u l a t e d for one goos e . The curve was generated as de s c r i b e d in f i g . 1A. The maximum phase shift (f(8)) of 0.24 occured when stimulation (single wing flap) was presented between 0.40 - 0.55 of the respiratory cycle (6). R e g i o n one (R-1) of the curve c o r r e s p o n d s to inspiratory stimulation and region two (R-2) to expiratory stimulation. P o s i t i v e v a l u e s of f(9) indicate a shortening of the respiratory c y c l e . 147 0.30 Figure 4-12. T y p i c a l p hase response curves from one goose showing the relationship between inspiratory stimulation (9,) and the phase shift of the ongoing inspiration (f9,; A) a n d the following expiration (f9E'; B), and of expiratory stimulation (9 E) on the p h a s e shift of the ongoing expiration (f9 E) a n d the following inspiration (f9,'). C u r v e s were generated a s d e s c r i b e d in fig. 1B. Positive v a l u e s of f(9) indicate a shortening of the respiratory c y c l e . 149 s h o w e d a positive s l o p e of f(9) v e r s u s 6 over this region, where f(9) was -0.15 at 9 = 0.05 a n d i n c r e a s e d to 0.00 at 9 = 0.20. Stimulation at the onset of the breath p r oduced small i n c r e a s e s in period (fv decreased). Th e magnitude of these c h a n g e s d e c r e a s e d a s the stimuli were delivered later in the c y c l e and had no effect at 9 = 0.25. T h u s, the overall response of the respiratory period to single wing flaps presented at varying times of the respiratory cycle c o n s i s t e d of, 1) a p hase independent positive p h a s e shift (f(9)) (increase in fv) over the first part of the cycle (region 1), a n d 2) a p hase dependent d e c r e a s e in the magnitude of p hase advancement (f(9)) o v er the latter portion of the respiratory c y c l e . In other words, stimulation during inspiration c a u s e d a small shortening of the breath. Stimulation during early expiration c a u s e d much larger d e c r e a s e s in breath duration, but the magnitude of the breath shortening d e c r e a s e d a s the stimulation was presented later into expiration. P h a s e response curves were also generated for 2 animals following denervation of the brachial plexus. T h e s e phase response curves were flat and grouped around f(9) = 0.00 indicating that single wing flaps do not advance or delay the respiratory c y c l e . The shortening of the respiratory period due to stimulation in region 1 (fig. 4-11) c ould be a c h i e v e d through a d e c r e a s e in the duration of the ongoing inspiration, the following expiration, both, or a d e c r e a s e in one that was greater than an increase in the other. To distinguish between these possibilities, plots of f(0), v s 9, a n d f(0) E' v s 9, were constructed. T h e s e graphs showed the effects of inspiratory stimulation on the length of the ongoing inspiration (f(9), v s 9,) a n d on the length of the following expiration (f(9) E v s 9,). Five of the nine birds examined showed an average 151 maximum ph a s e shift of 0.18 ± 0.01 when the stimuli were presented near the onset of inspiration. The magnitude of the phase shift d e c r e a s e d toward z e r o as 8, i n c r e a s e d toward 1.00. The responses of one bird are shown in fig. 4-12A. Inspiratory stimulation was found to have no effect on the length of the following expiration in five of nine birds, a s shown for one bird in fig. 4-12B. No other consistent effect of inspiratory stimulation on inspiratory or expiratory duration was found. Virtually all combinations of inspiratory and expiratory r e s p o n s e s were record e d . However, in spite of the inconsistent effects of inspiratory stimulation, w hen the effects of this stimulation on inspiration and expiration were summated, the net effect on overall respiratory period was a 5 - 1 5 % d e c r e a s e . The magnitude of this r e s p o n s e v a r i e d c onsiderably between animals, but e a c h individual s h o w e d a consistent shortening. P h a s e c h a n g e s occuring in response to stimulation in region 2 of the respiratory c y c l e (fig. 4-11), c o u l d only be a c h i e v e d through c h a n g e s in expiratory duration. Stimulation in the latter half of the breathing cycle resulted in a phase advance that d e c r e a s e d in magnitude a s 8 i n c r e a s e d in all 9 animals examined (fig. 4-11). By plotting f 8 E v s 8 E (fig. 12C), it was possible to determine the magnitude of the phase change in expiration required to produce the phase advance of respiration s e e n in region 2 of f i g . 4-11. On average, stimuli presented near the beginning of expiration (8 E = 0.00) produced maximum expiratory phase advance, f(8) E = 0.30 + 0.05. f(8) E then d e c l i n e d to f(8) E = -0.001 ± 0.028 at 8 E = 1.00 (end of expiration). It w a s a l s o p o s s i b l e that expiratory stimulation would affect the length of the following inspiration, thus f(8),' was plotted against 8 E (fig. 4-12D). From this plot, it ca n be s e e n that as stimuli were presented further into expiration, the following 152 inspiration was progressively shortened (f(e),* increased). On average, f(9),' i n c r e a s e d linearly from 0.03 + 0.03 at 0 E = 0.00 to 0.10 to 0.27 ± 0.03 at 0 E > 0.80. DISCUSSION It is c l e a r that afferent traffic produced through continuous passive wing flapping will entrain the respiratory rhythm over limited ranges. That this coordination represents an interaction between the two systems is indicated by two things. First, the range of breathing frequencies over which entrainment occurred was far greater than the normal variation in respiratory frequency (fig. 4-7). Thus the respiratory period w a s significantly altered through passive wing flapping. S e c o n d , perturbations of the wing flapping c y c l e produced corresponding changes in the respiratory c y c l e , s u c h that the p hase relationship between the two rhythms was maintained (fig. 4-8 & 4-9). R e g a r d l e s s of wing flap frequency, inspiration always started at or shortly after the peak of upstroke. If the coordination of respiration and wing beat was coincidental, p hase d e l a y s would not affect L A , but they would result in shifts in the p hase relationship (S/F) between f w and f v. The relevance of these findings to the control of locomotor-respiratory coordination, however, depends on whether the afferent traffic set up through passive wing motion is physiological. During free-flight, there will be c y c l i c changes in limb related feedback (Sevenn et a]., 1967; P r o c h a z k a , 1980), as well as c hanges in thoracic pressure/volume (Jenkins et a l . , 1988). P a s s i v e wing flapping also generates volume changes in the thorax (evident in the small amplitude oscillations superimposed on the larger 153 respiratory o s c i l l a t i o n s in fig. 4-5 and 4-8), and will activate a large number of the afferent pathways activated during free-flight. There will undoubtedly be differences in the afferent activity a s s o c i a t e d with the two conditions. T o begin with, p a s s i v e wing beat f r e q u e n c i e s were 4-5 times l e s s than free-flight f w's. In addition, the lack of muscie contraction will l e a d to d e c r e a s e s in the magnitude of some disch a r g e s a s well a s alterations of timing. Discharge from golgi tendon organs (GTO) remains constant with respect to limb motion during passive limb motion in cats, however, due to a d e l a y in sp i n d l e discharge relative to the locomotor c y c l e , GTO's discharge prior to the s p i n d l e s (Severin et al., 1967; P r o c h a z k a , 1980). Deformations of the thorax a s s o c i a t e d with contraction of the pectoralis muscles during flight (Jenkins et al., 1988) will a l s o be reduced during passive flapping (unpublished observations). M e c h a n o r e c e p t o r f e e d b a c k a s s o c i a t e d with these deformations will be reduced accordingly. E l e c t r i c a l stimulation of the flight muscles in phase with the pas s i v e wing motions c o u l d alleviate some of these problems, but this would introduce the potential for rhythmic activation of nociceptive afferent fibres. Nociceptor stimulation produces large c h a n g e s in ventilation ( M c C l o s k e y & Mitchell, 1972), but it is not clea r that nociceptors are rhythmically activated during actual locomotion. P r e v i o u s s t u d i e s have u s e d electrical stimulation of peripheral nerves to simulate the afferent traffic a s s o c i a t e d with rhythmic limb movements (Howard et al., 1969; Iscoe & P o l o s a , 1976; K a w a h a r a et a]., 1988). It is very difficult to reproduce the afferent neural flow s e e n during actual locomotion through electrical stimulation of discrete nerves. The afferent traffic set up through electrical stimulation of muscle or cutaneous nerves is very specific and not likely to represent the complex afferent signal a s s o c i a t e d with actual wing flapping. The best way to produce a 154 'physiological' afferent stimulus is to employ natural stimuli (Swett & B o u r a s s a , 1981). Activation of wing flapping through electrical stimulation of brainstem locomotor a r e a s would generate afferent traffic most c l o s e l y approximating the afferent traffic of wing flapping. However, this would also generate central, feedforward activity which may, in itself entrain respiration in the a b s e n c e of afferent f e e d b a c k (see chapter 5). In order to avoid these confounding influences, p a s s i v e wing flapping w a s u s e d . The finding that the phase relationship between f w and f v is very s i m i l a r during free-flight (inspiration starting just prior to peak upstroke; chapter 3, f i g . 3-4) a n d p a s s i v e wing flapping when ijiv is 3:1 (inspiration starting shortly at or shortly after peak upstroke; fig. 4-5, 4-8 & 4-9) s u g gests that the afferent pathways important in coordinating f w and f v during free-flight were activated during p a s s i v e wing flapping. A s already mentioned, p a s s i v e wing flapping will activate afferent activity not only from the wings, but from any source activated through depression and elevation of the wings. The signal responsible for entrainment of respiration to p a s s i v e wing flapping c o u l d therefore originate from locomotor related afferent activity or from a variety of respiratory related afferent pathways. C o n s i d e r a b l e attention has been paid to the role of afferent, neurogenic feedback from the working/moving muscles on the control of ventilation during e x e r c i s e . Part of the initial increase in ventilation at the onset of exercise has been attributed to the activation of a reflex pathway originating from the working muscles (Kao, 1963; S e n a p a t i , 1966; K a l i a et al., 1972; M c C l o s k e y & Mitchell, 1972; C r o s s et al., 1980; Mitchell, 1985; Waldrop et al., 1986a & b). There is also evidence that locomotor related afferent feedback can influence respiratory pattern. Repetitive 155 electrical stimulation of cutaneous leg (Iscoe & P o l o s a , 1976; K a w a h a r a et al., 1988) and muscle nerves (Iscoe & P o l o s a , 1976) will entrain respiration over limited ranges in the cat. However, it is difficult to determine the importance of this m e c h a n i s m in producing entrainment in the active animal. The problem with the s t u d i e s by Iscoe & P o l o s a (1976) and K awahara et a l . (1988) was that all animals were vag o t o m i z e d . One of the major peripheral systems involved in the breath by breath control of respiratory pattern was removed. Under suc h conditions respiratory rhythm may be influenced by other peripheral inputs that normally do not significantly influence respiratory rhythm. In addition, locomotor movements generate deformations in the s hape of the body wall and/or thorax in a variety s p e c i e s (wallabies: Baudinette et al., 1987; birds: • J e n k i n s et a l . , 1988; horses: Bramble & Carrier, 1983). C y c l i c c h a n g e s in respiratory related afferent pathways (lung mechanoreceptors, intercostal mechanoreceptors) generated by locomotion may therefore be predominantly responsible for the entrainment of locomotion and respiration. The afferent feedback from the limbs, although sufficient, may play only a minor role. The fact that nonvagotomized animals entrain to the frequency of a pump ventilator (vagally mediated pulmonary stretch receptor (PSR) input) over peripheral nerve stimulation supports this (Iscoe & P o l o s a , 1976; K a w a hara et al., 1988). In addition to c y c l i c , v agally mediated volume related feedback which, as recently shown in birds (Ballam et al., 1985), has long been known to entrain ventilation in cats (Adrian, 1933; Petrillo et al., 1983), periodic stimulation of a large number of respiratory reflex pathways will entrain respiration. Activation of afferent activity through repetitive, electrical stimulation of the vagus nerve in dogfish (Roberts & 156 Ballantijn, 1988), ca r o t i d s i n u s nerve (Eldridge, 1972a & b) and intercostal nerve ( R emmers & Marttila, 1975) in mammals also entrains respiration. In addition, os c i l l a t i o n s in intrapulmonary chemoreceptor discharge will pace respiration in birds (Kunz & Miller, 1974). It is unlikely that chemoreceptive feedback, whether humoral (Eldridge, 1 9 7 2 a & b) or airway related (Kunz & Miller, 1974) is involved in the entraiment of locomotion and respiration s i n c e neither is likely to fluctuate with the locomotor c y c l e . However, mechanoreceptive information from the lung and/or chest wall c o u l d be responsible for the entrainment of locomotion and respiration in g e e s e . T h i s s u g g e s t i o n is supported by the finding that removal of all afferent activity from the wings d o e s not affect the synchronization of wing beat and respiration during p a s s i v e wing flapping. The relationship between f w and f v was unaffected by denervation of the brachial plexus (fig. 4-9), inspiration occuring at or shortly after the peak of upstroke in both groups, regardless of f w . In addition, perturbations of the wing c y c l e did not affect the phase relationship in either group (fig. 4-9). The only difference between the two groups was that the denervated birds may have taken slightly longer to reestablish the phase relationship between f w and f v following d e l a y s (fig. 4-10). Thus, feedback from the moving wings, although it may be involved in coordinating f v to f w , d o e s not appear to be essential for the synchronization of respiration and wing motion. In a n attempt to further delineate which reflex pathway may be responsible for the synchronization of breathing and wing beat, a series of phase response curves were generated. T h e premise of these experiments was the observation by several workers that the effects of a given stimulus on respiratory pattern are dependent on the phase of the respiratory cycle in which they are presented, and on the nature 157 of the afferent pathway activated. The single wing flap stimulus u s e d here will activate a number of different afferent pathways. Comparison of the ventilatory r e s p o n s e s to single wing flaps with those s e e n in response to activation of sp e c i f i c pathways may help to define which pathway is responsible for the change in respiration s e e n during wing flapping. Presentation of single wing flaps during inspiration, although there was con s i d e r a b l e variation in results, produced a 5-10% shortening of the ongoing breath. The magnitude of this d e c r e a s e did not appear to change as the stimulus w as presented later into inspiration. In contrast, the effects of single wing flaps on expiratory duration were very consistent between animals and strongly phase dependent. T h e magnitude of the phase advance (breath shortening) d e c r e a s e d from 0.19 ± 0.05 at the beginning of expiration toward zero as the stimulus was presented later a n d later into expiration. C u t a n e o u s leg nerve stimulation (Iscoe & P o l o s a , 1976; K a w a h a r a et al., 1988) and muscle nerve stimulation in cats (Iscoe & P o l o s a , 1976) and femoral nerve stimulation in dogs produced similar reponses (Howard et al., 1969). Although varying in magnitude, stimulation during inspiration generally shortened inspiration. Stimulation during expiration produced a much more consistent d e c r e a s e in expiratory interval. T h e s e results agree with the finding that expiratory premotor neurons are inhibited by stimulation of muscle afferent fibres (Koizumi et al., 1961). M e c h a n i c a l stimulation of vagal activity (Petrillo et al., 1983), superior laryngeal (Larrabee & Hodes, 1948), intercostal (Remmers & Manilla, 1975), and carotid sinus nerves (Eldridge, 1 9 72a & b), also produced de c r e a s e s in inspiratory duration. Stimulation of thes e s a m e nerves during expiration, however, resulted in substantial 158 prolongation of the expiratory phase. The similarity of the ventilatory r esponses to limb afferent stimulation and single wing flaps suggest that some limb afferent rather than a chest wall or lung afferent system was responsible for the effects of single p a s s i v e wing flaps on respiratory rhythm. The flat p hase response curves obtained in two birds following denervation of the brachial plexus support this notion. Denervation of the brachial plexus, however, did not affect the entrainment of respiration by p a s s i v e wing flapping. In addition, although the effects of single wing flaps on respiration were somewhat phase dependent, they always c a u s e d i n c r e a s e s in respiratory frequency. To achieve coordination with wing beat during continuous wing flapping, respiratory frequency both increased and d e c r e a s e d . The two r e s p o n s e s , entrainment during continuous wing flapping and phase c h a n g e s during'single wing flaps, therefore, appear to be mediated v i a different mechanisms. The p hase r e s p o n s e curves appear to be a function of locomotor or muscle related feedback. Although this s a m e feedback may be involved in the coordination of wing beat a n d respiration during continuous wing flapping, feedback from some other s o u r c e (lung and/or chest wall) is sufficient for the coordination. The phase response c u r v e s did not help define which of the non wing afferent systems is responsible for entrainment, but they do indicate that wing related afferent activity, even if it d o e s not entrain respiration, d o e s affect respiratory pattern. Wing related afferent activity may be important in mediating rapid changes in respiratory rhythm. Following denervation of the brachial plexus, respiratory pattern d oes not respond to single wing flaps. Although the data are not at all conclusive, the time required to reestablish the phase relationship between f w and f v following delays may be slightly greater following denervation (fig. 4-10). Thus, although 159 highly s p e c u l a t i v e , during free-flight, when the wing beat rhythm is constantly modified to c o m p e n s a t e for environmental perturbation, feedback from the limbs may be important in the rapid adjustment of respiratory pattern and continual maintenance of entrainment. Although this study w a s not set up to directly examine the category of afferent fibres involved in the mediation of these reflexes, a number of conclusions can be drawn. T h e entrainment of respiration and wingbeat during p a s s i v e wing flapping a p p e a r e d to be due to activation of fibre types I and/or II. Although some type III fibres carrying high threshold pressure information may have been activated during p a s s i v e wing flapping, it s e e m s unlikely that type IV fibres, which typically carry nociceptive information (Swett & Bour a s s i , 1981), would be activated through p a s s i v e wing flapping. Furthermore, activation of fibre types III and IV is commonly a s s o c i a t e d with i n c r e a s e s in the overall level of minute ventilation in cats and dogs ( M c C l o s k e y & Mitchell, 1972). A s found during passive limb motion in rabbits (D'Angelo, 1984) a n d dogs (Russo et al., 1977), V E did not change during passive wing flapping (fig. 4-4), thus passive wing flapping does not appear to activate afferent types III a n d IV. Slowly adapting pulmonary receptor activity, which is primarily responsible for the entrainment of ventilation to respiratory pump frequency and electrical stimulation of the vagus, is con d u c t e d v i a type I and II afferent fibres (von Euler, 1986). The entrainment of respiration by intercostal nerve stimulation is also attributable to activation of group II afferent fibres (Remmers & Manilla, 1975). Conversely, entrainment of respiration is only obtained following activation of type III afferent fibres in the hamstring muscle nerve (Iscoe and P o l o s a , 1976) and type II, III and 160 p o s s i b l y type IV c u t a n e o u s afferents from the leg (Iscoe & P o l o s a , 1976; K a w a h a r a et a]., 1988). T h i s is consistent with the possibility that the non wing related afferent traffic is responsible for the entrainment of passive wing beat and respiration. E a c h of these afferent s y s t e m s is activated during locomotion (Severin et a]., 1967; K a u f m an & R y b i c k i , 1987; Rotto & Kaufman, 1989). However, in order to provide an entraining stimulus, the discharge of the afferent fibre must somehow con v e y information regarding locomotor frequency (rhythm). Activity in type IV afferent fibres from muscle i n c r e a s e s in response to certain chemical stimuli, but is insensitive to mechanical stimuli (probing and tendon stretch) (Kaufman & Rybicki, 1987). T h u s they are l e s s likely to be responsible for entrainment. Conversely, type II a n d III fibres, which are responsive to mechanical probing and tendon stretch, and whose d i s c h a r g e o s c i l l a t e s at the frequency of locomotion (Kaufman & R y b i c k i , 1988; S e v e r i n et al., 1967), are a more likely source of the entraining stimulus. Activation of type II fibres from the muscle, however, d o e s not entrain respiration (Iscoe & P o l o s a , 1976) a n d activation of type II! fibres would be a s s o c i a t e d with an increase in ventilation ( M c C l o s k e y & Mitchell, 1972). Therefore, of the wing related afferent f eedback activated through passive wing flapping, a type II cutaneous afferent pathway is the only likely contributor to entrainment in these experiments. 161 CONCLUSION A l l t y p e s of s y n c h r o n i z a t i o n observed between wing beat a n d respiration during free-flight c o u l d be reproduced v i a p a s s i v e wing flapping. Thus, afferent feedback a s s o c i a t e d with the p a s s i v e movement of the wings will entrain respiration and wing beat in the g o o s e . In addition, the phase relationship between the two rhythms was very si m i l a r during p a s s i v e and free-flight conditions, suggesting that the entrainment of f w a n d f v during p a s s i v e wing flapping is physiologically meaningful. T h e entraining stimulus appears to be mediated v i a a type I or II afferent pathway originating in the chest wall and/or respiratory system s i n c e denervation of the brachial plexus did not affect the entrainment nor the phase relationship between f w and f ¥.'Without showing that entrainment dissappears in the a b s e n c e of lung and chest wall f e e d b a c k in animals with wing related afferent feedback present, however, the participation of wing related feedback in the synchronization of wing beat a n d respiration cannot be ruled out. T h ese experiments have only shown that wing related f e e d b a c k is not essential for entrainment to occur. Wing afferent feedback d o e s a p p e a r to mediate the rapid c h a n g e s in respiratory timing produced through single wing f l a p s . A s suggested by the susceptibility of the respiratory system to pacing, it is likely that there is considerable redundancy in the system. Although f e e d b a c k from the chest wall/lung may be more important in producing entrainment, wing related feedback may be involved in mediating the rapid adjustments in respiratory pattern that would be required to maintain synchronization of wing a n d respiratory movements during free-flight. 162 CHAPTER 5 COORDINATION OF WINGBEAT AND RESPIRATION IN THE CANADA GOOSE III. "FICTIVE" FLIGHT 163 INTRODUCTION It w as shown in the previous chapter that afferent activity a s s o c i a t e d with p a s s i v e wing flapping will shift the respiratory period so that the two patterns are entrained o ver a limited range. The questions remains, however, is afferent f e edback e s s e n t i a l for the coordination of wing and respiratory movements? Is there a central program which c o u p l e s the outputs of the two motor s y s t e m s ? Zuntz & Geppert (1886; cited by Eldridge et a]., 1985) first invoked this concept of feedforward neurogenic drive to explain the rapid increase in ventilation at the onset of e x e r c i s e . S i m p l y stated, it s u ggested that neural impulses arising from the suprapontine brain that c o mmand muscle to exe r c i s e also 'irradiate' to the respiratory centres in the medulla. S u c h a command system could lead to a neurally driven i n c r e a s e in ventilation that is proportionate to the increase in metabolic rate without the intervention of afferent feedback mechanisms. S i n c e then, feedforward m e c h a n i s m s have been shown to regulate ventilatory drive during e x e r c i s e (Eldridge et al., 1985), and has been implicated in the synchronization of ventilatory pattern with locomotor pattern (Viala et al., 1987b; Pers e g o l et a]., 1988; K a w a h a r a et al., 1989a). V i a l a et a l . (1987b) and Pers e g o l et a l . (1988) working with p a r a l y z e d rabbits, a n d K a w a h a r a et a l . (1989a), using paralyzed cats, have shown that the neural outputs from the locomotor and respiratory nerves in their respective s y s t e m s are weakly coupled in paralyzed animals in the absence of afferent feedback. There were se v e r a l problems with these studies. Both studies used artificially ventilated animals, thus not all afferent feedback was removed. Although both 164 groups v a g o t o m i z e d their animals to remove the influence of pulmonary stretch receptor f e e d b a c k on respiratory pattern, intercostal (Remmer & Marttila, 1975; S h a n n o n et al., 1988) and diaphragmatic reflex pathways (Speck & Revelette, 1987; Speck, 1988) remained which could alter and even entrain (Remmers & Marttila, 1975) respiratory rhythm. This appeared to be a larger problem in the study of K a w a h a r a et a l . (1989a) s i n c e entrainment was only s e e n when the pump frequency approximated the locomotor frequency. The chief concern with the V i a l a s t u d i e s ( V i a l a et al., 1979: V i a l a , 1986; V i a l a et al., 1987b; P e r s e g o l et al., 1988) wa s that they u s e d unilateral hindlimb E N G recordings to a s s e s s locomotor pattern. A s rabbits have gaits in which the hindlimbs are both in and out of phase, unilateral recordings of hindlimb nerve activity do not adequately describe locomotor pattern in rabbits. T h u s it is v e r y difficult to a s s e s s the relationship between respiration and locomotion in these studies. Furthermore, V i a l a has only shown tight coupling between locomotion a n d spinal respiration (Viala et al., 1979; V i a l a , 1986; V i a l a et al., 1987b). Whether or not phrenic activity in the spinal animal truly represents respiration is still in question (von Euler, 1986). Th e purpose of this study was, therefore, to determine if the entrainment between wingbeat a n d respiration observed during free-flight in birds could be due to a central coupling of locomotor and respiratory rhythm generators. Birds offer unique advantages over the preparations previously employed. Due to the unique structure of the avian respiratory system, birds c an be unidirectionally ventilated (UDV), rather than pump ventilated during paralysis. By cannulating the trachea and an air s a c , a unidirectional flow of air c a n be set up that carries air in the trachea, a c r o s s the parabronchi, and out the cannulated air s a c . In s o doing, regular 165 mechanical deformation of the respiratory s y s t e m a n d chest w a l l , and the a s s o c i a t e d p h a s i c afferent feedback is avoided. In addition, the problem encountered by V i a l a and coworkers (Viala et al., 1979: V i a l a , 1986; V i a l a et al., 1987b; P e r s e g o l et al., 1988) in defining the gait and/or gait transition between walking a n d hopping is not a problem with birds since, once the wings are recruited into the locomotor pattern, they always beat synchronously (Sholomenko et a]., 1990a, b & c; J a c o b s o n & Hollyday, 1982). MATERIALS AND METHODS Experiments were carried out on 7 Pekin ducks (Anas platyrhynchos; body weight 2.63 + 0.08 kg) a n d 9 C a n a d a g e e s e (Branta canadensis; body weight 3.6 + 0.2 kg). T h e su r g i c a l procedures used in these experiments for d u c k s and g e e se are very similar to those d e s c r i b e d in chapter 1 for decerebrate geese. Briefly, the animals were tr a c h e o s t o m i z e d to allow for: anaesthetia, monitoring of ventilation v i a pneumotachography, an d for the subsequent administration of unidirectional ventilation (UDV) following paralysis. An air s a c was also cannulated to provide an outflow path for the air during UDV. The air s a c c a n n u l a was plugged during s p ontaneous breathing. A carotid artery and jugular vein were cannulated for monitoring blood pressure and for the later infusion of the paralytic agent (flaxedil = gallamine triethiodide). The birds were then supported in a sling overlying the treadmill. A thermometer was inserted approximately 25 cm down their esophagus to monitor body temperature (T B). T B was maintained constant using a heat lamp and a 15 cm, small diameter U-shaped copper tube inserted through the anus into 166 the intestine. Hot or c o l d water could be circulated through the tube a s required to maintain T B. T h e heads of the birds were placed in a stereotaxic headholder, and a craniotomy w a s performed, followed by a decerebration. T h e transection level e xtended from the c a u d a l border of the habenular nucleus dorsally to the rostral border of the anterior preoptic nucleus ventrally (fig. 5-1). Birds were more responsive following this level of transection than following lower level transection, which facillitated the development of locomotor activity following paralysis. After completion of the decerebration, the nerves innervating the external (inspiratory) or internal (expiratory) intercostal m uscles of the 6th costal s p a c e and the pectoralis major muscle were isolated for later recording of "fictive" respiration a n d wing flapping, respectively. In a s e r i e s of preliminary experiments, external intercostals and internal intercostals in the 5,h and 6th costal s p a c e s were found to be inspiratory a n d expiratory respectively. S i n c e there was the possibility that the intercostal m u s c l e s are recruited for some locomotor function during wing flapping, cranial nerve IX, which in mammals exhibits inspiratory related activity a s s o c i a t e d with airway dilation (Iscoe, 1988) and is much le s s likely to have a role in movement of the wings, was also isolated in three animals. Electromyographic ( E M G ) electrodes were also implanted into internal and/or external intercostal m u s c l e s of the 5th a n d 6th costal s p a c e s and the right and left pectoralis muscles to monitor intercostal muscle activity and wingbeat frequency during electrically induced wing flapping. The intercostal m u s c l e s form very thin adjacent bands between the ribs. The E M G electrodes originally used for monitoring the activity of these muscles 167 Figure 5-1. Diagram of the avian brain showing the transection level (A) u s e d in t h e s e experiments. The rostral transection level was u s e d here a s it facillitated "Active" locomotor activity. The diagram represents a near midline (lateral 0.5 mm) saggital s e c t i o n of the pigeon brain (Karten & Hodos, 1967). Abbreviations: A H P - a r e a hypothalami posterioris, A M - N u c l e u s anterior medialis hypothalami, A n l - N u c l e u s annularis, A P H - A r e a parahippocampalis, B O - olfactory bulb, C b , c e r e b e l l u m , C O - optic chiasm, C P - posterior commissure, C S - central superior nucleus, D B C - d e c u s s a t i o n brachium conjunctivum, D M A - dorsomedial anterior t halamic nucleus, D M P - dorsomedial posterior thalamic nucleus, E W -Edinger-Westphal nucleus, F D - dorsal funiculus, F L M - medial longitudinal f a s c i c u l u s , F V - ventral funiculus, G C - cuneate and gracile nuclei, Hb - habenular nucleus, Hp - hippocampus, H V - ventral hyperstiatum, IM - intermediate nucleus, IP - interpeduncular nucleus, L P O - parolfactory lobe, M N V - mesencephalic trigeminal nerve nucleus, N - neostriatum, N C - c a u d a l neostriatum, N l l l -occulomotor nerve, nIV - trochlear nucleus, nX - motor nuleus vagus, nXII -hypoglossal nucleus, O l - inferior olivary nucleus, O M d - dorsal part, occulomotor nucleus, O M v - ventral part, occulomotor nucleus, Ov - ovoid nucleus, P - pineal, P a M - paramedian nucleus, P O A - anterior preoptic nucleus, P O M - medial preoptic nucleus, R g c - gigantocellular reticular nucleus, R P - pontine caudal reticular nucleus, R P g c - gigantocellular part, caudal pontine reticular nucleus, Ru - red nucleus, S - nucleus solitarius, S C E - external cellular stratum, SCI - internal cellular stratum, S L - lateral septal nucleus, S M - medial septal nucleus, V -ventricle 168 169 (same a s tho s e u s e d for pectoralis E M G ) were too large and often picked up activity of the adjacent group. In addition, the electrodes were not anchored in place originally a n d were frequently d i s l o d g e d during wing flapping. Thus some traces obtained during the stimulation period contain a movement artifact. In later runs, all recordings were made with smaller electrodes that were sutured in place. The equipment u s e d to record the data in these experiments is the s a m e as that d e s c r i b e d in chapter 1. Anae s t h e t i c w as then discontinued a n d the birds were allowed a minimum of 1 hour to recover. Brainstem stimulation (pulse duration, 2 ms; pulse frequency, 60 Hz; stimulation intensity range, 30-150 |iA) was then initiated using procedures d e s c r i b e d by S t e e v e s et a l . (1987). High intensity stimulation (100 (iA) w as used to loc a l i z e a stimulation site that would elicit wing flapping. Stimulation intensity was then d e c r e a s e d to ze r o and slowly increased to establish the current threshold n e c e s s a r y to invoke wing flapping. This w as invariably higher than the threshold required to activate walking in chapter 1. Trials were then carried out using stimulation intensities approximately 2 0 % greater than threshold at an average of 137 + 23 uA. The suprathreshold stimulation avoided the delay between onset of stimulation a n d start of locomotion often a s s o c i a t e d with stimulation at threshold l e v e l s . Following completion of the protocol on the non paralyzed animals, the birds were unidirectionally ventilated with 4 % C 0 2 in air at flow rates of 1.5 to 2.0 L/min. Th i s composition and flow rate were chosen because blood gas data from preliminary experiments (n=3) indicated that this concentration of inspired gas maintained blood P a ^ (108.5 ± 4.0 mmHg), P a ^ (27.1 + 0.36 mmHg) and p H a 170 (7.41 ± 0.01) near normal levels s e e n in du c k s ( P a ^ , 99 ± 2 mmHg; P a c 0 2 , 28.5 + 0.5 mmHg; Bouverot, Hill & Jammes, 1974). Recent v a l u e s for domestic (Scheid et al., 1989; Pa,,, = 99.3 + 1.5, P a ^ = 32.2 + 1.5, p H a = 7.52 ± 0.01) and bar headed g e e s e (Fedde et al., 1989; Pao 2 = 92 + 3, P a ^ = 34 + 4, p H a = 7.44 + 0.02) indicate that the g e e s e in this study may have been slightly hyperventilated. Flow rates had to be high enough to maintain blood g a s e s (>1.2 L/min), but very high flow rates t e n d e d to inhibit breathing. W h en P a ^ was d e c r e a s e d or P ^ inc r e a s e d v e r y far from these levels in the paralyzed birds, the discharge of the intercostal motor neurons d e c r e a s e d both in rate and amplitude or dis a p p e a r e d altogether. In addition to this hyperventilation induced apnea, high flow rates (>2.5 L/min) also lowered T B. Thus, flows were c h o s e n that, when supplemented with 4 % C 0 2 , maintained blood g a s e s similar to those reported for ducks a n d did not alter T he birds were then paralyzed with an initial 1.0 mg/kg I.V. injection of flaxedil. S upplemental d o s e s were administered a s required. Tubocurarine w as tried in some preliminary experiments, but it was found to severely s u p p r e s s central respiratory activity. T h e left intercostal (either internal or external) and the left pectoralis nerves were then p l a c e d on bipolar platinum hook electrodes, submerged in mineral o i l . In three c a s e s , recordings were taken from a small branch of cranial nerve IX rather than from the intercostal nerve. S i g n a l s were amplified (10.000X), filtered and recorded in the sa m e manner as the E M G sign a l s . O ne channel of E M G was usually recorded to ensure that paralysis was complete during all trials. Stimulation trials (100-400 uA) were then resumed, this time recording E N G activity rather than E M G activity. H R and B P were also recorded. A s noted for other vertebrates 171 (Jordan et al., 1979; J a c o b s o n & Hollyday, 1982; Williams et al., 1984; Wallen & Will i a m s , 1984; Sh o l o m e n k o et al., 1990c), the current intensity required to activate locomotion in g e e s e (and ducks) was increased following paralysis from 137 + 23 to 340 ± 46 u A At the co n c l u s i o n of e a c h experiment, the position of the stimulating electrode w a s marked with an electrolytic lesion (3 mA/5 s e c ) . Histological identification of stimulation s i t e s w as performed a s des c r i b e d by S t e e v e s et a l . (1987). It s h o u l d be noted that only unilateral (left) recordings of intercostal and pectoralis nerves were taken. It was felt that bilateral recordings were not nec e s s a r y . S e v e r a l preliminary experiments were performed (n=5) to ensure, a s indicated by J o n e s & Bamford (1976), that the intercostal nerves provide a good index of respiration in birds. During the course of these experiments, several bilateral recordings were made, both before and following paralysis. The discharge of the right a n d left external intercostal nerves was always synchronous, and, before para l y s i s , in phase with inspiratory air flow (fig. 5-2). Right and left internal intercostal nerves d i s c h a r g e d synchronously in phase with expiration. In addition, the d i s c h a r g e of external a n d internal intercostal nerves, whether ipsi or contralateral, w a s always out of phase (fig. 5-3). Similarly, numerous bilateral pectoralis nerve recordings have been made (Sholomenko et al., 1990c) during spontaneous, a n d electrically and chemically induced "Active" wing flapping. A s y n c h r o n o u s discharge of left and right s i d e s was never observed. Thus I was confident that unilateral intercostal and pectoralis E N G recordings reliably indicated "fictive" respiration a n d flight. In contrast, bilateral recordings were never made from cranial nerve IX. 172 Figure 5-2. R e l a t i o n s h i p between right and left external intercostal E N G activity and respiratory air flow in a duck prior to paralysis. Exp indicates expiration and insp indicates inspiration. 173 Air Flow * * { W Y W W Y Y Y Y Y Y v R. External Int. ENG L. External Int. ENG 5 MC. Figure 5-3. "Fictive" respiration as indicated by E N G recordings taken from the left external (inspiratory activity) a n d internal (expiratory activity) intercostal nerves in a d u c k following p a r a l y s i s . 175 External int. ENG Internal int. ENG CD Figure 5-4. "Fictive" respiration as indicated by E N G recordings taken from the left c r anial nerve IX (inspiratory activity) and the left external intercostal nerve (inspiratory activity) in a duck following paralysis. 177 However, to ensure that it represented respiratory activity, its discharge was always c o m p a r e d with the dis c h a r g e of either an internal or external intercostal nerve prior to the initiation of experimental trials. Its activity was always found to be in phase with external intercostal nerve discharge (fig. 5-4) and out of phase with internal intercostal nerve d i s c h a r g e . Protocol Non paralyzed birds S e v e r a l stimulation trials were carried out to describe the relationship between air flow a n d intercostal a n d pectoralis muscle activity during actual electrically i n duced wing flapping. Minute ventilation (V E), ventilatory frequency (f ¥), tidal volume (V T), blood pressure (BP), heart rate (HR), internal or external E M G activity a s well a s right a n d left pectoralis E M G activity were all recorded for 30 s e c o n d s prestimulation, throughout stimulation and for up to 3 minutes post stimulation. The majority of stimulation periods were between 10 and 30 s e c o n d s . However, in two an i m a l s the duration w a s i n c r e a s e d to two minutes in order to de s c r i b e the kinetics of the ventilatory response to electrically induced wing flapping (fig. 5-8). Paralyzed birds Activity of the pectoralis nerve ("fictive" flight) and internal or external intercostal nerve or cranial nerve IX ("fictive" respiration) were recorded for 30 - 60 s e c o n d s prestimulation, throughout stimulation (range 10 - 30 s) and for up to 2 min post stimulation. P e c t o r a l i s E M G activity, H R and B P were also recorded. 179 TABLE 5-1 DIFFERENT WINGBEAT-RESPIRATORY COORDINATION SCHEMES IN DECEREBRATE BIRDS PRE-PARALYSIS POST-PARALYSIS BIRD# VL Stim. Site V f y stim. site Duck 1 * 1 : 1 TTD/Cnd 1:1 TTD/Cnd 2 4 : 1 TTD/Cnd 1:1 M R F 3 * 1 : 1 ICQ 1:1 ICQ 4 - 1:1 MRF 5 2:1 ICo 2:1/1:1 RP 6* 1:1 ICo 1:1 l C o 7 " 1:1 ICo Goose 1 * 2:1 RP 2:1/1:1 RP & ICo 2 1:1/3:1/4:1 RP 3 - - 1:1 ICo 4 1:1 TTD/Cnd 5 1:1 TTD/Cnd 6 2:1 Cnd 7 4:1 Cnd * - same stimulation site used before and after paralysis STIMULATION SITES (Stim. Site) TTD; nucleus and tract of the descending trigeminal nerve 180 C n d ; nucleus reticularis medullaris, pars dorsalis R P ; nucleus reticularis pontis c a u d a l i s , pars gigantocellularis M RF; m e s e n c e p h a l i c reticular formation ICo; nucleus intercollicularis The number of animals used in e a c h protocol is summarized in table 5-1. Not all birds were e x a m i n e d both before and after paralysis, however, there was no indication that t h o s e birds studied only before or only after paralysis differed from those studied under both conditions. A N O V A was used to test the difference between means. V a l u e s of P<0.05 were a s s u m e d to be significant. RESULTS NON-PARALYZED RESPONSES Brainstem stimulation s u c c e s s f u l l y initiated active wing flapping in all five d ucks and six of nine g e e s e . All effective stimulation sites fell within the a r e a s previously d e s c r i b e d by S t e e v e s et a l . (1987) and Sholomenko et a l . (1990a, b & c), and are s u m m a r i z e d in table 5-1, 5-2 and fig. 5-5. Two of the 5 ducks were stimulated in the mid medulla, between the nucleus reticularis medullaris, pars dorsalis (Cnd) ventromedially and the nucleus and tract of the trigeminal nucleus (TTD), d o r s o l a t e r a l ^ . The remaining three were stimulated in the nucleus intercollicularis of the midbrain. L e s i o n s were located to the TTD/Cnd (n=2), C n d (n=2), and the pontine reticular formation (RP; n=2) in the geese. The stimulation sites were similar for both d u c k s a n d g e e s e (see table 5-1 a n d fig. 5-5 for summary of stimulation si t e s ) . During the course of the experiments, a variety of flight patterns were observed. Figure 5-6 s h o w s an entire flight s equence from a duck. In this particular trace, rather than using the usual suprathreshold stimulus ( 2 0 % greater than 182 TABLE 5-2 RELATIONSHIP BETWEEN STIMULATION SITE AND 1Jtv A C T U A L F L A P P I N G "FICTIVE" F L A P P I N G S T I M S I T E t / f v i j i v 1:1 2:1 3:1 4:1 1:1 2:1 3:1 4:1 TTD/Cnd 4 1 - 2 1 R P 1 1 1 1 1 1 - -M R F . . . . 2 - - -ICo 2 1 - - 5 - - -threshold) which activates locomotion with little or no delay, threshold intensity w as use d . T h u s in this figure, it is possible to s e e the dramatic alteration of respiratory pattern a n d i n c r e a s e in B P and heart rate that occurs immediately with the onset of stimulation, prior to the development of wing activity. While V T d e c r e a s e d from 33.3 ml/kg to 16.3 ml/kg, the dramatic increase in f v from 18.2 to 135 breaths/min o b s e r v e d at the onset of stimulation resulted in a large increase in VE from 606 ml/min/kg to 2194 ml/min/kg. There were further small changes in pattern when the wings finally b e c a m e active. f v increased further to 144 (breaths/min), but a de c r e a s e in V T to 11.25 ml/kg resulted in a decr e a s e in V E to 1620 ml/min/kg. Similarly, H R i n c r e a s e d over the first 7 se c o n d s of stimulation from 200 beats/min at rest to a level of 513 which was maintained during the flapping period. Blood pressure a l s o i n c r e a s e d over the first s e v e n s e c o n d s from 122 mmHg at rest to a maintained maximum of 175 mmHg. With the termination of stimulation a nd wing flapping, all v a r i a b l e s gradually returned toward prestimulation levels. Similar c h a n g e s in respiratory pattern from rest to wing flapping c a n be s e e n in fig. 5-9 and 5-10. However, in these two c a s e s , suprathreshold stimulation was used. Thus the c h a n g e s in respiratory pattern were coincident with the onset of wing flapping. The r e s p o n s e s of all animals (ducks and geese) to electrically induced wing flapping were a v e r a g e d and are shown in fig. 5-7 (mean + S.E.). V E increased 2.5 fold on average a s a result of an approximate 10 fold increase in f v as V T d e c r e a s e d to half its resting l e v e l . Heart rate and blood pressure both increased approximately 40 to 4 5 % from rest to wing flapping. Although the data presented thus far is only for wing flapping periods of short duration, the responses observed in two g e e s e during two minutes of sustained wing flapping (fig. 5-8) are 184 Figure 5-5. C o m p o s i t e diagram of c r o s s sections through the avian brainstem showing the stimulation s i t e s u s e d to activate both actual (closed symbols) an d "fictive" (open symbols) wing flapping in ducks (triangles) and g e e s e (squares). Stimulation s i t e s effective in evoking both actual and "fictive" locomotion are d e s i g n a t e d with filled d iamonds (ducks) and solid s q uares surrounded by a diamond (geese). S e c t i o n s A through F run caudo-rostral (A = medulla; B = pons; E = midbrain). S e c t i o n A is magnified relative to the other sections to permit labelling. Abbreviations: A L - a n s a lenticularis, A Q - cerebral aqueduct, B C - brachium conjuctivum, C n d - nucleus reticularis medullaris pars dorsalis, C n v - nucleus reticularis medullaris pars ventralis, D B C - decussation of the brachium conjunctivum, E M - ectomammillary nucleus, E W - Edinger-Westphal nucleus, ICo -nucleus intercollicularis, III - oculomotor nucleus, 10 - inferior olivary nucleus, IP -interpeduncular nucleus, IV - trochlear nucleus, L C - locus ceruleus, M L d - lateral m e s e n c e p h a l i c nucleus, M L F - medial longitudinal fasciculus, M R F - mesencephalic reticular formation, MV - motor trigeminal nucleus, Mill - oculomotor nerve, NIV -trochlear nerve, N V - trigeminal nerve, N X - vagus nerve, O T - optic tectum, R -raphe nucleus, R P - pontine reticular formation, R p c - pontine nucleus, parvocellular part, R P O - pontine reticular nucleus, oral part, R u - red nucleus, S T -subtrigeminal nucleus, S V - trigeminal sensory nucleus, T P c - substantia nigra, pars c ompacta, T S - tractus solitarius, T T D - nucleus and tract of the descending trigeminal nerve, VII - facial nucleus, X - motor nucleus of the vagus, XII -hypoglossal nucleus 185 Figure 5-6. W i n g flapping s e q u e n c e initiated using threshold current (90 uA) in a decerebrate d u c k prior to paralysis. Prestimulation a n d recovery records are inclu d e d . Heart rate, blood pressure, left and right pectoralis E M G and left intercostal E M G records a n d respiratory air flow (inspiration up and expiration down) are shown. Note that the blood pressure trace is inverted. 187 Heart Rate stlm. off 2 sec CO 00 Figure 5-7. Mean levels (+ S.E.) of minute ventilation (V E), tidal volume (V T), breathing frequency (fv), heart rate (HR), blood pressure (BP) and wingbeat frequency (f j for all decerebrate birds (ducks and geese; sample size indicated by the number over each column) prior to paralysis at rest (open columns), during electrically induced wing flapping (solid columns), after paralysis at rest (-ve slope hatching) and during "fictive" wing flapping (+ve slope hatching). 189 190 Figure 5-8. Kinetics of the ventilatory response of two decerebrate geese to two minutes of electrically induced wing flapping, showing minute ventilation (VE), tidal v olume (V T) a n d breathing frequency (f v). Stimulation was turned on at t=0 and off t=120 (mean + S.D.). 191 Figure 5-9. Wing flapping s e q u e n c e in a decerebrate duck prior to paralysis s howing a 1:1 relationship between wingbeat frequency (left and right pectoralis E M G ) a n d respiratory frequency. Note the increase in chart s p e e d that o c c u r e d just before the onset of stimulation. 1 9 3 Figure 5-10. Re l a t i o n s h i p between right and left pectoralis E M G activity, left external intercostal E M G activity a n d respiratory air flow at rest (top) and during electrically i n d u c e d wing flapping in a decererate duck (bottom). Note the different time s c a l e s in the two panels. 195 R. Pact. EMG CO CO L. Poet. EMG L. External Int. EMG Air Flow Insp. t Exp. i 2 MC. R. Pect. EMG L Pect. EMG L. External Int. EMG Air Flow Insp. t Exp. I stjm. on t stfvn. off t Figure 5-11. Wing flapping s e q u e n c e in a decerebrate goose prior to paralysis s howing a 1:1 relationship between wingbeat frequency (left and right pectoralis E M G ) a n d respiratory frequency (Int. air flow = integrated air flow with expiration up and inspiration down). 197 similar. V T d e c r e a s e d with the onset of wing flapping, but a 4.5 fold increase in f v resulted in a 2.5 fold i n c r e a s e in V E. A c l o s e r examination of the data also indicates a very tight relationship between locomotor a n d respiratory patterns. Locomotion and respiration were syn c h r o n i z e d at one wing beat per breath (1:1) in 3 of 5 ducks (fig. 5-9 & 5-10) and 3 of 6 geese (fig. 5-11). Alternate relationships were a l s o observed. W hen the region of active wing flapping from f i g . 5-6 is expanded (fig. 5-12), although it is difficult to decipher the intercostal E M G trace (see methods), it is cle a r that wingbeat and respiratory f requency are coordinated at two wingbeats per breath. A coordination of two wingbeats per breath was also o b served during electrically induced wing flapping in two of the six C a n a d a geese examined. O n different o c c a s i o n s , a particular bird c o u l d s h o w different types of coordination. One goose showed entrainment at 1:1, 3:1 a n d 4:1. The remaining goose (fig. 5-13) and duck coordinated the two patterns at 4:1. From thes e results, s u mmarized in tables 5-1 and 5-2, it is cle a r that wingbeat and respiration are coordinated, even during electrically induced wing flapping. It is equally c l e a r that, although 1:1 coordination is the most prevalent, there are a number of alternative coordination s c h e m e s available. A c l o s e r examination of the figures 5-9, 5-10 and 5-11 also reveals a tight p h a s e relationship between the two patterns. When wing flapping a n d respiration are c o u p l e d at 1:1, inspiratory air flow always occurs during the upstroke (ie. no Pector a l i s E M G activity) and expiratory air flow always occurs during the downstroke (ie. active P e c t o r a l i s E M G ) in both the duck (fig. 5-9 & 5-10) and the goose (fig. 5-11). T h i s c a n best be se e n in fig. 5-10 which includes a trace of external intercostal muscle activity (indicating inspiratory efffort) with air flow. External intercostal E M G 199 activity is coincident with inspiration at rest and during wing flapping. Like inspiratory air flow, it is clearly out of phase with pectoralis E M G activity, indicating that inspiratory activity o c c u r s when the pectoralis is quiescent, i.e. during upstroke. The p h a s e relationship between respiration and wingbeat when the two patterns are not c o u p l e d at 1:1 is not as obvious. However, it d o e s appear from fig. 5-12 a n d 5-13, that a reversal of respiratory air flow, either from inspiration to expiration or the reverse, is c l o s e l y correlated with the onset of E M G activity in the pectoralis, whether the wingbeat to respiration ratio is 2:1 (fig. 5-12) or 4:1 (fig. 5-13). "FICTIVE" FLIGHT "Fictive" records of wing and respiratory activity were obtained from s e v e n ducks and 2 g e e s e . A s s e e n during actual wing flapping, the results obtained for ducks and g e e s e during "fictive" flight were virtually identical. A g a i n , all effective stimulation s i t e s fell within the regions described by S t e e v e s et a l . (1987) and S h o l o menko et a l . (1990a,b & c) and are summarized in tables 5-1, 5-2 and fig. 5-5. Although "fictive" flight c ould be initiated from the more caudal stimulation sites in the dorsolateral reticular formation of the medulla (TTD/Cnd; duck, n=1), the more rostral s i t e s , including the pontine reticular formation (RP; duck, n=1; goose, n=1), the m e s e n c e p h a l i c reticular formation (MRF; duck, n=2) and nucleus intercollicularis (ICo; duck, n=3; g e e s e , n=2), were most effective in initiating "fictive" activity. Stimulation sites u s e d to initiate flapping pre a n d post paralysis were not n e c e s s a r i l y the same. Of the 6 birds (5 ducks and 1 goose) examined during both actual flapping and paralysis, however, the same stimulation sites were used in four (3 d u c k s and 1 goose; table 5-1). 200 Paralysis itself did not significantly alter heart rate, blood pressure or respiratory frequency relative to the preparalyzed resting levels (fig. 5-7). The magnitude of the increases from resting levels in HR, BP and fv during "fictive" flight were also very similar to those seen during actual wing flapping (fig. 5-7). The main difference between the preparalyzed and paralyzed conditions was that following paralysis, fw was 40% lower in "fictive" flight than actual wing flapping. Since fv decreased only 25%, the relationship between fw and fv also changed during "fictive flight". During actual flapping, fw and fv were coordinated at 1:1 in 6 of 11 birds (table 5-11). The remaining birds showed 2:1, 3:1 and/or 4:1. Following paralysis, however, all 9 birds (7 ducks, 2 geese) that demonstrated "fictive" flight demonstrated 1:1 synchronization, although 2 also showed periods of coordination at 2:1 (table 5-1 & 5-2). An example of 1:1 coordination during "fictive" flight is shown in fig. 5-14. The top 2 traces indicate resting activity in the external intercostal and pectoralis nerves. The pectoralis nerve is silent and the external intercostal nerve, an inspiratory nerve indicating resting "fictive" fv, is discharging at 12 bursts/min. With the onset of stimulation and "fictive" flight, indicated in the bottom pair of traces, there is a dramatic increase in fv from 12 to 160 bursts/min as it increases to match the increase in fw 1:1. Note that the external intercostal nerve discharges out of phase with the pectoralis nerve. This indicates that inspiratory effort occurs during "fictive" upstroke (quiet pectoralis ENG) as seen during 1:1 coupling in actual wing flapping (fig. 5-9, 5-10 & 5-11). Similarly, inspiratory activity in cranial nerve IX was coordinated 1:1, and out of phase with pectoralis nerve activity during "fictive" flight (fig. 5-15). In this case, 201 resting "fictive" respiration i n c r e a s e d from 6.7 bursts/min to 87.4 with the onset of stimulation a n d "fictive" flapping. T h i s p h a s e relationship is also demonstrated in fig. 5-16. The top 2 traces indicate resting conditions. R esting discharge frequency (f„) of the internal intercostal nerve, an expiratory nerve, i n c r e a s e s from 22.1 to 205 bursts/min with the onset of "fictive" flight, a g a i n matching f w 1:1. A s expected from the air flow traces during actual flapping (fig. 5-9, 5-10 & 5-11) and the "fictive" records of fig. 5-14 & 5-15, the expiratory, internal intercostal nerve dis c h a r g e s in phase with the pectoralis nerve. Expiratory effort o c c u r s in phase with "fictive" downstroke (active pectoralis E N G , f i g . 5-16), and inspiratory effort o c c u r s in phase with "fictive" upstroke (quiet pectoralis E N G , fig. 5-14 & 5-15). T h e previous 3 recordings of "fictive" flight were all made on ducks. Similar r e s p o n s e s were s e e n in the 2 experiments performed on geese. One example is s hown in f i g . 5-17. Resting "fictive" f v increased from 6.2 bursts/min at rest (not shown) to 53.8 bursts/min during "fictive" flight. A s s e e n previously in ducks, f v and f w are c o u p l e d at 1:1, a n d the inspiratory activity (external intercostal E N G ) is out of p h ase with the pectoralis E N G . Although wingbeat and respiration were predominantly coupled at 1:1 during "fictive" wing flapping, 2 of the 9 birds examined (1 duck and 1 goose) also d i s p l a y e d 2:1 coordination. Interestingly, both of these birds had also displayed 2:1 coordination prior to paralysis. It is of further interest that one of these incidents o c c u r e d spontaneously. The "fictive" flight burst was not initiated through electrical stimulation of locomotor centres. This particular episode is shown in fig. 5-18. The top trace represents whole intercostal nerve discharge. S i n c e the whole 202 Figure 5-12. E x p a n d e d version of the section in fig. 5-6 in which the wings were moving to illustrate the precise 2:1 relationship between pectoralis E M G activity and respiratory rhythm. 203 Heart Rate 125 mmHg Blood Pressure 175 mmHg R. Pect EMG L. Pect EMG L. Intercostal EMG Insp. t Air Flow Exp. | i i 2 sec Figure 5-13. A 10 s e c o n d segment taken from a 2 minute wing flapping s e q u e n c e showing a 4:1 relationship between wingbeat frequency and respiratory frequency in a decerebrate g o o s e . 205 Int. A i r Flow Insp. t Exp. i L. Pect. EMG. Time (sec) Stim. t 206 Figure 5-14. Re c o r d i n g of left external intercostal E N G (inspiratory activity; "fictive" respiration) a n d left pectoralis E N G at rest (top) a n d during electrically induced "fictive" wing flapping (bottom) in a decerebrate, paralyzed duck showing precise 1:1 synchronization between the two patterns during "fictive" flight. Note the different time s c a l e s in the two tr a c e s . 207 L E x t e r n a l i n t . E N G L . P e c t . E N G ro o CO Figure 5-15. R e c o r d i n g of left cranial nerve IX (inspiratory activity; "fictive" respiration) a n d left pectoralis E N G at rest (top) and during electrically induced "fictive" wing flapping (bottom) in a decerebrate, paralyzed duck showing precise 1:1 synchronization between the two patterns during "fictive" flight. Note the different time s c a l e s in the two tr a c e s . 209 Cranial Nerve IX ENG Figure 5-16. R e c o r d i n g of left internal intercostal E N G (expiratory activity; "fictive" respiration) a n d left pectoralis E N G at rest (top) a n d during electrically induced "fictive" wing flapping (bottom) in a decerebrate, p a r a l y z e d duck showing precise' 1:1 synchr o n i z a t i o n between the two patterns during "fictive" flight. Note the different time s c a l e s in the two t r a c e s . 211 L. Internal int. ENG L. Pect. ENG L. Internal int. ENG L. Pect. ENG stim. ro ro Figure 5-17. R e c o r d i n g of left external intercostal E N G (inspiratory activity; "fictive" respiration) and left pectoralis E N G during electrically induced "fictive" wing flapping in a decerebrate, p a r a l y z e d goose showing precise 1:1 synchronization between the two patterns. 213 stim. on t stim. off 5 sec ro Figure 5-18. R e c o r d i n g of whole left intercostal E N G and left pectoralis E N G at rest (top) a n d during a s p o n t a n e o u s bout of "fictive" wing flapping in a decerebrate, p a r a l y z e d duck. T h e two patterns were coordinated at 2 wingbeats per breath. S i n c e "fictive" respiration was recorded using the whole intercostal nerve, e a c h breath is represented by two E N G bursts, a small amplitude burst followed by a large amplitude burst. This c a n be clearly s e e n when the time s c a l e was expanded (bottom t r a c e , right hand si d e ) . Note the different time s c a l e between top a n d bottom an d a l s o the time s c a l e change part way through the bottom trace. 215 Whole L int. ENG Figure 5-19. Effects of increasing inspired C 0 2 from 4 % (top) to 7 % (bottom) on the central c o u p l i n g of "fictive" respiration (indicated by whole left intercostal E N G ) and wing flapping (pectoralis E N G ) in a paralyzed, decerebrate goose. S i n c e "fictive" respiration w a s recorded using the whole intercostal nerve, e a c h breath is represented by two E N G bursts that closely follow one another. This can best be s e e n in the top trace prior to the onset of stimulation. 217 UDV: 4% C02/AIR Whole int. ENG i intercostal nerve contains motor fibres for both inspiratory and expiratory units, a single breath is often represented by a double burst, a smaller burst followed by a larger one. The bottom 2 traces show a 2:1 coordination between intercostal nerve dis c h a r g e (f¥=71.4) and pectoralis nerve discharge (f w=143). 219 DISCUSSION FREE-FLIGHT VS ELECTRICALLY INDUCED WING FLAPPING Although the decerebrate birds used to examine the relationship between wingbeat a n d respiratory rhythm in these experiments have undergone extensive surgery, the neural networks responsible for the generation of locomotor a n d respiratory patterns are essentially intact. Locomotion in decerebrate animals, whether produced spontaneously or by electrical/chemical stimulation, has been shown to be similar to locomotion in intact animals for a large number of s p e c i e s (Grillner, 1975; M c C l e l l a n , 1986), including g e e s e (Steeves, et al., 1987) and c h i c k s ( J a c o b s o n & Hollyday, 1982) during wing flapping. Similarly, the respiratory pattern produced by decerebrate animals is very similar to that s e e n in intact animals (Feldman et al., 1988), including g e e s e (chapter 1). Furthermore, the respiratory r e s p o n s e s of decerebrate g e e se to low level e xercise (treadmill walking) closely resemble the r e s p o n s e s of intact g e e se. A s the physiological r esponses to hindlimb v e r s u s forelimb v exercise c a n differ ( Asmussen & N i e l s e n , 1946), however, it is e s s e n t i a l to compare the responses of decerebrate g e e se during electrically induced wing flapping with the responses of intact animals during free-flight. A s d i s c u s s e d in chapter 3, very little information is available on the physiological r e s p o n s e s of birds to free-flight. From what is known about the C a n a d a goose and other birds, the most noticeable difference between free-flight responses and electrically induced responses in decerebrate birds, is one of magnitude. C h a n g e s in the c a r d i o v a s c u l a r and respiratory variables examined in this study were 220 TABLE 5-3 MAGNITUDE OF THE INCREASE IN SEVERAL VARIABLES DURING FREE-FLIGHT AND ELECTRICALLY INDUCED "FLIGHT" Ratio: Free-flight Electrical Stimulation Free/electrical V 0 2 10X 2.5X ? 1/4 ? V E 12.4X 2.5X 1/4 f v 5-6X 5-6X 1 V T 2.4X 1/2 1/4 f w 256 144 1/2 force/stroke ? ? 1/2 ? H R 2-5.8X 1.5 1/4 c o n s i d e r a b l y s m a l l e r during electrically induced flight than during free-flight (table 5-3). T h e magnitude of the r e s p o n s e s of decerebrate birds to electrically induced wing flapping were approximately one quarter of the average recorded during free-flight. Of the few s p e c i e s e xamined during free-flight, V E i n c r e a s e s 12.3 fold from rest to flight on average, in conjunction with a 10.7 fold increase in metabolic rate (see table 5-4a & b for details). The 2.5 fold i n crease in V E in decerebrate birds is approximately 1/4 of that s e e n during free-flight. Respiratory frequencies of free-flying barnacle (99 + 2; Butler & Woakes) and C a n a d a geese (84 + 1; Chapter 3), however, are similar to those obtained here for decerebrate g e e s e (103 + 11). This indicates, a s shown in table 5-3, that the differences in V E between electrically i n duced wing flapping and free-flight are due to differences in V T, where the i n c r e a s e in V T in decerebrate birds is approximately 1/4 that of intact birds during free-flight. T h i s also suggests, s i n c e f v appears preset to free-flight levels in the decerebrate birds, that V T is adjusted to match metabolic rate. A s suming that V E i n c r e a s e s similarly in g e e s e a s in s p e c i e s for which free-flight v alues of V E are a v a i l a b l e , V T must not d e c r e a s e , but remain constant or possibly increase along with f v during free-flight in the C a n a d a goose to achieve the average 12.4 fold i n c r e a s e in V E. From table 5-3, it is also c l e a r that f w is reduced to approximately 1/2 the level s e e n during free-flight in g e ese. A s s u m i n g that metabolic rate will be a function of both f w a n d the force of muscular contraction, a reduction in metabolic rate between free-flight and electrically induced wing flapping is expected. In turn, if the relationship between V E and V 0 2 s e e n during walking in decerebrate geese (chapter 222 1) also holds for wing flapping, then the 2.5 fold i n crease in V E would correspond to a similar i n c r e a s e in V 0 2, which again, is approximately 1/4 that s e e n during free-flight. If this were the c a s e , the difference in the magnitude of the responses of intact a n d decerebrate birds would not be due to a d e c r e a s e in the r e s p o n s i v e n e s s of the decerebrate birds, but would arise b ecause the decerebrate animals were performing at a lower metabolic rate. This conjecture is supported by the observation that the i n c r e a s e in HR was also diminished in the decerebrate birds. HR, which typically i n c r e a s e s between 2 and 5.8 fold during free-flight (Berger et al., 1970b; 12 s p e c i e s ; Butler et a]., 1977), only increased 1.5 fold in the decerebrate birds, again approximately 1/4 the increase s e e n during free-flight. Furthermore, the kinetics of the ventilatory response to wing flapping e xercise in decerebrate birds (fig. 5-8) are similar to those of intact birds (Berger et al., 1970b; black duck; Butler et al., 1977; pigeon). Both show a rapid increase in V E with the onset of flapping, followed by a plataeu which rapidly returns toward resting upon termination of flight. It was p o s s i b l e to increase the force of the wing flapping to levels c l o s e r to those s e e n during -free-flight by simply increasing the stimulus intensity used to evoke locomotion. The chief reason for not doing so in these experiments was that this also c a u s e d a dramatic increase in metabolic heat production. Non-evaporative, convective heat l o s s , which is responsible for between 65 and 1 0 0 % of the cooling in birds during flight (Baudinette et al., 1976; Bernstein, 1976: Torre-Bueno, 1978b; Hudson & Bernstein, 1981; Bi e s e l & Natchigall, 1987), does not occur in the lab. Due to the large mass of the g e e se and their a s s o c i a t e d thermal inertia, laboratory methods of controlling T B are rarely sufficient to match the large heat loads placed 223 TABLE 5-4a RESPIRATORY RESPONSES OF SEVERAL AVIAN SPECIES TO FLIGHT SPECIES (WT) NOTES R/F VE (ml/hr/g) (ml/g) fv 1 (min1) vD (ml/g) (ml/hr/g) TA (°C) Budgerigar (30-40g) 1 R F 80.5 398 0.033 199 --20 20 Pigeons (382g) (442g) 2 3 4 R F F R F 21.0 431 287 0.0137 0.0147 0.0098 26 487 487 19.7 411 0.005* n n n fi 13.5 283 132 25 25 Starling (78g) 6 R F 47.2 388 0.0086 0.0359 92 180 --Laughing Gull (300g) 8 F - - 150 0.0051 # - -Fish Crow (280g) 10 R F 47.8 384 0.029 0.053 27.2 120 0.0048# N 39.5 347 20 20 White-necked Raven (480g) 11 R F 37.5 223 - -0.0048# ft - 22 22 Grosbeak (59.3g) 13 R F 80.9 1022 0.0143 0.058 95 294 0.0043 0.0043 58.7 953 -Ring-Bill Gull (427g) 13 F .462 0.061 122 0.0051# 409 -Black Duck (1026g) 13 R F 36.6 504 0.023 0.055 27 158 0.005 n 29.2 474 -Barnacle Goose (1600g) 14 R F --8.5 99 - - -Canada Goose (3800g) 15 R F 25.0 0.028 14.8 84.1 0.011 15.1 -R = rest F = flight V^; = a i r capillary ventilation (similar to alveolar ventilation in mammals) * from Hinds & Calder (1971) # from Hinds & Calder (1971), values for particular species not available so values taken from bird of same genus of similar body weight 224 TABLE 5-4b MEATBOLIC AND CARDIOVASCULAR RESPONSES OF SEVERAL AVIAN SPECIES TO FLIGHT SPECIES (WT) NOTES R/F V™ Vc02 HR BP TA (ml/hr/g) (ml/hr/g) (min1) (mmHg) (°C Budgerigar 1 R 4.5 _ _ 20 (30-40g) F 21.9 - - - 20 Pigeons 2 R 0.89* 170 (382g) F - - 550 - -3 F 11.9 - ? - -(442g) 4 R 1.22 1.032 115 142 25 F 12 11.04 670 147 25 (325g) 5 F 17.2 - - - -Starling 6 R - 0.77 (78g) F - 14.6 - - -7 F - 15.2 - - -Laughing Gull 8 R-F 6-8X _ _ (300g) Fish Crow 9 R 2.32 - - -(275g) F 14.8 - - - -10 R 1.82 - - - 20 (280g) F 15.6 - - 20 White-necked 12 R 2.12 - • - 22 Raven (480g) F 11.41 - - - 22 Grosbeak 13 R 2.41 - 360 - -(59.3g) F 34.1 - 822 - -Ring-Bill Gull 13 F 9.20 - 612 - -(427g) Black Duck 13 R 1.09 - 144 - -(1026g) F 13.9 - 543 Barnacle Goose 14 R - - 72 - -(1600g) F - - 512 R = rest F= flight * Lasiewski & Dawson, 1967 225 Notes to table 5-4: 1: Tucker, 1968: open circuit wind tunnel with tethered mask. Flight durations were > 20 min. 2: Hart & Roy, 1966: "backpack" telemetry during tethered flight. Flight durations were 3-14 s e c . 3: L e F e b v r e , 1964. 4: Butler et al., 1977: wind tunnel with tethered mask and telemetry. Flight durations were > 10 min. A l s o d i d "implanted" telemetry during free flight. 5: Rothe et al., 1987: wind tunnel with tethered mask. Flight durations were > 60 min. 6: Torre-Bueno, 1978: s e a l e d wind tunnel. C a n n u l a e were attached for air s a c s a m p l e s . Flight durations were > 45 min. 7: Torre-Bueno & Larochelle, 1978: s e a l e d wind tunnel. Flight durations were > 90 min. 8: Tucker, 1972: open circuit wind tunnel with tethered mask. Flight durations were > 30 min. 9: Bernstein et al., 1973: open circuit wind tunnel with tethered mask. Flight durations were 15-20 min. 10: B e r n s t e i n , 1976: as in Bernstein et al., 1973. 11: H u d s o n & Bernstein, 1981: as in Bernstein et a]., 1973. Flight durations were 20 -30 min. 12: H u d s o n & Bernstein, 1983: a s in Hudson and Bernstein, 1981. 13: Berger et al., 1970b: A s in Hart & Roy, 1966. Flight durations were between 7 and 15 s e c . 14: Butler & W o a k e s , 1980: "implanted" telemetry during unrestrained flight using birds trained to follow a truck. Flight durations were 15 min on average. 15 T h i s t h e s i s , "backpack" telemetry with mask as well a s uninstrumented video a n a l y s i s of breathing and wing beat during unrestrained flight in birds trained to follow a motorcycle. Flight durations ranged from 0.5 to 5 min. 226 on the a n i m a l s at high work rates. This is also the reason that most stimulation periods were kept below 30 s e c o n d s . While the c h a n g e s in V E during free-flight are fairly consistent between s p e c i e s , the c h a n g e s in f v and V T employed to achieve the increase in V E vary considerably. Increases in V E during free-flight a ppear to be a c h i e v e d primarily through increases in f v s i n c e V T i n c r e a s e s 2.4 fold on average an d f v i n c r e a s e s 11.4 fold (table 5-4). However, s o m e s p e c i e s actually d e c r e a s e V T and produce the entire increase in V E through i n c r e a s e s in f v, while others increase V T more than f v. The spec i f i c pattern r e sponse may d epend to a large degree on the d e a d s p a c e volume (V D) of the s p e c i e s . T h o s e birds with high anatomical d e a d s p a c e may be limited in their ability to i n crease f v without compromising V A C (air capillary ventilation). However, from table 5-4a, there is no apparent relationship between V D and the birds' breathing pattern r esponse to flight. For example, V D is approximatley 0.005 ml_/g for both the pigeon an d the grosbeak, but the pigeon i n c r e a s e s V E entirely through a 20 fold i n c r e a s e in f v, while a 3 fold increase in f v and 4 fold increase in V T combine to produce the V E increase in the grosbeak. Like the pigeon, decerebrate g e e s e inc r e a s e V E entirely through increased f v. V T actually d e c r e a s e s about 50 % in decerebrate g e e se. However, since f v is the s ame in decerebrate birds during wing flapping a n d intact birds during free-flight, V T must also increase during free-flight to a c h i e v e the average 12.4 fold increase in V E a s s o c i a t e d with free-flight. That f„ is the s a m e in decerebrate birds and during free-flight, and that f w is reduced in the decerebrate birds implies that the relationship between f w and f v in the two groups is also different. During free-flight in the C a n a d a goose, ratios of 2 and 3 wingbeats per breath were observed. Similarly, the barnacle 'goose 227 demonstrates 2, 3 and 4 wingbeats per breath (Butler et al., 1977). All relationships o b s e r v e d during free-flight have been observed during electrically induced wing flapping. O n e to one synchronization, not recorded during free-flight, has also been o b s e r v e d . T h i s unique relationship may be a function of the differential effects of electrical stimulation on the relationships between f v - V T and f w and force of wingstroke. During stimulation, f v i n c r e a s e s to free-flight l e vels, thus V T appears to be adjusted to match metabolic rate. In contrast, the magnitude of the increase in f w is r e d u c e d during electrically i nduced wing flapping. The force of wing stroke is a l s o likely to be reduced. There are s e v e r a l other possible explanations for the appearance of this novel coordination s c h e m e s e e n during wing flapping in decerebrate birds. It is possible that the drastic alteration in breathing pattern, s e e n as a large increase in f v and d e c r e a s e in V T, that results in a 1:1 coupling, represents a movement artifact. The birds have all been tracheostomized. Thus pressure c h a n g e s within the thorax a s s o c i a t e d with contraction and relaxation of the pectoralis muscles could produce artifactual movements of air that do not represent actual respiration. However, a number of p i e c e s of e v i d e n c e suggest that this is not the c a s e . External intercostal m u s c l e s are active during, and supply part of the power for, inspiration at rest (fig. 5-10; F e d d e et al., 1 9 6 4a & b). Therefore the activity of t h e s e muscles should give an indication of central respiratory activity. From fig. 5-10 it can be s e e n that the intercostal E M G discharge remains in phase with the inspiratory air flow and completely out of phase with the pectoralis E M G during wing flapping. Thus it a p p e a r s that the c h a n g e s in air flow are respiratory in origin. It is also possible that the external intercostal muscles may be recruited for some locomotor function 228 during wing flapping, in which c a s e intercostal activity may not represent respiratory output. F i g . 5-6 s h o w s the development of the response of a bird to threshold stimulation. Note that the change in ventilatory pattern occurs long before there is any activity in the wings. Th e constant phase relationship between air flow a n d intercostal E M G at rest and during wing flapping (fig. 5-10), and the dissociation of breathing pattern c h a n g e s from wing motion (fig. 5-6) strongly s u g g e s t s that the 1:1 coordination of wingbeat and respiration during wing flapping in decerebrate birds is not an artifact. It is a l s o p o s s i b l e that the 1:1 pattern is typical only of take-off patterns and if the duration of the stimulation periods had been extended beyond 30 s e c o n d s , another type of coordination would have become evident. Most stimulation periods were kept shorter than 30 s e c o n d s to prevent a rise in body temperature, as mentioned earlier. T h e two flights that were extended for longer periods to describe the kinetics of the ventilatory reponse to wing flapping (fig. 5-8), did show 2:1 and 4:1 coordination. However, this was probably coincidence as there was no evidence in these animals of a transition from a 1:1 ratio to higher coupling ratios a s the wing flapping progressed. Unfortunately, the relationship between f w and f v during take off in free-flight (chapter 3) has not been described. In the free-flight experiments in this study, I had to be a minimum of 50 m from the birds before they would lift off; a range that greatly e x c e e d e d the range of the transmitter. Although the transmitter used in the study by Butler & W o a k e s (1980) had sufficient range for them to make these measurements, they never analyzed the relationship between f w and fv during the transition to flight. It is therefore difficult to a s s e s s the likelihood of the above 229 possibility. The observation that L / i is flexible during free-flight s h ows that it is responsive to modification. Thus, it is also possible that some combination of factors unique to electrically i nduced wing flapping interact with central motor programs to produce the novel 1:1 s y n c h r o n i z a t i o n . The 1.46 fold i n c r e a s e in H R in the decerebrate birds during wing flapping was as s o c i a t e d with an almost identical 1.5 fold i ncrease in B P from 110 to 164 mmHg. Most a n i m a l s (humans, Wolthius et al., 1977; dogs, Kirlin et al., 1987; ducks, Kiley & Fed d e , 1983; Kiley et al., 1979; B ech and Nomoto, 1982) demonstrate an inc r e a s e in B P that is proportional to work rate during e x e r c i s e . When the level of activity in the decerebrate birds is i n c r e a s e d by increasing the stimulus intensity, proportional i n c r e a s e s in B P are observed. B P also i n c r e a s e s in proportion with locomotor drive during locomotion in decerebrate cats (Eldridge et al., 1985; Di M a r c o et a l . , 1983; Millhorn et a]., 1987). Although the slope of this relationship, i.e. the inc r e a s e in B P for a given increase in work rate, appears to vary c o n s i d e r a b l y between s p e c i e s and often within s p e c i e s between different studies, i n c r e a s e s ' i n B P a s large a s those s e e n during wing flapping in decerebrate g e e s e have b e e n recorded in intact ducks running on a treadmill at 2.6 times resting V 0 2 (Bech & Nomoto, 1982). A s s u m i n g that the 2.5 fold increase in V E in the decerebrate g e e s e in this study corresponds roughly to a similar increase in V 0 2 (chapter 1), the inc r e a s e s in B P relative to V 0 2 are similar in both decerebrate g e e s e a n d intact ducks. Blood pressure has only been measured during free-flight in pigeons (Butler et al., 1977). While HR increased from 115 to 670 beats per minute in the pigeons, blood pressure only increased from 142 to 147 mmHg. The 230 a b s e n c e of an i n c r e a s e in B P from rest to flight in pigeons (Butler et aJ., 1977) s u g g e s t s that B P may have been elevated at rest. Butler et a l . (1977) made every effort to reduce s t r e s s in the pigeons. However, the resting B P for the pigeons was m e a s u r e d as they sat on a perch prior to take off a n d was slightly elevated relative to previously recorded "resting" v a l u e s (131 + 9, Butler, 1970; 127 + 6 mmHg, Butler & Taylor, 1974). T h u s there appear to be considerable differences in the arousal state of preflight pigeons a n d decerebrate birds. Although the magnitude of the pressor response during wing flapping is within the range e x p e c t e d for intact animals, this does not necessarily imply that the m e c h a n i s m by which B P i n c r e a s e s in the two s ystems is the same. Stroke volume and peripheral resistance were not measured in this study. However, the similarity of blood flow redistribution in c o n c i o u s and electrically stimulated decerebrate cats (Waldrop et a]., 1986c) s u g g e s t s that the mechanisms leading to the increase in B P are the s a m e under both conditions. Thus, the r e s p o n s e s of decerebrate birds during electrically induced wing flapping a ppear qualitatively similar to those anticipated to occur in intact g e e s e during free-flight. The c a r d i o v a s c u l a r adjustments observed during electrically i n d u c e d wing flapping approximate those s e e n during exe r c i s e . The kinetics of the ventilatory r esponse a n d the breathing pattern responses of g e e se to electrically i n duced flapping are also similar to those of free-flight. In addition, wingbeat and respiratory patterns are tightly entrained in both conditions. The larger responses s e e n during free-flight appear attributable to the greater metabolic rate a c h i eved under free-flight conditions. 231 C O M P A R I S O N O F A C T U A L W I NG F L A P P I N G T O " F I C T I V E " F L I G H T T h e ph y s i o l o g i c a l r e s p o n s e s of birds to "fictive" flight are remarkably similar to thos e of actual wing flapping. In spite of the complete lack of muscular activity, HR and B P in c r e a s e in a manner similar to that s e e n during actual wing flapping. L ikewise, the ca r d i o v a s c u l a r responses of paralyzed and active cats to "fictive' v e r s u s actual locomotion are very similar (Millhorn et al., 1987). Respiratory frequency i n c r e a s e s substantially during "fictive" flight, although this i ncrease is somewhat l e s s than during wing flapping (7.5 v s 4.8 times). Estimates of V T and, hence, overall levels of ventilatory effort (V E) co u l d not be made for the paralyzed birds during "fictive" flight. However, the integrated phrenic nerve activity in mammals, which is an index of V E, appears to increase similarly during "fictive" and actual locomotion in decerebrate cats (Eldridge et a]., 1985). Similar to f v, f w did not incr e a s e to the s a m e level during "fictive" flight (86 + 21 beats/min) a s during actual wing flapping (144 + 19 beats/min). The r e a s o n s for the reduced locomotor frequency during electrical stimulation in "fictive" relative to non-paralyzed preparations are not completely understood, but have b e e n consistently observed in a variety of s p e c i e s (stingray; Williams et a]., 1984; Lamprey; Wallen & Williams, 1984; chick; J a c o b s o n & Hollyday, 1982; duck and goose; Sholomenko et al., 1990c; cat; Jordan et al., 1979). It appears that p h a s i c afferent input provides an excitatory drive to the locomotor pattern generators. R e m o v a l of this drive through paralysis (Williams et a]., 1984; Wallen & Wil l i a m s , 1984; J a c o b s o n & Hollyday, 1982; Sholomenko et al., 1990c; cat; Jordan et al., 1979) results in a dec r e a s e in cycle frequency. This d e c r e a s e in excitability 232 of the locomotor s y s t e m is also evident in that the stimulation intensity required to activate locomotion in the paralyzed birds (340 ± 46 uA) was much greater than that required to activate wing flapping prior to paralysis (138 ± 23 uA). Similar i n c r e a s e s in the stimulation intensity required to activate locomotion following p a r a l y s i s have been o b s e r v e d in stingray (Williams et a]., 1984), lamprey (Wallen & W i l l i a m s , 1984), chic k ( J a c o b s o n & Hollyday, 1982), duck and goose (Sholomenko et al., 1990c) and cat (Jordan et al., 1979). O n c e locomotion is activated in the "fictive" preparation, however, the "fictive" preparation behaves very much like the u n p a r a l y z e d animal. Th e E M G and E N G patterns of the two preparation are virtually identical ( J a c o b s o n & Hollyday, 1982; Sholomenko et al., 1990c; Jordan et aJ., 1979). In addition, a s s e e n in non-paralyzed decerebrate preparations, c h a n g e s in respiration (intercostal nerve activity) precede the development of a "fictive" locomotor response at threshold stimulation intensities (this chapter; Millhorn et a]., 1987; Eldridge et aL, 1985). A s s e e n during wing flapping, wingbeat and respiration were always coupled during "fictive" flight. The way in which the two s y stems were coupled, however, changed.^ During wing flapping, 1:1 coupling was most prevalent, but other ratios, 2:1, 3:1 a n d 4:1, were frequently observed. During "fictive" flight, every animal s h o w e d 1:1 coordination, and only on two o c c a s i o n s were other coordination s c h e m e s observed. In addition to 1:1, two birds (1 duck and 1 goose) also showed periods of 2:1 coupling. S i n c e the stimulation sites used to activate wing flapping were not always the s a m e before an d after paralysis, it is possible that the difference in coordination s c h e m e is related to stimulation site. However, there did not appear to be any 233 relationship between stimulation site and the type of coordination o b served (table 5-2). The difference may therefore depend on the presence and type of afferent f eedback. MECHANISMS OF ENTRAINMENT T h e r e are two major control mechanisms involved in the gross c a r d i o v a s c u l a r a n d respiratory r e s p o n s e s to exercise; feedback from humoral a s well as neurogenic drives occuring after the start of exercise, and feedforward due to central coactivation of c a r d i o v a s c u l a r and respiratory a r e a s in the brainstem by the pathways involved in the initiation of locomotion. There is c l e a r evidence that feedforward m echanisms from locomotor centres are involved in the production of a proportionate drive to ventilation during exercise (Eldridge et a]., 1985). This feedforward drive to the respiratory and cardiovascular systems c a n also be s e e n in fig. 5-7 where B P , HR and respiratory frequency all increase during "fictive" locomotor activity. In addition, this chapter provides evidence that feedforward activity from the locomotor centres not only provides a proportionate drive to ventilation, but is a l s o strongly involved in establishing the pattern of respiratory output. All nine animals examined during "fictive" flight s howed a 1:1 synchronization (2 birds also s h o w e d periods of 2:1 coupling) between wing nerve discharge and intercostal nerve discharge. The external intercostal nerve (inspiratory, fig. 5-2) d i s c h a r g e d out of phase with the pectoralis nerve (fig. 5-14 & 5-17) and the internal intercostal nerve (expiratory, f i g . 5-3) discharged in phase with the pectoralis nerve 234 (fig. 5-16). T o ensure that the nerve activity of the intercostal nerves (internal a n d external) w a s truly representative of central respiratory activity, s i n c e these muscles may have been recruited for some locomotor function during wing flapping,' the relationship between Pect E N G and cranial nerve IX E N G was also examined. C r a n i a l nerve IX innervates the stylopharyngeus and glossopharyngeus muscles of the upper airway (Iscoe, 1988). The stylopharyngeus is an airway dilator, active during inspiration, a n d the glossopharyngeus is an airway constrictor, sometimes active during expiration in mammals. Thus, recruitment of thes e m u s c l e s for a locomotor function s e e m e d very unlikely. Inspiratory activity in cranial nerve IX, as determined by comparison with intercostal E N G activity at rest (fig. 5-4), was found to coordinate with P e c t E N G during "fictive" flight in exactly the s a m e manner as the external intercostal E N G (fig. 5-15) in all three animals from which recordings were made. Therefore, the neural networks responsible for the production of locomotor (wingbeat) a n d respiratory rhythms in birds interact centrally to produce a sy n c h r o n i z e d output in the a b s e n c e of phasic afferent feedback. There is evidence in the literature that feedforward mechanisms may also play a role in est a b l i s h i n g the synchronization of locomotor a n d respiratory rhythms in cats a n d rabbits. A s briefly outlined in the introduction to this chapter, however, there are se v e r a l problems with these studies. The animals in both studies were artificially ventilated by a respiratory pump following paralysis. Vagally mediated afferent f e edback has been shown to entrain respiration to the pattern of mechanical ventilation in cats (Petrillo et al., 1983). Appropriately, the animals examined by K a w a h a r a et a l . (1989a), V i a l a et a[. (1987b) and Pe r s e g o l et a l . (1988), were vagotomized to remove any vagally mediated respiratory entrainment. However, 235 phrenic a s well a s intercostal reflexes remained intact which are known to significantly alter (Speck, 1988; Shannon et al., 1988; Remmers & Marttila, 1975), a n d e v e n entrain respiratory rhythm (Remmers & Marttila, 1975). Thus, K a w a h a r a et a l . (1989a) only obtained coupling of locomotor and respiratory outputs at extremeley low P a ^ ' s , when the frequency of the pump u s e d for ventilation w as very high, i.e. similar to the elicited locomotor frequency. Entrainment d e c r e a s e d dramatically at higher P a ^ ' s when the pump frequency w as d e c r e a s e d . It is therefore possible that the respiratory nerve discharge in this c a s e was entrained by feed b a c k a s s o c i a t e d with the pump ventilator, rather than by a central interaction between locomotor a n d respiratory rhythm generating networks. The influence of pump related afferent feedback on the results of V i a l a et a l . (1979; 1987b), V i a l a (1986) a n d Pe r s e g o l et a l . (1988) d o e s not appear to be a co n c e r n b e c a u s e , in the studies in which traces of pump frequency are included (V i a l a et al., 1979; 1987b), phrenic nerve discharge is not entrained to the pump. A major w e a k n e s s of the these studies (Viala et al., 1987b; Per s e g o l et al., 1988) is that they examine the relationship between "spinal" respiration and locomotion. 'The animals were transected at the C 2 level of the spinal cord, then treated with D O P A to induce locomotor and respiratory movements. That locomotor movements c a n be generated in the isolated spinal cord is well established. However, the neural networks responsible for respiratory rhythmogenesis are predominantly located in the medulla and caudal pons. Respiratory rhythm, which o c c u r s spontaneously even in the isolated brainstem (Smith & Feldman, 1987), can not be obs e r v e d in the cord prior to pharmacological treatment. In addition, when neural activity is first pharmacologically activated in the phrenic nerves, right and left 236 phrenic nerves di s c h a r g e out of phase. Over time, their d i s c h a r g e s become sy n c h r o n o u s ( V i a l a et al., 1987a). Furthermore, the development of phrenic activity in t h e s e preparations follows, and perhaps is dependent on, the development of locomotor activity ( V i a l a et a]., 1987b). In preparations in which the pons and me d u l l a remain intact, the c h a n g e s in respiration precede, or o c c u r at a lower stimulation intensity than the locomotor c h a n g e s (fig. 5-6, Millhorn et al., 1987; Eldridge et al., 1985). The reason for this appears to be that the neural networks controlling the respiratory s y s t e m are well a bove threshold at rest. C onversely, the locomotor networks are subthreshold. A small activation of a common driving m e c h a n i s m would have a noticeable effect on the systems above threshold before a demonstrable effect w a s o b s e r v e d on the s y s t e m below threshold. O n c e the locomotor s y s t e m is above threshold, however, inc r e a s e s in activation, i.e. electrical stimulation, result in a proportional increase in the activity of all s ystems (Millhorn et al., 1987; Eldridge et al., 1985). R e v e r s a l of this s e q u e n c e in spinal rabbits (Viala et a]., 1987b), suggests either the removal of an excitatory input to a spinal "respiratory rhythm generator" following s p i n a l transection, or disconnection of the main pattern generator itself from the s p i n a l , phrenic motor neuron pools. In either c a s e , the respiratory related excitatory input to the phrenic motor neurons is greatly d e c r e a s e d in the spinal preparation and the influence of locomotor inputs is enhanced. While it is clear that sp i n a l locomotor networks c a n influence the discharge of phrenic motor neurons and therefore may be involved in the production of locomotor-respiratory coordation, the importance of this influence during actual locomotion, when phrenic motor neurons receive respiratory input, remains unclear. 237 V i a l a et a l . (1987b) and P e r s e g o l et a l . (1988) have also examined the relationship between respiratory nerve discharge (phrenic) and locomotor pattern in decorticate a n d decerebrate rabbits. T h e y found that the locomotor and respiratory patterns were o c c a s i o n a l l y coordinated in these preparations. However, V i a l a et a l . (1987b) only found coordination in 2 0 % of the animals examined. Similarly, P e r s e g o l et a l . (1988) found that sychronization only occurred if the locomotor and respiratory periods were very similar. If the locomotor pattern became too disimilar from the respiratory pattern, the coordination c e a s e d . The major w e a k n e s s of these studies w a s that bilateral recordings of leg and phrenic nerve d i s c h a r g e s were typically absent. During normal hopping, rabbits breathe once per leap. Unilateral recordings do not distinguish between walking (i.e. alternating hindlimb nerve activity) or hopping (i.e. s ynchronous hindlimb nerve activity). It is therefore possible that transitions from one gait to another occurred during their recordings, as gait c h a n g e s are commonly observed during pharmacological activation of locomotion (ducks & g e ese; Sholomenko et al., 1990a & b; cats; Garcia-Rill et al., 1985). C l e a r l y , it is essential to properly describe the locomotor pattern before the r e l a t i o n s h i p , between locomotor pattern a n d respiratory rhythm c a n be examined. T h i s is e s p e c i a l l y important in rabbits s i n c e they show gait dependent entrainment (Bramble & Carrier, 1983). Entrainment is very low in these animals during walking, but virtually 1 0 0 % upon the transition to trot or gallop. Thus it is not clear whether the intermittent central entrainment between locomotion and respiration observed in rabbits is due "fictive" gait c hanges or weak central coupling. The interaction between the neural s y stems generating the flight and respiratory rhythms in C a n a d a geese appears to be much stronger than that coupling 238 respiration a n d locomotion in cats or rabbits. In the complete a b s e n c e of phasic afferent feedback, the neural outputs indicating "fictive" flight and "fictive" respiration were a l w a y s c o u p l e d in birds. C O N C L U S I O N Although the synchronization between wing beat and respiratory frequencies in pa r a l y z e d birds differs from that o bserved when peripheral feedback is present, it is c l e a r that the networks responsible for the generation of locomotor and respiratory rhythms interact on a central l e v e l , brainstem or spinal cord, to produce a sy n c h r o n i z e d output in birds. Thus, both feedforward and feedback (Chapter 4) m e c h a n i s m s appear to be involved in the production and control of entrainment. Under certain conditions, either mechanism appears sufficient to entrain locomotion and respiration. During free-flight, the relationship between wing beat and respiration is likely to be the result of an interaction between these apparently redundant mechanisms. 239 GENERAL DISCUSSION Locomotor-respiratory synchrony has been d e s c r i b e d in a large number of vertebrates. However, the mechanisms responsible for this entrainment are not well understood. From the viewpoint that entrainment functions either to d e c r e a s e the cost of respiration and/or locomotion, or to minimize the antagonistic actions of locomotor and respiratory musculature, it follows that those mechanisms involved in the control of locomotion a n d ventilation during exercise/locomotion, are also likely to influence the relationship between locomotor a n d respiratory rhythms. This thesis has c o m p a r e d the relationship between locomotor a n d respiratory patterns in freely moving g e e s e with those s e e n in a number of reduced preparations in an attempt to delineate the potential m echanisms involved in the production and control of locomotor-respiratory synchrony. A s s e e n for the control of ventilation during exercise (see reviews by Dejours, 1964; D e m p s e y et al., 1985; W a s s e r m a n et al., 1986), this reductionist approach has y i e l d e d e vidence indicating that humoral and neurogenic feedback a s well as feedforward m echanisms are potentially involved in the production and control of locomotor-respiratory synchrony. T h e s e studies have laid the foundation from which many important questions c a n be a d d r e s s e d . T h e y also indicate that a more interactive approach will be required to achieve a complete understanding of how the s y s t e m operates outside the laboratory. That humoral drives affect the synchronization of hindlimb and respiratory movements is e v i d e n c e d by the d e c r e a s e in entrainment observed during hypoxic exposure in g e e s e (Chapter 2) and humans (Paterson et al., 1987). Shifts in locomotor respiratory coupling ratios have also been a s s o c i a t e d with hypoxia in fish (Webb, 1975) and hypercapnia in horses (Dempsey, personal communcation). 241 P e r i p h e r a l , neurogenic feedback also appears to be involved in the entrainment of locomotion an d respiration. During walking/running in g e e s e , i n c r e a s e s in entrainment were a s s o c i a t e d with i n c r e a s e s in stride frequency (Chapter 2) rather t han metabolic rate (Chapter 1). Thus, it a p p e a r s likely that the i n c r e a s e in entrainment was mediated v i a alterations in proprioceptive feedback. The entrainment of respiration by p a s s i v e wing flapping (Chapter 4) clearly indicates the potential importance of s e n s o r y afferent (neurogenic) feedback in the coordination of locomotion an d respiration. Attempts to delineate the entraining signal activated by p a s s i v e wing flapping indicated that afferent activity from the brachial plexus was not n e c e s s a r y for the entrainment. It appears that afferent feedback from the chest wall and/or respiratory s y s t e m was, as a result of the intimate mechanical a s s o c i a t i o n between the thorax an d the wings (Jenkins et a l . , 1988), cyclically activated at the frequency of locomotion and thus able to entrain respiration to wing movements. T h e s e studies do not, however, indicate that afferent activity from the wing is not involved in entraining wing beat and respiration. They only indicate that it is not e s s e n t i a l . Repetitive stimulation of cutaneous and muscle limb afferent fibres partially entrains respiration in cats (Iscoe & P o l o s a , 1976; K a w a h a r a et al., 1988). P r e v i o u s studies have shown only intermittent central entrainment of locomotor and respiratory neural output in the paralyzed rabbit (Viala et a]., 1987b; Persegol et a l . , 1988) a n d cat (Kawahara et a l . , 1989a). Entrainment during "fictive" walking/running was not examined in g e e se. However, during "fictive" flight in the g o o s e (and duck), wing and respiratory motor nerve discharges were tightly s y n c h r o n i z e d in the a b s e n c e of phasic afferent input. This clearly indicates, for the 242 first time, that central interactions between the two motor systems are important, not only in establishing respiratory drive (Eldridge et al., 1985), but also in establishing respiratory pattern. The details of these three groups of mechanisms have not been fully worked out in this t h e s i s . F o r example, although it was s h o wn in chapter 4 that feedback a s s o c i a t e d with p a s s i v e wing flapping from the chest wall and/or lung would entrain respiration, the role of c h est wall v e r s u s lung afferent activity was not examined. V agotomy co u l d be u s e d to remove the input from lung mechanoreceptors, without affecting f e e d b a c k from the chest wall. T h i s procedure, however, l e a d s to s u c h a d r a s t i c alteration of breathing pattern in birds (Fedde & Burger, 1963) that interpretation of results from s u c h experiments is difficult. Alternatively, dorsal c o lumn se c t i o n rostral to the brachial plexus could be u s e d to interupt most of the c h e st wall related afferent activity allowing one to examine the effects of vagal f e e d b a c k during p a s s i v e wing flapping in isolation. L ikewise, it is not completely c l e a r from the results whether afferent activity from the brachial plexus is involved in the entrainment of f w and f v during passive wing flapping. A s limb muscle (Iscoe & P o l o s a , 1976) a n d cutaneous (Iscoe & P o l o s a , 1976; K a w a h a r a et al., 1988) afferent activity will entrain respiration to s o me degree in c a t s , entrainment needs to be re-examined in birds with chest wall afferent activity a n d v a g a l l y mediated mechanoreceptive information blocked. Alternatively, afferent fibres from the wing could be activated separately. It is a l s o not c l e a r at which level(s) of the central nervous system (CNS) the v a r i o u s afferent and efferent interactions are taking place. It has been suggested that entrainment may be entirely due to the c o n c ious (ie. cortical) control of 243 respiratory pattern in humans (Yonge & P e t e r s e n , 1983). However, entrainment still o c c u r s during bipedal walking/running (Chapter 1), actual a n d "fictive" wing flapping in d ecerebrate birds (Chapter 5), a s well as quadrupedal locomotion in decerebrate c a t s ( K a w a h a r a et al., 1989b). Thus, although concious control of respiration may be involved in the production of entrainment in humans a n d other vertebrates, interactions at the level of the brainstem and/or spinal cord a p p e a r sufficient to generate much of the locomotor-respiratory synchrony. Synchronization of spinal locomotor rhythm and "spinal respiration" (as indicated by phrenic nerve discharge) in pa r a l y z e d , C 2 sp i n a l i z e d rabbits (Viala et al., 1979; 1987b) implicates s p i n a l interactions in the development of entrainment. However, the influence of locomotor rhythm on respiratory motor nerve output is likley to be substantially reduced when the normal respiratory related input to respiratory motor neuron pools, originating within the brainstem, is not interupted v i a spin a l i z a t i o n . T h u s the importance of spinal interactions in the production of entrainment is unclear. The locomotor a n d respiratory s y s t e m s may also interact through C N S axon collateral projections at the level of the brainstem/midbrain. In the cat, the brainstem contains the major respiratory centres a s well a s the major nuclei involved with integration of respiratory related afferent information (see review by von Euler, 1986; Fel d m a n , 1986). It also contains several a r e a s which are involved in the initiation a n d control of locomotion ( M c C l e l l a n , 1986; S t e e v e s et al., 1987; Sholomenko et al., 1 9 9 0 a & b). Brainstem interactions are supported by the feedforward effects of the subthalamic a n d mese n c e p h a l i c locomotor regions on respiration (Eldridge et al., 1985), the development of partial entrainment in decorticate, paralyzed rabbits (Viala 244 et a]., 1987b; P e r s e g o l et a]., 1988), and tight synchrony during brainstem/midbrain stimulated "fictive" flight in birds (Chapter 5). However, feedforward interactions in the brainstem/midbrain have not been directly e x amined in isolation from the spinal c o r d s i n c e s p i n a l i z a t i o n (C1) d i s c o n n e c t s the s p i n a l nerve roots that are commonly u s e d to indicate locomotor pattern from the brainstem, making it difficult to follow locomotion. Examination of the relationship between the activity of respiratory neurons (or respiratory related cranial nerve activity) a n d reticulospinal neurons, which s h o w c y c l i c a l activity with locomotion (Garcia-Rill & Skinner, 1988), in high s p i n a l a nimals would be required to determine the presence of brainstem/midbrain interactions. O n a s y s t e m l e v e l , the respiratory s y s t e m could a c c e s s s o me part of the locomotor s y s t e m a n d thus influence locomotion. C o n v e r s e l y , the locomotor s y s t e m c o u l d directly a c c e s s the rhythm generator for respiration or c o u l d influence respiration through any of the respiratory rhythm generator's afferent or efferent pathways. T h e former possibility is supported by the finding that some reticulospinal neurons di s c h a r g e in phase with respiration a n d c a n be driven by electrical stimlation of cranial nerves as well a s tactile stimulation of the pharynx in the dogfish, S q u a l u s acanthias (Satchell, 1968). However, as s e e n during hypoxia in a teleost, C v m a t o a a s t e r aaareqata, where the coupling ratio shifts between 1:1 and 2:1 (Webb, 1975), and in g e e s e during free-flight where more than one coupling ratio is o b s e r v e d (Chapter 3), respiration usually shifts to match a constant locomotor rhythm, not the reverse. Thus it appears that respiration is entrained by locomotion. 245 S p e c i f i c interneurons have been identified a s mediating the central interaction b etween flight a n d respiratory rhythm generators in locusts (Ramirez & P e a r s o n , 1989). However, the sp e c i f i c neural substrates involved in coordinating the two motor s y s t e m s in vertebrates are unknown. Anatomical and central recording studies, s i m i l a r to those u s e d in describing many of the connections and interactions between the various respiratory centres in cats (see review by von Euler, 1986) would be helpful in determining how the two sy s t e m s interact on a neuronal l e v e l . Although the central organization of the locomotor s y s t e m in birds is virtually the s a m e a s mammals (Sholomenko & Steev e s , 1987; S t e e v e s et a]., 1987; M c C l e l l a n , 1986), very little is known about the central control of respiration in birds (see reviews by G l e e s o n & Molony, 1989; Davey & Seller, 1987). Studies identifying the avian C N S regions involved in generating respiration would have to be completed prior to the examination of the C N S locomotor-respiratory coordinating m e chanisms. C l e a r l y this reductionist approach, examining the effects of e a c h these m e c h a n i s m s in isolation of the others, i s of limited v a l u e in determining how the relationship between locomotion and respiration is modulated. The reductionist a p p r o a c h , however, is essential for defining the mechanisms potentially involved in producing entrainment. The demonstration that one mechanism, in isolation, is involved in the control of a particular process d o e s not preclude others from also playing a role. Outside a controlled laboratory environment, very few of these m e c h a n i s m s operate in isolation. Thus, an interactive approach must also be utilized to understand how the various mechanisms identified in isolation, are integrated to produce and adjust locomotor respiratory synchrony in freely moving animals. 246 Examination of t / f v during electrically induced wing flapping in decerebrate, non-p a r a l y z e d a n d p a r a l y z e d animals, indicated that, although entrainment is maintained in t h e s e preparations, the relationship between f w a n d f v during "fictive" flight (almost e x c l u s i v l e y 1:1) differs from that s e e n prior to paralysis (several patterns ranging from 1:1 to 4:1) which in turn differs from that in free-flight (predominantly 3:1). T h i s s u g g e s t s there is a central 1:1 synchronization that is s o m e h o w modulated to produce alternate patterns under various conditions. Similar modulation of locomotor-respiratory relationships a l s o o c c u r s in other free-ranging animals. In humans, where locomotor-respiratory entrainment is intermittent, variations in locomotor-respiratory relationships have been well documented (Bramble & Carrier, 1983; Bramble, 1983). The flexibility of the relationship between locomotor and respiratory patterns has e v e n been demonstrated in animals where the relationship between locomotion a n d respiration was, until recently, felt to be the result of inflexible m e c hanical interactions (horses; Dempsey, personal communication; birds; table 3-2). Peripheral modulation of this centrally derived synchronization is also supported by the finding that f w and f v were synchronized 1:1 during "fictive" flight in two s p e c i e s that s h o w different relationships between f w a n d f v during free-flight. f w i s c o u p l e d 2:1 with f v in ducks (Anas platvrhvnchos: Lord et al., 1962) and 3:1 in g e e s e (Branta c a n a d e n s i s ; Chapter 3; Branta leucopsis: Butler & W o akes, 1980). How the centrally derived 1:1 synchronization in birds is modulated is an intriguing question. Afferent feedback from a number of s o u r c e s , including the limbs, respiratory s y s t e m (chapter 4) and chemoreceptors could modify the central coupling ratio to ensure that the two s ystems remain entrained and still meet the metabolic requirements of exercise (flight). In the paralyzed preparation where 247 a n d V 0 2 do not change, chemoreceptor drive is maintained constant, and there is no proprioceptive feedback, the locomotor and respiratory s y s t e m s appear to produce outputs that are s y n c h r o n i z e d 1:1. During electrically induced wing flapping, metabolic rate i n c r e a s e s to varying d e g rees (2.5 times on average) which may in turn produce a range of chemoreceptor drives. Varying d e g r e e s of proprioceptive f e edback from the limbs, chest wall and lungs are also produced depending on the frequency a n d strength of the induced wing flapping. If the relationship between f w a n d f v is the result of an interaction of peripheral f eedback with a central program that c o u p l e s the two patterns at 1:1, the variation in peripheral f eedback may a c c o u n t for the range of ratios o b s e r v e d (1:1 to 4:1) during wing flapping in decerebrate birds. That intact g e e s e do not s h o w 1:1 synchronization during free-flight may simply be an extension of this interaction. In a s s e s s i n g the relative importance of chemoreceptive v e r s u s mechanoreceptive f e e d b a c k in this modulation, it is important to keep in mind the time course of the relationship between f w a n d f v during free-flight. Although no measurements of i j i v were made from take-off, three s e c o n d s after take off (earliest measurement made), i j i v w a s 3:1. T h e s p e e d of this response rules out a humoral (chemoreceptive) sig n a l in modifying the centrally derived 1:1 coupling to the e x p r e s s e d 3:1 relationship. It s u g g e s t s that, during free-flight, the particular combination of peripheral feedback, in conjunction with the central program, immediately produces 3:1 synch r o n i z a t i o n . S i n c e the wing beat pattern v a r i e s in response to environmental perturbations during free-flight, mechanoreceptive feedback is also likely to be very important in mediating the rapid respiratory adjustments that would be required to maintain synchronization between f w and f v under these conditions. It is important to 248 remember that the forebrain, although not essential for the production of locomotor-respiratory entrainment in laboratory situations, may be involved in the coordination of wing beat and respiration during free-flight. A number of mechanical factors are also likely to be important in determining the different coupling ratios e x p r e s s e d by different s p e c i e s . In order to meet the metabolic d e m a nds of flight, V E, either through i n c r e a s e s in V T, f v or both, must inc r e a s e . T h e type of coordination will d e pend on the factors limiting respiratory frequency and those determining wingbeat frequency. The relationship between wing a s p e c t ratio (length of wing/breadth) and body weight provides an index of how fast a bird must flap its wings to achieve sufficient lift for flight. High wing beat frequencies l e a d toward higher numbers of wingbeats per breath. Of course, the actual number of wing beats per breath will also d epend on how fast a bird c a n breathe without compromising alveolar ventilation. Airway resistance, the magnitude of the force (inspiratory a n d expiratory) the birds c a n generate to move air against the resistance of the airway, as well a s anatomical d ead s p a c e of the respiratory s y s t e m will set a ceiling on how fast respiratory movements c a n occur. For example, a g o o s e in free-flight performing work at levels 10 times resting metabolic rate may not be able to couple f w and f v at 1:1. The time available for inspiration c o u l d reduce V T to a level that was inadequeate to maintain V A C. Although the increase in f v would produce a substantial increase in V E, the amount of d e a d s p a c e ventilation would also increase, and the level of alveolar ventilation, that g a s available for gas exchange, may not meet the requirements of free-flight. A n extension of the respiratory period (i.e. d e c r e a s e in f v so that y f v increases from 1:1 to 2:1 or 3:1) would allow greater volumes of air to be inspired per breath and 249 alleviate the problem of in c r e a s e d d e a d s p a c e ventilation. If f „ / f ¥ continued to in c r e a s e , eventually a point would be r e a c h e d where the time allowed for inspiration would b ecome greater than the time required to inspire maximally. At this point V A C c o u l d only be i n c r e a s e d through i n c r e a s e s in f v and therefore d e c r e a s e s in i j i v . T h i s c a n be illustrated for other s p e c i e s from data in table 5-4a. In pigeons, where f w a n d f v are coordinated at 1:1, d e a d s p a c e ventilation c omprises 3 5 % of minute ventilation, both at rest and during flight. In contrast, dead s p a c e ventilation, which c o m p r i s e s 27 and 2 0 % of resting V E in grosbeaks and black ducks respectively, only c o m p r i s e s 6.8 a n d 6.0% of V E during free-flight when f w a n d f v are coordi n a t e d at 3:1 and 4:1 in the grosbeak and 2:1, 3:1 and 4:1 in black ducks. C h e m o r e c e p t o r drives to respiration a lso affect the relationship between locomotor a n d respiratory patterns. That respiration normally follows locomotor rhythm d o e s not imply that the activity of the respiratory rhythm generating networks is unimportant. Increases in inspired C 0 2 shift the relationship between locomotor frequency and respiratory frequency from 1:1 to 2:1 in galloping ponies (Dempsey, personal communication). There are also indications that locomotor-respiratory entrainment during "fictive" walking in cats d e c r e a s e s a s end tidal C 0 2 i n c r e a s e s ( K awahara et al., 1989a). Hypoxia has also been shown to c a u s e a d e c r e a s e in the entrainment of locomotor and respiratory frequencies during running/walking in g e e s e (chapter 2) a n d in humans (Paterson et al., 1987). Hypoxia also shifts breathing frequency in Cymatoaaster aggreqata s u c h that the relationship between breathing movements and pectoral fin beats changes from 1:1 to 2:1 (Webb, 1975). That locomotor frequency remains constant while respiratory pattern changes in response to chemoreceptor inputs, indicates that, while locomotor rhythm strongly 250 affects respiratory timing in s o me animals, respiration is still responsive to c h emoreceptive inputs. It also s u g g e s t s that altered non-exercise related respiratory drive affects entrainment, not by changing locomotor rhythm, but, by altering the activity level of the respiratory rhthym generator and thus affecting the way that locomotor input affects respiration. W h ether chemoreceptor feedback has an effect on entrainment during flight in birds is not c l e a r . It is possible that shifts in blood gas status account for the brief a p p e a r a n c e of 2:1 coordination between f w a n d f v in 2 of the 9 p a r a l y z e d birds. Without blood gas data, however, it is not possible to determine whether this is due to c h a n g e s in central coupling, or whether the 2:1 coupling was due to an uncontrolled c h a n g e in feedback. In a single experiment (fig. 5-19), a mild 3 % i n c r e a s e in inspired C 0 2 did not affect the central coupling of f w and f v in the C a n a d a goo s e . It is p o s s i b l e , however, that this increase in C 0 2 was too s m a l l . It is a l s o p o s s i b l e that the increase in C 0 2 produced an increase in V T rather than f v, a n d therefore the relationship between f w and f v was not affected. In c o n c l u s i o n , although the data presented here provide no evidence of the central neural substrates involved in the entrainment of locomotion an d respiration, both f e e d b a c k and feedforward mechanisms appear to be involved. The entrainment of respiration by limb (Iscoe & P o l o s a , 1976) and cutaneous afferent stimulation (Iscoe & P o l o s a , 1976; K a w a h a r a et al., 1988) and by passive limb motion (Chapter 4; P a l i s s e s et al., 1988) suggests a role for afferent feedback (from limbs and/or chest wall/lung) in entrainment. Feedforward mechanisms also appear to be involved, as s u g g e s t e d by the intermittent synchronization of locomotion and respiration in paralyzed cats (Kawahara et al., 1989a) and rabbits (Viala et a]., 251 1987b; P e r s e g o l et al., 1988) a n d the tight synchronization of respiration and wing beat in p a r a l y z e d birds (Chapter 5). In addition, chemoreceptor drive also affects entrainment (Chapter 2; Paterson et al., 1987; Webb, 1975; Dempsey, personal communication). Thus, it s e e m s likely that afferent and efferent locomotor input as well a s afferent respiratory input, following one or several levels of integration in the midbrain, brainstem, and/or sp i n a l cord, produce entrainment through a collective interaction on s o me portion of the respiratory pattern oscillator. T h i s c a n be s e e n in g e e s e during walking/running where the factors affecting respiratory pattern a p p e a r to be arranged in a hierarchy. The influence of locomotor pattern on respiration is e x p r e s s e d under certain permissive conditions. However, i n c r e a s e d respiratory drives a s s o c i a t e d with hypoxia and possibly body temperature, override the effects of locomotion on respiration and entrainment d e c r e a s e s . The stronger relationship between locomotion and respiration during flight s u g g e s t s - that locomotor inputs may be higher in the hierarchy under these conditions. Thus, e n h a n c e d non-ventilatory drives may not override the effects of locomotion on respiration, but produce an increase in f v that, in combination with locomotor input, results in a shift in i j i v as s e e n in horses (Dempsey personal communication) and C v matoaaster (Webb, 1975). The shifts in recorded during free-flight in a number of birds (table 3-2) suggest this type of interaction. 252 REFERENCES 253 A d a m s , L , Datta, A.K. & G u z , A. (1989). Synchronization of motor unit firing during different respiratory a n d postural t a s k s in human sternocleidomastoid muscle. J . P h y s i o l . 413, 213-231. A d a m s , R.D. (1980). P a i n . In Harrison's principles of internal medicine. 9th edition, (ed. K.J. Isselbacher, R.D. A d a m s , E. Braunwald, R.G. Petersdorf & J.D Wilson), pp. 13-20. New York: McGraw-Hill. A d r i a n , E.D. (1933). Afferent impulses in the va g u s and their effect on respiration. J . P h y s i o l . 79, 332-358. Ainsworth, D.M., Smith, C.A., Eicker, S.W. Henderson, K.S. & Dempsey, J.A. (1989). The effects of locomotion on respiratory muscle activity in the awake dog. Re s p i r . P h y s i o l . 78, 145-162. A s m u s s e n , E. & N i e l s e n , M. (1946). Studies on the regulation of respiration in heavy work. A c t a P h y s i o l . S c a n d . 16, 270-285. B a l l a m , G.O., Cl a n t o n , T.L & Kunz, A.L. (1982). Ventilatory phase duration in the ch i c k e n : role of mechanical a n d C 0 2 feedback. J . A p p l . Physiol.: Respirat. Environ. E x e r c i s e . P h y s i o l . 53(6), 1378-1385, 1982. Ba l l a m , G.O., C l a n t o n , T.L, Kaminski, R.P. & Kunz, A.L (1985). Effect of sinusoidal forcing of ventilatory volume on avian breathing frequency. J . A p p l . P h y s i o l . 59(3), 991-1000. Barnas,/ G.M. & Burger, R.E. (1984). Effect of chronic a c i d o s i s on extra- and intrpulmonary chemoreceptor control of ventilatory movements in the chick e n . Comp. Bi o c h e m . P h y s i o l . 79A(2), 235-240. Baudinette, R.V., Loveridge, J.P., W i l s o n , K.J., Mills, C D . & Schmidt-Nielsen, K. (1976). Heat l o s s from feet of herring gulls at rest a n d during flight. Am. J . P h y s i o l . 230(4), 920-924. Baudinette, R.V., G a n n o n , B.J., Runciman, W.B., Wel l s , S. & Love, J.B. (1987). Do cardiorespiratory frequencies show entrainment with hopping in the tammar wallaby? J . exp. B i o l . 129, 251-263. B e c h , C. & Nomoto, S. (1982). C a r d i o v a s c u l a r c h a n g e s a s s o c i a t e d with treadmill running in the pekin duck. J . exp. Bi o l . 97, 345-358. B e c h b a c h e , R.R. & Duffin, J . (1977). The entrainment of breathing frequency by ex e r c i s e rhythm. J . P h y s i o l . 272, 553-561. Berger, M., Roy, O.Z. & Hart, J.S. (1970a). The co-ordination between respiration and wing beats in birds. Z. vergl. Physiologies 66, 190-200. Berger, M., Hart, J.S. & Roy, O.Z. (1970b). Respiration, oxygen consumption a n d heart rate during rest and flight. Z. verg l . Physiologie. 66, 201-214. 254 B e r n s t e i n , M.H. (1976). Ventilation and respiratory evaporation in the flying crow, C o r v u s o s s i f r a a u s . R e s p i r . P h y s i o l . 26, 371-382. Be r n s t e i n , M.H., Thomas, S.P. & Schmidt-Nielsen, K. (1973). P o w e r input during flight in the fish crow, C o r v u s os s i f r a a u s . J . exp. B i o l . 58, 401-410. B i e s e l , W. & Natc h i g a l l , W. (1987). P i g e o n flight in a wind tunnel. IV. Thermoregulation and water homeostasis. J . Comp. Ph y s i o . B, 157, 117-128. Bouverot, P., H i l l , N. & J a m m e s , Y. (1974). Ventilatory r e s p o n s e s to C 0 2 in intact a n d chron i c a l l y c hemodenervated peking ducks. Respir. P h y s i o l . 22, 137-156. Brackenbury, J.H., G l e e s o n , M. & Avery, P. (1981a). Effects of sustained running e x e r c i s e on lung air-sac g a s composition and respiratory pattern in domestic fowl. C omp. Bioc h e m . P h y s i o l . 69A, 449-453. Brackenbury, J.H., Avery, P. & G l e e s o n , M. (1981b). Respiration in exercising fowl. I. O x y g e n consumption, respiration rate and respired g a s e s . J. exp. B i o l . 93, 317-325. Brackenbury, J.H., G l e e s o n , M. & Avery, P. (1982a). Respiration in exercising fowl. III. Ventilation. J . exp. B i o l . 96, 315-334. Brackenbury, J.H., G l e e s o n , M. & Avery, P. (1982b). Control of ventilation in running birds; effects of hypoxia, hyperoxia, a n d C 0 2 . J. A p p l . P h y s i o l . 53(6), 1397-1404. Brackenbury, J.H. & G l e e s o n , M. (1983). Effects of P ^ on respiratory pattern during thermal and exe r c i s e hyperventilation in domestic fowl. Respir. P h y s i o l . 54, 109-119. Brackenbury, J.H, (1986). B l o o d g a s e s and respiratory pattern in exercising fowl: c o m p a r i s o n in normoxic a n d hypoxic conditions. J. exp. B i o l . 126, 423-431. Bramble, D.M. (1983). Respiratory patterns a n d control during unrestrained human running. In Modelling and control of breathing (ed. B.J. Whipp & D.M. Wiberg), pp. 213-220. New York: E l s e v i e r B iomedical. Bramble, D.M. & Carrier, D.R. (1983). Running and breathing in mammals. S c i e n c e . 219, 251-256. Butler, P.J. (1970). The effect of progressive hypoxia on the respiratory and car d i o v a s c u l a r s y s t e m s of the pigeon and duck. J. P h y s i o l . 201, 527-538. Butler, P.J. & Tayler, E.W. (1974). R e s p o n s e s of the respiratory and cardiovascular s y s t e m s of ch i c k e n s and pigeons to c h a n g e s i n P a ^ and P a ^ . Respir. P h y s i o l . 21, 351-363. 255 Butler, P.J., West, N.H. & J o n e s , D.R. (1977). Respiratory and car d i o v a s c u l a r r e s p o n s e s of the pigeon to su s t a i n e d level flight in a wind tunnel. J . exp. B i o l . 7 1 , 7-26. Butler, P.J. & W o a k e s , A.J. (1980). Heart rate, respiratory frequency and wing beat frequency of free flying barnacle g e e s e Branta l e u c o p s i s . J . exp. B i o l . 85, 213-226. Carrier , D.R. (1987). Lung ventilation during walking and running in four s p e c i e s of li z a r d s . Exp. B i o l . 47, 33-42. C e l l i , B., Criner, G. & R a s s u l o , J . (1988). Ventilatory muscle recruitment during unsupported arm exe r c i s e in normal subjects. J . A p p l . P h y s i o l . 64(5), 1936-1941. C r o s s , B.A., Davey, A., G u z , A., Katona, P.G., Ma c l e a n , M., Murphy, K., Semple, S.J.G. & Stidwill, R. (1982). The role of spinal cord transmission in the ventilatory r e s p o n s e to electrically i n d u c e d e x e r c i s e in the anaesthetized dog. J . P h y s i o l . 329, 37-55. C o m p e n d i u m of pharmaceuticals a n d spe c i a l t i e s . XVII edition, (1982). (ed. C.M.E. Krogh, C B . Schneider, C. S h a u g n e s s y & L. Welbanks), p. 242. Toronto, Southam Murray. D'Angelo, E. (1984). Effects of body temperature, passive limb motion and level of a n a e s t h e s i a on the activity of the inspiratory muscles. Respir. P h y s i o l . 56, 105-129. Davey, N.J. & S e l l e r , T.J. (1987). Brain m echanisms for respiratory control. In Bi r d respiration (ed T.J. Selier), pp. 169-188. B o c a Raton, Florida, C R C P r e s s , Inc. Dejours, P. (1964). Control of respiration in muscular e x e r c i s e . In Handbook of Physiology. S e c t i o n 3, Respiration. V o l . I, (ed. W.O. Fenn & H. Rahn), pp. 631-648. Washington, D.C, A m e r i c a n Phy s i o l o g i c a l Society. D empsey, J.A., Vidruk, E.H. & Mitchell, G.S. (1985). Pulmonary control s y stems in exe r c i s e : update. Federation Proc. 44, 2260-2270. Di M a r c o , A.F., Romaniuk, J.R., Euler, C v o n & Yamamoto, Y. (1983). Immediate c h a n g e s in ventilation and respiratory pattern a s s o c i a t e d with onset and cess a t i o n of locomotion in the cat. J . P h y s i o l . 343, 1-16. Eldridge, F.L (1972a). The importance of timing on the respiratory effects of intermittent carotid s i n u s nerve stimulation. J . P h y s i o l . 222, 297-318. Eldridge, F.L. (1972b). The importance of timing on the respiratory effects of intermittent carotid body chemoreceptor stimulation. J . P h y s i o l . 222, 319-333. Eldridge, F.L. (1989). P h a s e resetting of respiratory rhythm: effect of changing respiratory "drive". Am. J . P h y s i o l . 257 (Reg. Int. Comp. P h y s i o l . 26), R271-R277. 256 Eldridge, F .L, Millhorn, D.E., Kiley, J.P. & Waldrop, T.G. (1985). Stimulation by central c o m m a n d of locomotion, respiration and circulation during e x e r c i s e . R e s p . P h y s i o l . 59, 313-337. Euler, C. v o n . (1986). Brainstem mechanisms for generation and control of breathing pattern. In Handbook of physiology. Section 3, The respiratory system. V o l . II, Control of breathing. Part I (ed. S.R. geiger, A.P. Fishman, N.S. Ch e r n i a c k & J.G. Widdicombe), pp. 1-67. Baltimore Maryland, Waverly P r e s s , Inc. Fedde, M.R. & Burger, R.E. (1963). Death and pulmonary alterations following bilateral c e r v i c a l vagotomy in the fowl. Poultry S c i . 42, 1236-1246. Fedde, M.R., Burger, R.E. & Kitchell, R.L. (1964a). Electromyographic studies of effects of bodily position a n d a n e s t h e i a on the activity of the respiratory muscles of the d o m e s t i c cock. Poultry S c i . 43, 839-846. Fe d d e , M.R., Burger, R.E. & Kitchell, R.L. (1964b). Electromyogrphic studies of the effects of bilateral, c e r v i c a l vagotomy on the action of the respiratory m u s c l e s of the dome s t i c cock. Poultry S c i . 43, 1119-1125. Fedde, M.R., Burger, R.E. & Kitchell, R.L. (1964c). Anatomic and electromyographic st u d i e s of the costopulmonary muscles in the cock. Poultry S c i . 43, 1177-1184. Fedde, M.R., Burger, R.E., G e i s e r , J., Gratz, R.K., Estavillo, J.A. & S c h e i d , P. (1986). Effects of altering d e a d s p a c e volume on respiratory and air s a c g a s e s in ge e s e . R e s p i r . P h y s i o l . 66, 109-122. Fedde, M.R., Orr, J.A., Sham, H. & S c h e i s , P. (1989). Cardiopulmonary function in exe r c i s i n g bar-headed g e e s e during normoxia and hypoxia. Respir. P h y s i o l . 77, 239-262. F e l d m a n , J . L (1986). Neurophysiology of breathing in mammals. In Handbook of physiology. S e c t i o n 1, The nervous system. V o l . IV, Intrinsic regulatory systems of the brain (ed. F.E. Bloom), pp. 463-524. Bethesda, American Ph y s i o l o g i c a l Society. F e l dman, J . L , Smith, J . C , McCrimmon, D.R., Ellenberger, H.H. & Speck, D.F. (1988). Generation of respiratory pattern in mammals. In Neural control of rhythmic movements in vertebrates (ed. A.H. Co h e n , S. Rossi g n o l & S. Grillner), pp. 73-100. New York: J o h n Wiley & So n s . G a r c i a - R i l l , E. & Skinner (1985). Modulation of rhythmic function in the posterior midbrain. N e u r o s c i . 27(2), 639-654. G a r c i a - R i l l , E., Skinner, R.D. & Fitzgerald, J.A. (1985). C h e m i c a l activation of the me s e n c e p h a l i c locomotor region. Brain. R e s . 330, 43-54. G l e e s o n , M. & Molony, V. (1989). Control of breathing. In Form and function in birds. V o l . 4 (ed. A.S. King & J . McLelland), pp. 439-484. S a n Diego, A c a d e m i c P r e s s . 257 G r i l l n e r , S. (1975). Locomotion in vertebrates: central m echanisms and reflex interaction. P h y s i o l . Rev. 55, 247-304. Hart, J.S. & Roy, O.Z. (1966). Respiratory and c a r d i a c r e s p o n s e s to flight in pigeons. P h y s i o l . Z o o l . 39, 291-306. Hi l l , A.R., Adams, J.M., Parker, B.E. & Rochester, D.F. (1988). Short-term entrainment of ventilation to the walking cycle in humans. J . A p p l . P h y s i o l . 65(2), 570-578. Hinds, D.S. & C a l d e r , W.A. (1970). T r a c h e a l d e a d s p a c e in the respiration of birds. Evolution 25, 429-440. Hof, V. Im., West, P. & Y o u n e s , M. (1986). Steady state r e s p o n s e s of normal sub j e c t s to inspiratory resistive load. J . A p p l . P h y s i o l . 60(5), 1471-1481. Howard, P., Bromberger-Barnea, B., Fitzgerald, R.S. & Bane, H.N. (1969). Ventilatory r e s p o n s e s to peripheral nerve stimulation at different times in the respiratory c y c l e . R e s p i r . P h y s i o l . 7, 389-398. Hudson, D.M. & Bernstein, M.H. (1981). Temperature regulation and heat balance in flying white n e c k e d ravens, C o r v u s cryptoleucus. J . exp. Bi o l . 90, 267-281. Ikegami, Y., Hiruta, S., Ikegami, H. & Miyamura, M. (1988). Develpoment of a telemetry s y s t e m for measuring oxygen uptake during sports activities. Eur. J . P h y s i o l . 57, 622-626. Iscoe, S.D. (1988). Central control of the upper airway. In Respiratory function of the upper airway (ed. O.P. Mathew & G. Sant'Ambrogio), pp. 125-192. New York, M a r c e l Decker, Inc. Iscoe, S. & P o l o s a , C. (1976). Synchronization of respiratory frequency by somatic afferent stimulation. J . A p p l . P h y s i o l . 40(2): 138-148. J a c o b s o n , R.D. & Hollyday, M. (1982). Electrically e v o k e d walking and fictive locomotion in the chick. J . Neurophysiol. 48(1), 257-270. J a s i n s k a s , C L , W i l s o n , A.B. & Hoare, J . (1980). Entrainment of breathing rate to movement frequency during work at two intensities. R espir. P h y s i o l . 42, 199-209. J e n k i n s , F.A.Jr., D i a l , K.P. & Goslow, G.E.Jr. (1988). A cineradiographic analysis of bird flight: the wishbone in starlings is a spring. S c i e n c e 241, 1495-1498. J o n e s , D.R. & Bamford, O.S. (1976). Open-loop respiratory chemosensitivity in c h i c k e n s a n d ducks. Am. J . P h y s i o l . 230(4), 861-867. J o n e s , D.R. & Purves, M.J. (1970). T h e effect of carotid body denervation upon the respiratory response to hypoxia and hypercapnia in the duck. J . P h y s i o l . 211, 295-309. 258 J o r d a n , L.M., Pratt, C.A. & M e n z i e s , J.E. (1979). Locomotion e v o k e d by brainstem stimulation: o c c u r e n c e without p h a s i c segmental afferent input. Brain R e s . 177, 204-207. K a l i a , M., S e n a p a t i , J.M., P a r i d a , B. & P a n d a , A. (1972). Reflex increase in ventilation by muscle receptors with nonmedullated fibres (C fibres). J . A p p l . P h y s i o l . 32, 189-193. Kao, F.F. (1963). A n experimental study of the pathways involved in exercise h y p erpnea employing cross-circulation techniques. In The regulation of human respiration (ed. D.J.C. C u n n i n g h a m & B.B. Lloyd), pp. 461-502. Oxford, UK., B l a c k w e l l . Karten, H.J. & Hodos, W. (1967). A stereotaxic atlas of the brain of the pigeon ( C o l u m b i a livia). Baltimore, J o h n s Hopkins P r e s s . K aufman, M.P. & R y b i c k i , K.J. (1987). Discharge properties of group III a n d IV muscle afferents: their r e s p o n s e s to mechanical and metabolic stimuli. C i r c . res. 61 (suppl), I 60-I 65. K a w a h a r a , K., Kumagai, S., Nakazono, Y. & Miyamoto, Y. (1988). A n a l y s i s of entrainment of respiratory rhythm by somatic afferent stimulation in c a t s using p h a s e r e s p o n s e c u r v e s . B i o l . C y b e r n . 58, 235-242. K a w a h a r a , K., N a k a zono, Y. Y a m a u c h i , Y. & Miyamoto, Y. (1989a). C o u p l i n g between respiratory a n d locomotor rhythms during fictive locomotion in decerebrate c a t s . N e u r o s c i . Lett. 103, 326-332. K a w a h a r a , K., K u m a g a i , S., N a k a zono, Y. & Miyamoto, Y. (1989b). C o u p l i n g between respiratory and stepping rhythms during locomotion in decerebrate cats. J . A p p l . P h y s i o l . 67(1), 110-115. Kay, J.D.S., P e t e r s e n , E.S. & Vejby-Christensen, H. (1975). Breathing in man during steady-state e x e r c i s e on the bicycle at two pedalling frequencies, and during treadmill walking. J . P h y s i o l . 251, 645-656. K e l m a n , G.R. & Watson, A.W.S. (1973). Effect of a d ded d e a d s p a c e on pulmonary ventilation during sub-maximal, steady-state e x e r c i s e . Q. J . Exp. P h y s i o l . 58, 305-313. Kiley, J.P. & Fedde, M.R. (1983). Cardiopulmonary control during exercise in the duck.' J . A p p l . Physiol.: Respirat. Environ. E x e r c i s e P h y s i o l . 55(5), 1574-1581. Kiley, J.P., Kuhlmann, W.D. & Fedde, M.R. (1979). Respiratory and cardiovascular r e s p o n s e s to e x e r c i s e in the duck. J . A p p l . P h y s i o l . 47(4), 827-833. Kiley, J.P., Kuhlmann, W.D. & Fedde, M.R. (1982). Ventilatory and blood gas adjustments in e x e r c i s i n g isothermic ducks. J . Comp. P h y s i o l . 147, 107-112. 259 K i r l i n , P.C., Kittleson, M.D. & J o h n s o n , L E . (1987). Neurohumoral and cardiopulmonary response to s u s t a i n e d submaximal exercise in the dog. J . A p p l . P h y s i o l . 62(3), 1040-1045. Ko h l , J., Koller, E.A. & J a g e r , M. (1981). Relation between pedalling a n d breathing rhythm. Eur. J . A p p l . P h y s i o l . 47, 223-237. K o i z u m i , K., U s h i y a m a , J . & Brooks, C.McC. (1961). M u s c l e afferents and activity of respiratory neurons. Am. J . P h y s i o l . 200(4), 679-684. Kunz, A.L (1987). Peripheral mechanisms in the control of breathing. In Bird respiration (ed. T.J. Se l l e r ) , pp. 129-167. B o c a Raton, Flo r i d a , C R C P r e s s , Inc.. Kunz, A.L & Miller, D.A. (1974). P a c i n g of avian respiration with C 0 2 oscillation. R e s p i r . P h y s i o l . 22, 167-177. Larra b e e , M.G. & Hodes, R. (1948). C y c l i c c h a n g e s in the respiratory centers, revealed by the effects of afferent impulses. Am. J . P h y s i o l . 155, 147-164. L a s i e w s k i , R.C. & Dawson, W.R. (1967). A re-examination of of the relation between standard metabolic rate and body weight in birds. C o n dor 69, 13-23. L e F e b v r e , E.A. (1964). The use of D 2 01 8 for measuring energy metabolism in C o l u m b i a livia at rest and in flight. A u k 81, 403-416. Loeser, J.D. & Black, R.G. (1975). A taxonomy of pain. P a in 1(1), 81-84. Lord, R.D., Bel l r o s e , F.C. & Cochraw, W.W. (1962). Radiotelemetry of respiration of a flying duck. S c i e n c e 137, 39-40. L u msden, T. (1923). Observations on the respiratory centres in the cat. J . P h y s i o l . 59, 153-160. M c C l e l l a n , A.D. (1986). C o m m a n d systems for initiating locomotion in fish and amphibians: paralells to initiation s ystems in mammals. In Neural control of vertebrate locomotion (ed. S. Grillner, R.M. Herman, P.S.G. St e i n , H. Forsberg & D.G. Stuart), pp. 3-20. London, Macmillan P r e s s . M c C l o s k e y , D.I. & Mitchell, J.H. (1972). Reflex c a r d i o v a s c u l a r and respiratory r e s p o n s e s originating in exercising muscle. J . P h y s i o l . 224, 173-186. Millhorn, D.E., Eldridge, F.L, Waldrop, T.G. & Kiley, J.P. (1987). Diencephalon regulation of respiration and arterial pressure during actual and fictive locomotion in cat. C i r c . R e s . (suppl I), I 53-I 59. Mitchell, J.H. (1985). C a r d i o v a s c u l a r control during exercise: central and .reflex neural mechanisms. Am. J . Cardiology. 55, 34D-41D. 260 Mitchell, G.S., G l e e s o n , T.T. & Bennett, A.F. (1981). Ventilation and acid-base b a l a n c e during graded activity in lizards. Am. J . P h y s i o l . 240 (Reg. Int. Comp. Physfol.9), R29-R37. Nomoto, S., Rautenberg, W. & iriki, M. (1983). Temperature regulation during e x e r c i s e in the J a p a n e s e Quail (Coturnix coturnix iaponica). J . Comp. P h y s i o l . 149, 519-525. Obrecht, H.H.III., Pennycuick, C.J. & Fuller, M.R. (1988). Wind tunnel experiments to a s s e s s the effect of back-mounted radio transmitters on bird body drag. J . exp. B i o l . 135, 265-273. P a t e r s o n , D.J., Wood, G.A., Morton, A.R. & Henstridge, J.D. (1986). The entrainment of ventilation frequency to exercise rhythm. Eur. J . P h y s i o l . O ccup. P h y s i o l . 55, 530-537. P a t e r s o n , D.J., Wood, G.A., Mars h a l l , R.A., Morton, A.R. & Harrison, A.B.C. (1987). Entrainment of respiratory frequency to ex e r c i s e rhythm during hypoxia. J . A p p l . P h y s i o l . 62(5), 1767-1771. P a v l i d i s , T. (1973). Biological oscillations: Their mathematical a n a l y s i s . New York: A c a d e m i c P r e s s . Petrillo, G.A., G l a s s , L. & Trippenbach, T. (1983). P h a s e locking of respiratory rhythm to a mechanical ventilator. C a n . J . P h y s i o l . P h a r macol. 61, 599-607. P r o c h a z k a , A. (1980). M u s c l e spindle activity during walking and during free f a l l . In Spinal a n d supr a s p i n a l mechanisms of voluntary motor control and locomotion, P r o g r e s s in neurophysiology. V o l . 8 (ed. J.E. Desmedt), pp. 282-293. Karger, B a s e l . R a m i r e z , J.M. & P e a r s o n , K.G. (1989). Alteration of the respiratory system at the onset of locust flight. J . exp. B i o l . 142, 401-424. Remmers, J.E. & Marttila, (1975). Action of intercostal muscle afferents on the respiratory rhythm of anaesthetized cats. Respir. P h y s i o l . 24, 31-41. Roberts, B.L & Ballintijn, C M . (1988). S e n s o r y interaction with central 'generators' during respiration in the dogfish. J . Comp. P h y s i o l . A. 162, 695-704. Roberts, J.L. & Rowell, D.M. (1987). Periodic respiration of gill-breathing fishes. C a n . J . Z o o l . 66, 182-190. Rotto, D.M. & Kaufman, M.P. (1988). Effect of metabolic products of muscular contraction on discharge of group III and IV afferents. J . A p p l . P h y s i o l . 64(6), 2306-2313. R u s s o , A.K., Tara s a n t c h i , J . & Griggio, M.A. (1977). Oxygen consumption and ventilation of dogs during passive and active exercise. J . A p p l . Physiol.: Respirat. Environ. E x e r c i s e P h y s i o l . 42(6), 923-927. 261 S a t c h e l l , G.H. (1968). A neurological basis for the co-ordination of swimming with respiration in f i s h . Comp. Biochem. P h y s i o l . 27, 835-841. S c h e i d , P., Fedde, M.R. & Piiper, J . (1989). G a s exchange and air-sac composition in the u n a n a e s t h e t i z e d , spontaneously breathing goose. J . exp. B i o l . 142, 373-385. Schmidt-Nielsen, K. (1979). A n imal Physiology. Adaptation a n d environment. 2nd ed'n. pp. 162-163. New York: Cambridge University P r e s s . S e n a p a t i , J.M. (1966). Effect of stimulation of muscle afferents on ventilation of dogs. J . A p p l . P h y s i o l . 21, 242-246. S e v e r i n , F.V., O r l o v s k i , G.N. & Shik, M.L (1967). Activity of muscle receptors during controlled locomotion. B i o f i z i k a 12, 502. S h a n n o n , R. (1986). R e f l e x e s from respiratory muscles and costovertebral muscle. In H a n dbook of physiology. S e c t i o n 3, The respiratory s y stem. V o l . II, Control of breathing. Part I (ed. S.R. Geiger, A.P. Fishman, N.S. C h e r n i a c k & J.G. Widdicombe), pp. 431-447. Baltimore Maryland, W a v e r l y P r e s s , Inc. S h a n n o n , R., Bolser, D.C. & Lindsey, B.G. (1988). Medullary neurons mediating the inhibition of inspiration by intercostal muscle tendon organs. J . A p p l . P h y s i o l . 65(6), 2498-2505. Sherrington, C.S. (1910). Flexion-reflex of the limb, c r o s s e d extension reflex, and reflex stepping and standing. J . P h y s i o l . 40, 28-121. Sherrington, C.S. (1915). Simple apparatus for obtaining a decerebrate preparation of the cat. J . P h y s i o l . 49, 1ii-1iv. S hik, M. L , S e v e r i n , F.V. & Orio v s k i i , G.N. (1966). Control of walking and running by means of electrical stimulation of the midbrain. B i o p h y s i c s ( U S S R ) ( E n g l i s h translation) 11, 756-765. Sholomenko, G.N. & S t e e v e s , J.D. (1987). Effects of spinal cord lesions on hind limb locomotion in birds. Exp. Neurol. 95, 403-418. Sholomenko, G.N., Funk, G.D. & S t e e v e s , J.D. (1990a). A v i a n locomotion activated by brainstem infusion of neurotransmitter agonists a n d antagonists. I. Acetylcholine, excitatory amino ac i d s and Substance P. Exp. Brain R e s . (accepted for publication). S holomenko, G.N., Funk, G.D. & S t e e v e s , J.D. (1990b). A v i a n locomotion activated by brainstem infusion of neurotransmitter agonists and antagonists. II. Gamma-Aminobutyric a c i d . Exp. Brain R e s . (accepted for publication). S h olomenko, G.N., Funk, G.D. & S t e e v e s , J-D. (1990c). Locomotor activities in the decerebrate bird without phasic afferent input. Brain R e s . (accepted for publication). 262 S p e c k , D.F. (1989). B o t z i n g e r complex region role in phrenic-to-phrenic inhibitory reflex of cat. J . A p p l . P h y s i o l . 67(4), 1364-1370. Speck, D.F. & Revelette, W.R. (1987). Attenuation of phrenic motor discharge by phrenic nerve afferents. J . A p p l . P h y s i o l . 62(3), 941-945. S t e e v e s , J.D., Sholomenko, G.N. & Webster, D.M.S. (1987). Stimulation of the pontomedullary reticular formation initiates locomotion in decerebrate birds. Brain R e s e a r c h 4 0 1, 205-212. Suthers, R.A., Thomas, S.P. & Suthers, B.J. (1972). Respiration, wing-beat and ultrasonic p u l se e m i s s i o n in a n echo-locating bat. J . exp. B i o l . 56, 37-48. Swett, J.E. & B o u r a s s a , C M . (1981). Electrical stimulation of peripheral nerve. In Electrical stimulation techniques (ed. M. Patterson & R. Kesner), pp. 243-295. T h omas, S.P. (1981). Ventilation and oxygen extraction in the bat, Pteropus gouldii during rest & ste a d y flight. J . exp. Bi o l . 94, 231-250. T o m l i n s o n , J.T. (1957). P i g e o n wing beats s y n c h r o n i z e d with breathing. C o n d o r 59, 401. T o m l i n s o n , J.T. (1963). Breathing of birds in flight. C o n d o r 65, 514-516. Torre-Bueno, J.R. (1978). Respiration during flight in birds. In Respiratory function in birds. Adult and embryonic (ed. J . Piiper), pp. 89-94. New York, Springer-Verlag. Torre-Bueno, J.R. (1978b). Evaporative cooling and water balance during flight in birds. J . exp. B i o l . 75, 231-236. Torre-Bueno, J.R. & Larochelle, J . (1978). T h e metabolic cost of flight in unrestrained birds. J . exp. B i o l . 75, 223-229. Tucker, V.A. (1968). Respiratory exchange a n d evaporative water loss in the flying budgerigar. J . exp. B i o l . 48, 67-87. Tucker, V.A. & Schmidt-Koenig, K. (1971). Flight s p e e d s in relation to energetics and wind directions. A u k 88, 97-107. V i a l a , D. (1986). E v i d e n c e for direct reciprocal interactions between the central rhythm generators for spinal "respiratory" and locomotor activities in the rabbit. Exp. Brain R e s . 63, 225-232. V i a l a , D., V i a l a , G., P e r s e g o l , L. & P a l i s s e s , R. (1987a). C h a n g e o v e r from alternate to s ynchronous bilateral pattern of phrenic bursts entrained by fictive locomotion in the s p i n a l rabbit prepration. N e u r o s c i . Lett. 78, 318-322. V i a l a , D., P e r s e g o l , L & P a l i s s e s , R. (1987b). Relationship between phrenic and extensor activities during fictive locomotion. Neurosci. Lett. 74, 49-52. 263 V i a l a , D., V i d a l , C. & Freton, E. (1979). Coordinated rhythmic bursting in respiratory a n d locomotor muscle nerves in the s p i n a l rabbit. N e u r o s c i Lett. 11, 155-159. Waldrop, T.G., Mullins, D.C. & Henderson, M.C. (1986a). Effects of hypothalamic l e s i o n s on the cardiorespiratory responses to muscular contraction. Respir. P h y s i o l . 66, 215-224. Waldrop, T.G., Mullins, D.C. & Millhorn, D.E. (1986b). Control of respiration by the hypothalamus an d by f e e d b a c k from contracting muscles in cats. Respir. P h y s i o l . 64, 317-328. Waldrop, T.G. H e n d e r s o n , M.C, Iwamoto, G.A. & Mitchell, J.H. (1986c). Regional blood flow r e s p o n s e s to stimulation of the subthalamic locomotor region. Respir. P h y s i o l . 64, 93-102. W a s s e r m a n , K., Whipp, B.J. & C a s a b u r i , R. (1986). Respiratory control during e x e r c i s e . In Handbook of physiology. Section 3, The respiratory system, V o l . II, Control of breathing. Part 2 (ed. S.R. Geiger, A.P. F ishman, N.S. Cherniak & J.G. Widdicombe), pp. 595-619. Baltimore, Waverly P r e s s . W a l l , P.D. (1975). Editi o r i a l . P a i n 1(1), 1-2. W a l l e n , P. & W i l l i a m s , T.L. (1984). Fictive locomotion in the lamprey spinal cord in vitro c o mpared with swimming in the intact and spinal animal. J . P h y s i o l . 347, 225-239. Webb, P.W. (1975). Synchrony of locomotion and ventilation in Cvmatogaster aggregata. C a n . J . Z o o l . 53, 904. W i l l i a m s , B.J., Livingston, C.A. & Leonard, R.B. (1984). Spi n a l cord pathways involved in initiation of swimming in the stingray, Dasyatis sabina: spinal cord stimulation a n d les i o n s . J . Neurophysiol. 51(3), 578-591. Wolthius, R.A., Froelicher, V.F.Jr., Fischer, J . & Triebwasser, J.H. (1977). The response of healthy men to treadmill e x e r c i s e . C i r c . 55(1), 153-157. Y o n g e , R.P. P e t e r s e n , E.S. (1983). Entrainment of breathing in rhythmic exercise. In Modelling a n d control of breathing (ed. B.J. Whipp & D.M. Wiberg), pp. 197-204. New York: E l s e v i e r B i o m e d i c a l . Y o u n e s , M. & Polachek, J . (1981). Temporal c h a n g e s in effectiveness of a constant inspiratory-terminating vagal stimulus. J . A p p l . Physiol.: Respirat. Environ. Exercise P h y s i o l . 50(6), 1183-1192. 264 

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0106749/manifest

Comment

Related Items