## Control of Breathing in Ectothermic Vertebrates

### Abstract

The ectothermic vertebrates are a diverse group that includes the Fishes (Agnatha, Chondrichthyes, and Osteichthyes), and the stem Tetrapods (Amphibians and Reptiles). From an evolutionary perspective, it is within this group that we see the origin of air‐breathing and the transition from the use of water to air as a respiratory medium. This is accompanied by a switch from gills to lungs as the major respiratory organ and from oxygen to carbon dioxide as the primary respiratory stimulant. This transition first required the evolution of bimodal breathing (gas exchange with both water and air), the differential regulation of O2 and CO2 at multiple sites, periodic or intermittent ventilation, and unsteady states with wide oscillations in arterial blood gases. It also required changes in respiratory pump muscles (from buccopharyngeal muscles innervated by cranial nerves to axial muscles innervated by spinal nerves). The question of the extent to which common mechanisms of respiratory control accompany this progression is an intriguing one. While the ventilatory control systems seen in all extant vertebrates have been derived from common ancestors, the trends seen in respiratory control in the living members of each vertebrate class reflect both shared‐derived features (ancestral traits) as well as unique specializations. In this overview article, we provide a comprehensive survey of the diversity that is seen in the afferent inputs (chemo and mechanoreceptor), the central respiratory rhythm generators, and the efferent outputs (drive to the respiratory pumps and valves) in this group. © 2022 American Physiological Society. Compr Physiol 12: 3869–3988, 2022.

 Figure 1. Differences in the physical characteristics of water and air result in dramatic differences in breathing patterns and arterial blood gases in water and air‐breathing ectotherms. The basis of this is explored in the introduction utilizing the panels of this figure. (A) The relationship between the partial pressure and content of O2 and CO2 in water and air. (B) The air convection requirement [the ratio of ventilation (V̇E) to O2 consumption ($V˙O2$)] in the different vertebrate classes. (C) Arterial partial pressure of O2 (PaO2), CO2 (PaCO2) and pH (pHa) in representatives of the different vertebrate classes. (D) Relationship between partial pressures of O2 (PO2) and CO2 (PCO2) in water and air breathers, solid lines are respiratory quotient (RQ) lines showing possible combinations of PO2 and PCO2 in unimodal air (human, hen) or water (dogfish) breathers in steady state. Values for bimodal breathers (eel, lungfish, salamander, bullfrog) fall in between the two lines. (E) Representative breathing traces of a fish, reptile, and mammal (left) and the consequences for levels of arterial O2 and CO2 (right). Note that all key abbreviations used in this article appear in Table 1. Figure 2. (A) Gills and gill coverings in (left to right) lamprey, elasmobranchs, teleosts, and larval salamanders. Note the shift from pouched gills in agnatha to septal gills in elasmobranchs to opercular gills in bony fishes and amphibians. (B‐E) Simplified drawings of the basic structural features of the elasmobranch (B and C) and teleost (D and E) gill. (C) and (E) are enlarged views of the boxes in (B) and (D), respectively. The path of water flow through the gills is indicated by arrows. Note that while the total surface area, the ultrastructure, and the thickness of the water/blood barrier are similar, the efficiency with which the actual gas exchange surface, the secondary lamellae, are exposed to the water flowing through the gills, is greater in teleosts. (A) Modified, with permission, from Kardong KV, 2006 624. (B‐E) Reused, with permission, from Wegner NC, 2015 1229. Figure 3. Evolution of aerial gas exchange structures and pumps in vertebrates. A simplified cladogram indicating where changes originated. Groups with air sacs/air‐breathing organs (highlighted in blue) have a single structure that arises from the dorsal surface of the digestive tract. They may have evolved independently or may have been modified from earlier lungs. The air sacs of teleosts and sturgeon are nonrespiratory (indicated by dotted lines) while the air‐breathing organs of bowfin and gar retain a respiratory function. Groups with lungs (highlighted in yellow) have paired structures that arise from the ventral surface of the digestive tract. Above the cladogram showing the evolutionary origin of each group are sagittal (left) and cross‐sectional (right) views of the organs and their connections to the digestive tract. The origins of major changes in the respiratory pumps are indicated on the cladogram (*see subsection titled “10Air‐Breathing” under section titled “4Respiratory Organs, Pump, and Valves”). Diagram by Jacelyn Shu. Figure 4. Schematic diagrams of sagittal sections of a hagfish, and a larval (ammocoetes larvae) and adult lamprey, showing key respiratory structures. In hagfish and ammocoetes larva, the velum scrolls up and down and along with branchial pouch contractions, drives water through the branchial pouches and out the branchial pores. The inset shows the velum scrolling up and down. In the adult lamprey, the velum serves to occlude the opening to the branchial tube while the animal is feeding, and ventilation is tidal in and out through the branchial pores. Exhalation (branchial compression) is active while inhalation (branchial expansion) is passive. Diagram by Jacelyn Shu. Figure 5. Schematic diagrams of coronal sections through the buccal/pharyngeal regions of a lamprey, an elasmobranch, and a teleost. (A) Comparison of the paths of water flow in larval and adult lamprey. In the larvae, water flow is continuous and unidirectional, in through the mouth and out through the branchial pores. In the adult, it is tidal, in and out through the branchial pores. (B,C) Comparison of the paths of water flow in elasmobranch and teleost fishes. In both cases, water flow over the gills is continuous during both buccal expansion and compression due to the dual pumping action of the buccal pump and the parabranchial/opercular chambers. Diagram by Jacelyn Shu. Figure 6. Schematic diagram illustrating the ventilatory movements in an anuran tadpole. Gray arrows indicate movement of the buccal and/or pharyngeal walls, white arrows indicate movement of the velum, and black arrows indicate the path of water flow—see text for details. Diagram by Jacelyn Shu. Figure 7. Schematic diagrams of various accessory air‐breathing organs: (A) Buccal diverticulae in the climbing perch (Anabas, testudinosus), (B) Pharyngeal cavity in the Asian catfish (Heteropneustes fossilis), (C) Pharyngeal diverticulae in the African catfish (Clarius lazera), and (D) Respiratory intestine in the armored catfish (Hoplosternum littorale). Accessory air‐breathing organs are found in different species throughout the oro‐branchial‐pharyngeal cavities. Diagram by Jacelyn Shu. Figure 8. Schematic diagrams illustrating the generalized circulation to (A) the air‐breathing organ of a lungfish and (B) a teleost fish (see text for details). Insets illustrate the general structure of the hearts of the two groups illustrating the differences in the venous return to the heart from the air‐breathing organ. In lungfish, the efferent vessels leaving the lung return blood directly to the heart (not the vena cava) and to some degree remain separate from systemic venous blood as it flows through the heart. In teleosts, the venous return drains into the vena cava and mixes with the systemic venous blood. Diagram by Jacelyn Shu. Modified, with permission, from Kardong KV, 2006 624. Figure 9. Schematic diagrams illustrating the four‐stoke pump found in most actinopterygian fishes (A) and the two‐stroke buccal pump found in sarcopterygian fishes (B). Large white arrows indicate movement of the buccal and/or pharyngeal walls, white arrows indicate movement of the glottis, and black arrows indicate the path of air flow—see text for details. Diagram by Jacelyn Shu. Figure 10. (A) Schematic diagram illustrating the ventilatory movements in an adult anuran. The ventilation cycle is divided into four phases as described in the text: (i) initial expansion (the first stroke) of the buccal cavity brings fresh air into the mouth; (ii) the glottis opens and gas from the lungs enters the buccal cavity where it mixes with the fresh air to varying degrees as it exits via the mouth and nares that remain open; (iii) the nares and mouth then close and buccal compression (the second stroke) forces buccal gas into the lungs; (iv) the glottis then closes and any excess gas left in the buccal cavity is expelled through the nares or mouth at the end of the buccal compression phase. (B) Summary diagrams illustrating the changes in lung volume, airflow at the nostrils, and pressures in the lungs and buccal cavity, along with the timing of nostril and glottis movement during an inflation and a deflation breath. Note that the determination of whether a breath will lead to further lung inflation or deflation is a function of the timing of the opening and closing of the different valves (nostrils and glottis). Diagram by Jacelyn Shu. Reused, with permission, from Vitalis TZ and Shelton G, 1990 1195. Figure 11. (A) Respiratory activities recorded from the right (RX) and left (LX) vagal motor nuclei in a semi‐intact lamprey preparation showing both rhythmic fast events as well as single‐ and double‐burst slow events (indicated by the arrows). (B) Expanded view of the area highlighted in gray in (A). (C) Schematic drawing showing the motor nuclei and their nerves. (D) Schematic diagram showing the connections between the left pTRG and putative “cCRG” and both ipsilateral and contralateral nuclei of the Vth, VIIth, IXth, and Xth motor nuclei. Panel (D) shows, schematically, the locations of the motor nuclei, and panel (E) shows their connections to the different gill pouches. (A,B) Taken, with permission, from Martel B, et al., 2007 733. (C,E) Adapted, with permission, from Missaghi K, et al., 2016 795. Figure 12. (A) Schematic diagram of the left side of a catshark showing details of the cranial nerves innervating the respiratory system, together with the location of their motor and sensory nuclei in the brainstem. The nerves and nuclei are color coded and are listed from the most rostral to the more caudal. Red are motor nuclei: Vm, trigeminal nucleus supplying; Vmand, mandibular; Vmax, maxillary branches of Vth to jaws; VIIm, facial nucleus supplying VIIth nerve to spiracle; IXm, glossopharyngeal nucleus supplying IXth to 1st gill slit; Xm1‐4, vagal (Xth) motor nuclei supplying branches of X1‐4 to gill slits 2 to 5; XmL, lateral nucleus of Xth supplying the heart; Xvisc, visceral vagus. Blue are sensory nuclei supplying trigeminal (Vth), facial (VIIth), glossopharyngeal (IXth), and vagus (Xth) nerves. Yellow is the reticular formation (RF). Brown represents midbrain nuclei (MD), including the tegmentum. Green is occipital (occ) and anterior spinal (sn) nerves supplying the hypobranchial nerve. CER, cerebellum; GS, gill slit; MO, medulla oblongata; SC, spinal cord; SP, spiracle. (B) Schematic dorsal view of the catshark brainstem and anterior spinal cord to show the location of motor nuclei innervating respiratory and feeding muscles. Motor nuclei to the facial (VIIth), glossopharyngeal (IXth), and vagus (Xth) nerves (red shaded areas) have an overlapping rostro‐caudal, sequential distribution that may facilitate the spread of excitation from the more rostral nucleus of the trigeminal Vth nerve (the arrow suggests this connection). The target organs for these motor nuclei are listed in the caption to panel (A). The hypobranchial motor nuclei (green shaded areas) named for the muscles they innervate also have an overlapping sequential distribution: cm, coraco‐mandibular; ch, coraco‐hyal; cb, coracobranchiales; ca, coraco‐arcual. Nerves y and z are branches of the occipital nerve, which, together with the anterior spinal nerves, contribute to the hypobranchial nerve. TS indicates the position of section shown in Panel C. Arrows indicate possible interactions; details are given in the text. (C) Schematic diagram of a transverse section through the brainstem of the catshark at the point labeled in panel (B) (TS) to show spatial and possible functional relationships between nuclei supplying respiratory and cardiac nerves [color code and labels as in panel (A)]. Labels: 4th vent., 4th ventricle; XmL, lateral vagal preganglionic neurons; XmM, preganglionic neurons in the dorsal vagal motor nucleus; Xs, dorsal sensory nucleus of vagus; X4, 4th branchial branch of vagus. Arrows indicate possible functional connections. (A,C) Redrawn, with permission, from Taylor EW, 2011 1124; Adapted, with permission, from Taylor EW, 2011 1125. (B) Redrawn, with permission, from Taylor EW, et al., 2006 1127; Taylor EW, et al., 1999 1128. Figure 13. Schematic diagrams of the frog hindbrain (pons and medulla) with cerebellum and tectum removed. (A) Sites of the buccal priming (red) and lung power stroke (blue) oscillators as well as the buccal oscillator (green) as described by Baghdadwala et al. 43. (B) sites of the transections (red lines) and rough location of the multiple lung breath oscillators as described by Klingler and Hedrick 649. See text for details. (A) Based on Baghdadwala MI, et al., 2015 43. (B) Based on, with permission, Klingler MJ and Hedrick MS, 2013 649. Figure 14. (A) Integrated traces of neural discharge in cranial nerves V (CNV), VII (CNVII), X (CNX), and XII (CNXII) of an in vitro preparation of a juvenile bullfrog illustrating bursts of nerve activity associated with fictive buccal and lung inflations. Activity associated with buccal expansion attributed to a lung priming oscillator appears in the lung traces in CNVII and CNXII (highlighted in red). Also note the progressive increase in amplitude of the buccal breaths preceding the lung burst episode (blue arrows). (B) Expanded view of the first two breaths of the lung breath episode shown in (A). See text for details. Modified, with permission, from Baghdadwala MI, et al., 2015 43. Figure 15. Integrated traces of neural discharge in cranial nerves V (CNV), X (CNX), and both the main branch CNXIIm) and sternohyoid branch (CNXIIsh) of XII of an in vivo preparation of an adult bullfrog. (A) Top traces illustrate the activity associated with a sequence of buccal bursts and a single lung burst. Lower traces are an overlay of several single lung bursts illustrating the components of the lung breath. (B) Top traces illustrate the activity associated with an episode of six lung bursts. Lower traces are an overlay showing the progressive increases in the discharge in CNV as well as both branches of CNXII in the episode of six lung breaths shown above. In both (A) and (B), the blue bars separate three phases of the lung burst. The time between the first two blue bars is the period during which the activity associated with buccal expansion attributed to a lung priming oscillator appears in the lung traces in CNVII and CNXII in Figure 14. Note the absence of any activity other than very modest discharge in CNXIIm during this period but a large progressive increase in early discharge in CNXIIsh during the lung breath episode. The time between the second and third blue bars is associated with normal buccal expansion as indicated by the appearance of activity in all traces except CNXIIm, Note the consistency of the discharge in progressive breaths in the lung breath episode. The period between the third and fourth blue bars indicates the period of buccal compression associated with lung inflation. See text for details. Modified, with permission, from Kogo N and Remmers JE, 1994 651. Figure 16. Fictive lung burst activity in a premetamorphic (left) and postmetamorphic (right) brainstem preparation prior to transection (A), after a transection above CN IX (B), and below CN XI (C). Bursts recorded after transections are asynchronous and have a lower amplitude and frequency than bursts recorded from the intact brainstem. Vertical scale bars indicate voltage (V) for CN V, X, and XII cranial nerves prior to transection. The scale bars are the same for each channel before and after transection for pre‐ and postmetamorphic brainstems. Note the presence of small amplitude, high frequency fictive gill bursts in the intact premetamorphic brainstem and buccal bursts in the intact postmetamorphic brainstem. Both types of bursts are abolished following transection. Reused, with permission, from Klingler MJ and Hedrick MS, 2013 649. Figure 17. Traces illustrating episodic breathing in which breaths are clustered into episodes containing a varying number of breaths, interspersed by apneic periods of different lengths. Episodic breathing is found in some species of every class of ectotherm as illustrated for different species here. Figure 18. (A‐F) Traces illustrating various parts of the anuran respiratory pattern continuum under progressively increasing tonic drive (due to slowly elevating CO2) recorded in situ from the trigeminal nerve of a bullfrog. Small amplitude oscillations represent buccal ventilation while the larger amplitude oscillations in this figure represent lung breaths. Due to the time scale used for this figure, periods of prolonged apnea punctuated by isolated single breaths are not shown. (A) Buccal oscillations with no lung ventilation. (B) Evenly spaced single lung breaths. (C) Single and doublet lung breaths. (D) Small episodes of variable length. (E) Large episodes of variable length. (F) Continuous lung ventilation. (G) Traces of airflow at the nares and the changes in lung gas PO2 and PCO2 during episodic breathing in Xenopus illustrating the wide swings in respiratory gases associated with episodic breathing. (F) Chatburn and Milsom, unpublished. (G) Redrawn, with permission, from Boutilier RG and Shelton G, 1986 105 by Jacelyn Shu. Figure 19. (A) Traces of the changes in lung (top trace) and buccal (bottom trace) pressure associated with lung ventilation cycles in the toad (Bufo marinus). Ventilation cycles end with an apnea with the lungs inflated. This is followed by a series of deflation breaths, followed by a series of balanced breaths, and then a series of inflation breaths ending in another apnea. (B) Expanded traces of the segments labeled ➀ and ➁ in panel (A). ➀ illustrates three progressive inflation breaths in the inflation sequence while ➁ illustrates the first two breaths in a deflation sequence. (C) The changes in lung pressure (green), buccal pressure (blue), and the discharge in the Xth cranial nerve innervating the glottis (Xl) and the Vth cranial nerve innervating buccal compressors (Vm) during an inflation breath ➀ and a deflation breath ➁. Note that with the inflation breath the buccal compressors are activated before the glottis opens while with the deflation breath, the glottis opens before the buccal compressors are activated. Sanders and Milsom, unpublished. Figure 20. Fictive breathing recorded from in vitro brainstem‐spinal cord preparations of the frog show that while transection at the base of the cerebellum appears to eliminate episodic motor output (A vs. C), when viewed over a longer time scale, episodic behavior is still present, albeit less organized (B vs. D). Panels (E‐G) are Poincaré plots of the interbreath intervals (in seconds) from preparations with the midbrain intact (E) and with the midbrain removed (F and G—different time scales) illustrating the manner in which breaths cluster into episodes both before and after the transection. Chatburn and Milsom, unpublished. Figure 21. Distribution of serotonin‐containing neuroepithelial cells in various species of fish. Serotonergic neuroepithelial cells (NECs, stained green for goldfish and rainbow trout and purple for ricefish) in the filament and lamellae of goldfish (Carasius auratus), rainbow trout (Onchorhynchus nerka) and ricefish (medaka; Oryzias latipes). Tissues were prepared for immunohistochemistry and imaged using confocal microscopy following procedures similar to Jonz et al. 612. Modified from Porteus CS, et al., 2012 920. Figure 22. (A) Schematic sketch of dissected palate of the catfish, Clarias batrachus, showing position of the pseudobranchial neurosecretory cell masses (NSM) in relation to the gill cavity, muscle fascicles (M), aortic arches (I‐III) and carotid labyrinth (CL). E1, first epibranchial arch; P1, first pharyngobranchial arch; GC, gill cavity; E2, second epibranchial arch; P2, second pharyngobranchial arch; E3, third epibranchial arch; P3, third pharynobranchial arch; LDA, lateral dorsal aorta. (B) Anastomotic origin of internal carotid (IC) from labyrinthine vessels. SEM × 9. (C) Spray type origin of small olfactory artery (OA). Buds on OA form additional spray‐type vessels. SEM × 20. (D) Scanning electron micrograph of a vascular resin cast of the carotid labyrinth of Rana catesbeiana. View shows the vascular maze (mz) and origin of the internal carotid artery. (E) The interior surface area of the tegu lizard (Tupinambis merianae) carotid artery is expanded by a lattice‐like array of tissue cords. A bright‐field image (×12 magnification) of the interior of the carotid artery shows folds and cords of tissue of different thicknesses (arrows). The artery distal to the heart is oriented toward the upper left of the image. (A) Modified, with permission, from Gopesh A, 2009 437. (B,C) Reproduced from Olson KR, et al., 1981 850. (D) Reproduced from Reyes C, et al., 2014 976. (E) Reproduced from Reichert MN, et al., 2015 965. Figure 23. Diagram of the central vasculature of the frog (A), turtle (B), and snake (C) showing the location and distribution of putative O2‐sensing cells. (A) Central vasculature of the frog. 5‐HT is shown in red, TH in green, and cells where both neurotransmitters colocalized are shown in yellow. (B) Central vasculature of the turtle showing the location of VAChT cell clusters (putative O2‐sensing cells) in blue and 5HT containing cells in red. Polygonal VAChT cells are shown in purple. (C) Central vasculature of the snake, showing AChT cells in blue and 5‐HT containing cells in red. cl, carotid labyrinth; LAo, left aorta; LCCa, left common carotid artery; LPca, left pulmocutaneous artery; LPA, left pulmonary artery; LSca, left subclavian artery; PA, pulmonary artery; Lica, left internal carotid artery; Leca, left external carotid artery. Reyes and Milsom, unpublished. Figure 24. Multiple factors model of chemosensitive signaling based on findings in LC neurons. In the multiple factors model, an acid stimulus such as hypercapnic acidosis (HA), results in multiple signals, including decreased pHi, decreased pHo, and increased CO2 level. These signals are proposed to affect multiple ion channel targets, including various K+ channels [TASK and tetraethylammonium (TEA)‐sensitive K+ channels] as well as Ca2+ channels. It is the overall effect of these multiple signals affecting multiple channels that determines the magnitude of the final response of the neuron, that is, increased firing rate. Modified from Putnam RW, et al., 2004 931. Figure 25. Traces of respiratory airflow, and inspired CO2 levels from a caiman illustrating the changes in ventilation that occurred during the return to air after breathing 7% CO2. The switch back to breathing air occurred at time 0. Note the increase in ventilation frequency that occurs immediately when CO2 is removed from the inspired gas. Reused from Tattersall GJ, et al., 2006 1114. Figure 26. Representative recordings of opercular displacement, a measure of ventilation amplitude or stroke volume, in adult zebrafish (Danio rerio) before and during exposure to acute hypoxia (water PO2 = 40 mmHg). Panels (A) and (B) show the response pattern of fish exhibiting continuous breathing in normoxia, Panels (C) and (D) show the response pattern of fish exhibiting episodic breathing in normoxia. Note that in both cases the response is an increase in frequency. Zebrafish do not normally increase ventilation amplitude during hypoxia (Table 4). In these examples, breathing amplitude is reduced. Perry, Milsom, Porteus, and Kumai, unpublished. Figure 27. Representative recordings from a single zebrafish (Danio rerio) larva at 4 days postfertilization showing the changes in ventilation frequency (fV; black circles) and O2 consumption (ṀO2; grey circles) during progressive hypoxia. The data indicate that ventilation increases more‐or‐less linearly with decreasing PO2 while ṀO2 remains constant until the critical PO2 (Pcrit) is reached at 19 mmHg (vertical line). Note that the hyperventilation persists after the Pcrit is reached and that the zone of maximal ventilation (unfilled circles) spans the Pcrit. These results demonstrate that ventilation does not necessarily decrease with the onset of the decreased ṀO2 associated with Pcrit. Unpublished data of Mandic M, et al., 2020 727. Figure 28. Changes in ventilation frequency (fV) in adult zebrafish (Danio rerio) exposed to continuous hypoxia (PO2 = 90 mmHg; grey circles) or normoxia (PO2 = 153 mmHg; black circles) for 72 h. At 72 h, all fish were exposed acutely to a more severe level of hypoxia (PO2 = 40 mmHg). Data are shown as means ±1 SEM, N = 6. Significant differences from values at T = 0 (prehypoxia) are indicated by the horizontal line or asterisks (P < 0.05). The initial acute HVR was short‐lived and by 6 h, fV returned to baseline levels. The 3‐day exposure to moderate hypoxia did not affect the ability of zebrafish to hyperventilate when challenged with a more severe level of acute hypoxia. Previously unpublished data from M. Mandic, A. Bailey, and S.F. Perry. Figure 29. The effects of acute hypoxia (PO2 = 55 mmHg; represented by the gray box) on ventilation frequency (fV) in larval zebrafish (Danio rerio) at 4 (N = 10) and 10 (N = 10) days postfertilization (dpf). In larvae at 4 dpf, the hyperventilation [significant differences (P < 0.05) indicated by the horizontal line] was transient whereas the hyperventilation [significant differences (P < 0.05) indicated by the horizontal line] persisted for the entire period of hypoxia in larvae at 10 dpf. Data are shown as means ±1 SEM. Data, with permission, from Mandic M, et al., 2019 728. Figure 30. Representative measurements of buccal (black circles) and fin (grey circles) ventilation frequencies in individual zebrafish (Danio rerio) larvae at (A) 4 days postfertilization (dpf) and (B) 21 dpf. At 4 dpf, acute hypoxia (25 mmHg; horizontal line) caused synchronized increases in buccal and fin ventilation; at 21 dpf, acute hypoxia caused an increase in buccal breathing only. Data, with permission, from Zimmer AM, et al., 2020 1294. Figure 31. Representative recordings of gill ventilation (assessed by measuring opercular displacement) in a single adult zebrafish (Danio rerio) under (A) normocapnic and (B) hypercapnic (PCO2 = 7.5 mmHg) conditions. Note that this particular fish exhibited episodic breathing during normocapnia that persisted during hypercapnia as opercular displacement (a measure of ventilation amplitude) increased with no change in breathing frequency. Unpublished data from S.F. Perry, C. Porteus, Y. Kumai and W.K. Milsom. Figure 32. Representative recordings of gill ventilation (assessed by measuring opercular displacement) in three different adult zebrafish (Danio rerio) before (A,C,E) and during (B,D,F) exposure to high external ammonia [500 μM (NH4)2SO4]. These recordings represent three distinct patterns of ventilatory responses to high external ammonia in zebrafish that shared the common response of increased opercular displacement (a measure of ventilation amplitude). In a first example (panels A and B), episodic breathing persisted with no change in breathing frequency (fV). In a second example (panels C and D), ammonia caused a switch from episodic to continuous breathing. In a third example (panels E and F), fish already exhibiting continuous breathing continued to do so. Unpublished data from S.F. Perry, C. Porteus, Y. Kumai, and W.K. Milsom. Figure 33. Effects on ventilation frequency (fV) of acute increases (A) and decreases (B) of water temperature for a variety of water‐breathing fish species. For references, see Table 5. Note the range of responses and that for many species the change with increasing temperature is larger than the fall with decreasing temperature (hysteresis). Figure 34. Relationships between fold changes in ventilation frequency (fV) and fold changes in the rate of oxygen consumption (ṀO2) for water‐breathing fishes exposed to acute changes in water temperature (A) or acclimated to different temperatures (B). The solid line in each figure is the line of equality. Most points fall below this line, indicating that changes in fV alone are not sufficient to increase ventilation volume to match the change in ṀO2. Thus, ventilatory stroke volume must also increase. (A) Data were drawn, with permission, from Aguiar LH, et al., 2002 14; Azuma T, et al., 1998 37; Berschick P, et al., 1987 74; Fernandes MN, et al., 1995 338; Gehrke PC and Fielder DR, 1988 402; Heath AG and Hughes GM, 1973 476; Johansen K, et al., 1973 591; Keen AN and Gamperl AK, 2012 627; Maricondi‐Massari M, et al., 1998 731; Moffitt BP and Crawshaw LI, 1983 800; Randall DJ and Cameron JN, 1973 945; Zhao Z, et al., 2011 1290; Zhao Z, et al., 2017 1292. (B) Data were drawn, with permission, from Aguiar LH, et al., 2002 14; Campagna CG and Cech Jr. JJ, 1981 176; Cech Jr. JJ and Wohlschlag DE, 1982 188; Fernandes MN and Rantin FT, 1989 340; Maricondi‐Massari M, et al., 1998 731; Spitzer KW, et al., 1969 1081; Zhao Z, et al., 2011 1290; Zhao Z, et al., 2018 1291; Zhao Z, et al., 2017 1292. Figure 35. Panel (A) presents the effects of acclimation temperature on ventilation frequency (fV) for a variety of water‐breathing fishes (for references, see Table 7). The effects on fold change in ventilation volume (V̇w; panel A) and ventilation frequency (fV; panel B) of acute temperature changes versus acclimation to different temperatures were explored for five species for which data were available; Amur sturgeon Acipenser schrenckii 1292, grass carp Ctenopharyngodon idellus 1291, Piaractus mesopotamicus 14, Nile tilapia Oreochromis niloticus 731, and bluegill sunfish Lepomis macrochirus 475,1081. See text for details. Data, with permission, from Zhao Z, et al., 2017 1292; Zhao Z, et al., 2018 1291; Aguiar LH, et al., 2002 14; Maricondi‐Massari M, et al., 1998 731; Heath AG, 1973 475; Spitzer KW, et al., 1969 1081. Figure 36. (A) Parasagittal section of the bullfrog brain approximately 0.63 mm from the midline showing the location of nucleus isthmi (Is) and locus coeruleus (LC). Left is rostral and right is caudal. (B) Ventilation (V̇E) of control, vehicle, and ibotenic acid lesioned (IBO) groups exposed to hypoxia or hypercarbia. *Indicates a significant effect of hypercarbia compared to the normocarbic value, #indicates a significant difference between the control and IBO groups, +indicates a significant difference between the vehicle and IBO groups. (C1) V̇E of the control, vehicle, peri, and 6‐OHDA groups exposed to normocarbia and hypercarbia. *Indicates significant effect of hypercarbia compared with the normocarbic value, +Indicates significant differences between 6‐OHDA and all other groups during hypercarbia. (C2) V̇E after microinjection of mock CSF of pH 7.2, 7.4, 7.6, 7.8 (control pH value), and 8.0. NG means V̇E before the injection (no injection group). *Significant difference from pH 7.8 group, #significant difference from pH 8.0 group; +significant difference from NG group. (C3) Firing rate dose‐response curve during extracellular acidification was achieved by increasing the percentage of CO2 from control (1.3%; ΔpH 0) to HA (1.7%, n = 4; 1.8%, n = 6; 1.9%, n = 6; 2%, n = 8; 2.1%, n = 6; 3%, n = 8; 5%, n = 21; ΔpH 0.07, 0.085, 0.10, 0.115, 0.165, 0.285, and 0.42, respectively). Hypercapnia induced significant firing rate increases across the range of ΔpH (one‐sample t‐test; P < 0.05). Small changes in extracellular pH caused half‐maximal firing at Δ 0.12 pH units; h = 2.1. (A) Adapted, with permission, from Gargaglioni LH, et al., 2007 389; Hoffman A, 1973 518. (B) Adapted, with permission, from Gargaglioni LH, et al., 2002 388. (C1,C2) Based on Noronha‐de‐Souza CR, et al., 2006 841. (C3) Based on, with permission, Santin JM, et al., 2013 1021. Figure 37. (A) The effect of changes in the inspired O2 concentration on total inspired ventilation of Rhinella scheneideri at temperatures of 15, 25, and 32°C, * and ** indicates P < 0.05 and P < 0.001, respectively. (B) The effect of changes in pH of mock CSF on total inspired ventilation of Rhinella scheneideri at temperatures of 15, 25, and 35°C. Note the reduction in response to both stimuli at decreasing temperature. (A) Reused, with permission, from Kruhoffer M, et al., 1987 663. (B) Reused, with permission, from Branco LG, et al., 1993 121. Figure 38. Ventilatory responses to hypoxia compared to the positions of O2 dissociation curves for Chrysemys blood at 10, 20, and 30°C. Values for P50 and in vivo points are indicated. Note that the blood is somewhat desaturated for normoxic turtles at all temperatures. Also note that ventilatory responses occur in hypoxic turtles when arterial PO2 approaches P50 of the blood which is reduced with decreasing temperature. As temperature falls, blood remains saturated at lower levels of PO2, and ventilation is reduced. Based on data from Glass ML, Boutilier RG and Heisler N, 1983 418. Figure 39. Traces depicting the tidal volume [Tv (mL)] and airflow (mL/s) at the nares and the PO2 and PCO2 of end‐expiratory gas in a tegu lizard at 35 and 40°C. At 40°C, this lizard exhibits very high frequency buccal movements that serve to flush the pharyngeal cavity, but that are not associated with gas exchange as can be seen from the end‐expired gases. Reused, with permission, from Abe AS, 1991 4.