Oxygen and carbon dioxide sensitivity of ventilation in amphibious crabs, Cardisoma guanhumi, breathing air and water

Comparative Biochemistry and Physiology, Part A 138 (2004) 111 117 www.elsevier.com/locate/cbpa Oxygen and carbon dioxide sensitivity of ventilation in amphibious crabs, Cardisoma guanhumi, breathing air and water Andrew T. Gannon a, *, Raymond P. Henry b a Department of Biology, Birmingham-Southern College, Box 549022, Birmingham, AL 35254, USA b Department of Biological Sciences, Auburn University, Auburn, AL, USA Received 2 October 2003; received in revised form 9 March 2004; accepted 11 March 2004 Abstract Amphibious crabs, Cardisoma guanhumi, were acclimated to breathing either air or water and exposed to altered levels of oxygen and/or carbon dioxide in the medium. Hypercapnia (22, 36 and 73 torr CO 2 ) stimulated a significant hypercapnic ventilatory response (HCVR) in both groups of crabs, with a much greater effect on scaphognathite frequency (Df SC = + 700%) in air-breathing crabs than water-breathing crabs (Df SC = + 100%). In contrast, hyperoxia induced significant hypoventilation in both sets of crabs. However, simultaneous hyperoxia and hypercapnia triggered a greater than 10-fold increase in f SC in air-breathing crabs but no change in water-breathing crabs. For water-breathing crabs hypoxia simultaneous with hypercapnia triggered the same response as hypoxia alone bradycardia ( 50%), and a significant increase in f SC at moderate exposures but not at the more extreme levels. The response of air-breathing crabs to hypoxia concurrent with hypercapnia was proportionally closer to the response to hypercapnia alone than to hypoxia. Thus, C. guanhumi were more sensitive to ambient CO 2 than O 2 when breathing air, characteristic of fully terrestrial species, and more sensitive to ambient O 2 when breathing water, characteristic of fully aquatic species. C. guanhumi possesses both an O 2 - and a CO 2 -based ventilatory drive whether breathing air or water, but the relative importance switches when the respiratory medium is altered. D 2004 Elsevier Inc. All rights reserved. Keywords: Hypercapnia; Hyperoxia; Hypoxia; Ventilation; Bimodal breathing; Crustacean 1. Introduction Fluctuations in oxygen and carbon dioxide levels in the respiratory medium can be potent stimulators of ventilation in many animals. A generally accepted principle of comparative respiratory physiology is that ventilation of aquatic animals is more sensitive to oxygen than carbon dioxide, but ventilation of terrestrial animals is primarily CO 2 -driven. This is assumed to be due to the differences in solubility of these two gases in air and water. Oxygen and CO 2 have equal solubility in air, but O 2 is 28 times less soluble in water than CO 2 (Dejours, 1975; Cameron, 1989). Because of this, water is considered to be an O 2 -poor medium, and the ventilatory drive of aquatic animals is set at a relatively high rate to extract O 2. Due to the high ventilatory drive and high solubility of CO 2 in water, aquatic CO 2 excretion occurs at maximal rates and is not limited by changes in * Corresponding author. Tel.: +1-205-226-4899; fax: +1-205-226-3078. E-mail address: agannon@bsc.edu (A.T. Gannon). ventilation. Air is a relatively O 2 -rich medium (Dejours, 1975) and ventilatory rates in terrestrial animals are generally lower, as a result of O 2 availability and partly to reduce evaporative water loss. Although the solubility of CO 2 in air and water is approximately the same, because of the lower ventilation rates, CO 2 excretion from blood to air is considered physically more difficult. Air-breathing animals consequently have higher levels of CO 2 in their blood, and have evolved a CO 2 -sensitive ventilatory drive to control extracellular fluid PCO 2 /ph through changes in CO 2 excretion. Oxygen uptake remains maximal and unaffected by these changes in ventilation (Dejours, 1975). This general pattern has been confirmed in crabs. Ventilation in aquatic crabs is primarily controlled by O 2 with less sensitivity to CO 2 levels (reviewed in Taylor, 1982; McMahon and Wilkens, 1983). For example, the fully aquatic blue crab (Callinectes sapidus) does not alter ventilation significantly in response to elevated CO 2 but increases ventilation 37% in response to a reduction in O 2 (Batterton and Cameron, 1978). Ventilation in terrestrial crabs, however, is highly CO 2 -sensitive (see review by 1095-6433/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.cbpb.2004.03.008

112 A.T. Gannon, R.P. Henry / Comparative Biochemistry and Physiology, Part A 138 (2004) 111 117 McMahon and Burggren, 1988). For example, Gecarcinus lateralis and Birgus latro, two fully terrestrial species, hyperventilate strongly when exposed to hypercapnia but have a reduced or non significant response to hypoxia (Cameron and Mecklenburg, 1973; Cameron, 1975). Interestingly, there are some exceptions to this general paradigm. Crayfish have been shown to have a ventilatory response to aquatic hypercapnia (Massabuau and Burtin, 1985), and the terrestrial hermit crab, Coenobita clypeatus, showed a strong ventilatory response to hypoxia but responded to hypercapnia only at severe (50% CO 2 ) levels (McMahon and Burggren, 1979). This anomalous response may occur because this crab uses water, retained in its shell, for ventilating in air (Burggren, 1992) in such a way that it may act as an intermediate respiratory medium between the air and the gills, especially for CO 2 excretion (Henry, 1994). Bimodal breathing crabs, those that breathe both air and water, have been shown to have both O 2 - and CO 2 -sensitive ventilation. It has been suggested that the importance of CO 2 -sensitive ventilatory drive may correlate with the degree of dependence on air as a respiratory medium (McMahon and Burggren, 1988). The bimodally breathing crab, Cardisoma guanhumi (Latreille) alters ventilation in response to hypercapnia (Pinder and Smits, 1993) and hypoxia (Herreid et al., 1979). The Australian freshwater/ land crab, Holthuisana transversa responds to smaller changes in CO 2 than O 2 during air breathing, leading Greenaway et al. (1983b) to conclude that ventilation in this crab during its terrestrial phase is normally controlled by PCO 2. This crab also shows sensitivity to hypoxia when breathing water (Greenaway et al., 1983a). The extant bimodally breathing crabs could be viewed as reflections of stages in the evolution of terrestriality, representing various degrees of efficiency of air-breathing structures and devolution of the gills (McMahon and Burggren, 1988). In this context, O 2 -sensitive ventilatory drive would be considered primitive and CO 2 -sensitive drive more derived. Bimodally breathing crabs might also be considered a distinct group with specific adaptations for existing at the interface of air and water (Henry, 1994; Morris, 2002). In either case, this group might be expected to have retained both O 2 - and CO 2 -sensitive control mechanisms, with the added adaptation of being able to switch between them when switching media. The responses of bimodally breathing crabs to hypoxia and hypercapnia have been measured separately (Greenaway et al., 1983a,b; O Mahoney and Full, 1984), but they have not been compared quantitatively or simultaneously. Altering ambient CO 2 and O 2 levels simultaneously such that the two gases should stimulate a change in ventilatory frequency in the opposite direction (e.g., hypercapnia and hyperoxia in air) provides a framework for examining the role of the ventilatory control mechanism in air and water. We sought to determine the relative importance of fluctuations in oxygen versus carbon dioxide to the scaphognathite rate of the bimodally breathing crab C. guanhumi, while it was breathing air or water, and thus determine the importance and contribution of each ventilatory control mechanism in each medium. 2. Materials and methods 2.1. Animal care Adult, intermolt male and female C. guanhumi (mean mass = 139.9 g, range = 84.0 215.51 g) were obtained from South Florida in May and maintained at Auburn University, Alabama for up to 3 months in large sand-filled terraria with PVC burrows and 15 ppt seawater for drinking. Crabs were held under normoxic and normocapnic conditions on a 12:12 h light/dark schedule and were acclimated to breathing either air or water for a minimum of 4 (most >7) days before experimentation. Air-acclimated crabs were placed in a sand-filled aquarium with access to only enough 15 ppt water for hydration but not submersion. We did not attempt to forcibly remove residual water from the crab gill chambers during air acclimation because initial attempts to do this resulted in injury to crabs. These crabs normally carry small amounts of water in their gill chambers while breathing air, so this is the natural state. Water-acclimated crabs were forced to breathe only water for at least 1 week by submersion in an aquarium containing 15 ppt water. Air-acclimated crabs were fed mixed vegetables ad lib during acclimation, but food was withheld for at least 24 h prior to experimentation. Crabs were not given food during acclimation to water-breathing because they rarely feed while in water. Some crabs were used in both parts of the experiment with a 7-day recovery in between. Crabs were kept at temperatures of 24 26 jc prior to and during experimentation. 2.2. Cardioventilatory measurements The impedance technique was used to measure heart rate ( f H ) and scaphognathite rate ( f SC ) (Taylor, 1976). For f H, 28-gauge copper wires were implanted through holes drilled in the dorsal carapace on either side of the heart and held in place with cyanoacrylate adhesive and rubber patches. To measure ventilation rate (i.e., the rate of pumping of the scaphognathite f SC ), one wire lead was inserted into the opening of the excurrent canal of each of the gill chambers and a second lead was implanted through a hole drilled in the dorsal carapace directly above the first lead in each gill chamber. These leads were connected to impedance converters (UFI model 2991) and the signal was recorded on a recording oscillograph (Astro Med Dash IV). Crabs were given 16 24 h to recover from surgery before measurements were taken. During recovery the crabs were held in 4-l Plexiglas chambers supplied with a continuous stream of either aerated 15 ppt water or humidified air. Scaphognathite rate could be recorded for only one gill

A.T. Gannon, R.P. Henry / Comparative Biochemistry and Physiology, Part A 138 (2004) 111 117 113 chamber at a time. Since left and right scaphognathites are usually coupled in their movements, patterns recorded for one side can be assumed to be representative of both sides (Burggren et al., 1985). We have reported the values we recorded for left or right scaphognathite as f SC. We have used f SC as an index of ventilation rate since there is an approximately linear relationship between these two parameters over normal physiological ranges (McMahon and Wilkens, 1983). 2.3. Protocol Prior to experimentation, crabs were placed in 500-ml Plexiglas respirometers for a 12-h acclimation period with either normoxic, normocapnic air or water passed through. In these respirometers, the crabs were not restrained but had little room for movement. Burggren et al. (1985) reported that f H and f SC measured by impedance were not significantly altered between restrained and free-ranging crabs. At the end of the acclimation period, f H and f SC values were recorded for 1 h. The gas levels of the air or water passing through the respirometer were then manipulated so that each crab was subjected to a series of exposures to altered CO 2 and O 2 levels, with recovery periods of normoxia and normocapnia interspersed between. Recovery periods, almost all 60 min (range 60 110 min), lasted until f H and f SC values were estimated to have returned to initial levels ( F 10%). In an earlier study (O Mahoney and Full, 1984), recovery of C. guanhumi respiratory parameters from hypercapnic and hypoxic exposures was complete within an hour. Gas mixtures were presented in the following order: hypercapnia (7, 22, 36 and 73 torr O 2 ; all with 147 torr O 2, balance N 2 30 min each), recovery, hypoxia (73 and 22 torr O 2 ; balance N 2 30 min each), recovery, hypoxic hypercapnia (73 torr O 2 with 22 torr CO 2 and 22 torr O 2 with 73 torr CO 2 ; both balanced with N 2 ), recovery, hyperoxia (284 torr O 2 ; balance N 2 30 min), recovery and hyperoxic hypercapnia (284 torr O 2 and 73 torr CO 2 ; balance N 2 30 min). Gas mixtures, regulated with a Gas Mixing Flowmeter (Cameron Instruments), were humidified as they were pumped directly through the respirometer at 300 ml/min for the air acclimated crabs. For waterbreathing crabs, the mixed gas was pumped to a bubble diffuser in a 1m-tall Plexiglas equilibration column to allow water passing through at about 600 ml/min to approach saturation at each gas partial pressure; however, it was not possible to achieve the same precision with water samples as with gas. Oxygen and CO 2 levels in samples from the respirometers were checked with O 2 (Radiometer E5046-0) or CO 2 (Cameron Instruments E201) electrodes coupled to a Radiometer PHM 72 Digital Acid Base Analyzer. The levels of hypercapnia and hypoxia used were within the range measured in a field study of C. guanhumi burrow gas tension extremes by Pinder and Smits (1993). Heart and scaphognathite rates were counted as the mean of the last 5 10 min of each period. Values for crabs exposed to different gas mixes were compared to the control or initial rates recorded from the same crabs exposed to normoxic, normocapnic air at the end of the previous recovery period with a General Linear Model One-Way Repeated-Measures ANOVA using a priori tests of withinsubjects contrasts (SPSS 11.0 for Windows). The assumption of homogeneity of variances of differences in the data was validated with Mauchly s test of sphericity. 3. Results 3.1. Air-breathing crabs The initial (normoxic, normocapnic) scaphognathite rates for air-breathing crabs (Figs. 1 3) were consistently around 10 bpm. Mean normoxic, initial (control) heart rates showed a little more variability (Figs. 1 3), ranging from 75 to 82 bpm. Exposure to hypercapnia had no effect on the heart rate of air-breathing crabs, but stimulated significant ( p < 0.05) hyperventilation at all but the lowest CO 2 exposure (Fig. 1). In a similar fashion, heart rate was not significantly affected by moderate hypoxia (Fig. 2), but at the more extreme level of hypoxia (22 torr O 2 ), ventilation rate increased dramatically ( p < 0.001). During the simultaneous hypoxia and hypercapnia exposures (Fig. 2), ventilation rate decreased significantly ( p < 0.05) at the moderate exposure (73 torr O 2 /22 torr CO 2 ) and increased significantly ( p < 0.01) at the more extreme exposure (22 torr O 2 /73 torr CO 2 ). Heart rate at the moderate exposure was significantly less than the initial values ( p < 0.05), but at the more extreme hypoxia/ hypercapnia exposure, heart rate was not significantly different from initial ( p = 0.053). The already low initial scaphognathite rates (10 bpm) seen in air-breathing crabs decreased significantly ( p < 0.05) during hyperoxia (Fig. 3). Exposure to simultaneous hyperoxia and hypercapnia stimulated a 12-fold increase in scaphognathite rate ( p < 0.05), but had no effect on heart rate (Fig. 3). 3.2. Water-breathing crabs Mean normoxic heart rates for water-breathing crabs were consistent, ranging from 76 to 80 bpm (Figs. 4 6) and similar to the resting values of air-breathing crabs (range 75 82 bpm; Figs. 1 3). However, the mean normoxic scaphognathite rates for the water-breathing crabs (Figs. 4 6) were much greater than for air-breathing crabs (Figs. 1 3), and more variable, ranging from 66 to 88 bpm. Hypercapnia had no effect on heart rate of water-breathing crabs but stimulated hyperventilation ( p < 0.05), at all levels of increased CO 2 (Fig. 4). Exposure to hypoxia

114 A.T. Gannon, R.P. Henry / Comparative Biochemistry and Physiology, Part A 138 (2004) 111 117 Fig. 1. Mean heart rate and scaphognathite rate ( + 1 S.E.M.) for airbreathing crabs (C. guanhumi) during 30-min exposures to increasing levels of CO 2 with 146 torr O 2, balance N 2. Asterisks indicate values significantly ( p < 0.05) different from initial values recorded during normocapnic, normoxic exposure (repeat-measures ANOVA); n = 8. stimulated hyperventilation (Fig. 5) at 73 torr O 2 ( p < 0.05), but there was no significant effect at 22 torr O 2. The combination of hypoxia and hypercapnia produced the same response as hypoxia. During hypoxia there was significant bradycardia ( p < 0.001) only at the lowest level of hypoxia, Fig. 3. Mean heart rate and scaphognathite rate ( + 1 S.E.M.) for airbreathing crabs (C. guanhumi) during 30-min exposures to hyperoxia (284 torr O 2 ) and combined hyperoxia/hypercapnia (284 torr O 2 and 73 torr CO 2 ). Asterisks indicate values significantly ( p < 0.05) different from initial values recorded during normoxic, normocapnic exposure (repeat-measures ANOVA); n =7. but during the combined hypoxia/hypercapnia exposures there was significant bradycardia ( p < 0.05) at both the low and moderate exposures. Exposure to hyperoxia and simultaneous hyperoxia/hypercapnia did not affect heart rate of the water-breathing Fig. 2. Mean heart rate and scaphognathite rate ( + 1 S.E.M.) for airbreathing crabs (C. guanhumi) during 30-min exposures to decreasing levels of O 2 and simultaneous hypoxia/hypercapnia. Moderate hypoxia indicates 73 torr O 2. Extreme hypoxia indicates 22 torr O 2. Moderate hypoxia/hypercapnia indicates 73 torr O 2 and 22 torr CO 2. Extreme hypoxia/hypercapnia indicates 22 torr O 2 and 73 torr O 2. Asterisks indicate values significantly ( p < 0.05) different from initial values recorded during normoxic, normocapnic exposure (repeat-measures ANOVA); n = 8. Fig. 4. Mean heart rate and scaphognathite rate ( + 1 S.E.M.) for waterbreathing crabs (C. guanhumi) during 30-min exposures to increasing levels of CO 2. Asterisks indicate values significantly ( p < 0.05) different from initial values recorded during normocapnic exposure (repeat-measures ANOVA); n =6.

A.T. Gannon, R.P. Henry / Comparative Biochemistry and Physiology, Part A 138 (2004) 111 117 115 Fig. 5. Mean heart rate and scaphognathite rate ( + 1 S.E.M.) for waterbreathing crabs (C. guanhumi) during 30-min exposures to decreasing levels of O 2 and simultaneous hypoxia/hypercapnia. Moderate hypoxia indicates 73 torr O 2. Extreme hypoxia indicates 22 torr O 2. Moderate hypoxia/hypercapnia indicates 73 torr O 2 and 22 torr CO 2. Extreme hypoxia/hypercapnia indicates 22 torr O 2. and 73 torr O 2. Asterisks indicate values significantly ( p < 0.05) different from initial values recorded during normoxic exposure (repeat-measures ANOVA); n = 6. crabs (Fig. 6); however, hyperoxia did induce a significant hypoventilation ( p < 0.005) that was abolished by the simultaneous presence of hypercapnia. Fig. 7. HCVR of air- and water-breathing crabs (C. guanhumi) expressed as percent change in scaphognathite rate at different levels of CO 2 exposure; n =6 8. 3.3. Comparison of water-breathing and air-breathing crabs Ventilatory responses to hypercapnia have often been expressed as the hypercapnic ventilatory response (HCVR) or the percent change in ventilation rate relative to the % CO 2 of the inspired air (Fig. 7). The air-breathing crabs had a much steeper slope for this relationship, indicating greater sensitivity to CO 2. 4. Discussion Fig. 6. Mean heart rate and scaphognathite rate ( + 1 S.E.M.) for waterbreathing crabs (C. guanhumi) during 30-min exposures to hyperoxia (284 torr O 2 ) and combined hyperoxia/hypercapnia (284 torr O 2 and 73 torr CO 2 ). Asterisks indicate values significantly ( p < 0.05) different from initial values recorded during normoxic, normocapnic exposure (repeat-measures ANOVA); n =6. Air breathing in crabs is generally associated with reduced gill surface area, ostensibly to reduce evaporative water loss (Gray, 1957). Air breathing is also correlated with slower ventilatory frequencies, elevated hemolymph O 2 carrying capacity, elevated hemolymph PCO 2 and total CO 2 and a CO 2 -sensitive ventilatory drive (Innes and Taylor, 1986; Burggren, 1992). Additionally, the branchiostegite lining of the gill chamber of air-breathing crabs has developed into well-vascularized lungs that function primarily in the uptake of O 2 (Greenaway and Farrelly, 1984; Farrelly and Greenaway, 1987). C. guanhumi possesses the most primitive and simplest anatomical lung among airbreathing crabs (reviewed by Henry, 1994). These structural and functional adaptations can be linked to the differences in O 2 and CO 2 solubility between air and water, primarily the relative ease of extracting O 2 from air and the relative difficulty of excreting CO 2 from blood to air. In contrast, water-breathing crabs have been shown to have an O 2 -sensitive ventilatory drive (Taylor, 1982). Gannon et al. (2001) reported that the bimodally-breathing crab, C. guanhumi, exhibited respiratory and acid base adaptations typically found in both fully aquatic and fully terrestrial

116 A.T. Gannon, R.P. Henry / Comparative Biochemistry and Physiology, Part A 138 (2004) 111 117 crabs. In this study, we have focused on the effect of altered levels of inspired O 2 and CO 2 on ventilatory drive in C. guanhumi breathing air and water and contrasted the magnitude of the HCVR and the responses to simultaneous hyperoxia and hypercapnia, and simultaneous hypoxia and hypercapnia. Although both air- and water-breathing crabs exhibited a significant HCVR, it was much stronger in air-breathing crabs (DV = 700% at 73 torr CO 2 and 500% at 36 torr or 5% CO 2 ) than in water-breathing crabs (DV = 100% at 73 torr CO 2 and 50% at 36 torr CO 2 ) as indicated by the slopes of the HCVR curves (see Fig. 7). The magnitude of the HCVR in response to 5% CO 2 in C. guanhumi is the largest reported for air-breathing animals, making it the most CO 2 -sensitive ventilatory drive known (Williams et al., 1995). Furthermore, C. guanhumi, which spends most of its time in relatively shallow, water-filled burrows, does not show the blunted ventilatory response to elevated CO 2 that is ubiquitous among burrowing endotherms (Boggs et al., 1984) despite burrow CO 2 tensions of up to 90 torr (Pinder and Smits, 1993). It would be interesting to measure the HCVR of Cardisoma that have sealed themselves in their burrows for months during the dry season. The pattern of hyperventilation reported here is similar to that found by Pinder and Smits (1993) in response to small (2%) stepwise increases in CO 2 with an approximate 20 bpm increase in f SC with each 2% increase in CO 2. In that study, a 10% increase in CO 2 tension caused apnea which may be related to an irritant response, by so-called defense-receptors commonly found in lower vertebrates (Jones and Milsom, 1982). The results presented here clearly show that both O 2 - and CO 2 -sensitive ventilatory drives exist in C. guanhumi. ACO 2 -sensitive drive is present in water-breathing crabs as evidenced by an increase in scaphognathite frequency at all hypercapnic CO 2 tensions. A basal CO 2 -sensitive ventilatory drive appears to be common among aquatic invertebrates, especially crustaceans (Massabuau and Burtin, 1985). The more typical O 2 -sensitive drive is also present and may be more dominant in crabs breathing water. In the present study, the response of water breathing crabs to simultaneous hypoxia and hypercapnia was virtually identical to the response to hypoxia alone. In these same crabs hyperoxia reduced f SC by 33%, indicating the importance of the O 2 drive and during simultaneous hyperoxia and hypercapnia, the 100% increase in f SC seen in hypercapnia alone did not occur. The O 2 -sensitive drive persists in crabs breathing air, as hypoxia results in an increase in f SC, but the CO 2 -sensitive drive appears to be dominant in this respiratory medium. The increase in f SC in response to hypercapnia is dramatic and unaffected by hyperoxia. The expected pattern for water-breathing animals exposed to hypoxia is hyperventilation (e.g., Batterton and Cameron, 1978). We found a significant hyperventilation only at the moderate hypoxic exposure. The lack of response to severe hypoxia may be more indicative of the onset of general metabolic depression, which has been shown to be a response to extreme levels of physiological stress in this species (Wood et al., 1986). Other reports have presented similar responses to hypoxia in C. guanhumi and other species (O Mahoney and Full, 1984; Greenaway et al., 1983a). Metabolic depression may also explain the observations of bradycardia under some hypoxic treatments. Terrestrial crabs are generally expected to have CO 2 - sensitive ventilatory drive, and some studies have found no ventilatory response to O 2 (hypoxia) in land crabs (C. guanhumi and G. lateralis; O Mahoney and Full, 1984; Cameron, 1975). Other studies have found smaller magnitude ventilatory responses to hypoxia (B. latro, Cameron and Mecklenburg, 1973; C. clypeatus, McMahon and Burggren, 1979; H. transversa, Greenaway et al., 1983b). This is consistent with the hyperventilation that we recorded during hypoxia in air-breathing crabs and the proportionally smaller increase in ventilation volume of air-breathing C. guanhumi in hypoxia reported by Herreid et al. (1979). It is not clear how much O 2 sensitivity has generally been retained as terrestrial crabs have evolved highly CO 2 sensitive ventilatory drive; however, possession of both drives would be beneficial for crabs normally exposed to fluctuations in both O 2 and CO 2. Radiotelemetry monitoring of C. guanhumi in the field in their natural burrows (Pinder and Smits, 1993) indicated that the crabs could be exposed to extreme hypercapnia [PCO 2 up to 60 (gas) and 90 (water) torr) and extreme hypoxia (PO 2 <20 torr (water)]. The most adaptive response for bimodally breathing crabs such as C. guanhumi might be to adjust the sensitivity of ventilatory drive to be most sensitive to CO 2 when the crab is acclimated to breathing air and to be most sensitive to O 2 when the crab is acclimated to breathing water. In summary, C. guanhumi fits the profile of a bimodally breathing crab (Henry, 1994), having both an O 2 - and a CO 2 -sensitive ventilatory drive that allow it to regulate its cardioventilatory responses to changing gas tensions in both air and water. The presence of both mechanisms of ventilatory control allows this species to exploit a unique niche at the interface of air and water. Acknowledgements We are grateful to Michelle Bilger and Stacy Mote for care and feeding of crabs. Thanks to Sherry Reed and Jeff Shimonski for providing the crabs and thanks to Francesca Gross, Pamela Hanson and Katie Williamson for reading the manuscript and to Rick McCallum for statistical advice. Supported by NSF IBN 93-04844 and IBN 97-27835 to RPH and an NSF Research Opportunity Award to RPH and ATG.

A.T. Gannon, R.P. Henry / Comparative Biochemistry and Physiology, Part A 138 (2004) 111 117 117 References Batterton, C.V., Cameron, J.N., 1978. Characteristics of resting ventilation and response to hypoxia, hypercapnia and emersion in the blue crab, Callinectes sapidus (Rathbun). J. Exp. Zool. 203, 403 418. Boggs, D.F., Kilgore Jr., D.L., Birchard, G.F., 1984. Respiratory physio logy of burrowing mammals and birds. Comp. Biochem. Physiol. A77, 1 7. Burggren, W.W., 1992. Respiration and circulation in land crabs: novel variations on the marine design. Am. Zool. 32, 417 427. Burggren, W.W., Pinder, A., McMahon, B., Wheatly, M., Doyle, M., 1985. Ventilation, circulation and their interactions in the land crab, Cardisoma guanhumi. J. Exp. Biol. 117, 133 154. Cameron, J.N., 1975. Aerial gas exchange in the terrestrial Brachyura Gecarcinus lateralis and Cardisoma guanhumi. Comp. Biochem. Phy siol. A52, 129 134. Cameron, J.N., 1989. The Respiratory Physiology of Animals. Oxford Univ. Press, New York. Cameron, J.N., Mecklenburg, T.A., 1973. Aerial gas exchange in the coconut crab, Birgus latro with some notes on Gecarcoidea lalandii. Res. Physiol. 19, 245 261. Dejours, P., 1975. Principles of Comparative Respiratory Physiology. Elsevier/North Holland Press, New York, USA. Farrelly, C.A., Greenaway, P., 1987. The morphology and vasculature of the lungs and gills of the soldier crab, Mictyris longicarpus Latreille. J. Morphol. 193, 285 304. Gannon, A.T., Arunakul, N., Henry, R.P., 2001. Respiratory, cardiovascular, and hemolymph acid base changes in the amphibious crab, Cardisoma guanhumi, during immersion and emersion. Mar. Freshw. Behav. Physiol. 34, 73 92. Gray, I.E., 1957. A comparative study of the gill area of crabs. Biol. Bull. 112, 34 42. Greenaway, P., Farrelly, C.A., 1984. The venous system of the terrestrial crab Ocypode cordimanus (Desmarest 1825) with particular reference to the vasculature of the lungs. J. Morphol. 181, 133 142. Greenaway, P., Bonaventura, J., Taylor, H.H., 1983a. Aquatic gas exchange in the freshwater/land crab, Holthuisana transversa. J. Exp. Biol. 103, 225 236. Greenaway, P., Taylor, H.H., Bonaventura, J., 1983b. Aerial gas exchange in Australian freshwater/land crabs of the genus Holthuisana. J. Exp. Biol. 103, 237 251. Henry, R.P., 1994. Morphological, behavioral and physiological characterization of bimodal-breathing crustaceans. Am. Zool. 34, 205 215. Herreid, C.F., O Mahoney, P.M., Shah, G.M., 1979. Cardiac and respiratory response to hypoxia in the land crab, Cardisoma guanhumi (Latreille). Comp. Biochem. Physiol. A63, 145 151. Innes, A.J., Taylor, E.W., 1986. The evolution of air-breathing in crustaceans: a functional analysis of branchial, cutaneous and pulmonary gas exchange. Comp. Biochem. Physiol. A85, 621 637. Jones, D.R., Milsom, W.K., 1982. Peripheral receptors affecting breathing and cardiovascular function in non-mammalian vertebrates. In: Butler, P.J. (Ed.), Control and Co-ordination of Respiration and Circulation. J. Exp. Biol., vol. 100. The Company of Biologists, Cambridge, UK, pp. 51 59. Massabuau, J.C., Burtin, B., 1985. Ventilatory CO 2 reflex response in hypoxic crayfish Astacus leptodactylus acclimated to 20 jc. J. Comp. Physiol. B156, 115 118. McMahon, B.R., Burggren, W.W., 1979. Respiration and adaptation to the terrestrial habitat in the land hermit crab, Coenobita clypeatus. J. Exp. Biol. 79, 265 281. McMahon, B.R., Burggren, W.W., 1988. Respiration. In: Burggren, W.W., McMahon, B.R. (Eds.), Biology of the Land Crabs. Cambridge Univ. Press, New York, pp. 249 297. McMahon, B.R., Wilkens, J.L., 1983. Ventilation, perfusion and oxygen uptake. In: Mantel, L.H., Bliss, D. (Eds.), The Biology of Crustacea, vol. 5. Academic Press, New York, pp. 290 372. Morris, S., 2002. The ecophysiology of air-breathing crabs with special reference to Gecarcoidea natalis. Comp. Biochem. Physiol. B131, 559 570. O Mahoney, P.M., Full, R.J., 1984. Respiration of crabs in air and water. Comp. Biochem. Physiol. A79, 275 282. Pinder, A.W., Smits, A.W., 1993. The burrow microhabitat of the land crab Cardisoma guanhumi: respiratory/ionic conditions and physiological response of crabs to hypercapnia. Physiol. Zool. 66, 216 236. Taylor, A.C., 1976. The respiratory responses of Carcinus maenas to declining oxygen tension. J. Exp. Biol. 65, 309 322. Taylor, E.W., 1982. Control and co-ordination of ventilation and circulation in crustaceans: responses to hypoxia and exercise. J. Exp. Biol. 100, 289 319. Williams Jr., B.R., Boggs, D.F., Kilgore Jr., D.L., 1995. Scaling of hypercapnic ventilatory responsiveness in birds and mammals. Respir. Physiol. 99, 313 319. Wood, C.M., Boutilier, R.G., Randall, D.J., 1986. The physiology of dehydration stress in the land crab, Cardisoma carnifex: respiration, ionoregulation, acid base balance and nitrogenous waste excretion. J. Exp. Biol. 126, 271 296.