Daily and seasonal rhythms in the respiratory sensitivity of red-eared sliders (Trachemys scripta elegans)

Transcription

1 3339 The Journl of Experimentl iology 212, Pulished y The Compny of iologists 29 doi:1.1242/je Dily nd sesonl rhythms in the respirtory sensitivity of red-ered sliders (Trchemys script elegns) Ctlin Reyes nd Willim K. Milsom Deprtment of Zoology, University of ritish Columi, Vncouver, C, Cnd V6T 1Z4 uthor for correspondence (reyes@zoology.uc.c) ccepted 16 July 29 SUMMRY The purpose of the present study ws to determine whether the dily nd sesonl chnges in ventiltion nd rething pttern previously documented in red-ered sliders resulted solely from dily nd sesonl oscilltions in metolism or lso from chnges in chemoreflex sensitivity. Turtles were exposed to nturl environmentl conditions over one yer period. In ech seson, oxygen consumption, ventiltion nd rething pttern were mesured continuously for 24 h while turtles were rething ir nd for 24 h while they were rething hypoxic hypercpnic gs mixture (H H). We found tht oxygen consumption ws reduced eqully during the dy nd night under H H in ll sesons except spring. Ventiltion ws stimulted y H H ut the mgnitude of the response ws lwys less t night. On verge, it ws lso less in the winter nd greter in the reproductive seson. The dt indicte tht the dy night differences in ventiltion nd rething pttern seen previously resulted from dily chnges in chemoreflex sensitivity wheres the sesonl chnges were strictly due to chnges in metolism. Regrdless of mechnism, the chnges resulted in longer pnes t night nd in the winter t ny given level of totl ventiltion, fcilitting longer sumergence t times of the dy nd yer when turtles re most vulnerle. Key words: circdin rhythms, circnnul rhythms, respirtory sensitivity, chemoreflexes, turtles. INTRODUCTION Circdin nd circnnul rhythms time iologicl processes so tht specific events tke plce t pproprite times of the dy or yer (Underwood, 1992; Tosini et l., 21; Mortol nd Seifert, 22). Such temporl chnges in physiologicl vriles my e prticulrly crucil for the survivl of temperte species, which re sujected to lrge dily nd sesonl vritions in their environment (chnges in dy length nd temperture cycles). We previously showed tht endogenous circdin nd circnnul oscilltions were present in the metolism nd ventiltion of red-ered sliders (Trchemys script elegns Wied) (Reyes nd Milsom, 29), nd were ccompnied y chnges in rething pttern tht resulted in longer pnes t night nd during colder sesons. We speculted tht y reducing trips to the surfce to rethe, turtles could reduce the cost of locomotion, risk of predtion (Cgle, 195) nd, potentilly, the cost of rething (Vitlis nd Milsom, 1986) during dormncy. Little is known out the mechnisms tht generte the long pnes tht re common to rrhythmic rething in turtles. Endogenous circdin nd circnnul oscilltions in pne durtion could reflect oscilltions in metolic rte (the need to rethe) or in chemoreflex sensitivity (the drive to rethe). Chemoreflexes re thought to e importnt in inititing nd terminting periods of pne (Shelton et l., 1986; Milsom, 199), ut the few studies of circdin nd circnnul vrition in chemoreflex responses, performed in mmmls nd mphiins, hve not lwys oserved rhythms in chemoreflex sensitivity (Mcrthur nd Milsom, 1991; Milsom et l., 1993; Peever nd Stephenson, 1997; Roch nd rnco, 1998; icego-nhs nd rnco, 1999; Mortol nd Seifert, 2; Stephenson et l., 2; icego-nhs et l., 21; Mortol nd Seifert, 22; Seifert nd Mortol, 22; Seifert nd Mortol, 22; Mortol, 24). In the present study, we sought to determine whether dily nd sesonl oscilltions in chemosensitivity per se [defined s the increse in ventiltion of turtles rething hypoxic hypercpnic (H H) gs, fter correcting for ny chnges in metolism, i.e. the chnge in ir convection requirement; CR= V E / V O2, where V E is ventiltion nd V O2 is oxygen consumption] could e involved in producing the dily nd sesonl chnges in ventiltion nd rething pttern previously reported in red-ered sliders (Reyes nd Milsom, 29). We hypothesized tht endogenous rhythms in chemosensitivity contriute to the dily nd sesonl oscilltions oserved in the rething pttern of red-ered sliders, llowing turtles to remin sumerged longer t night nd in the winter, which my hve implictions on their survivl. MTERILS ND METHODS Red-ered slider turtles (Trchemys script elegns Wied) (men mss=1.9±.35 kg, N=8) were otined from commercil suppliers (Lemerger Compny, Wisconsin nd Sullivn Compny, TN, US) nd were housed in semi-nturl pond ( m, contining 3.3 m 3 of wter), where they experienced the environmentl temperture nd nturl photoperiod of Vncouver (C, Cnd) throughout the yer. Two temperture loggers (DS1921 Thermocron iutton, Dlls Semiconductor Corportion, Sunnyvle, C, US) were plced on the sking re ( m) in the shde nd sumerged in the pond to record mient nd wter tempertures, respectively, every hour for one yer. Prior to ech tril, tempertures recorded y the dt loggers were verged to determine the experimentl sesonl temperture. Dily photoperiod nd rinfll were otined from the Meteorologicl Service of Cnd wesite. The holding nd experimentl procedures followed Cndin Council on niml Cre guidelines nd were pproved THE JOURNL OF EXPERIMENTL IOLOGY

2 334 C. Reyes nd W. K. Milsom Tle 1. Temperture nd photoperiod used in the experimentl series Sesonl cues Tretment Seson Photoperiod Temperture 9 h:15 h L:D 9 C 14 h:1 h L:D 14.6 C 16 h:8 h L:D 2.8 C utumn 1 h:14 h L:D 14.7 C Constnt drkness (outdoor turtles) h:24 h L:D 13.6 C y the University of ritish Columi niml Cre Committee (niml cre certificte No. 416). Experiments were crried out over 12 month period (Octoer to Octoer) on unrestrined turtles tht were fsted for seven dys. In ech seson turtles were removed from the semi-nturl pond nd plced in n experimentl tnk ( m) for four dys, t the men sesonl temperture (sme dy nd night wter tempertures) nd photoperiod previling in tht seson (Tle 1). The tnk ws plced inside wooden ox to insulte the turtle from externl stimuli (noise nd light). Full spectrum lights were set with timer to control the light nd drk cycle s we hve previously descried (Reyes nd Milsom, 29), nd turtles were cclimted for two dys to djust to the hndling stress, the experimentl tnk nd the lck of therml cycle. On the first dy following cclimtion turtles were given humidified ir to rethe for 24 h. On the second dy fter cclimtion turtles were given H H gs mixture (8% oxygen 3% CO 2 ) to rethe for 24 h. Wter temperture in the experimentl tnk ws recorded every hour for the length of the experimentl tretment with temperture logger (DS1921 Thermocron iutton, Dlls Semiconductor Corportion) to ensure tht temperture remined stedy during the experimentl tril. To determine the effects of removing ll cues tht could give ny indiction of the time of dy, the sme protocol ws repeted in the utumn, except tht turtles were mintined under constnt drkness for four dys. The wter temperture used for this experiment ws 13.6 C. Dy nd night differences mesured in the utumn under the nturl photoperiod were compred with the corresponding dy nd night vlues mesured under constnt drkness. Mesurement of metolic rte nd ventiltion V O2 nd V E were mesured simultneously using open-flow respirometry (for detils, see Reyes nd Milsom, 29). ir or the H H gs mixture ws delivered through port in the side of rething funnel t rte of 35mlmin 1. The flow rte ws Tle 2. Men ± s.e.m. temperture coefficients (Q 1 ) used to correct vlues of oxygen consumption, ventiltion nd ir convection requirement mesured in the summer nd winter to 14.7 C for comprison with other sesons (see Mterils nd methods for detils) Condition Seson Q 1 (V O 2 ) Q 1(V E ) Q 1 (CR) ir 2.85± ±.76.85± ± ±.38.96±.46 Hypoxi Hypercpni 3.3±.6 1.6±.2.36± ± ±.61.46±.11 V O 2, oxygen consumption; V E, ventiltion nd CR, ir convection requirement. THE JOURNL OF EXPERIMENTL IOLOGY regulted with gs mixing flow meter (Cmeron Instrument Compny, Port rnss, TX, US). The ir/gs mixtures entering nd leving the funnel were dried nd smpled y gs nlyzer (model 222 version 1.2, Rytech Instruments, Vncouver, C, Cnd). V O2 rtes were clculted from the flow rte nd the difference in frctionl concentrtions of oxygen etween inflow nd outflow gs. Rtes of V O2 re expressed s ml O 2 STPDmin 1 kg 1 (where STPD is stndrd temperture nd pressure, dry). V E ws monitored t the outflow of the rething funnel with pneumotchogrph (Fleisch, Richmond, V, US) ttched to differentil pressure trnsducer (Vlidyne, Northridge, C, US) (Funk et l., 1986). The rething trce ws nlyzed for rething frequency (f R, rethsmin 1 ) nd men tidl volume (V T, mlkg 1 ). Totl ventiltion (V E : f R V T, mlmin 1 kg 1 ) ws clculted from these vlues. The components of f R : numer nd frequency of rething episodes, numer of reths in ech episode, the instntneous frequency (the frequency of the reths within n episode) nd the percentge time spent pneic were lso clculted. pne ws defined s respirtory puse tht exceeded the durtion of two reths. Dt nlyses ll dt were continuously recorded using WINDQ cquisition softwre (version 2.19, Dtq Instruments, kron, OH, US). utumn nd winter dt were nlyzed first for every one-hour segment of the 24h experiment. sed on this, we determined tht nlyses of dt every other hour ws sufficient to ccurtely descrie dily rhythms; thus, spring, summer nd constnt drk runs were nlyzed every other hour. Dt were verged over oth the dytime nd night-time mesurements, nd over the full 24 h for ll turtles (N=6 8) in ech seson. Segments of rething trces were removed from the nlyses when turtles were ctive to ensure tht rething nd metolism vlues used for the nlyses were resting rtes. Turtle ctivity produced wter movement tht induced pressure chnges tht were sensed y the pneumotcogrph (noise). This method ws verified y compring the noise in the rething trce with 24 h-long video recordings of the turtles during the initil utumn trils. Clcultion of the ir convection requirement nd chemosensitivity V O2 nd V E were used to clculte the CR (ml irml 1 O 2 ) for ech hour, where CR=V E /V O2. The H H ventiltory response ws clculted s the solute chnge in totl ventiltion when turtles went from rething ir to rething the H H gs mixture ( V E ). ecuse our gol ws to determine whether ventiltory sensitivity showed endogenous dily nd sesonl rhythms (i.e. independent of temperture nd metolism), we eliminted the effects of temperture vrition etween sesons on our clcultions of chemosensitivity using temperture coefficient (Q 1 ) vlues to correct ll dt to 14.7 C (the men nnul wter temperture). Q 1 vlues for V E, V O2 nd CR were clculted for ech turtle under oth the ir nd the H H tretments. Q 1 vlues were clculted y inserting the V O2, V E nd CR mesured t two different tempertures in this seson (19.6 C nd 8.8 C, indoor nd outdoor tempertures, respectively) into Eqn1 (Tle2). MR 2 ws the vrile to e determined (corrected winter V O2,

3 Dily nd sesonl rhythms in chemosensitivity 3341 VE (ml min 1 kg 1 ) P=.3 Hypoxi hypercpni D P=.1 ir Dy Night Fig. 1. Dy nd night vlues of resting ventiltion (V E ), oxygen consumption (V O2 ) nd ir convection requirement (CR) ±s.e.m of turtles rething hypoxic hypercpnic (H H) gs ( C) nd ir (D F) [modified from Reyes nd Milsom (Reyes nd Milsom, 29)] in different sesons (N=6 for summer nd spring, N=8 for winter nd utumn) mesured t men sesonl tempertures nd nturl photocycle (summer: 2.8 C, 16 h:8 h L:D; utumn: 14.7 C, 1 h:14 h L:D; winter: 9 C, 9 h:15 h L:D; spring: 14.6 C, 14 h:1 h L:D). Denotes difference etween dy nd night within sesons under the H H mixture nd denotes differences etween dy nd night within sesons on ir (Holm Sidk pirwise comprison). VO2 (ml O2 min 1 kg 1 ).4 P= E P=.12 C 2 P<.1 F P=.7 CR (VE/VO2) utumn utumn V E nd CR t T 2 =14.7 C). MR 1 ws the winter vlue mesured t 8.8 C (T 1 ). s the summer outdoor temperture (2.8 C) ws similr to the indoor temperture (19.6 C), we clculted the Q 1 vlues y inserting V O2, V E nd CR vlues mesured t the summer nd utumn outdoor tempertures (2.8 C nd C, respectively). utumn vlues were selected ecuse chnges etween summer nd utumn for ll physiologicl vriles were solely the result of chnges in temperture (Reyes nd Milsom, 29). Q 1 vlues nd summer vlues (MR 2 ) mesured t 2.8 C (T 2 ) were inserted into Eqn1 to clculte V O2, V E nd CR vlues (MR 1 ) t 14.7 C (T 1 ). Temperture-corrected chnges in ventiltion (ΔV E ), oxygen consumption (ΔV O2 ) nd ir convection requirement (ΔCR) were otined from the temperturecorrected vlues clculted for V E, V O2 nd CR under ir nd the H H gs: Q 1 = (MR 2 / MR 1 ) 1 / (T 2 T 1 ). (1) We eliminted the effect of chnges in metolism (dy night nd sesonl) on chemosensitivity y clculting temperture-corrected chnges in CR when turtles went from rething ir to rething the H H gs ( CR). Thus, in this study we use chnges in CR clculted from temperture-corrected vlues of ΔV E nd ΔV O2 s mesure of chnges in chemosensitivity. Sttisticl nlyses Dt re expressed s mens ± s.e.m. Differences etween dy nd night, s well s the effects of seson on the dily chnges were ssessed with two-wy repeted-mesures nlysis of vrince (RM NOV). Differences etween sesons (24 h verge) were ssessed using one-wy RM NOV. Holm Sidk multiple comprisons tests were used to determine pirwise differences. Dt tht did not meet the ssumptions of norml distriution or equl vrinces were nturl log (ln) trnsformed. Sigm Stt (version 3.11, Systt Softwre, IL, US) ws used for ll sttisticl nlyses. RESULTS Dily rhythms in the metolism nd rething of turtles during exposure to H H We hve previously shown tht there re circdin rhythms in V O2 nd V E in red-ered sliders rething ir (Reyes nd Milsom, 29) (Fig. 1D,E). Here we show for the first time tht red-ered sliders lso hve dily rhythms in V O2 nd V E while rething H H gs mixture (V O2 : P=.8; V E : P=.3, dy night, two-wy RM NOV) (Fig. 1,). Dytime f R ws higher thn night-time f R in ll sesons (P<.1, dy night, two-wy RM NOV) (Tle3), while V T remined THE JOURNL OF EXPERIMENTL IOLOGY

4 3342 C. Reyes nd W. K. Milsom Tle 3. Dy nd night vlues (±s.e.m.) of tidl volume (ml kg 1 ), rething frequency (reths min 1 ), frequency of rething episodes (episodes h 1 ) nd the percentge time spent in pne for turtles rething hypoxic hypercpnic (H H) gs nd ir (Reyes nd Milsom, 29) in different sesons (N=6 for summer nd spring, N=8 for winter nd fll) nd mesured t the men sesonl tempertures nd nturl photocycle (see Tle 1) Tidl volume rething frequency Episodes h 1 % Time in pne Seson Gs/ir Dy Night Dy Night Dy Night Dy Night H H 11.3± ±.9 9.6± ± ± ± ±3 8.9±1.6 ir 5.5±.6 6.3±.7 3.1±.3 2.1± ± ± ± ±.4 utumn H H 13.8± ±.8 1.6± ± ± ± ± ±2 ir 6.7±.8 5.7±.4 2.9±.3 1.5± ± ± ± ±.5 H H 1.2± ± ±.7 1.6±.4 2.1± ± ± ±1.9 ir 5.4±.9 3.3±.7 1.5±.1.5±.1 14±2.3 5± ±.6 98±.6 H H 2.3± ± ± ± ± ± ± ±2.3 ir 6.1± ±.6 1.8±.2 1.2± ± ± ± ±.5 Denotes differences etween dy nd night within sesons under hypoxi hypercpni (H H) nd denotes differences etween dy nd night within sesons on ir (Holm Sidk pirwise comprison). reltively constnt throughout the 24 h in ll sesons (P=.236, dy night, two-wy RM NOV) (Tle3). Chnges in f R were solely the result of more frequent rething episodes during the dy (P=.3, dy night, two-wy RM NOV) (Tle 3); the numer nd frequency of reths in ech episode did not vry with the time of dy [P=.373 nd P=.222, numer nd frequency of reths, respectively, two-wy RM NOV (dt not shown)]. Thus, turtles were pneic longer t night thn during the dy in ll sesons (P=.2, dy night, two-wy RM NOV) (Tle 3). While dy night differences in the rething pttern of turtles exposed to the H H gs were similr to those oserved in nimls rething ir (Tle 3) (Reyes nd Milsom, 29), the dytime increses in rething under H H were lrger thn the increses in V O2, giving rise to dily rhythms in CR (P<.1, dy night, twowy RM NOV) tht were not present in nimls rething ir (Fig.1C). Dily rhythms in chemosensitivity H H reduced V O2 in the utumn (dy: 37%, night: 11%), summer (dy: 17%, night: 3%) nd winter (dy: 4%, night:.5%). In spring, however, oth dytime nd night-time metolism incresed (13% nd 56%, respectively) when turtles were rething the H H gs. The mgnitude of these chnges, in solute terms, ws the sme during the dy nd the night (P=.165, dy night, two-wy RM NOV) (Fig.2). H H lso cused n increse in V E (summer dy: 536%, summer night: 538%; utumn dy: 629%, utumn night: 592%; winter dy: 341%, winter night: 568%; spring dy: 1496%, spring night: 99%). This increse ws reduced t night in utumn nd spring (P<.5, Holm Sidk) (Fig. 2). The CR incresed when nimls were rething the H H gs (summer dy: 799%, summer night: 6%; utumn dy: 1627%, utumn night; 92%; winter dy: 749%, winter night: 953%; spring dy; 1194%, spring night: 871%) nd this increse in overll CR ( CR: V E / V O2 ) vried etween dy nd night cross ll sesons (P=.2, dy night, two-wy RM NOV) (Fig. 2C). Thus, dily chnges in V E were not simply due to chnges in metolism ut were the result of dily cycles in the sensitivity of chemoreflexes to respirtory stimuli. Exposure to four dys of constnt drkness lunted dily oscilltions in the chemoreflex response of red-ered sliders. Dytime chnges in ventiltion ( V E ) under constnt drkness were reduced y 63% from dy vlues under the nturl photocycle wheres the decrese in the night ventiltory response ws less (21% decrese under constnt drkness). s result no dy night differences in respirtory responses were oserved fter four dys of constnt drkness ( V E : P=.118; CR: P=.48, Holm Sidk) (Fig.3). Sesonl rhythms in the metolism nd rething of turtles during exposure to H H s we hve shown for red-ered sliders held in normoxi (Reyes nd Milsom, 29), when turtles were given the H H gs mixture to rethe they lso showed sesonl cycle in metolism. We mesured the highest rtes of V O2 in the summer nd metolism decresed sesonlly s expected with temperture chnges (P<.1, one-wy RM NOV) (Fig.4C). V E showed the sme generl pttern etween sesons s metolism, except tht V E ws not s elevted in summer (P=.13, one-wy RM NOV) (Fig. 4). Overll differences in metolism etween sesons seemed to result solely from the effects of temperture on V O2, ecuse these differences were not pprent fter temperture correction (P=.266, one-wy RM NOV, 14.7 C) (Fig. 4D). y contrst, V E remined elevted in the spring nd reduced in the winter fter correction for temperture (P<.1, one-wy RM NOV) (Fig.4). Sesonl chnges in the V O2 nd V E of turtles rething the H H gs showed similr trends s those oserved in turtles rething ir. Sesonl differences in rething during H H were cused y smll chnges in V T nd lrge chnges in f R (V T : P=.13; f R : P<.1, one-wy RM NOV) (Tle4). f R in H H chnged due to ltertions in the numer nd frequency of reths within rething episodes (P<.1, one-wy RM NOV) (Tle 4) not due to chnges in the frequency of rething episodes (P=.76, one-wy RM NOV) (Tle 4). These chnges in the rething pttern resulted in considerle increse in the time spent in pne during winter (P<.1, one-wy RM NOV) (Tle 4). ecuse V E nd V O2 chnged lrgely in prllel, CR chnged very little etween sesons. However, CR ws reduced in the summer compred with other sesons (P<.5, Holm Sidk) (Fig. 4E). No sesonl differences were found fter correcting for temperture (Fig. 4F). Sesonl rhythms in chemosensitivity V O2 fell when turtles were given the H H gs insted of ir in ll sesons (13, 26 nd 28% reduced in the summer, utumn nd winter, THE JOURNL OF EXPERIMENTL IOLOGY

5 Dily nd sesonl rhythms in chemosensitivity 3343 Chnge in CR (VE/VO2) Chnge in VE (ml min 1 kg 1 ) Chnge in VO2 (ml O2 min 1 kg 1 ) P=.165 P=.5 2 C P= respectively) (P=.3, one-wy RM NOV) except spring when metolism incresed y 95% (Fig.5). y contrst, V E incresed in ll sesons during H H (summer: 539%; utumn: 615%; winter: 416%; spring: 132%; P<.1) (Fig.5) nd this increse ws gretest in the spring nd lowest in the winter. H H incresed the CR in ll sesons (summer: 728.6%; utumn: 124%; winter: 64%; spring: 141%; P=.1) ut the mgnitude of this chnge ws similr in ll sesons except summer when it incresed the lest (Fig.5C). To remove the effects of temperture on the metolism nd ventiltion of turtles, we then temperture corrected the V E nd V O2 vlues mesured in ll sesons under oth H H nd ir to 14.7 C using temperture coefficients (Q 1 vlues) nd used these vlues utumn Fig. 2. Men dytime (open rs) nd night-time (closed rs) chnges in oxygen consumption (ΔV O 2 ) (), ventiltion (ΔV E) () nd ir convection requirement (ΔCR, the hyperventiltory response, C) s turtles went from rething ir to rething hypoxic hypercpnic (H H) gs. P vlues reported correspond to overll differences in men dy nd night vlues (two-wy RM NOV). Denote dy nd night differences within sesons (Holm Sidk pirwise comprison) (N=6 for summer nd spring, N=8 for winter nd utumn). Chnge in CR (VE/VO2) Chnge in VE (ml min 1 kg 1 ) P= P= Photocycle Constnt drkness Fig. 3. Comprison of dy (open rs) nd night (closed rs) chnges in ventiltion (ΔV E ) () nd ir convection requirement (ΔCR) () s turtles went from rething ir to rething hypoxic hypercpnic (H H) gs under the utumn photocycle nd constnt drk. P vlues reported correspond to overll differences in men dy nd night vlues (two-wy RM NOV). Denote dy nd night differences within the photocycle nd constnt drk tretments (Holm Sidk pirwise comprison) (N=6 for summer nd spring, N=8 for winter nd utumn). to clculte the V E, V O2 nd CR. The solute chnge in V O2 in the spring ws significntly different from tht in the utumn nd summer (P<.5, Holm Sidk test). Differences etween spring nd winter were lost, however, fter correcting for temperture (P>.5, Holm Sidk test) (Fig. 6). Red-ered sliders still showed elevted ventiltory responses to H H in the spring nd reduced ventiltory responses in the winter when temperture effects were removed (P<.1, one-wy RM NOV) (Fig. 6). However, fter temperture correction, the increse in ventiltion reltive to metolism ( CR) did not vry etween sesons (P=.428, onewy RM NOV) (Fig.6C). DISCUSSION The min gol of our present study ws to determine whether respirtory chemosensitivity vried dily nd/or sesonlly in redered sliders. Respirtory chemosensitivity is normlly defined in mmmls s unit chnge in V E rising from unit chnge in rteril gs prtil pressure ( V E / P O2/CO2 ). In the present study we did not mesure lood gses due to the difficulty of mintining cnnule open for over yer. s result we report only the V E etween nimls rething ir nd the H H gs (i.e. no units of chnge in stimulus intensity). In ectotherms, metolism nd temperture cn lter the reltionship etween V E nd P O2/CO2. We therefore corrected for these fctors y stndrdizing V E to V O2 (CR), nd lso djusting these vlues for chnges in temperture etween sesons. fter tking these precutions, we found tht the dy night differences in V E resulted from dily chnges in chemoreflex sensitivity ut tht the sesonl chnges were directly relted to chnges in metolism. THE JOURNL OF EXPERIMENTL IOLOGY

6 3344 C. Reyes nd W. K. Milsom CR (VE/VO2) VO2 (ml O2 min 1 kg 1 ) VE (ml min 1 kg 1 ) C Sesonl temperture H H: P=.13 ir: P<.1, H H: P<.1 ir: P<.1 c E H H: P<.1 ir: P=.11 utumn C,c,, H H ir Temperture corrected (14.7 C) c H H: P<.1 ir: P=.29, D H H: P=.266 ir: P<.1 F H H: P=.294 ir: P<.1, utumn Fig. 4. Ventiltion (V E ), oxygen consumption (V O 2) nd ir convection requirement (CR) of turtles rething hypoxic hypercpnic gs (H H) nd ir [modified from Reyes nd Milsom (Reyes nd Milsom, 29)] in different sesons. Vlues recorded t sesonl tempertures re given s well s summer nd winter vlues corrected to 14.7 C. Sesonl differences etween vlues mesured under H H re indicted y different lowercse letters. Sesonl differences etween vlues mesured in ir re indicted y different uppercse letters (N=6 for summer nd spring, N=8 for winter nd utumn; one-wy RM NOV). Dily rhythms in ventiltory sensitivity Red-ered sliders in our study showed reduced respirtory response to the H H stimulus t night compred with the dy. Time of the dy hs een shown to ffect the ventiltory response to hypoxi, hypercpni or oth stimuli comined in mmmls (Stephenson et l., 2; Jrsky nd Stephenson, 2; Mortol nd Seifert, 22; Mortol, 24) nd irds (Woodin nd Stephenson, 1998), nd now we hve shown this in reptiles. The mechnisms underlying dy night chnges in ventiltory responses hve not yet een elucidted, lthough interesting speciesspecific differences hve een found. Rts hd higher ventiltory response to hypoxi nd hypercpni t night, when they were ctive, thn during the dy. These oscilltions in rething were correlted to dily chnges in metolism nd not to chnges in chemosensitivity ecuse the hyperventiltory response did not vry etween dy nd night (i.e. there ws no % CR etween night nd dy) (Peever nd Stephenson, 1997; Mortol nd Seifert, 2; Seifert nd Mortol, 22; Seifert nd Mortol, 22). In humns, however, the ventiltory response to hypercpni ws higher in the dy nd ws independent of metolism or sleep (Stephenson et l., 2; Mortol nd Seifert, 22; Mortol, 24), indictive of n endogenous dily cycle in chemosensitivity. In the present study we did not ttempt to determine to wht extent the dy night differences seen in chemosensitivity were result of chnges in sleep stte. Stte of rousl is known to ffect oth the hypoxic nd hypercpnic ventiltory responses in mmmls. lthough, the vriility etween species is lrge, the generl consensus is tht the hypercpnic ventiltory response is lunted during sleep nd the hypoxic ventiltory response is reduced or unchnged (Phillipson et l., 1978; owes et l., 1981; Hunter et l., 1998; Mortol, 24). Sleep ptterns hve een descried in turtles nd lizrds (Flnign, 1973; Flnign et l., 1974) nd periodic reductions in f R hve een reported in pythons nd lizrds during sleep (Donnelly nd Woolcock, 1977; Crgg, 1978). Thus, our dt support the notion tht turtles hve dily rhythm in respirtory sensitivity tht is independent of circdin chnges in metolism, ctivity nd temperture ut we cnnot rule out chnges in stte. Mortol (Mortol, 24) suggested tht the species difference etween rts THE JOURNL OF EXPERIMENTL IOLOGY

7 Dily nd sesonl rhythms in chemosensitivity 3345 Tle 4. Men ± s.e.m. tidl volume (ml kg 1 ), rething frequency (reths min 1 ), frequency of rething episodes (episodes h 1 ), reths per episode, instntneous rething frequency (reths min 1 in n episode) nd percentge time spent in pne of turtles rething hypoxic hypercpnic gs (H H) nd ir (Reyes nd Milsom, 29) nd exposed to sesonl conditions (see Tle 1) Seson Gs/ir Tidl volume rething frequency Frequency of episodes reths per episode Inst. frequency % Time pne H H 11.8±.7 8.6± ± ±2.6.6± ±2.4,, c ir 5.8±.6 2.8±.3 2.6± ±.8.7± ±.7, utumn H H 13.4±.7 7.2± ± ±5.4.5± ±2.9, ir 6.1±.5 2.1± ± ±2.3.6± ±.7 C H H 8.1± ± ± ±1.4.3± ±1.9 ir 4.1±.6.8±.1 8.4± ±.4.2± ±.3 C C H H 19.9± ± ±2.3 13±1.1.5± ±2.6 c ir 6.8±.3 1.5±.1 13± ±1.1.6± ±.3, C Significnt differences etween sesons (P<.5, one-wy RM NOV) re indicted y different lowercse letters (H H) nd uppercse letters (ir) (N=6 for summer nd spring, N=8 for winter nd fll). nd humns could reflect the fct tht circdin chnges in ventiltion re coupled to chnges in metolism in rts where these chnges re rpid nd sleep stte is frgmented ut endogenous in humns where the chnges in metolism re smll nd slow nd the sleep cycle is prolonged. Metolism in turtles lso chnges slowly, nd is further constrined y chnges in environmentl temperture, perhps incresing selection pressure on lternte mechnisms to reduce ventiltion during periods of inctivity. Free-running rhythms (constnt temperture nd D D or L L photocycles) in metolism nd ventiltion hve een shown in the ox turtle, grter snke, Europen green lizrd nd red-ered slider (Glss et l., 1979; Hicks nd Riedesel, 1983; Rismiller nd Heldmier, 1991; Reyes nd Milsom, 29). In the present study we found tht four dys of constnt drkness ws sufficient to olish the dy night difference in chemosensitivity. Interestingly, this ws cused y reduction in the dytime vlue, while night-time ventiltory sensitivity did not chnge from the levels seen under the nturl photocycle. The sme turtles continued to show dy night differences in metolism nd ventiltion fter three dys in constnt drkness while rething ir, lthough the cycles were lunted nd the period ws slightly shifted (Reyes nd Milsom, 29). The loss in the dy night difference in chemosensitivity fter four dys of constnt drkness does not imply tht the dily chnges in the physiologicl vriles were not driven y circdin system. Removl of ll environmentl cues typiclly reduces the mplitude of dily rhythms, until the rhythms re lost ltogether (schoff et l., 1982; Kengy nd Vleck, 1982; Underwood, 1992; Tosini et l., 21). Tegu lizrds hve lso een shown to lose circdin oscilltions in metolism over time (Milsom et l., 28). There ws lrge vriility in the response of these lizrds to constnt drkness, with some individuls losing the rhythms fter dy in constnt drkness nd others mintining the rhythms up to 14 dys. Unfortuntely, it is impossile for us to determine where in the reflex rc tht underlies the ventiltory response to H H, chnges occurred or even whether the chnges were in the hypoxic, hypercpnic or oth components of the response. While hypoxi is only sensed y peripherl chemoreceptors, hypercpni stimultes oth centrl nd peripherl chemoreceptors, nd the centrl rinstem sites involved in processing oth stimuli re different (Shelton nd outilier, 1982; Shelton et l., 1986; Milsom, 199). We used H H gs mixture, however, to mimic the respirtory stimuli tht turtles experience during sumergence (urggren nd Shelton, 1979; Shelton nd outilier, 1982). Regrdless of the mechnism ehind the dily chnges in chemosensitivity of turtles, it is cler tht night-time reductions in chemosensitivity should reduce sensitivity to fll in oxygen stores nd n ccumultion of CO 2 in the lood, nd thus llow for longer dives. This ppers to e true for other diving species s well. Thus, cnvsck ducks showed reduced chemosensitivity to progressive sphyxi during the night-time tht ws independent of metolism. These nimls forge underwter t night nd incresing the level of lood gses t which chemoreflexes re elicited t this time ws suggested to llow them to prolong their dives nd forging times (Wooding nd Stephenson, 1998). In red-ered sliders, the endogenous circdin rhythms in metolism (lower t night) (Reyes nd Milsom, 29) long with dily oscilltions in chemoreflex responses (independent of metolism) llow these turtles to reduce the locomotion costs nd predtion risks ssocited with surfcing to rethe t night. Sesonl rhythms in ventiltory sensitivity Red-ered sliders showed sesonl differences in their ventiltory response to H H: it ws reduced in the winter nd elevted in the spring. This my not seem surprising ecuse it is well estlished tht sensitivity to hypoxi or hypercpnic cidosis increse with temperture (Shelton nd outilier, 1982; Glss et l., 1983; Glss et l., 1985; Funk nd Milsom, 1987). In our study, however, sesonl chnges in the H H response remined fter the dt were temperture corrected, nd ppered to hve resulted from the effects of the respirtory stimulus on metolism nd not from chnges in chemosensitivity, ecuse the chnge in CR ( CR) did not vry with seson. The reduced ventiltory response during winter coincides with the dormncy period. Temperte reptiles, including red-ered sliders, often undergo ctive metolic suppression, when metolism THE JOURNL OF EXPERIMENTL IOLOGY

8 3346 C. Reyes nd W. K. Milsom Chnge in CR (VE/VO2) Chnge in VE (ml min 1 kg 1 ) Chnge in VO2 (ml O2 min 1 kg 1 ).2 P= C 16 P< C 8 18 C 16 P= C C 2 utumn 14.6 C 8.8 C 14.7 C 14.7 C 8.8 C 14.6 C 8.8 C c 14.6 C Fig. 5. Chnges in oxygen consumption (ΔV O 2) (), ventiltion (ΔV E ) () nd ir convection requirement (ΔCR) (C) s turtles went from rething ir to rething hypoxic hypercpnic (H H) gs in different sesons. Significnt differences etween sesons (P<.5, one-wy RM NOV) re indicted y different letters (N=6 for summer nd spring, N=8 for winter nd utumn). Chnge in CR (VE/VO2) Chnge in VE (ml min 1 kg 1 ) Chnge in VO2 (ml O2 min 1 kg 1 ) C c 14 P< P=.5 18 C 14.7 C P=.428 utumn, 14.7 C Fig. 6. Temperture-corrected (14.7 C) chnges in oxygen consumption (ΔV O 2) (), ventiltion (ΔV E ) () nd ir convection requirement (ΔCR) (C) s turtles went from rething ir to rething hypoxic hypercpnic (H H) gs in different sesons. Significnt differences etween sesons (P<.5, one-wy RM NOV) re indicted y different letters (N=6 for summer nd spring, N=8 for winter nd utumn). cnnot e sustined under winter environmentl conditions (Myhew, 1965; ennett nd Dwson, 1976; Gregory, 1982; Ultsch, 1989; Zri, 1999). Tods nd ullfrogs lso show reduced metolism nd lower ventiltory response to hypoxi [tods (icego-nhs et l., 21)] nd hypercpni [ullfrogs (icego- Nhs nd rnco, 1999)] in the winter. It would pper tht sesonl reductions in oth, the temperture-corrected resting ventiltion nd in the ventiltory response to H H of red-ered sliders were due to the orchestrted fll in metolism. Short photoperiods in the utumn trigger ehviors such s fsting nd selection of lower tempertures (Cgle 195; Gregory, 1982; Rismiller nd Heldmier, 1982), which will slowly reduce metolism in preprtion for dormncy (ennett nd Dwson, 1976). Turtles in our study underwent norexi in the lte utumn nd did not et in the winter. For mny reptiles, the highest metolic rtes occur fter feeding, with metolism flling drmticlly during fsting. Furthermore, during long periods of fsting the digestive system trophies removing the dily cost of regenertion (Wng et l., 21). lthough reduced gut mss nd lck of feeding my hve influenced winter metolism, the sesonl trends in V O2 cnnot e solely explined y chnges in feeding regime. Turtles in our study stopped eting voluntrily in mid lte utumn nd this continued throughout the winter. Thus, if feeding were importnt, temperture-corrected metolism in the utumn (in nimls THE JOURNL OF EXPERIMENTL IOLOGY

9 Dily nd sesonl rhythms in chemosensitivity 3347 rething ir) should lso hve een prtilly reduced nd this ws not the cse (Fig. 4C,D). V E incresed in ll sesons with exposure to H H, which grees with other studies on reptiles (Jckson, 1973; Jckson et l., 1974; Glss et l., 1985; Shelton et l., 1986). However, the increse in spring V E ws lrger thn tht in ll other sesons even fter temperture correction. Red-ered sliders mte during the spring (Cgle, 195; Duvll et l., 1982; Ernst nd rour, 1989; Kuchling, 1999) nd hormones involved in reproduction my e the source of the elevted metolism nd the higher ventiltory response during this seson (Ttsumi et l., 1997). Higher metolic rtes during oogenesis nd mting hve lso een reported in Lcert viridis (Rismiller nd Heldmier, 1991). Prdoxiclly, while the ventiltory response to H H ws gretest in the spring, resting metolism nd ventiltion in nimls rething ir were very low (Fig.4) (Reyes nd Milsom, 29) remining t winter levels. In ll cses, however, the chnges in ventiltion nd metolism occurred in prllel, suggesting tht the spring time increses in the H H ventiltory response were result of spring time increses in the effects of H H on metolism nd not the result of sesonl chnges in chemosensitivity. The resons for this re not cler. Hypoxi nd/or hypercpni generlly suppress metolism, if nything, nd this ws the trend tht ws seen in ll other sesons. It is possile tht cuse nd effect re reversed here nd tht metolism increses ecuse of the very lrge increse in ventiltion rther thn vice vers, ut estimtes of the oxidtive cost of ventiltion in turtles remin controversil. Estimtes of the reltive metolic cost of rething sed on the elimintion of rething y unidirectionl ventiltion (Kinney nd White, 1977) or sed on chnges in ventiltion nd metolism ssocited with hypoxic exposure (Wng nd Wrurton, 1995; Skovgrd nd Wng, 24) suggest tht the costs could e s high s 15 2% of resting metolism. Estimtes sed on chnges in ventiltion nd metolism ssocited with hypercpnic exposure, however, suggest the costs re extremely smll nd often overshdowed y n cidosisinduced metolic suppression leding to prdoxicl decrese in metolism with n increse in rething (Wng nd Wrurton, 1995; Skovgrd nd Wng, 24). This issue remins to e resolved, s does the source of the increse in metolism descried here. Conclusion Wht we hve mesured in this study is the response to fixed stimulus pplied t the lungs. The stimuli for the chemoreceptors involved in ventiltory regultion re the rteril lood gses (P O2, P CO2 nd ph) t the peripherl receptors nd the P CO2 /ph of the cererl spinl fluid (CSF); however, these were not mesured. s result, the chnges in chemosensitivity tht we descrie my hve een due to chnges in the sensitivity of the peripherl rteril chemoreceptors per se or to chnges in the P O2 /P CO2 /ph t receptor sites. The ltter could occur due to dily nd/or sesonl chnges in the mgnitude of the crdic shunt tht is known to e lrge in turtles. There were sesonl chnges in temperture tht re lso known to give rise to chnges in the CR nd hence to resting lood gs tensions. To dd to the complexity, similr chnges in seline lood gs tensions initited from different seline lood gs levels my give rise to different responses due to the nonlinerity of the hypoxic ventiltory response (the sme chnge in lood gses my not mount to the sme chnge in stimulus when strting from different selines) (Glss, 1992). Finlly, even when no chnges were seen in chemosensitivity s function of seson, this could hve een due to sesonl chnges in crdic shunting cting to mintin constnt lood gs tensions (Wng nd Hicks, 1996; Wng et l., 1997). ny such chnges, however, re prt of dily nd sesonl rhythms. In this study we re looking t the whole niml output for given environmentl input nd determining the mechnistic sis of these chnges will e the next chllenge. Mechnism side, red-ered sliders showed reduced respirtory chemosensitivity t night nd in the winter, nd enhnced chemosensitivity in the spring. Dy nd night differences resulted from dily oscilltions in the sensitivity of chemoreflexes wheres sesonl differences could e explined y the effects of the H H stimuli on metolism. Regrdless of the different mechnisms, dily nd sesonl chnges in the ventiltory response together with circdin nd circnnul rhythms in metolism (Reyes nd Milsom, 29) enefit turtles y fcilitting longer pnes. These physiologicl chnges will reduce time t the surfce nd my e importnt for minimizing the cost of locomotion t times when the cost of surfcing or risk of predtion my compromise their survivl. LIST OF REVITIONS CR ir convection requirement f R rething frequency H H hypoxic hypercpnic P CO 2 rteril CO 2 pressure P O 2 rteril O 2 pressure Q 1 temperture coefficient quotient RM NOV repeted-mesures nlysis of vrince V E totl ventiltion V O 2 oxygen consumption tidl volume V T The uthors re grteful to Kim org nd Crystl runer for their contriution with dt nlyses. Dr Dve Jones, Dr Colin runer, Dr Jonthn Shurin, Grhm Scott nd Dr ngelin Fong provided useful comments nd constructive criticism. The uthors lso wnt to thnk ruce Gillespie, Vincent Grnt nd Don rndys for the construction of the experimentl set-up nd Ry Dhly for the mnufcture of the gs nlyzer used to otin the metolic rte dt. This study ws funded y NSERC (Cnd). REFERENCES schoff, J., Gerecke, U., von Goetz, C., Groos, G.. nd Turek, F. W. (1982). Phse responses nd chrcteristics of free-running ctivity rhythms in the golden hmster: independence of the pinel glnd. In Verterte Circdin Systems (ed. J. schoff, S. Dnn nd G. Groos), pp erlin: er-verlg. ennett,. F. nd Dwson, W. R. (1976). Metolism. In iology of the Reptili, vol. 5 (ed. C. Gns nd W. R. Dwson), pp New York: cdemic Press. icego-nhs, K. C. nd rnco, L. G. S. (1999). Sesonl Chnges in the crdiorespirtory responses to hypercri nd temperture in the ullfrog, Rn ctesein. Comp. iochem. Physiol. Mol. Integr. Physiol. 124, icego-nhs, K. C., Grgglioni, L. H. nd rnco, L. G. S. (21). Sesonl chnges in the preferred ody temperture, crdiovsculr nd respirtory responses to hypoxi in the tod, ufo prcnemis. J. Exp. Zool. 289, owes, G., Townsend, E. R., Kozr, L. F., romley, S. M. nd Phillipson, E.. (1981). Effect of crotid ody denervtion on rousl response in sleeping dogs. J. ppl. Physiol. 51, urggren, W. W. nd Shelton, G. (1979). Gs exchnge nd trnsport during intermittent rething in chelonin reptiles. J. Exp. iol. 82, Cgle, F. R. (195). The life history of the slider turtle Pseudemys script troostii (Holrook). Ecol. Monogr. 2, Crgg, P.. (1978). Ventiltory ptterns nd vriles in rest nd ctivity in the lizrd, Lcert. Comp. iochem. Physiol 6, Donnelly, P. nd Woolcock,. J. (1977). Ventiltion nd gs exchnge in the crpet python, Moreli spilotes vriegte. J. Comp. Physiol. 122, Duvll, D., Guillette, L. J., Jr nd Jones, R. E. (1982). Environmentl control of reptilin reproductive cycles. In iology of the Reptili, vol. 13D (ed. C. Gns nd F. H. Pough), pp New York: cdemic Press. Ernst, C. H. nd rour, R. W. (1989). Turtles of the World. Wshington, DC: Smithsonin Institution Press. Flnign, W. F. (1973). Sleep nd wkefulness in igunid lizrds, Ctenosur pectint nd Igun igun. rin. ehv. Evol. 8, Flnign, W. F., Knight, C. P., Hrtse, K. M. nd Rechtschffen,. (1974). Sleep nd wkefulness in chelonin reptiles. I. The ox turtle, Terrpene crolin. rch. Itl. iol. 112, Funk, G. D. nd Milsom, W. K. (1987). Chnges in ventiltion nd rething pttern produced y chnging ody temperture nd inspired CO 2 concentrtion in turtles. Respir. Physiol. 67, Funk, G. D., We, C. L. nd Milsom, W. K. (1986). Non-invsive mesurement of respirtory tidl volume in qutic, ir-rething nimls. J. Exp. iol. 126, THE JOURNL OF EXPERIMENTL IOLOGY

10 3348 C. Reyes nd W. K. Milsom Glss, M. L. (1992). Ventiltory responses to hypoxi in ectothermic vertertes. In Physiologicl dpttions in Vertertes: Respirtion, Circultion nd Metolism (ed. C. Lenfnt), pp New York: Mrcel Dekker. Glss, M. L., Hicks, J. W. nd Riedesel, M. L. (1979). Respirtory responses to longterm temperture exposure in the ox turtle, Terrpene ornte. J. Comp. Physiol. 131, Glss, M. L., outilier, R. G. nd Heisler, N. (1983). Ventiltory control of rteril PO 2 in the turtle Chrysemys pict ellii: effects of temperture nd hypoxi. J. Comp. Physiol. 151, Glss, M. L., outilier, R. G. nd Heisler, N. (1985). Effects of ody temperture on respirtion lood gses nd cid se sttus in the turtle Chrysemis pict ellii. J. Exp. iol. 114, Gregory, P. T. (1982). Reptilin hierntion. In iology of the Reptili, vol. 13 (ed. C. Gns nd F. H. Pough), pp New York: cdemic Press. Hicks, J. W. nd Riedesel, M. L. (1983). Diurnl ventiltory ptterns in the grter snke, Thmnophis elegns. J. Comp. Physiol. 149, Hunter, J. D., McLeod, J. Z. nd Milsom, W. K. (1998). Corticl ctivtion sttes in sleep nd nethesi. II. Respirtory reflexes. Respir. Physiol. 112, Jckson, D. C. (1973). Ventiltory response to hypoxi in turtles t vrious tempertures. Respir. Physiol. 18, Jckson, D. C., Plmer, S. E. nd Medow, W. L. (1974). The effects of temperture nd cron dioxide rething on ventiltion nd cid se sttus of turtles. Respir. Physiol. 2, Jrsky, T. M. nd Stephenson, R. (2). Effects of hypoxi nd hypercpni on circdin rhythms in the golden hmster (Mesocricetus urtus). J. ppl. Physiol. 89, Kengy, G. J. nd Vleck, D. (1982). Dily temporl orgniztion of metolism in smll mmmls: dpttion nd diversity. In Verterte Circdin Systems (ed. J. schoff, S. Dnn nd G. Groos), pp erlin: er-verlg. Kinney, J. L. nd White, F. N. (1977). Oxidtive cost of ventiltion in turtle, Pseudemys floridn. Respir. Physiol. 31, Kuchling, G. (1999). Reproductive iology of the Chelonin. New York: er. Myhew, W. W. (1965). Hierntion in the horned lizrd, Phrynosom m clli. Comp. iochem. Physiol. 16, Mcrthur, M. D. nd Milsom, W. K. (1991). Chnges in ventiltion nd respirtory sensitivity ssocited with hierntion in Columin (Spermophilus columinus) nd Golden-mntled (Spermophilus lterlis) ground squirrels. Physiol. Zool. 64, Milsom, W. K. (199). Control nd co-ordintion of gs exchnge in ir rethers. In dvnces in Comprtive nd Environmentl Physiology, vol. 6 (ed. R. Gilles), pp erlin: er-verlg. Milsom, W. K., Osorne, S., Chn, P. F., Hunter, J. D. nd McLeod, J. Z. (1993). Sleep, hypothermi, nd hierntion: metolic rte nd the control of rething pttern in golden-mntled ground squirrels. In Life in the Cold: Ecologicl, Physiologicl nd Moleculr Mechnisms (ed. C. Crey nd G. L. Flornt), pp Sn Frncisco, C: Westview Press. Milsom, W. K., ndrde, D. V., rito, S. P., Toledo, L. F., Wng, T. nd e,. S. (28). Sesonl chnges in dily metolic ptterns of tegu lizrds (Tupinmis merine) plced in the cold (17 C) nd drk. Physiol. iochem. Zool. 81, Mortol, J. P. (24). rething round the clock: n overview of the circdin pttern of respirtion. Eur. J. ppl. Physiol. 91, Mortol, J. P. nd Seifert, E. L. (2). Hypoxic depression of circdin rhythms in dult rts. J. ppl. Physiol. 88, Mortol, J. P. nd Seifert, E. L. (22). Circdin ptterns of rething. Respir. Physiol. Neuroiol. 131, Peever, J. H. nd Stephenson, R. (1997). Dy-night differences in the respirtory response to hypercpni in wke dult rts. Respir. Physiol. 19, Phillipson, E.., Sullivn, C. E., Red, D. J. C., Murphy, E. nd Kozr, L. F. (1978). Ventiltory nd wking responses to hypoxi in sleeping dogs. J. ppl. Physiol. 44, Reyes, C. nd Milsom, W. K. (29). Circdin nd circnnul rhythms in the metolism nd ventiltion of red-ered sliders (Trchemys script elegns). Physiol. iochem. Zool. doi:1.186/ Rismiller, P. D. nd Heldmier, G. (1982). The effect of photoperiod on temperture selection in the Europen Green Lizrd, Lcert viridis. Oecologi 53, Rismiller, P. D. nd Heldmier, G. (1991). Sesonl chnges in dily metolic ptterns of Lcert viridis. J. Comp. Physiol. 161, Roch, P. nd rnco, L. G. S. (1998). Sesonl chnges in the crdiovsculr, respirtory nd metolic responses to temperture nd hypoxi in the ullfrog Rn ctesein. J. Exp. iol. 21, Seifert, E. L. nd Mortol, J. P. (22). Circdin pttern of ventiltion during cute nd chronic hypercpni in conscious dult rts. m. J. Physiol. Regul. Integr. Comp. Physiol. 282, R244-R251. Seifert, E. L. nd Mortol, J. P. (22). Circdin pttern of ventiltion during prolonged hypoxi in conscious rts. Respir. Physiol. Neuroiol. 133, Shelton, G. nd outilier, R. G. (1982). pnoe in mphiins nd reptiles. J. Exp. iol. 1, Shelton, G., Jones, D. R. nd Milsom, W. K. (1986). Control of rething in ectothermic vertertes. In Hndook of Physiology, Section 3: The Respirtory System, Volume 2: The Control of rething, Prt 2 (ed. N. S. Fishmn, J. G. Chernick, J. G. Widdicome nd S. R. Geiger), pp ethesd, MD: mericn Physiologicl Society. Skovgrd, N. nd Wng, T. (24). Cost of ventiltion nd effect of digestive stte on the ventiltory response of the tegu lizrd. Respir. Physiol. Neuroiol. 141, Stephenson, R., Mohn, R. M., Duffin, J. nd Jrsky, T. M. (2). Circdin rhythms in the chemoreflex control of rething. m. J. Physiol. Regul. Integr. Comp. Physiol. 278, R Ttsumi, K., Pickett, C. K., Jcoy, C. R., Weil, J. V. nd Moore, L. G. (1997). Role of endogenous femle hormones in hypoxic chemosensitivity. J. ppl. Physiol. 83, Tosini, G., ertolucci, C. nd Fo,. (21). The circdin system of reptiles: multioscilltory nd multiphotoreceptive system. Physiol. ehv. 72, Ultsch, G. R. (1989). Ecology nd physiology of hierntion nd overwintering mong freshwter fishes, turtles nd snkes. iol. Rev. 64, Underwood, H. (1992). Endogenous rhythms. In iology of the Reptili: Hormones, rin nd ehvior, vol. 18 (ed. C. Gns nd D. Crews), pp Chicgo, IL: University of Chicgo Press. Vitlis, T. Z. nd Milsom, W. K. (1986). Mechnicl nlysis of spontneous rething in the semi-qutic turtle, Pseudemys script. J. Exp. iol. 125, Wng, T. nd Hicks, J. W. (1996). The interction of pulmonry ventiltion nd the right-left shunt on rteril oxygen levels. J. Exp. iol. 199, Wng, T. nd Wrurton, S. T. (1995). rething pttern nd cost of ventiltion in the mericn lligtor. Respir. Physiol. 12, Wng, T., Krosniuns, E. H. nd Hicks, J. W. (1997). The role of crdic shunts in the regultion of rteril lood gses. m. Zool. 37, Wng, T., usk, M. nd Overgrd, J. (21). The respirtory consequences of feeding in mphiins nd reptiles. Comp. iochem. Physiol. 128, Woodin, M. nd Stephenson, R. (1998). Circdin rhythms in diving ehviour nd ventiltory response to sphyxi in cnvsck ducks. m. J. Physiol. Regul. Integr. Comp. Physiol. 274, R686-R693. Zri, T.. (1999). Sesonl cclimtiztion in metolic rte of the fn-fingered gecko, Ptyodctylus hsselquistii (Reptili: Gekkonide). J. Therm. iol. 24, THE JOURNL OF EXPERIMENTL IOLOGY