ThermorégulationGabriel Blouin-Demers

Professeur titulaire, Département de biologie

BIO3503 Écologie des déserts

Écologie thermique• Les ectothermes ont un contrôle très limité sur leur température

corporelle

• La Tc dicte la performance via ses effets sur la physiologie

• Les ectothermes utilisent la thermorégulation comportementale pour contrôler leur Tc et ainsi éviter la diminution de performance

• Il y a de la variation dans la précision de la thermorégulation entre les individus et entre les espèces

• Il y a un continuum de stratégies entre la thermoconformité et la thermorégulation précise

Température

Perf

orm

ance

To

TCmin TCmax

P80

Hypothèse de la coadaptation

• Correspondance entre To et Tc en nature

• Correspondance entre la largeur de la plage de haute performance et la variation de Tc en nature

• « Bon à tout faire, mais maître en rien »

• Formes isomériques des enzymes et compromis

Température

Perf

orm

ance

To To

Température

Perf

orm

ance

P80

P80

Thermorégulation Méthode scientifique I

• Motif : la forme des courbes de performance varie d’une espèce à l’autre

• Hypothèse de la coadaptation : les espèces ont évolué des courbes de performance qui correspondent à leur Tc en nature

• Prédiction : grande plage de haute performance pour les espèces qui ont de grandes variations de Tc en nature

ORIGINAL PAPER

Thermal strategies and energetics in two sympatric colubridsnakes with contrasted exposure

Herve Lelievre • Maxime Le Henanff •

Gabriel Blouin-Demers • Guy Naulleau •

Olivier Lourdais

Received: 21 September 2009 / Revised: 5 November 2009 / Accepted: 9 November 2009 / Published online: 29 November 2009! Springer-Verlag 2009

Abstract The thermoregulatory strategy of reptilesshould be optimal if ecological costs (predation risk and

time devoted to thermoregulation) are minimized while

physiological benefits (performance efficiency and energygain) are maximized. However, depending on the exact

shape of the cost and benefit curves, different thermoreg-

ulatory optima may exist, even between sympatric species.We studied thermoregulation in two coexisting colubrid

snakes, the European whipsnake (Hierophis viridiflavus,

Lacepede 1789) and the Aesculapian snake (Zamenis lon-gissimus, Laurenti 1768) that diverge markedly in their

exposure, but otherwise share major ecological and mor-

phological traits. The exposed species (H. viridiflavus)selected higher body temperatures (*30"C) than the

secretive species (Z. longissimus, *25"C) both in a labo-

ratory thermal gradient and in the field. Moreover, thisdifference in body temperature was maintained under

thermophilic physiological states such as digestion and

molting. Physiological and locomotory performances wereoptimized at higher temperatures in H. viridiflavus com-

pared to Z. longissimus, as predicted by the thermalcoadaptation hypothesis. Metabolic and energetic

measurements indicated that energy requirements are atleast twice higher in H. viridiflavus than in Z. longissimus.

The contrasted sets of coadapted traits between H. viridi-flavus and Z. longissimus appear to be adaptive correlatesof their exposure strategies.

Keywords Ectotherm ! Thermal preference !Coadaptation ! Metabolic reaction norm ! Energy budget !Thermoregulatory strategy

Introduction

The body temperature of ectotherms affects most traits,

from physiological performance (Angilletta et al. 2002;Blouin-Demers et al. 2003; Stevenson et al. 1985) to

growth rates (Goldsbrough et al. 2004), and reproductive

success (Shine et al. 1997). Ultimately, therefore, tem-perature has a significant impact on fitness in ectotherms

(Huey and Kingsolver 1989). Terrestrial reptiles can buffer

to some extent the ubiquitous impacts of body temperatureon performance by thermoregulation. Thermoregulation is

achieved mainly via habitat selection (Blouin-Demers andWeatherhead 2001a, 2002; Huey 1991; Reinert 1984). The

optimal investment in thermoregulation depends on the

ratio between associated costs and benefits (Huey andSlatkin 1976). Costs are mainly ecological because ther-

moregulatory behavior can increase predation risk (Hertz

et al. 1982; Pianka and Pianka 1970) and reduce the timeavailable for other activities (Gregory et al. 1999; Huey

and Slatkin 1976). Benefits are mainly physiological

because locomotory performance and digestive efficiencyare temperature-dependant (Angilletta et al. 2002; Blouin-

Demers et al. 2003; Stevenson et al. 1985). The different

cost–benefit ratios result in a gradient of optimal

Communicated by I. D. Hume.

H. Lelievre (&) ! M. Le Henanff ! G. Naulleau ! O. LourdaisCentre d’Etudes Biologiques de Chize,CNRS UPR 1934, 79360 Villiers en Bois, Francee-mail: lelievre@cebc.cnrs.fr

H. Lelievre ! M. Le HenanffUniversite de Poitiers, 40 Avenue du Recteur Pineau,86022 Poitiers, France

G. Blouin-DemersDepartement de Biologie, Universite d’Ottawa,Ottawa, ON K1N 6N5, Canada

123

J Comp Physiol B (2010) 180:415–425

DOI 10.1007/s00360-009-0423-8

Table 2) and in Z. longissimus (12.2 ± 5.1 vs. 21.1 ±

2.4!C; linear model, F1,30 = 39.03, P \ 0.0001; Table 2).

B80 were not statistically different between the two speciesfor crawling speed (10.6 ± 1.4!C in H. viridiflavus vs.

12.2 ± 5.1!C in Z. longissimus; linear model, F1,37 =

2.15, P = 0.15; Table 2), but differed for tongue-flickingfrequency (15.7 ± 1.5!C in H. viridiflavus vs. 21.1 ±

2.4!C in Z. longissimus; linear model, F1,37 = 73.06,

P \ 0.0001; Table 2).

Thermal sensitivity of metabolism and energy gain

Oxygen consumption was positively influenced by tem-perature (linear model; F3,86 = 115.3; P \ 0.0001; Fig. 5).

Male H. viridiflavus and Z. longissimus exhibited similar

metabolic reaction norms (no interaction between speciesand temperature; linear model, t3,86 = 0.59, P = 0.56).

H. viridiflavus had higher SMR (linear model, t3,86 = 2.29,

P = 0.02), especially at high Tb, but this difference wasonly significant at 30!C (linear model, t2,15 = 3.14,

P = 0.007). For the two species, the best fit was a

quadratic regression (O2 consumption = 0.04687 -0.005203 Tb ? 0.0002283 Tb

2 in Z. longissimus, adjusted

R2 = 0.988 and O2 consumption = 0.04425 - 0.004726

Tb ? 0.0002264 Tb2 in H. viridiflavus, adjusted R2 =

0.985). Using metabolic reaction norms, we calculated the

predicted oxygen consumption within Tset, and we esti-

mated the energy expenditure at Tb selected in the fieldduring hot days (Table 3). Energy expenditure was sig-

nificantly higher in H. viridiflavus (33.86 ± 1.66

J g-1 day-1) than in Z. longissimus (21.77 ± 1.13 J g-1

day-1; linear mixed effects model, t1,15 = 7.29,

P \ 0.0001). The difference was more pronounced during

Fig. 4 Mean fitted thermalperformance curves (lines) andindividual responses (circles) ofH. viridiflavus (EWS) andZ. longissimus (AS). The grayboxes indicate the meanpreferred temperature range inthe thermal gradient

Table 2 Mean (!C ± SD) To and B80 for crawling speed and tongue-flicking frequency performances for H. viridiflavus (EWS) andZ. longissimus (AS)

Species Performance To (!C) B80 (!C) N Curve

AS Crawling 27.6 ± 4.0 12.2 ± 5.1 16 Log-exponentiala

Tongue-flicking

24.9 ± 3.3 21.1 ± 2.4 16 Log-exponentiala

EWS Crawling 30.6 ± 1.0 10.6 ± 1.4 23 Quarticb

Tongue-flicking

26.6 ± 2.4 15.7 ± 1.5 23 Quarticb

a Performance = S 1= 1þ k1e"k2 Tb"CTminð Þ! "# $1" ek3 Tb"CTmaxð Þ! "

b Performance = aT4 ? bT3 ? cT2 ? xT ? z

J Comp Physiol B (2010) 180:415–425 421

123

A stringent test of the thermal coadaptation hypothesis in flourbeetles

William D. Halliday n, Gabriel Blouin-DemersDepartment of Biology, University of Ottawa, 30 Marie Curie, Ottawa, ON, Canada K1N 6N5

a r t i c l e i n f o

Article history:Received 27 February 2015Received in revised form11 May 2015Accepted 7 June 2015Available online 16 June 2015

Keywords:OvipositionRighting timeSprint speedThermal reaction normTribolium castaneumTribolium confusum

a b s t r a c t

Whole-organism performance depends on body temperature and ectotherms have variable body tem-peratures. The thermal coadaptation hypothesis posits that thermal reaction norms have coevolved withthermal preference such that organisms attain optimal performance under a narrow range of bodytemperatures commonly experienced in the wild. Since thermal reaction norms are often similar, re-searchers interested in the effects of temperature on fitness often use one easily measured thermal re-action norm, such as locomotor performance, and assume it is a good proxy for fitness when testing thethermal coadaptation hypothesis. The extent to which this assumption holds, however, is often untested.In this study, we provide a stringent test of the thermal coadaptation hypothesis in red and in confusedflour beetles by comparing the thermal reaction norm for reproductive output to the preferred bodytemperature range. We also test the assumption that locomotor performance can serve as a proxy for thethermal reaction norm for reproductive output, a more ultimate index of fitness. In both species, wemeasured the number of eggs laid, righting time, and sprint speed at eight temperatures, as well as thethermal preference in a thermal gradient. The number of eggs laid increased with female sprint speedand with male righting time, and all three performances had similar thermal reaction norms, with 80% ofthe maximum achieved between 23 and 37 °C. Red flour beetles had preferred body temperatures thatmatched the optimal temperature for performance; confused flour beetles had lower preferred bodytemperature than the optimal temperature for performance. We found support for the assumption thatlocomotor performance can serve as a proxy for reproductive output in flour beetles, but we only foundevidence for thermal coadaptation in one of the two species.

& 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Ectotherms have variable body temperature and body tem-perature affects whole-organism performance (Angilletta, 2009;Angilletta et al., 2002a,b; Bennett, 1980; Huey and Kingsolver,1989). Most ectotherms thus use behavioral thermoregulation tomaintain their body temperature within a narrow range and avoidreduced performance associated with body temperatures far fromthe optimal temperature (Angilletta, 2009; Angilletta et al., 2002a,b; Bennett, 1980; Huey and Kingsolver, 1989). For instance, ec-totherms can maintain body temperatures that maximize meta-bolic rate (e.g., Dubois et al., 2009; Gillooly et al., 2001), growthrate (e.g., Angilletta et al., 2004), locomotion (e.g., Blouin-Demersand Weatherhead, 2008), and reproduction (e.g., Berger et al.,2008). Thermal reaction norms for whole-organism performancetend to have similar shapes, with a gradual increase in

performance with increasing temperature below the optimaltemperature (To) followed by a sharp decrease in performancewhen body temperature exceeds To (Angilletta, 2006; Bulté andBlouin-Demers, 2006; Dell et al., 2011). According to the thermalcoadaptation hypothesis, these similar shapes are due to coevo-lution of the thermal reaction norm for fitness and thermal pre-ference, where the optimal temperature for fitness should evolveto closely match temperatures commonly experienced in the wild(Angilletta, 2009; Angilletta et al., 2002a, 2006; Bennett, 1980;Blouin-Demers et al., 2003; Dorcas et al., 1997; Hertz et al., 1983;Huey and Bennett, 1987; Huey and Kingsolver, 1989). Organismswith strongly left skewed thermal reaction norms (i.e. thermalspecialists), however, may have sub-optimal body temperaturepreference (Martin and Huey, 2008).

Researchers often use easily obtained thermal reaction norms,such as that for locomotor performance (e.g., Blouin-Demers andWeatherhead, 2008), as proximate measures of the effect of tem-perature on fitness instead of using more ultimate measures offitness, such as reproductive success, that can be more difficult toobtain. Indeed, the vast majority of tests of the thermal

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/jtherbio

Journal of Thermal Biology

http://dx.doi.org/10.1016/j.jtherbio.2015.06.0020306-4565/& 2015 Elsevier Ltd. All rights reserved.

n Corresponding author.E-mail address: whall075@uottawa.ca (W.D. Halliday).

Journal of Thermal Biology 52 (2015) 108–116

castaneum and T. confusum living together in a thermal gradientsegregated according to temperature, where T. castaneum wasaround 30 °C while T. confusum was around 25 °C. The similarvalues of Tset obtained 40 years apart for the two species suggest itis a conserved trait.

3.2. Egg laying potential

T. castaneum laid a similar number of eggs in the two 48-h egglaying periods at 30 °C (days 1–2 mean7S.E.¼19.672.7, days 7–8¼21.974.0; t15¼0.3759, p¼0.71), and the same was true of T.confusum (days 1–2¼9.472.0, days 7–8¼11.972.6; t14¼2.0176,p¼0.06), although T. confusum laid approximately half the eggslaid by T. castaneum. Overall, 58% of T. castaneum pairs and 53% ofT. confusum pairs increased egg output between the first andsecond period, 35% of T. castaneum pairs and 37% of T. confusumpairs decreased egg output between the first and second period,while 7% of T. castaneum pairs and 10% of T. confusum pairs did notchange their egg output between the first and second period.Therefore, the changes in egg laying rate as a function of tem-perature we observed cannot be attributed to changes in egglaying potential as the experiment progressed.

3.3. Thermal reaction norms

All performances, except egg laying rate in T. confusum, were fitbest by a Ratkowsky curve (Table 3). For this reason and to easecomparison, we fit Ratkowsky curves to all individual data. T.castaneum achieved 80% of its maximum performance (B80) be-tween 20 and 34 °C for righting time and for egg laying, and be-tween 26 and 38 °C for sprint speed (Table 4, Fig.1A). Similarly, T.confusum achieved 80% of its maximum performance (B80) be-tween 23 and 37 °C for righting time and sprint speed, and be-tween 21 and 37 °C for egg laying (Table 4, Fig.1B). Optimal tem-peratures and B80 did not differ significantly between the twospecies (p¼0.12). In both species, the values of To (po0.01), lowerB80 (po0.01), and upper B80 (p¼0.04) were lower for egg layingthan for sprint speed, while the values of To (p¼0.01) and lowerB80 (p¼0.02) were lower for righting time than for sprint speed.

We compared the thermal reaction norms of performance tothe thermal preference of T. castaneum (Table 4, Fig. 1A). Forrighting time and for egg laying, the B80 was broader than Tset: thelower B80 was colder than the lower bound of Tset and the upperB80 was hotter than the upper bound of Tset while the medianselected temperature matched To. For sprint speed, To exceededthe median selected temperature while the upper B80 exceededthe upper bound of Tset. For egg laying, the lower B80 was colderthan lower Tset and the upper B80 was warmer than upper Tsetwhile the median selected temperature matched To. We alsocompared the thermal reaction norms of performance to thethermal preference of T. confusum (Table 4, Fig. 1B). For all threereaction norms, B80 extended into warmer temperatures than Tset:To and upper B80 were warmer than the median selected tem-perature and the upper bound of Tset, while the lower B80 matchedthe lower bound of Tset.

3.4. Concordance of locomotor and reproductive performances

T. castaneum laid more eggs than T. confusum (t1,219¼2.80,po0.01). The number of eggs laid by both species was positivelycorrelated with male righting time (t1,219¼2.87, po0.01; Table 5,Fig. 2a) and with female sprint speed (t1,219¼3.66, po0.001;Fig. 2b), although in both species the variance in egg laying rateexplained by locomotor performance was relatively small. Withineach sex, righting time and sprint speed were moderately corre-lated (female: r¼0.25; male: r¼0.28).

Among the variables considered, path analysis (Fig. 3) de-monstrated that the combined effect of temperature and tem-perature squared had the strongest effect on the number of eggslaid (partial R2¼0.12), with female sprint speed having the nextstrongest effect (partial R2¼0.06), and male righting time (partialR2¼0.02) and species (partial R2¼0.02) having the smallest con-tributions to the number of eggs laid. Temperature also had aneffect on female sprint speed (partial R2¼0.11) and on male

Fig. 1. Thermal reaction norms of red flour beetles (Tribolium castaneum) (A) andconfused flour beetles (Tribolium confusum) (B) for righting time, sprint speed, andeggs laid over 24 h. All three performance measures were transformed to a percentof their maximum value. The histograms at the bottom of the figures represent thetemperatures selected by the species in a thermal gradient ranging from 20 to40 °C, and the gray boxes represent the thermal preference (Tset; interquartilerange) for the species. Curves represent the lines of best fit for a non-linear curvewith a linear increase and exponential decay from Ratkowsky et al. (1983). Thehorizontal line at 80% is to help visualize the performance breadth (B80). The op-timal temperature (To) for each performance measure is marked by a dot at thepeak of the curve.

W.D. Halliday, G. Blouin-Demers / Journal of Thermal Biology 52 (2015) 108–116112

castaneum and T. confusum living together in a thermal gradientsegregated according to temperature, where T. castaneum wasaround 30 °C while T. confusum was around 25 °C. The similarvalues of Tset obtained 40 years apart for the two species suggest itis a conserved trait.

3.2. Egg laying potential

T. castaneum laid a similar number of eggs in the two 48-h egglaying periods at 30 °C (days 1–2 mean7S.E.¼19.672.7, days 7–8¼21.974.0; t15¼0.3759, p¼0.71), and the same was true of T.confusum (days 1–2¼9.472.0, days 7–8¼11.972.6; t14¼2.0176,p¼0.06), although T. confusum laid approximately half the eggslaid by T. castaneum. Overall, 58% of T. castaneum pairs and 53% ofT. confusum pairs increased egg output between the first andsecond period, 35% of T. castaneum pairs and 37% of T. confusumpairs decreased egg output between the first and second period,while 7% of T. castaneum pairs and 10% of T. confusum pairs did notchange their egg output between the first and second period.Therefore, the changes in egg laying rate as a function of tem-perature we observed cannot be attributed to changes in egglaying potential as the experiment progressed.

3.3. Thermal reaction norms

All performances, except egg laying rate in T. confusum, were fitbest by a Ratkowsky curve (Table 3). For this reason and to easecomparison, we fit Ratkowsky curves to all individual data. T.castaneum achieved 80% of its maximum performance (B80) be-tween 20 and 34 °C for righting time and for egg laying, and be-tween 26 and 38 °C for sprint speed (Table 4, Fig.1A). Similarly, T.confusum achieved 80% of its maximum performance (B80) be-tween 23 and 37 °C for righting time and sprint speed, and be-tween 21 and 37 °C for egg laying (Table 4, Fig.1B). Optimal tem-peratures and B80 did not differ significantly between the twospecies (p¼0.12). In both species, the values of To (po0.01), lowerB80 (po0.01), and upper B80 (p¼0.04) were lower for egg layingthan for sprint speed, while the values of To (p¼0.01) and lowerB80 (p¼0.02) were lower for righting time than for sprint speed.

We compared the thermal reaction norms of performance tothe thermal preference of T. castaneum (Table 4, Fig. 1A). Forrighting time and for egg laying, the B80 was broader than Tset: thelower B80 was colder than the lower bound of Tset and the upperB80 was hotter than the upper bound of Tset while the medianselected temperature matched To. For sprint speed, To exceededthe median selected temperature while the upper B80 exceededthe upper bound of Tset. For egg laying, the lower B80 was colderthan lower Tset and the upper B80 was warmer than upper Tsetwhile the median selected temperature matched To. We alsocompared the thermal reaction norms of performance to thethermal preference of T. confusum (Table 4, Fig. 1B). For all threereaction norms, B80 extended into warmer temperatures than Tset:To and upper B80 were warmer than the median selected tem-perature and the upper bound of Tset, while the lower B80 matchedthe lower bound of Tset.

3.4. Concordance of locomotor and reproductive performances

T. castaneum laid more eggs than T. confusum (t1,219¼2.80,po0.01). The number of eggs laid by both species was positivelycorrelated with male righting time (t1,219¼2.87, po0.01; Table 5,Fig. 2a) and with female sprint speed (t1,219¼3.66, po0.001;Fig. 2b), although in both species the variance in egg laying rateexplained by locomotor performance was relatively small. Withineach sex, righting time and sprint speed were moderately corre-lated (female: r¼0.25; male: r¼0.28).

Among the variables considered, path analysis (Fig. 3) de-monstrated that the combined effect of temperature and tem-perature squared had the strongest effect on the number of eggslaid (partial R2¼0.12), with female sprint speed having the nextstrongest effect (partial R2¼0.06), and male righting time (partialR2¼0.02) and species (partial R2¼0.02) having the smallest con-tributions to the number of eggs laid. Temperature also had aneffect on female sprint speed (partial R2¼0.11) and on male

Fig. 1. Thermal reaction norms of red flour beetles (Tribolium castaneum) (A) andconfused flour beetles (Tribolium confusum) (B) for righting time, sprint speed, andeggs laid over 24 h. All three performance measures were transformed to a percentof their maximum value. The histograms at the bottom of the figures represent thetemperatures selected by the species in a thermal gradient ranging from 20 to40 °C, and the gray boxes represent the thermal preference (Tset; interquartilerange) for the species. Curves represent the lines of best fit for a non-linear curvewith a linear increase and exponential decay from Ratkowsky et al. (1983). Thehorizontal line at 80% is to help visualize the performance breadth (B80). The op-timal temperature (To) for each performance measure is marked by a dot at thepeak of the curve.

W.D. Halliday, G. Blouin-Demers / Journal of Thermal Biology 52 (2015) 108–116112

Thermorégulation Méthode scientifique II

• Motif : les activités de thermorégulation varient entre les sexes d’une même espèce

• Hypothèse : les femelles gravides doivent maintenir une haute Tc pour favoriser le développement de leurs oeufs

• Prédiction : plus de temps en thermorégulation pour les femelles gravides que pour les femelles non-gravides et les mâles

Cold climate specialization: Adaptive covariation between metabolicrate and thermoregulation in pregnant vipers

Olivier Lourdais a,b,⁎, Michaël Guillon a, Dale DeNardo b, Gabriel Blouin-Demers c

a Centre d'Études Biologiques de Chizé, CNRS, 79 360, Villiers en Bois, Franceb School of Life Sciences, Arizona State University, 85287-4501 Tempe, AZ, USAc Département de Biologie, Université d'Ottawa, Ottawa, Ontario K1N 6N5, Canada

H I G H L I G H T S

• Climatic specialization is a critical question in ectotherms.• We studied thermoregulation and metabolic rate during pregnancy in two vipers.• The cold specialist is a more efficient thermoregulator with higher metabolic rate.• Covariation between physiology and behavior may be a key component of climatic adaptations.

a b s t r a c ta r t i c l e i n f o

Article history:Received 18 February 2013Accepted 24 May 2013

Keywords:EctothermGestationThermoregulationBoreal climateMetabolism

We compared thermoregulatory strategies during pregnancy in two congeneric viperid snakes (Vipera berusand Vipera aspis) with parapatric geographic ranges. V. berus is a boreal specialist with the largest known dis-tribution among terrestrial snakes while V. aspis is a south-European species. Despite contrasted climatic af-finities, the two species displayed identical thermal preferences (Tset) in a laboratory thermal gradient. Underidentical natural conditions, however, V. berus was capable of maintaining Tset for longer periods, especiallywhen the weather was constraining. Consistent with the metabolic cold adaptation hypothesis, V. berusdisplayed higher standard metabolic rate at all temperatures considered. We used the thermal dependenceof metabolic rate to calculate daily metabolic profiles from body temperature under natural conditions. Theboreal specialist experienced higher daily metabolic rate and minimized gestation duration chiefly because ofdifferences in the metabolic reaction norms, but also superior thermoregulatory efficiency. Under cold climates,thermal constraints should make precise thermoregulation costly. However, a shift in the metabolic reactionnormmay compensate for thermal constraints andmodify the cost–benefit balance of thermoregulation. Covari-ation betweenmetabolic rate and thermoregulation efficiency is likely an important adaptation to cold climates.

© 2013 Elsevier Inc. All rights reserved.

1. Introduction

Environmental temperature is especially important in shaping thedistribution of ectotherms because of their reliance on behavioralmechanisms for thermoregulation [1–4]. Unraveling the proximatemechanisms that enable climate specialization is vital to better un-derstand thermoregulatory strategies. Such an understanding is im-portant because the response of particular species to climate changewill likely depend on their physiological tolerances [5]. For example,tropical ectotherms are supposedly particularly at risk from climatechange because of their narrow thermal tolerances, which confer a re-duced “thermal safety margin” [4,6]. Cold-adapted species, however,

may also be particularly vulnerable to anticipated climate changefor at least two reasons: (i) the magnitude of predicted changes intemperature increases with latitude, and (ii) adaptations specific tocold temperature (i.e., higher metabolic rate) will be disadvantageousbecause the predicted temperature changes will result in further in-creased maintenance costs [7].

Thermoregulatory strategies of ectotherms vary from active ther-moregulation to thermal conformity, and theoretical models predictthat the optimal investment in thermoregulation should balance costsand benefits [8]. The cost–benefit model of thermoregulation predictslimited thermoregulatory effort (thermal conformity) in habitats withlow thermal quality [8], but empirical data provide only ambiguoussupport for this prediction [9,10]. One reason for the limited supportmight be that specific shifts in thermal requirementsmay enable the ex-ploitation of a given habitat [11]. Specifically, the thermal sensitivity ofmetabolic rate constitutes an important aspect of thermal adaptation

Physiology & Behavior 119 (2013) 149–155

⁎ Corresponding author at: CEBC-CNRS UPR 1934, 79 360 Villiers en Bois, France.Tel.: +33 5 49 09 96 16; fax: +33 5 49 09 65 26.

E-mail address: Lourdais@cebc.cnrs.fr (O. Lourdais).

0031-9384/$ – see front matter © 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.physbeh.2013.05.041

Contents lists available at SciVerse ScienceDirect

Physiology & Behavior

j ourna l homepage: www.e lsev ie r .com/ locate /phb

Thermorégulation Méthode scientifique III

• Motif : les activités de thermorégulation varient entre les individus d’une même espèce et d’un même sexe

• Hypothèse : les individus ayant un estomac plein doivent maintenir une haute Tc pour favoriser la digestion

• Prédiction : plus de temps en thermorégulation pour les individus ayant une grande consommation de nourriture

Journal of Animal Ecology

2001

70

, 1006–1013

© 2001 British Ecological Society

Blackwell Science Ltd

An experimental test of the link between foraging, habitat selection and thermoregulation in black rat snakes

Elaphe obsoleta obsoleta

GABRIEL BLOUIN-DEMERS* and PATRICK J. WEATHERHEAD†

Department of Biology, Carleton University, 1125 Colonel By Drive, Ottawa (ON), K1S 5B6, Canada

Summary

1.

Most physiological processes are temperature-dependent. Thus, for ectotherms,behavioural control of body temperatures directly affects their physiology. Ectothermsthermoregulate by adjusting habitat use and therefore thermoregulation is probably thesingle most important proximate factor influencing habitat use of terrestrial reptiles, atleast in temperate climates.

2.

Snakes have been shown to raise their body temperature following feeding in alaboratory thermal gradient, presumably to enhance digestion. This experiment wasexported to the field to explore the link between feeding, habitat selection and thermo-regulation in free-ranging snakes.

3.

Experimental feeding was conducted in the laboratory and in the field on black ratsnakes (

Elaphe obsoleta obsoleta

) that had temperature-sensitive radio-transmitterssurgically implanted.

4.

Snakes had higher mean body temperatures following feeding than prior to feedingin a laboratory thermal gradient.

5.

Some, but not all evidence, indicated that black rat snakes increased their meanbody temperature following feeding in the field. Indices of thermoregulation indicatedthat the snakes thermoregulated more carefully and more effectively after they hadeaten.

6.

Forest edges provided the best opportunities for thermoregulation in the study area.Black rat snakes were less likely to move following feeding when fed in edges than whenfed in the forest and were more likely to be found in edges following feeding, whetherthey had been fed in the forest or in an edge.

7.

Results of this study and one previous study suggest that thermoregulatory behaviourof snakes following feeding in the laboratory is a reliable predictor of their behaviour inthe field. A review of 13 studies of the thermoregulatory behaviour of snakes followingfeeding in the laboratory revealed that not all species behave similarly. However, thequality and number of studies currently available is not adequate for testing hypothesesabout which species should change thermoregulatory behaviour in response to eatingand which should not.

Key-words

:

Black rat snake, movements, postprandial thermophily, radio-telemetry,supplemental feeding.

Journal of Animal Ecology

(2001)

70

, 1006–1013

Correspondence: Gabriel Blouin-Demers, Department of Evolution, Ecology and Organismal Biology, Botany and ZoologyBuilding, Ohio State University, 1735 Neil Avenue, Columbus, OH 43210-1293, USA. E-mail: gblouind@ccs.carleton.ca*Present address and correspondence: Department of Evolution, Ecology and Organismal Biology, Botany and Zoology Building,Ohio State University, 1735 Neil Avenue, Columbus, OH 43210-1293, USA. †Present address: Program in Ecology and Evolutionary Biology, University of Illinois, Shelford Vivarium, 606 E. Healey St.Champaign, IL 61820, USA.

JAE_554.fm Page 1006 Friday, November 2, 2001 5:24 PM

1010G. Blouin-Demers & P.J. Weatherhead

© 2001 British Ecological Society, Journal of Animal Ecology, 70,1006–1013

the trial, these 37 trials produced only 10 completeobservations on eight individuals (an observation wasconsidered complete if temperatures were recorded for≥ 90% of the 24-h period prior to and following feed-ing. The 10 complete trials consisted of 2052 Tb read-ings that reduced to 355 individual hourly mean Tbs.The values for Tset (26·5 °C to 29·8 °C) used to calculatethermoregulation indices in the following analysescome from data for 32 post-absorptive individuals testedin the thermal preference chamber (Blouin-Demers &Weatherhead 2001b). For each of the eight snakes, amean Tb, Te, db, de and Ex (for the whole day and fordaytime only) were calculated for the 24 h prior to feed-ing and the 24 h following feeding. To test whether ornot snakes increased their Tb following feeding in the field,mean Te and feeding status were used as independentvariables and mean Tb as the dependent variable in anancova. Least square mean Tbs (corrected for Tes) ofsnakes following feeding for the whole day or daytimeonly were not significantly different from Tbs main-tained prior to feeding at α = 0·05 (Table 2). However,the indices of thermoregulation were higher followingfeeding. The index de – db was ≈1 °C higher followingfeeding and Ex was ≈18% higher following feeding.However, paired t-tests indicated that none of thesedifferences was statistically significant at α = 0·05(Table 2).

Post-hoc analyses indicated that, despite large effectsizes (all ds were ≥ 0·75), power was ≤ 0·25 for all testsand that a sample size of ≥ 55 would have been necessaryto achieve acceptable power (≈ 0·80: Stevens 1996). Giventhe difficulties involved in collecting this type of data, itis highly unlikely that such a substantial data set willbecome available in the near future. Under conditionsof large effect sizes and low power due to small samplesizes, Cohen (1977) and Stevens (1996) have suggestedadopting a more lenient α level (0·10 or 0·15) to improvepower sharply, particularly if the consequences ofmaking a Type I error are not serious, such as in thepresent case. At α = 0·15, several effects are significant(de – db whole day and Ex day only: Table 2).

Another source of data to assess the effect of feedingon thermoregulation comes from unfed snakes trackedduring the same time period as part of a larger telem-etry study (Blouin-Demers & Weatherhead 2001b).This analysis lacks the paired feature of the previousanalysis, but provides more extensive data from snakes

that we assume were primarily doing things other thandigesting meals. Snakes that had been fed thermore-gulated more carefully and at higher Tbs. Fed snakeshad significantly higher Tb (26·7 ± 1·1 °C vs. 23·9 ±0·4 °C, t59 = 2·43, P = 0·018), significantly higher de – db

(4·4 ± 0·8 °C vs. 2·6 ± 0·3 °C, t59 = 1·96, P = 0·050) andsignificantly higher Ex (51·8 ± 4·5% vs. 24·8 ± 1·7%,t59 = 5·60, P < 0·001) than unfed snakes.

habitat use and movement by snakes before and after feeding

A total of 23 free-ranging individuals was fed a total of37 times (23 times in edges and 14 times in the forest).We considered that a snake was faced with a binomialchoice following feeding: to be in an edge or not. Afterbeing fed in an edge (23 times), snakes were relocatedin an edge the following day 21 times (Fig. 1), whichdeviates significantly from random (P < 0·001) basedon the proportions of the binomial distribution. After

Table 2. Summary of the thermoregulatory behaviour of black rat snakes prior to and after feeding in the field

Thermoregulation index Period Before feeding After feeding Significance

Mean Tb* Whole day 26·5 ± 0·5 °C 26·7 ± 0·5 °C F1,13 = 0·013 P = 0·91Day only 26·8 ± 0·6 °C 27·2 ± 0·6 °C F1,13 = 0·239 P = 0·63

de – db Whole day 3·3 ± 0·7 °C 4·4 ± 0·8 °C Paired t7 = 1·255 P = 0·12Day only 3·1 ± 0·8 °C 4·1 ± 0·6 °C Paired t7 = 0·891 P = 0·20

Ex Whole day 32 ± 12% 52 ± 5% Paired t7 = 0·775 P = 0·23Day only 31 ± 13% 51 ± 8% Paired t7 = 1·220 P = 0·13

*Mean Tb was adjusted for differences in mean Te by using least-square means.

Fig. 1. Number of observations of black rat snakes in forestor edges before and after being fed in either edge or foresthabitat. Statistically significant differences based on theproportions of the binomial distribution are indicated by anasterisk.

JAE_554.fm Page 1010 Friday, November 2, 2001 5:24 PM

1010G. Blouin-Demers & P.J. Weatherhead

© 2001 British Ecological Society, Journal of Animal Ecology, 70,1006–1013

the trial, these 37 trials produced only 10 completeobservations on eight individuals (an observation wasconsidered complete if temperatures were recorded for≥ 90% of the 24-h period prior to and following feed-ing. The 10 complete trials consisted of 2052 Tb read-ings that reduced to 355 individual hourly mean Tbs.The values for Tset (26·5 °C to 29·8 °C) used to calculatethermoregulation indices in the following analysescome from data for 32 post-absorptive individuals testedin the thermal preference chamber (Blouin-Demers &Weatherhead 2001b). For each of the eight snakes, amean Tb, Te, db, de and Ex (for the whole day and fordaytime only) were calculated for the 24 h prior to feed-ing and the 24 h following feeding. To test whether ornot snakes increased their Tb following feeding in the field,mean Te and feeding status were used as independentvariables and mean Tb as the dependent variable in anancova. Least square mean Tbs (corrected for Tes) ofsnakes following feeding for the whole day or daytimeonly were not significantly different from Tbs main-tained prior to feeding at α = 0·05 (Table 2). However,the indices of thermoregulation were higher followingfeeding. The index de – db was ≈1 °C higher followingfeeding and Ex was ≈18% higher following feeding.However, paired t-tests indicated that none of thesedifferences was statistically significant at α = 0·05(Table 2).

Post-hoc analyses indicated that, despite large effectsizes (all ds were ≥ 0·75), power was ≤ 0·25 for all testsand that a sample size of ≥ 55 would have been necessaryto achieve acceptable power (≈ 0·80: Stevens 1996). Giventhe difficulties involved in collecting this type of data, itis highly unlikely that such a substantial data set willbecome available in the near future. Under conditionsof large effect sizes and low power due to small samplesizes, Cohen (1977) and Stevens (1996) have suggestedadopting a more lenient α level (0·10 or 0·15) to improvepower sharply, particularly if the consequences ofmaking a Type I error are not serious, such as in thepresent case. At α = 0·15, several effects are significant(de – db whole day and Ex day only: Table 2).

Another source of data to assess the effect of feedingon thermoregulation comes from unfed snakes trackedduring the same time period as part of a larger telem-etry study (Blouin-Demers & Weatherhead 2001b).This analysis lacks the paired feature of the previousanalysis, but provides more extensive data from snakes

that we assume were primarily doing things other thandigesting meals. Snakes that had been fed thermore-gulated more carefully and at higher Tbs. Fed snakeshad significantly higher Tb (26·7 ± 1·1 °C vs. 23·9 ±0·4 °C, t59 = 2·43, P = 0·018), significantly higher de – db

(4·4 ± 0·8 °C vs. 2·6 ± 0·3 °C, t59 = 1·96, P = 0·050) andsignificantly higher Ex (51·8 ± 4·5% vs. 24·8 ± 1·7%,t59 = 5·60, P < 0·001) than unfed snakes.

habitat use and movement by snakes before and after feeding

A total of 23 free-ranging individuals was fed a total of37 times (23 times in edges and 14 times in the forest).We considered that a snake was faced with a binomialchoice following feeding: to be in an edge or not. Afterbeing fed in an edge (23 times), snakes were relocatedin an edge the following day 21 times (Fig. 1), whichdeviates significantly from random (P < 0·001) basedon the proportions of the binomial distribution. After

Table 2. Summary of the thermoregulatory behaviour of black rat snakes prior to and after feeding in the field

Thermoregulation index Period Before feeding After feeding Significance

Mean Tb* Whole day 26·5 ± 0·5 °C 26·7 ± 0·5 °C F1,13 = 0·013 P = 0·91Day only 26·8 ± 0·6 °C 27·2 ± 0·6 °C F1,13 = 0·239 P = 0·63

de – db Whole day 3·3 ± 0·7 °C 4·4 ± 0·8 °C Paired t7 = 1·255 P = 0·12Day only 3·1 ± 0·8 °C 4·1 ± 0·6 °C Paired t7 = 0·891 P = 0·20

Ex Whole day 32 ± 12% 52 ± 5% Paired t7 = 0·775 P = 0·23Day only 31 ± 13% 51 ± 8% Paired t7 = 1·220 P = 0·13

*Mean Tb was adjusted for differences in mean Te by using least-square means.

Fig. 1. Number of observations of black rat snakes in forestor edges before and after being fed in either edge or foresthabitat. Statistically significant differences based on theproportions of the binomial distribution are indicated by anasterisk.

JAE_554.fm Page 1010 Friday, November 2, 2001 5:24 PM

Thermorégulation Méthode scientifique IV

• Motif : les activités de thermorégulation varient entre les individus d’une même espèce et d’un même sexe pendant la journée

• Hypothèse : les individus passent plus de temps en thermorégulation lorsque Ta est loin de To

• Prédiction : plus de temps en thermorégulation en début et en fin de journée qu’en milieu de journée

Urosaurus ornatus

Sceloporus virgatus

Sceloporus jarrovii

Collecte de données

• Pour chaque espèce, notez si l’individu était à l’ombre ou au soleil au moment où vous l’avez aperçu

• Nous pourrons calculer la proportion des individus qui faisaient de la thermorégulation par blocs de deux heures entre 7h00 et 17h00