Thermoregulation and aggregation in neonatal bearded dragons (Pogona vitticeps)

Physiology & Behavior 100 (2010) 180–186

Contents lists available at ScienceDirect

Physiology & Behavior j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / p h b

Thermoregulation and aggregation in neonatal bearded dragons (Pogona vitticeps) Jameel J. Khan, Jean M.L. Richardson, Glenn J. Tattersall ⁎ Department of Biological Sciences, Brock University, St. Catharines, ON, Canada L2S 3A1

a r t i c l e

i n f o

Article history: Received 19 October 2009 Received in revised form 19 February 2010 Accepted 19 February 2010 Keywords: Thermoregulation Reptile Sociality Thermal preference

a b s t r a c t Ectothermic vertebrates, such as reptiles, thermoregulate behaviorally by choosing from available temperatures in their environment. As neonates, bearded dragons (Pogona vitticeps) are often observed to aggregate in vertical strata. A proximate mechanism for this behavior is the thermal advantage of heat storage (i.e., grouped lizards benefit through a decreased surface area to volume ratio), although competition for limited thermal resources, or aggregation for social reasons are alternative explanations. This study was designed to gain an understanding of how aggregation and thermoregulation interact. We observed that both isolated and grouped individuals achieved a similar level of thermoregulation (mean Tb over trial) within a thermal gradient, but that individuals within a group had lower thermoregulatory precision. An experimental design in which light and ambient temperature (Ta) (20 versus 30 °C) were altered established that a light bulb (source of heat) was a limited and valuable resource to both isolated and grouped neonatal lizards. Lizards aggregated more when the light was on at both temperatures, suggesting that individuals were equally attracted to or repelled from the heat source, depending on the ambient temperature. These data suggest aggregation occurs in neonatal bearded dragons through mutual attraction to a common resource. Further, increased variability in thermal preference occurs in groups, demonstrating the potential for agonistic behaviors to compromise optimal thermoregulation in competitive situations, potentially leading to segregation, rather than aggregation. Crown Copyright © 2010 Published by Elsevier Inc. All rights reserved.

1. Introduction Many lizard species rely on their environment for the regulation of body temperature, sustaining a relatively constant preferred body temperature (PBT) through a suite of behaviors [4,19,40]. In fact, the PBT of lizards can be estimated by providing them with a choice of ambient temperatures to select from in a thermal gradient. Maintaining an optimally elevated body temperature (Tb) is essential for numerous physiological processes including metabolism, endurance capacity, locomotor performance, enzyme function, and digestion; all of these ultimately affect development and survival [1,13,24,29,30]. The thermal biology of reptiles can also influence their foraging success [2], with behaviors such as perception, coordination, and muscle function being enhanced at higher temperatures. Thus, factors that impede the ability to thermoregulate are expected to be of importance to thermal biology and fitness of reptiles [34]. For example, the precision of thermoregulation of the desert iguana (Dipsosaurus dorsalis) was shown to be affected by both the presence of predators and territorial fights [14], resulting in lizards tolerating body temperatures upwards of 46.5 °C (versus normal upper threshold levels of 38.5 °C in a thermal gradient). Huey and Slatkin [24] devised a mathematical cost–benefit model for species-level

⁎ Corresponding author. Tel.: + 1 905 688 5550x4815; fax: + 1 905 688 1855. E-mail address: [email protected] (G.J. Tattersall).

lizard thermoregulatory patterns, predicting that the precision of thermoregulation is affected by extrinsic factors that interrupt thermoregulation for other behaviors such as sociality, predator avoidance, territoriality, and/or feeding. The active Tb of lizards in nature, however, is not always the same as their PBT in laboratory thermal gradients [5]. Indeed, thermoregulatory behavior is quite context and species-specific; lizards may elect not to thermoregulate when the costs associated are too high [24] or preferred temperatures are unavailable, instead conforming to their environmental conditions. The potential for social interactions to impose or alleviate constraints and thereby influence thermoregulation, however, warrants consideration [25]. Numerous studies, however, have demonstrated that some reptilian groups exhibit rather complex social behaviors [8,11,12,27,31,44]. Hierarchy formation can be induced in reptiles due to crowding in captive laboratory environments, and can be eliminated under environments with unlimited resources [3,7,15,30,33,36,38]. It is likely the spatial constraints, limited access to desired microclimatic features, and the transient nature of food provisioning that produce these hierarchies. In fact, raising captive male green iguanas (Iguana iguana) with limited access to resources will cause two distinct groups to develop: rapid growing dominant and slow growing subordinate individuals [30]. Dominant males maintain access to supplemental heat sources and perch sites twice as often as subordinate males, which is associated with higher digestive efficiencies [30]. Thus, aggressive tendencies may be linked to the

0031-9384/$ – see front matter. Crown Copyright © 2010 Published by Elsevier Inc. All rights reserved. doi:10.1016/j.physbeh.2010.02.019

J.J. Khan et al. / Physiology & Behavior 100 (2010) 180–186

competitive context of the environment, which leads to the prediction that hierarchy establishment is intensified depending on the availability and importance of a resource. Due to their use of behavioral thermoregulation and the potential competition for optimal thermal resources, squamate reptiles offer a valuable opportunity to explore the basic principles of aggregation (social or otherwise) with respect to thermal motives. It is possible that aggregations of reptiles serve a physiological purpose (e.g. thermoregulation), which may have set the stage for the evolution of more complex sociality [26]. Indeed, previous research has found mutual attraction to conspecifics in the cordylid lizard, Cordylus cataphractus, where 85% of individuals were aggregated within retreat sites, despite the unlimited availability of these refugia [43]. However, thermoregulation could also be facilitated by mutual attraction to habitat features [42] coincidentally resulting in aggregation that would not necessarily be evidence for sociality [18]. When neonatal squamates hatch from large clutches, their proximity to one another allows for social interaction [18]. Indeed, in captivity neonatal bearded dragons (Pogona vitticeps) are often observed to sit atop one another on basking perches, similar to neonatal aggregations of other reptiles [32]. Neonatal aggregation is thought to function primarily to reduce predation rates through predator saturation, dilution, or contagion, but may also provide physiological benefits [17,18]. The possible interaction between environmental pressures (and subsequent thermoregulation) and inter-individual behaviors (including aggregation) in lizards of this life stage is an area that has not been explored in great detail. These aggregations seem more curious considering that neonates exhibit highly agonistic behaviors when food is provided [see also 38]. The apparent aggregative behaviors that occur in neonatal lizards intimate a physiological and/or social significance to neonatal fitness, yet the potential benefits and proximate mechanisms involved in aggregation need to be clarified. The focus of this study is the interplay between sociality and thermoregulation in neonatal bearded dragons. Aggregations may reflect social tendencies or may exist as a behavioral mechanism involved in modulating heating and cooling rates [6,29,35], suggesting them to be an adaptive response to variable climatic environments. Yet, aggregation may also occur simply due to the fact that these lizards are attracted to similar microclimatic features. The latter scenario creates the potential for intraspecific competition that may affect thermoregulatory capabilities and subsequently lead to dominance hierarchy establishment [30]. Specifically, we were interested in determining how neonatal bearded dragons (P. vitticeps) interact for a limited resource, and how their thermoregulatory capabilities (i.e., level and precision of thermoregulation) are affected by the presence of conspecifics. We hypothesized that: 1) individual thermoregulatory behavior in neonatal groupings will be affected by social interactions, thus influencing their attainment of optimal, precise temperature regulation, and 2) neonates will aggregate to a greater extent in response to a limited resource (a source of heat). 2. Materials and methods 2.1. Animals and husbandry During the course of this study a total of 79 neonatal bearded dragons (P. vitticeps) were used in experimentation (average mass 3.16 ± 0.86 g). They were acquired from eggs raised from multiple pairings of P. vitticeps at Brock University. All lizards were studied at a developmental stage at which sex could not be determined. Animals were housed in size-matched groupings within opaque plastic containers (460 mm L × 250 mm W × 200 mm H), and were therefore already ‘socialized’. The floor inside was covered with paper towel (changed weekly) and twigs were provided as perch sites. A 12:12 h light:dark cycle was maintained, where lizards had access to a 100 W

181

light bulb for heat during times of light. A mixed diet was fed including live crickets dusted with vitamin powder (fed daily) and leafy green vegetables with carrot shavings (fed 2–3 times per week). Water was sprayed into each housing unit daily in order to ensure proper hydration. Each individual was marked with a unique pattern of non-toxic paint blotches for identification. The marking procedure caused minimal harm or distress to the lizards and they were observed to engage in normal feeding behaviors only minutes afterwards. If a lizard underwent a bout of ecdysis, it was necessary to re-mark that individual; these individuals were not immediately examined for thermoregulatory preference since ecdysis may have influenced this behavior. Marking individuals ensured that we could track which animal went into each experiment, however, within group trials the marking was not sufficiently visible to track individuals with our video analysis. Animal husbandry and experimental protocols were approved by Brock University's Committee on Animal Care and Use (ACUC) and are consistent with the ethics of animal experimentation as set out by the Canadian Council on Animal Care (CCAC). 2.2. Experiment 1: thermal gradient trials A thermal gradient was used to answer questions regarding thermoregulatory capabilities (i.e., ability to achieve optimal temperature regulation) in the presence and absence of other individuals. For this, a gradient of floor temperatures was established by fastening a sheet of copper (550 mm L × 250 mm W) across two adjacent heating/ cooling plates. One plate cooled one side of the copper, while the adjacent plate heated the opposite side creating a smooth gradient of floor temperatures between ~ 19 °C and ~ 45 °C. Floor temperatures (using a thermally non-reflective material with emissivity N0.95) were measured using a thermal imaging camera (Mikron Instruments Model 7515) and software (MikroSpec R/T Version 2.1394 by Mikron Infrared Inc.). The copper plate was divided horizontally into 4 sections (62.5 mm/section), allowing for multiple simultaneous assessments of thermoregulatory behavior. Data were collected from a total of 40 lizards. At the commencement of each trial, either isolated (N = 10) or grouped (N = 10 groups; 3 individuals/group) lizards were placed into the warm side of each section of the gradient and a 1 h habituation period was provided. Images of the gradient using the thermal imaging software were collected at 30 s intervals, sufficiently rapid to track a focal individual over time. The second hour of recording was considered for analysis whereby the selected temperature (T s; collected as the temperature of the gradient floor at the dorsal mid-line location) and body temperature (Tskin; collected as temperature of the skin at the dorsal mid-line location) of each individual was extracted from images collected every 60 s, as described previously [41]. Selection of this anatomical site is supported in the literature, as it is easy to consistently locate and most accurately approximates the locale for internal body temperatures [39]. In separate experiments where we monitored T b and T skin simultaneously across a range of ambient temperatures (25 to 40 °C), a very high and near unitary correlation was observed (T skin = 0.0431 + 0.954 × T b , r 2 = 0.999). Being slightly anterior, this site also accounts for evidence suggesting that cephalic temperatures may be the primary regulated variable in reptiles rather than core body temperatures [40]. The level of thermoregulation was determined by an individual's mean T skin over the 1 h period. Thermoregulatory precision was quantified by the standard deviation (SD) of an individual's T skin over the 1 h period. T s data were used primarily in characterizing the behavioral pattern of thermoregulation, whereas the T skin data were used to describe the resultant effects of that behavior on an estimate of body temperature. Experiments were conducted between 1100 and 1400 h.

182

J.J. Khan et al. / Physiology & Behavior 100 (2010) 180–186

2.3. Experimental 2: light orientation and aggregation trials During experimentation, neonatal lizards were placed into a white plastic rectangular container (30 cm L × 26 cm W × 13 cm H) that remained within a temperature controlled incubator (ThermoForma Diurnal Growth Chamber, Thermo Electron Corp.), allowing for the preservation of a constant ambient temperature (Ta = 20 °C or 30 °C ± 0.1 °C) set prior to each trial. For the purposes of thermal and light attraction, three treatments were used: 1) a 50 W halogen light bulb (hot light, wavelengths from 400 to 800 nm), 2) a 15 W fluorescent light bulb (cold light, wavelengths from 400 to 800 nm, although less intense and with reduced levels above 650 nm), and 3) a lamp switched to off (cold dark). Light sources were placed so that the light bulb was directed downwards into the container at the lip of one of its corners. For each experimental trial the light source was randomly placed at one of the four corners. In the case of the halogen light, the temperature on the floor under the light was always 5–6 °C higher than the set Ta in the incubator and the spread of its radiating heat on the floor of the container was always b5 cm away from the corner. Air flow throughout the chamber ensured that the floor and air temperatures in the viewing container were maintained at either 20 or 30 °C (depending on treatment), forcing basking under the hot light as the only means to achieve temperatures above 30 °C. An ambient temperature of no greater than 30 °C, which is below the thermal preference of these animals, was intentionally chosen to ensure the need for thermoregulatory behavior during the trial, as this was the response we wished to measure. The fluorescent light source (cold light) produced no heat and acted as a control to test whether lizards were attracted to light or to the heat provided from a light source. A camera (Flex-a-Vision, Ken-a-Vision©) was also fixed above the container at a height that enabled the whole container to be in view (0.85 m). The camera software (Applied Vision v2.0.2, Ken-a-Vision©) allowed for the collection of images at five-minute intervals over the course of each trial period. Two thermocouples were taped onto the edges of the container at opposite corners to monitor Ta. A diffuse, background light within the incubator was on under all experimental conditions to allow for consistent image acquisition and to provide visual cues for the lizards. All trials lasted for a period of 2 h. The first hour was assigned as a habituation period, beginning after the initial placement of the lizard(s) into the container. Lizards were weighed immediately prior to each trial and then placed directly into the container's center (i.e., perpendicular to the light). All experiments were conducted between 1100 and 1800 h. To assess how neonatal bearded dragons interact for a limited resource, required first establishing that a light source emitting heat is a resource that individuals use (i.e., are attracted towards) for thermoregulatory purposes, as well as to control for the influence of attraction to a perceived heat source (cold light). Thus, it was necessary to obtain information on how neonates use this thermal resource when absent from other individuals. Therefore, each light (hot light, cold light, cold dark) and temperature (20 and 30 °C) treatment described above was crossed with two further treatments: an isolated individual versus a group of three. Groupings were assigned using a random number generator to select individuals for each trio, while constraining each group to individuals from size- and age-matched cohorts. A unique grouping of lizards was used for each treatment combination (i.e., replicate). Treatment combinations were run in a random order, with a sample size of 8 per combination (N = 48 for the individual trials, N = 48 unique combinations of individual lizards for the group trials). The number of group combinations (48 × 3 = 144) necessitated the re-assignment of individual lizards into multiple groupings, however, care was taken to ensure that no lizard was studied more than once within a 3 day period of time, and used in no more than three group combinations. Data for all trials came from images taken during the second hour of each trial period (13 pictures analyzed per trial). SigmaScan

Pro© Image Analysis Software (Version 5.0.0) was employed to plot the x,y-coordinates of the lizard(s) in each image with respect to the coordinates of the light source (i.e., individual distance to the light). A conversion using known distances within the images was used to transform pixel distance into true distance. The position of the lizard was plotted as a point between the shoulders on the dorsal mid-line. Proximity to the light was measured as the shortest line distance between the lizard and the light. In the group treatment with three lizards present, this was measured in two different ways: 1) as proximity of the nearest individual to the light, and 2) as the average proximity to the light of the three individuals. Both measures were initially considered because we were unsure as to which would be most relevant. If interactions among individuals are such that one lizard dominates the other two, then the distance of the closest individual is a better indicator of preferred behavior. However, if no individual dominates and each individual's position reflects a compromise between being in the ideal position and interaction with other group members, then mean distance of all three to the light is a better indicator of preferred position. In grouped trials we also measured aggregation behavior, using the lizard closest to the light as the focal individual; the average distance between this individual and each of the other two lizards was used as a measure of nearest neighbor distance (NND), an indicator of the aggregation tendency of lizards, with smaller NND indicating greater aggregation. 2.4. Statistical analyses For experiment 1, both the mean level (Tb) and precision of thermoregulation (SD of Tb) were compared between grouped and isolated treatments using two-sample t-tests or Mann Whitney Rank Sum Tests in cases where normality was violated. For experiment 2, statistical analysis of proximity to the light values was done in SAS v. 9.1 (SAS Institute Inc., Cary, NC, USA) to test the effects of treatment (isolated or group), light (hot light, cool light, cool dark), and temperature (20 or 30 °C) on distance from the light source to the neonate(s). Within the group treatment, aggregation behavior was assessed using NND as the dependent variable, with light and temperature as independent variables. In all cases, an α = 0.05 was utilized. 3. Results 3.1. Experiment 1: thermal gradient Two of 10 isolated lizards and 7 of 30 lizards in the group trials moved to the cold side of the gradient and remained there for the 1 h period, resulting in an average T skin b 30 °C for that individual. This behavior was not linked to mass variation within the group (regression analysis of variance in T skin versus variance in mass across the three grouped lizards, p = 0.992, r 2 = 0.004). The level of thermoregulation (mean T skin ) between grouped and Table 1 Surface temperature ± SE for isolated and grouped neonatal bearded dragons. Category Isolated (neonate) Grouped (neonate) Adulta

+ conformers − conformers + conformers − conformers

N

Level of thermoregulation (°C)

10 8 30 23 10

32.3 ± 1.6 34.5 ± 0.7 32.5 ± 0.8 34.7 ± 0.3 34.7 ± 0.2

Note: Both categories were sampled with thermoconformers (i.e., nonthermoregulating lizards) included (+) and removed (−). a Also presented is the PBT or level of thermoregulation for adult bearded dragons as collected in our laboratory in another study [10].

J.J. Khan et al. / Physiology & Behavior 100 (2010) 180–186

183

ungrouped lizards did not differ (t-test, t(18) = − 0.12, p = 0.898; Table 1; Fig. 1A). When removing the effect of non-thermoregulating individuals from the data set, the level of thermoregulation was increased approximately 2 °C in both group and isolated treatments; again isolated and grouped treatments did not differ (t-test, t(16) = − 0.107, p = 0.916). Variation (SD) in each individual Tskin throughout a single trial was determined to compare thermoregulatory precision between isolated and grouped lizards. Tskin SD (N = 10, mean = 1.28) of isolated individuals was significantly different from lizards in groups (N = 10, mean = 2.36; Mann–Whitney, H = 3.86, p = 0.0494; Fig. 1B). Similar statistical results were obtained using a randomization and re-sampling procedure to test for significance.

Table 2 ANOVA summary table for analysis of lizard proximity to light.

3.2. Experiment 2: orientation and aggregation

light were not significant for the other four treatment combinations of cool dark/cool light by isolated/group (Fig. 2). However, at 20 °C proximity to the light was close to significantly greater in the cool dark condition than in the cool light condition for isolated individuals (Tukey adjusted t = 3.2929, p = 0.0604). This was not the case for grouped individuals, using either average or nearest individual measures (average distance Tukey adjusted t = 0.0065, p = 1.000), leading to the observed 3-way interaction. Analysis of nearest neighbor distance in the group treatment showed a clear effect of light treatment on NND, with hot light leading to smaller NND, indicating greater aggregation (Fig. 3). This was equally true for both temperature treatments (main effect of light, F(2,42) = 9.54, p b 0.001, with Tukey post-hoc revealing that hot light differed from both cold light and cold dark). In combination with the above proximity to the light data, this indicates that lizards at 20 °C lizards aggregate near the hot light and at 30 °C lizards aggregate far from the hot light.

The appropriate measure for proximity to light in the group treatment was not clear in advance of the study, so we considered both the mean proximity to the light of all three lizards and the proximity to the light of the individual closest to the light (the latter a more appropriate measure if one dominant lizard usurps the ideal position and keeps other lizards away). Using either measure, proximity to light revealed a significant 3-way interaction among isolated versus group treatment, light treatment, and temperature (F(2,84) = 4.62, p = 0.012 using mean proximity; F(2,84) = 3.37, p = 0.039 using nearest individual; Table 2 shows analysis for mean proximity, the metric we felt more appropriate based on our results). An isolated individual was positioned closer to the hot light on average when Ta was 20 °C than when Ta was 30 °C; the same trend was observed for temperature effect on the average lizard distance to the hot light in the group treatment, but the difference was not significant (Fig. 2). The differences between the two temperature treatments in proximity to

Fig. 1. Levels (A) and variability (B) of behavioral thermoregulation (Tskin) in neonatal Pogona vitticeps tested within a thermal gradient while solitary (I) and while in groups (G) of three. Shown are box plots depicting median (line), 25th and 75th percentiles (box) and the range of the data set (bars). Group values are determined based on the mean of the 3 lizards. Bars with the same letters above indicate similar values (based on t-tests), comparing individual to group mean values.

Effect

df

MS

F

p

Treatment (isolated versus grouped) Temperature Light Treatment × Temperature Treatment × Light Temperature × Light Treatment × Temperature × Light Error

1 1 2 1 2 2 2 84

1.1948 2.0692 1.2995 0.0541 0.1242 1.8254 0.6669 0.1442

8.29 14.35 9.01 0.38 0.86 12.66 4.62

0.0051 0.0003 0.0003 0.5418 0.4263 b0.0001 0.0124

Note: Grouped data were derived from the mean values of the 3 lizards in the group. Analysis was performed on ln-transformed data, to meet assumption of equal variances.

Fig. 2. Mean (±SE) proximity to light for individual neonatal Pogona vitticeps (A) and in groups of three (B; mean of all 3 lizard distances). A significant treatment by light by temperature interaction effect occurred. Bars with the same letter above them are not significantly different in a Tukey post‐hoc multiple comparison test.

184

J.J. Khan et al. / Physiology & Behavior 100 (2010) 180–186

The individual performance of neonates within the thermal gradient was likely influenced by pre-existing conditions. Aggression and dominant behavioral attributes in male Pseudemoia entercasteauxii lizards were correlated to higher PBTs [39]. At present, we have not detected any visible differences between individuals that could be correlated with their PBT, however differences in thermoregulatory patterns are likely attributed to an individual's performance ability. Since the microclimate occupied by an animal influences its physiological capacities and ultimately its ecological performance [22], it would be interesting to directly explore how these basic thermoregulatory strategies are related to individual success and whether they are linked to a lizard's status on a shy-bold continuum [37,39]. 4.2. Level and precision of thermoregulation

Fig. 3. Mean (±SE) nearest neighbor distance from a focal lizard (Pogona vitticeps) in the group treatment. Only the main effect of light treatment was significant, with hot light differing significantly from both cool light and cool dark.

4. Discussion The goal of this study was to examine the interplay between thermoregulation and aggregation to better understand social behavior in neonatal lizards. This was achieved by first establishing how thermoregulatory precision was influenced by groupings, and subsequently by assessing whether lizards would alter attraction to a limited resource (i.e., a heat source) required for regulation of body temperature. When providing heat, the light was established as an attractive resource. When the importance of this resource was increased (i.e., at a lower Ta), both isolated and grouped lizards were attracted to a significantly greater extent. Attraction to the light cannot be based on light cues alone, since lizards were not as attracted to non-heating light source. When the importance of the basking resource was decreased (i.e., Ta = 30 °C), however, both isolated and grouped lizards showed minimal attraction towards any light source when it was on, remaining at distances similar to those seen when the resource was not present (i.e., light off). These results imply that thermoregulatory needs are met primarily through individual drive to achieve a PBT, but that under group situations, this ability is influenced by interactions with conspecifics.

4.1. Patterns of behavioral thermoregulation Neonatal bearded dragons subjected to a thermal gradient were observed to select temperatures. However, while the upper extreme temperatures were often avoided, some lizards (~20%) remained in the cold end. In theory, the PBT of lizards was at a certain level (range), and in order to achieve that PBT they could move and select the corresponding temperature. Herczeg et al. [20] demonstrated that when given an opportunity to reach PBT, common lizards (Zootoca vivipara) that thermoregulated precisely maintained their daily level of activity and improved their body condition. However treatment groups that were unable to reach PBTs exhibited thermoconformation, decreased daily activity and did not improve body condition. The results of the present study suggest that lizards can indeed change their thermoregulatory strategy, based on the physiological benefits of maintaining optimal Tskin — when the PBT is available, lizards can thermoregulate with high precision (Fig. 1), but when it is unavailable and, thus, the physiological benefits decrease, lizards elect not to thermoregulate [20].

On average, neonatal lizards in groups were able to achieve the same Tskin as those that were isolated. Therefore, the overall level of thermoregulation was not affected by the presence of conspecifics. Interestingly, however, precision of thermoregulation was reduced in lizards that were grouped, suggesting that interactions between individuals impinged on optimal temperature regulation in some lizards. The fact that the proportion of different thermoregulatory patterns was not different between the group and individual trials suggests that the changes observed reflect interactions between individuals that altered the achievement of PBT rather than a switch in thermoregulatory strategy of each individual. Interestingly, when non-thermoregulating individuals were removed in our analysis, the level of neonatal thermoregulation was the same as that found in conspecific adults [9]. This information may be important in our understanding of lizard life history, particularly relating the in ovo environment contributions to thermoregulatory preference and/or the early development of thermosensation [21]. Nevertheless, caution should be excised in translating laboratory derived PBTs into the field. Given the various potential ecological constraints, it is important to note that field PBTs could be quite divergent from those obtained in laboratory situations where costs and benefits are generally not being realized. In contrast to our results with grouped lizards, Labra [25] showed in two Pristidactylus lizard species that the average Tb of grouped lizards was always about 1.2 °C lower than isolated individuals, and independent of group size. Labra [25] attributed the lower grouped Tb to a reduction in thermoregulatory ‘set-point’ due to the high costs associated with thermoregulation in crowded scenarios. This Tb reduction would result in a mean daily energy savings of about 6.5%, a potentially profitable strategy when crowded situations bring about other stressful situations such as a decrease in food availability [25]. That the mean Tskin of grouped lizards was not lower than isolated lizards in our study may reflect less severe agonistic interactions between neonate individuals in our study. Furthermore, Labra [25] studied lizards that were at maturity, larger in size, and not typically associated with group living. These are factors that may have enhanced conspecific competition for access to thermally optimal microhabitats. Indeed, in adult bearded dragons, when the environment is costly (i.e., requires locomotory efforts to attain PBT), the variability in PBT may increase, with or without a change in the level of thermoregulation [10]. Similar mechanisms are likely at work in neonatal bearded dragons; in other words, decreases in themoregulatory precision in grouped trials may be linked to increased energy expenditure during efforts to achieve their PBT. 4.3. Mutual attraction to a thermal resource or social aggregation The thermal gradient experiments suggest that competitive interactions are occurring within the groups, leading to alterations in thermoregulatory precision. If competition for an optimal thermal resource is occurring in neonatal lizards, how do they cope with these

J.J. Khan et al. / Physiology & Behavior 100 (2010) 180–186

situations, and do they exhibit ‘tolerance’ for each other? We attempted to address these questions by providing a limited thermal resource. The hot light proved to be an attractant to lizards, at least at 20 °C. At 30 °C, solitary lizards maintained a significantly greater distance from the hot light relative to 20 °C in the same light treatment, but not relative to 20 °C for the cool light or cool dark treatments. Lizards within groups also stayed closer to the hot light at 20 °C and were also more strongly aggregated than lizards in groups within other treatment combinations. Solitary individuals at 20 °C would associate more closely to the cool light than to cool dark, suggesting some attraction to the light even without heat. It is possible that an association is made with light and heat on the basis of experience, however, this association disappears under group dynamics. Overall, these results show some attraction to a cool light at 20 °C, with more random behavior occurring with respect to the cool dark source. At a Ta of 30 °C, closer to their PBT, lizards were observed to aggregate far from the hot light, suggesting that this aggregation was based on avoidance of a source of heat that was warmer than preferred. As hypothesized, the presence of a limited resource, a hot light, induced significantly greater aggregation than when it was not present during cool light and cool dark treatments. This implies that the lizards were mutually attracted to the light because of the microclimate it provided. The extent to which aggregation occurred however was not influenced by the importance of the resource, as NND was the same whether Ta was 20 or 30 °C (with lights on or off). What may be occurring is that the relative group cohesion is maintained whether grouped lizards aggregate toward or aggregate away from a basking source; indeed, aggregation thus reflects a mutual tolerance of conspecifics. It is plausible that being far from the heat source at 30 °C is equally important a resource to the lizards as being near to the heat source is at 20 °C, particularly if prolonged exposure to the hot light resulted in aversive temperatures. Nevertheless, these observations confirm that intraspecific interactions affect the thermoregulatory capabilities of a neonate lizard [23,25]. The advantages of thermal inertia manifest a possible explanation for behavioral aggregations in small to medium-sized ectotherms exposed to variable climatic conditions [6,16,29,35]. In static thermal environments, however, aggregations did not seem to underlie any mutual benefit in our studies — they do not show significantly greater aggregation in response to environments that present high thermoregulatory challenges (i.e., cold, dark conditions). Due to their low thermal inertia and minimal heat production, it is unlikely that ‘huddling’ at cold temperatures would augment the body temperature of the lizards in the steady state thermal conditions. A more plausible explanation for aggregation in neonatal P. vitticeps is that it is due to mutual attraction to microclimatic features and not a social behavior per se. Nevertheless, previous studies have shown that huddling serves a thermoregulatory advantage in ectotherms most likely because the increased thermal inertia enhances control over the rate of thermal exchange [35]. Whether lizards detect conspecific temperatures and “huddle” for thermoregulatory purposes, or aggregate prior to assessing temperatures of one another remains unknown. 5. Summary and conclusions The results of this study demonstrate that social interactions in neonatal bearded dragons play a role in individual thermoregulatory capability. In a thermal gradient, grouped lizards achieved a similar level of thermoregulation (PBT) as isolated individuals, but exhibited lower thermoregulatory precision. This may be linked to higher locomotory demands in grouped scenarios. Both isolated and grouped neonatal bearded dragons were attracted to a limited resource of heat when Ta permitted; lizards in groups aggregated due to mutual attraction to this limited microclimate, but did not actively aggregate

185

in response to cold Ta when no resource was available. Furthermore, the degree of aggregation was not dependent upon the importance of the basking resource. Aggregations observed in neonatal bearded dragons, therefore, are likely a mutual response to a desired, limited resource. When this resource is a microclimatic feature, individuals must find a balance between the costs and benefits of thermoregulation in the presence of others. These findings are relevant in consideration of the potential effects individuals may have on each other when interacting for thermally stable microenvironments in nature. Experiments of this sort have implications to our understanding of the challenges faced by lizards in the neonatal life stage — important information when attempting to rehabilitate endangered reptilian species in both captive and natural environments [28]. Further research into the interplay of aggregation and thermoregulation will help shed light on theories regarding the evolution of complex social systems in the animal kingdom.

Acknowledgements We would like to acknowledge Christopher Loewen and Viviana Cadena for assistance with experiments and Tom Eles for providing essential animal care. The research was conducted as part of Brock University's Biological Sciences BSc Honours Thesis program (JK), and was funded by the Natural Sciences and Engineering Research Council of Canada (NSERC) to GJT.

References [1] Autumn K, Jindrich D, DeNardo D, Mueller R. Locomotor performance at low temperature and the evolution of nocturnality in geckos. Evolution 1999;53: 580–99. [2] Ayers DY, Shine R. Thermal influences on foraging ability: body size, posture and cooling rate of an ambush predatory, the python Morelia spilota. Funct Ecol 1997;11:342–7. [3] Barker DG, Murphy JB, Smith KW. Social behavior in a captive group of Indian pythons, Python molurus (Serpentes, Boidae) with formation of a linear social hierarchy. Copeia 1979;1979:466–71. [4] Bartholomew GA, Tucker VA. Control of changes in body temperature, metabolism, and circulation by the agamid lizard, Amphibolurus barbatus. Physiol Zool 1963;36:199–218. [5] Blouin-Demers G, Nadeau P. The cost–benefit model of thermoregulation does not predict lizard thermoregulatory behavior. Ecology 2005;86:560–6. [6] Boersma PD. The benefits of sleeping aggregations in marine iguanas, Amphlyrhynchus cristatus. In: Burghardt GM, Rand AS, editors. Iguanas of the world: their behavior, ecology, and conservation. Park Ridge, New Jersey: Noyes; 1982. p. 292–9. [7] Brattstrom BH. Evolution of reptilian social behavior. Am Zool 1974;14:35–49. [8] Burghardt GM, Greene HW, Rand AS. Social behavior in hatchling green Iguanas — life at a reptile rookery. Science 1977;195:689–91. [9] Cadena V, Tattersall GJ. Decreased precision contributes to the hypoxic thermoregulatory response in lizards. J Exp Biol 2009;212:137–44. [10] Cadena V, Tattersall GJ. The effect of thermal quality on the thermoregulatory behavior of the bearded dragon Pogona vitticeps: influences of methodological assessment. Physiol Biochem Zool 2009;82:203–17. [11] Chapple DG. Ecology, life-history, and behavior in the Australian scinid genus Egernia, with comments on the evolution of complex sociality in lizards. Herpetol Monogr 2003;17:145–80. [12] Chapple DG, Keogh S. Group structure and stability in social aggregations of white's skink, Egernia whitii. Ethology 2006;112:247–57. [13] Christian KA, Tracy CR. The effect of the thermal environment on the ability of hatchling Galapagos land iguanas to avoid predation during dispersal. Oecologia 1981;49:218–23. [14] Dewitt CB. Precision of thermoregulation and its relation to environmental factors in desert iguana Dipsosaurus dorsalis. Physiol Zool 1967;40:49–66. [15] Drummond H. Dominance in vertebrate broods and litters. Q Rev Biol 2006;81:3–32. [16] Espinoza RE, Quinteros S. A hot knot of toads: aggregation provides thermal benefits to metamorphic Andean toads. J Therm Biol 2008;33:67–75. [17] Gautier P, Olgun K, Uzum N, Miaud C. Gregarious behaviour in a salamander: attraction to conspecific chemical cues in burrow choice. Behav Ecol Sociobiol 2006;59:836–41. [18] Graves MG, Duvall D. Aggregation of squamate reptiles associated with gestation, oviposition, and parturition. Herpetol Monogr 1995;9:102–19. [19] Heath JE. Behavioural thermoregulation of body temperature in poikilotherms. Physiologist 1970;13:399–410. [20] Herczeg G, Gonda A, Saarikivi J, Merila J. Experimental support for the cost–benefit model of lizard thermoregulation. Behav Ecol Sociobiol 2006;60:405–14.

186

J.J. Khan et al. / Physiology & Behavior 100 (2010) 180–186

[21] Hori T, Shinohara K. Hypothalamic thermoresponsive neurones in the new born rat. J Physiol (Lond) 1979;294:541–60. [22] Huey RB. Physiological consequences of habitat selection. Am Nat 1991;137: S91–S115. [23] Huey RB. Temperature, physiology, and the ecology of reptiles. In: Gans C, Pough FH, editors. Biology of Reptilia. New York: Academic Press; 1982. p. 25–91. [24] Huey RB, Slatkin M. Cost and benefits of lizard thermoregulation. Q Rev Biol 1976;51:363–84. [25] Labra A. Thermoregulation in Pristidactylus lizards (Polycridae) — effects of group size. J Herpetol 1995;29:260–4. [26] Lancaster JR, Wilson P, Espinoza RE. Physiological benefits as precursors of sociality: why banded geckos band. Anim Behav 2006;72:199–207. [27] Mora JM. Comparative grouping behavior of juvenile ctenosaurs and iguanas. J Herpetol 1991;25:244–6. [28] Morafka DJ, Karen S, Valentine AL. Neonatology of Reptiles. Herpetol Monogr 2000;14:353–70. [29] Myres BC, Eells MM. Thermal aggregation in Boa constrictor. Herpetologica 1968;24:61–6. [30] Phillips JA, Allison CA, Pratt NC. Differential resource use, growth, and the ontogeny of social relationships in the green iguana. Physiol Behav 1993;53:81–8. [31] Rand AS. A nesting aggregation of iguanas. Copeia 1968;1968:552–61. [32] Reiserer RS, Schuett GW, Earley RL. Dynamic aggregations of newborn sibling rattlesnakes exhibit stable thermoregulatory properties. J Zool 2008;274:277–83. [33] Schutz L, Stuart-Fox D, Whiting MJ. Does the lizard Platysaurus broadleyi aggregate because of social factors? J Herpetol 2007;41:354–9.

[34] Seebacher F, Franklin CE. Physiological mechanisms of thermoregulation in reptiles: a review. J Comp Physiol B 2005;175:533–41. [35] Shah B, Shine R, Hudson S, Kearney M. Sociality in lizards: why do thick-tailed geckos (Nephrurus milii) aggregate? Behaviour 2003;140:1039–52. [36] Stamps JA. Conspecific attraction and aggregation in territorial species. Am Nat 1988;1:329–47. [37] Stamps JA. Growth-mortality tradeoffs and ‘personality traits’ in animals. Ecol Lett 2007;10:355–63. [38] Stamps JA, Tanaka S. The relationship between food and social behavior in juvenile lizards (Anolis aeneus). Copeia 1981;2:422–34. [39] Stapley J. Individual variation in preferred body temperature covaries with social behaviours and colour in male lizards. J Therm Biol 2006;31:362–9. [40] Tattersall GJ, Cadena V, Skinner MC. Respiratory cooling and thermoregulatory coupling in reptiles. Resp Physiol Neurobiol 2006;154:302–18. [41] Tattersall GJ, Milsom WK, Abe AS, Brito SP, Andrade DV. The thermogenesis of digestion in rattlesnakes. J Exp Biol 2004;207:579–85. [42] Ten Hwang Y, Lariviere S, Messier F. Energetic consequences and ecological significance of heterothermy and social thermoregulation in striped skunks (Mephitis mephitis). Physiol Biochem Zool 2007;80:138–45. [43] Visagie L, Mouton PLN, Bauwens D. Experimental analysis of grouping behaviour in cordylid lizards. Herpetological Journal 2005;15:91–6. [44] Werner DI, Baker EM, Gonzalez EDC, Sosa IR. Kinship recognition and grouping in hatchling green iguanas. Behav Ecol Sociobiol 1987;21:83–9.