THE TEMPORAL ORGANIZATION OF DAILY TORPOR AND HIBERNATION: CIRCADIAN AND CIRCANNUAL RHYTHMS

Mammals and birds have evolved the ability to maintain a high and constant body temperature T_b over a wide range of ambient temperatures T_a using endogenous heat production. In many, especially small endotherms, cost for thermoregulatory heat production can exceed available energy; to overcome these energetic bottlenecks, they enter a state of torpor (a regulated reduction of T_b and metabolic rate). Since the occurrence of torpor in many species is a seasonal event and occurs at certain times of the day, we review whether circadian and circannual rhythms, important in the timing of biological events in active animals, also play an important role during torpor when T_b is reduced substantially and may even fall below 0°C. The two distinct patterns of torpor, hibernation (prolonged torpor) and daily torpor, differ substantially in their interaction with the circadian system. Daily torpor appears to be integrated into the normal circadian rhythm of activity and rest, although torpor is not restricted only to the normal rest phase of an animal. In contrast, hibernation can last for several days or even weeks, although torpor never spans the entire hibernation season, but is interrupted by periodic arousals and brief normothermic periods. Clearly, a day is no longer divided in activity and rest, and at first glance the role of the circadian system appears negligible. However, in several hibernators, arousals not only follow a regular pattern consistent with a circadian rhythm, but also are entrainable by external stimuli such as photoperiod and T_a. The extent of the interaction between the circadian and circannual system and hibernation varies among species. Biological rhythms of hibernators for which food availability appears to be predictable seasonally and that hibernate in deep and sealed burrows show little sensitivity to external stimuli during hibernation and hence little entrainability of arousal events. In contrast, opportunistic hibernators, which some times use arousals for foraging and hibernate in open and accessible hibernacula, are susceptible to external zeitgebers. In opportunistic hibernators, the circadian system plays a major role in maintaining synchrony between the normal day-night cycle and occasional foraging. Although the daily routine of activity and rest is abandoned during hibernation, the circadian system appears to remain functional, and there is little evidence it is significantly affected by low T_b. (Chronobiology International, 17(2), 103–128, 2000)     

The temporal organization of daily torpor and hibernation: circadian and circannual rhythms

Mammals and birds have evolved the ability to maintain a high and constant body temperature T_b over a wide range of ambient temperatures T_a using endogenous heat production. In many, especially small endotherms, cost for thermoregulatory heat production can exceed available energy; to overcome these energetic bottlenecks, they enter a state of torpor (a regulated reduction of T_b and metabolic rate). Since the occurrence of torpor in many species is a seasonal event and occurs at certain times of the day, we review whether circadian and circannual rhythms, important in the timing of biological events in active animals, also play an important role during torpor when T_b is reduced substantially and may even fall below 0°C. The two distinct patterns of torpor, hibernation (prolonged torpor) and daily torpor, differ substantially in their interaction with the circadian system. Daily torpor appears to be integrated into the normal circadian rhythm of activity and rest, although torpor is not restricted only to the normal rest phase of an animal. In contrast, hibernation can last for several days or even weeks, although torpor never spans the entire hibernation season, but is interrupted by periodic arousals and brief normothermic periods. Clearly, a day is no longer divided in activity and rest, and at first glance the role of the circadian system appears negligible. However, in several hibernators, arousals not only follow a regular pattern consistent with a circadian rhythm, but also are entrainable by external stimuli such as photoperiod and T_a. The extent of the interaction between the circadian and circannual system and hibernation varies among species. Biological rhythms of hibernators for which food availability appears to be predictable seasonally and that hibernate in deep and sealed burrows show little sensitivity to external stimuli during hibernation and hence little entrainability of arousal events. In contrast, opportunistic hibernators, which some times use arousals for foraging and hibernate in open and accessible hibernacula, are susceptible to external zeitgebers. In opportunistic hibernators, the circadian system plays a major role in maintaining synchrony between the normal day-night cycle and occasional foraging. Although the daily routine of activity and rest is abandoned during hibernation, the circadian system appears to remain functional, and there is little evidence it is significantly affected by low T_b. (Chronobiology International, 17(2), 103-128, 2000)     

Sexual differences in over-winter torpor patterns of Richardson's ground squirrels in natural hibernacula

Summary Over-winter torpor patterns of Richardson's ground squirrels hibernating in southern Alberta were monitored with temperature-sensitive radiocollars to determine if these patterns differed between males and females in a manner related to the greater costs of mating effort by males than females. The hibernation season (from immergence to emergence) was composed of three periods: post-immergence euthermy, heterothermy, and pre-emergence euthermy. The hibernation season was shorter for juveniles than adults both among males (< 150 versus 234 days) and females (185 versus 231 days), a reflection of the later immergence into hibernation by juveniles. However, regardless of the absolute duration of hibernation, heterothermy accounted for a smaller proportion of the hibernation season of males (93±5%) than females (98±1%) and, within the heterothermal period, males had shorter torpor bouts and longer inter-torpor arousals. Overall, males spent a smaller proportion of the hibernation season in torpor (85±6%) than females (92±1%). This sexual difference was largely attributable to the longer duration of preemergence euthermy for males than females. Males terminated torpor in January and February, when hibernacula were at their coldest, then remained euthermic for 8.8 days (range 0.5–25.0 days) before emergence. In contrast, females terminated torpor in March, when hibernaculum temperatures were increasing, then remained euthermic for only 1.1 days (range 0.5–2.0 days) before emergence. Males lost less mass per euthermic day during hibernation than females (7.0 versus 9.3 g/day). Males and females hibernated at similar depths (56 cm), but males had larger chambers than females (18 versus 16 cm³/g). Many males, but no females, cached seeds in the hibernaculum. Males met the costs of thermogenesis and euthermy from a combination of fat reserves and food caches, whereas females relied solely on fat. Access to food caches permitted males to terminate torpor several weeks in advance of emergence, during which time they recouped mass and developed sperm in preparation for the forthcoming mating season.     

The role of energy availability in Mammalian hibernation: a cost-benefit approach

Hibernation is widely regarded as an adaptation to seasonal energy shortage, but the actual influence of energy availability on hibernation patterns is rarely considered. Here we review literature on the costs and benefits of torpor expression to examine the influence that energy may have on hibernation patterns. We first establish that the dichotomy between food- and fat-storing hibernators coincides with differences in diet rather than body size and show that small or large species pursuing either strategy have considerable potential scope in the amount of torpor needed to survive winter. Torpor expression provides substantial energy savings, which increase the chance of surviving a period of food shortage and emerging with residual energy for early spring reproduction. However, all hibernating mammals periodically arouse to normal body temperatures during hibernation. The function of these arousals has long been speculated to involve recovery from physiological costs accumulated during metabolic depression, and recent physiological studies indicate these costs may include oxidative stress, reduced immunocompetence, and perhaps neuronal tissue damage. Using an optimality approach, we suggest that trade-offs between the benefits of energy conservation and the physiological costs of metabolic depression can explain both why hibernators periodically arouse from torpor and why they should use available energy to minimize the depth and duration of their torpor bouts. On the basis of these trade-offs, we derive a series of testable predictions concerning the relationship between energy availability and torpor expression. We conclude by reviewing the empirical support for these predictions and suggesting new avenues for research on the role of energy availability in mammalian hibernation.     

Reversible depression of oxygen consumption in isolated liver mitochondria during hibernation

The biochemical mechanisms by which hibernators cool as they enter torpor are not fully understood. In order to examine whether rates of substrate oxidation vary as a function of hibernation, liver mitochondria were isolated from telemetered ground squirrels (Spermophilus lateralis) in five phases of their annual hibernation cycle: summer active, and torpid, interbout aroused, entrance, and arousing hibernators. Rates of state 3 and state 4 respiration were measured in vitro at 25 degrees C. Relative to mitochondria from summer-active animals, rates of state 3 respiration were significantly depressed in mitochondria from torpid animals yet fully restored during interbout arousals. These findings indicate that a depression of ADP-dependent respiration in liver mitochondria occurs during torpor and is reversed during the interbout arousals to euthermia. Because this inhibition was determined to be temporally independent of entrance and arousal, it is unlikely that active suppression of state 3 respiration causes entrance into torpor by facilitating metabolic depression. In contrast to the observed depression of state 3 respiration in torpid animals, state 4 respiration did not differ significantly among any of the five groups, suggesting that alterations in proton leak are not contributing appreciably to downregulation of respiration in hibernation.     

Daily torpor and hibernation in birds and mammals

Many birds and mammals drastically reduce their energy expenditure during times of cold exposure, food shortage, or drought, by temporarily abandoning euthermia, i.e. the maintenance of high body temperatures. Traditionally, two different types of heterothermy, i.e. hypometabolic states associated with low body temperature (torpor), have been distinguished: daily torpor, which lasts less than 24 h and is accompanied by continued foraging, versus hibernation, with torpor bouts lasting consecutive days to several weeks in animals that usually do not forage but rely on energy stores, either food caches or body energy reserves. This classification of torpor types has been challenged, suggesting that these phenotypes may merely represent extremes in a continuum of traits. Here, we investigate whether variables of torpor in 214 species (43 birds and 171 mammals) form a continuum or a bimodal distribution. We use Gaussian‐mixture cluster analysis as well as phylogenetically informed regressions to quantitatively assess the distinction between hibernation and daily torpor and to evaluate the impact of body mass and geographical distribution of species on torpor traits. Cluster analysis clearly confirmed the classical distinction between daily torpor and hibernation. Overall, heterothermic endotherms tend to be small; hibernators are significantly heavier than daily heterotherms and also are distributed at higher average latitudes (～35°) than daily heterotherms (～25°). Variables of torpor for an average 30 g heterotherm differed significantly between daily heterotherms and hibernators. Average maximum torpor bout duration was >30‐fold longer, and mean torpor bout duration >25‐fold longer in hibernators. Mean minimum body temperature differed by ～13°C, and the mean minimum torpor metabolic rate was ～35% of the basal metabolic rate (BMR) in daily heterotherms but only 6% of BMR in hibernators. Consequently, our analysis strongly supports the view that hibernators and daily heterotherms are functionally distinct groups that probably have been subject to disruptive selection. Arguably, the primary physiological difference between daily torpor and hibernation, which leads to a variety of derived further distinct characteristics, is the temporal control of entry into and arousal from torpor, which is governed by the circadian clock in daily heterotherms, but apparently not in hibernators.     

Myosin isoform expression and MAFbx mRNA levels in hibernating golden-mantled ground squirrels (Spermophilus lateralis)

Hibernating mammals present many unexplored opportunities for the study of muscle biology. The hindlimb muscles of a small rodent hibernator (Spermophilus lateralis) atrophy slightly during months of torpor, representing a reduction in the disuse atrophy commonly seen in other mammalian models. How torpor affects contractile protein expression is unclear; therefore, we examined the myosin heavy-chain (MHC) isoform profile of ground squirrel skeletal muscle before and after hibernation. Immunoblotting was performed first to identify the MHC isoforms expressed in this species. Relative percentages of MHC isoforms in individual muscles were then measured using SDS-PAGE (sodium dodecyl sulfate-polyacrylamide gel electrophoresis). The soleus and diaphragm did not display differences in isoforms following hibernation, but we found minor fast-to-slow isoform shifts in MHC protein in the gastrocnemius and plantaris. These subtle changes are contrary to those predicted by other models of inactivity but may reflect the requirement for shivering thermogenesis during arousals from torpor. We also measured mRNA expression of the Muscle Atrophy F-box (MAFbx), a ubiquitin ligase important in proteasome-mediated proteolysis. Expression was elevated in the hibernating gastrocnemius and the plantaris but was not associated with atrophy. Skeletal muscle from hibernators displays unusual plasticity, which may be a combined result of the intense activity during arousals and the reduction of metabolism during torpor.     

Major histocompatibility complex variation is similar in little brown bats before and after white‐nose syndrome outbreak

White‐nose syndrome (WNS), caused by the fungal pathogen Pseudogymnoascus destructans (Pd), has driven alarming declines in North American hibernating bats, such as little brown bat (Myotis lucifugus). During hibernation, infected little brown bats are able to initiate anti‐Pd immune responses, indicating pathogen‐mediated selection on the major histocompatibility complex (MHC) genes. However, such immune responses may not be protective as they interrupt torpor, elevate energy costs, and potentially lead to higher mortality rates. To assess whether WNS drives selection on MHC genes, we compared the MHC DRB gene in little brown bats pre‐ (Wisconsin) and post‐ (Michigan, New York, Vermont, and Pennsylvania) WNS (detection spanning 2014–2015). We genotyped 131 individuals and found 45 nucleotide alleles (27 amino acid alleles) indicating a maximum of 3 loci (1–5 alleles per individual). We observed high allelic admixture and a lack of genetic differentiation both among sampling sites and between pre‐ and post‐WNS populations, indicating no signal of selection on MHC genes. However, post‐WNS populations exhibited decreased allelic richness, reflecting effects from bottleneck and drift following rapid population declines. We propose that mechanisms other than adaptive immunity are more likely driving current persistence of little brown bats in affected regions.     

The impact of cold acclimation and hibernation on antioxidant defenses in the ground squirrel (Spermophilus citellus): An update

Any alteration in oxidative metabolism is coupled with a corresponding response by an antioxidant defense (AD) in appropriate subcellular compartments. Seasonal hibernators pass through circannual metabolic adaptations that allow them to either maintain euthermy (cold acclimation) or enter winter torpor with body temperature falling to low values. The present study aimed to investigate the corresponding pattern of AD enzyme protein expressions associated with these strategies in the main tissues involved in whole animal energy homeostasis: brown and white adipose tissues (BAT and WAT, respectively), liver, and skeletal muscle. European ground squirrels (Spermophilus citellus) were exposed to low temperature (4 ± 1 °C) and then divided into two groups: (1) animals fell into torpor (hibernating group) and (2) animals stayed active and euthermic for 1, 3, 7, 12, or 21 days (cold-exposed group). We examined the effects of cold acclimation and hibernation on the tissue-dependent protein expression of four enzymes which catalyze the two-step detoxification of superoxide to water: superoxide dismutase 1 and 2 (SOD 1 and 2), catalase (CAT), and glutathione peroxidase (GSH-Px). The results showed that hibernation induced an increase of AD enzyme protein expressions in BAT and skeletal muscle. However, AD enzyme contents in liver were largely unaffected during torpor. Under these conditions, different WAT depots responded by elevating the amounts of specific enzymes, as follows: SOD 1 in retroperitoneal WAT, GSH-Px in gonadal WAT, and CAT in subcutaneous WAT. Similar perturbations of AD enzymes contents were seen in all tissues during cold acclimation, often in a time-dependent manner. It can be concluded that BAT and muscle AD capacity undergo the most dramatic changes during both cold acclimation and hibernation, while liver is relatively unaffected by either condition. Additionally, this study provides a basis for further metabolic study that will illuminate the causes of these tissue-specific AD responses, particularly the novel finding of distinct responses by different WAT depots in hibernators.     

Temporal organisation of hibernation in wild-type and tau mutant Syrian hamsters

The temporal pattern of hibernation was studied in three genotypes of Syrian hamsters with different circadian periodicity to assess a potential circadian control of alternating torpor and euthermy. We recorded the pattern of hibernation by measuring activity in continuous dim light and constant environmental temperature (6 +/- 1 degrees C). In spite of differences in the endogenous circadian period of three genotypes (tau +/+: approximately equals 24 h, tau +/-: approximately equals 22 h, and tau -/-: approximately equals 20 h) torpor bout duration was statistically indistinguishable (tau +/+: 86.9+/-5.3 h; tau +/-: 94.2+/-3.3 h; tau -/-: 88.8+/-6.2 h). The time between two consecutive arousals from torpor showed unimodal distributions not significantly different between genotypes. The first entry into torpor occurred within the active phase of the circadian cycle in all genotypes whereas the first arousal from torpor appeared to be timed randomly with respect to the prior circadian cycle. The amplitude of the activity rhythm was lower after hibernation compared with the amplitude before hibernation. The results suggest that in the Syrian hamster the circadian system does not control periodicity of torpor and arousal onsets in prolonged hibernation at 6 degrees C.     

Hibernation induces glutathione redox imbalance in ground squirrel intestine

Glutathione (GSH) is the major thiol-disulfide redox buffer in cells and is a critical component of antioxidant defense. Here we examined GSH redox balance in the intestinal mucosa during the annual cycle of 13-lined ground squirrels (Spermophilus tridecemlineatus). The ratio of reduced GSH to its oxidized form (glutathione disulfide, GSSG), which is an index of oxidative stress, was five-fold lower in hibernating compared with summer-active squirrels, an effect due primarily to elevated GSSG concentration in hibernators. During hibernation the total pool of GSH equivalents was lowest in squirrels undergoing arousal and highest in squirrels during interbout arousals. Hibernation decreased intestinal GSSG reductase activity by approximately 50%, but had no effect on activities of glutathione peroxidase or glucose-6-phosphate dehydrogenase. Within the hibernation season, expression of the stress protein HSP70 in intestinal mucosa was highest in squirrels entering torpor and early in a torpor bout, and lowest in squirrels arousing from torpor and during interbout euthermia. The results suggest that hibernation in ground squirrels is associated with a shift in intestinal GSH redox balance to a more oxidized state. Higher levels of HSP70 during the early phases of torpor may reflect induction of the stress response due to aberrations in protein folding or may be a mechanism to increase enterocyte tolerance to subsequent stress imposed by extended torpor or the arousal process.     

Ecology of natural hibernation in the marsupial mountain pygmy-possum (Burramys parvus)

The hibernating marsupial mountain pygmy-possum (Burramys parvus, 40 g) has to raise its slow-growing offspring during a short alpine summer. Only females provide parental care, while after mating males emigrate to marginal habitats often at lower altitudes which can sustain only low possum densities. We predicted that the hibernation strategies in mountain pygmy-possums are distinct from those of similar-sized placental hibernators, because of the developmental constraints in marsupials and because hibernation differs between the sexes. Using temperature-sensitive radio transmitters, we studied the hibernation patterns of free-living male and female mountain pygmy-possums living in a north- and a south-facing boulder field (Kosciusko National Park) for two consecutive winters. Individual possums commenced hibernation several months before the snow season. As in other hibernators, torpor in the mountain pygmy-possum was interrupted by periodic arousals which occurred most often during the late afternoon. Torpor bouts initially lasted a few days when the hibernacula temperature (T _hib) ranged from 4 to 7°C. As the hibernation season progressed, torpor bouts became longer and possum body temperatures (T _b) approached 2°C. The T _bs of females were significantly lower and torpor bouts were longer in the second half of the hibernation season than in males. Between torpor bouts, both sexes were often active and left hibernacula for periods of up to 5 days. Especially during the first months of the hibernation season, possums also frequently changed hibernacula sites probably in an attempt to select a site with a more suitable microclimate. Emergence from hibernation was closely coupled with the disappearance of snow from the possum habitat (September 1995, October 1996) and the limited fat stores probably dictate an opportunistic spring emergence. However, in 1995, spring was early and males emerged significantly earlier than females. In 1996, when snow melt was delayed, this difference vanished. Testes are regressed in males during hibernation and the time needed for testes growth and spermatogenesis favours an earlier emergence for males which was probably achieved by their preference for the more sun exposed north-facing boulder field. A sexual dimorphism in hibernation strategies and spring emergence therefore enables mountain pygmy-possums to cope with their harsh alpine environment. Received: 22 May 1997 / Accepted: 21 August 1997     

Hibernation in the tropics: lessons from a primate

The Malagasy primate Cheirogaleus medius hibernates in tree holes for 7 months, although ambient temperatures during hibernation rise above 30°C in their natural environment. In a field study we show that during hibernation the body temperature of most lemurs fluctuates between about 10°C and 30°C, closely tracking the diurnal fluctuations of ambient temperature passively. These lemurs do not interrupt hibernation by spontaneous arousals, previously thought to be obligatory for all mammalian hibernators. However, some lemurs hibernate in large trees, which provide better thermal insulation. Their body temperature fluctuates only little around 25°C, but they show regular arousals, as known from temperate and arctic hibernators. The results from this study demonstrate that maximum body temperature is a key factor necessitating the occurrence of arousals. Furthermore, we show that hibernation is not necessarily coupled to low body temperature and, therefore, low body temperature should no longer be included in the definition of hibernation.     

Cholesterol and lipoprotein dynamics in a hibernating mammal

Hibernating mammals cease feeding during the winter and rely primarily on stored lipids to fuel alternating periods of torpor and arousal. How hibernators manage large fluxes of lipids and sterols over the annual hibernation cycle is poorly understood. The aim of this study was to investigate lipid and cholesterol transport and storage in ground squirrels studied in spring, summer, and several hibernation states. Cholesterol levels in total plasma, HDL and LDL particles were elevated in hibernators compared with spring or summer squirrels. Hibernation increased plasma apolipoprotein A-I expression and HDL particle size. Expression of cholesterol 7 alpha-hydroxylase was 13-fold lower in hibernators than in active season squirrels. Plasma triglycerides were reduced by fasting in spring but not summer squirrels. In hibernators plasma β-hydroxybutyrate was elevated during torpor whereas triglycerides were low relative to normothermic states. We conclude that the switch to a lipid-based metabolism during winter, coupled with reduced capacity to excrete cholesterol creates a closed system in which efficient use of lipoproteins is essential for survival.     

The parasympathetic nervous system: its role during torpor in the fat-tailed dunnart (Sminthopsis crassicaudata)

This study investigated the effect of parasympathetic inhibition on the cardio-ventilatory interaction during torpor in the fat-tailed dunnart (Sminthopsis crassicaudata). Studies on the influence of the autonomic nervous system on cardiac function during torpor have focused on deep hibernation in eutherians. S. crassicaudata was used as a representative of the Metatheria that exhibits shallow, daily torpor as a comparison for the patterns of cardiac function found in other mammalian heterotherms. During torpor, parasympathetic inhibition removed the cardio-ventilatory interaction, eliminated heart rate variability and increased the overall heart rate; these are responses that have been shown to be typical of eutherian hibernators under the same conditions. Similarly, there was evidence to suggest that as the bout of torpor progressed, the variation in instantaneous heart rate decreased as a result of the progressive removal of parasympathetic tone. It has been suggested that the ability to enter a "steady state" during torpor, which is characterised by a regular heart rate, is limited to deep hibernators. On the basis of this, and the results of previous physiological studies, it was proposed that there is little evidence to suggest that there is any physiological difference between shallow, daily torpor and deep hibernation.     

Incapacitating effects of fungal coinfection in a novel pathogen system

As globalization lowers geographic barriers to movement, coinfection with novel and enzootic pathogens is increasingly likely. Novel and enzootic pathogens can interact synergistically or antagonistically, leading to increased or decreased disease severity. Here we examine host immune responses to coinfection with two closely related fungal pathogens: Batrachochytrium dendrobatidis (Bd) and Batrachochytrium salamandrivorans (Bsal). Both pathogens have had detrimental effects on amphibian populations, with Bd now largely enzootic, while Bsal is currently spreading and causing epizootics. Recent experimental work revealed that newts coinfected with Bd and Bsal had significantly higher mortality than those infected with either pathogen alone. Here we characterize host immunogenomic responses to chytrid coinfection relative to single infection. Across several classes of immune genes including pattern recognition receptors, cytokines, and MHC, coinfected host gene expression was weakly upregulated or comparable to that seen in single Bd infection, but significantly decreased when compared to Bsal infection. Combined with strong complement pathway downregulation and keratin upregulation, these results indicate that coinfection with Bd and Bsal compromises immune responses active against Bsal alone. As Bsal continues to invade naïve habitats where Bd is enzootic, coinfection will be increasingly common. If other Bd‐susceptible species in the region have similar responses, interactions between the two pathogens could cause severe population and community‐level declines.     

The role of energy availability in Mammalian hibernation: an experimental test in free-ranging eastern chipmunks

Reduced torpor expression by hibernating mammals is often attributed to physiological constraints that limit their hibernation ability but may instead reflect adaptive, plastic responses to surplus energy availability. We evaluated this hypothesis by supplementing the food hoards of free-ranging eastern chipmunks (Tamias striatus) before hibernation and then documenting their use of torpor during the subsequent winter. In both years of study, chipmunks that received additional food were euthermic more than twice as frequently as nonsupplemented individuals. Furthermore, when food-supplemented individuals did express torpor, their minimum collar temperature was 5 degrees -10 degrees C warmer than nonsupplemented animals. These results indicate that reduced torpor expression by hibernators can result from an absence of energetic necessity rather than a lack of physiological capability and suggest that even endotherms sequestered in a hibernaculum may benefit from maintaining an elevated body temperature whenever possible.     

Hibernation by tree-roosting bats

In summer, long-eared bats (Nyctophilus spp.) roost under bark and in tree cavities, where they appear to benefit from diurnal heating of roosts. In contrast, hibernation is thought to require a cool stable temperature, suggesting they should prefer thermally insulated tree cavities during winter. To test this prediction, we quantified the winter thermoregulatory physiology and ecology of hibernating tree-roosting bats, Nyctophilus geoffroyi and N. gouldi in the field. Surprisingly, bats in winter continued to roost under exfoliating bark (65%) on the northern, sunny side of trees and in shallow tree cavities (35%). Despite passive re-warming of torpid bats by 10-20 degrees C per day, torpor bouts lasted up to 15 days, although shorter bouts were also common. Arousals occurred more frequently and subsequent activity lasted longer on warmer nights, suggesting occasional winter foraging. We show that, because periodic arousals coincide with maximum roost temperatures, when costs of rewarming and normothermic thermoregulation are minimal, exposure to a daily temperature cycle could largely reduce energy expenditure during hibernation. Our study provides further evidence that models of torpor patterns and energy expenditure from hibernators in cold temperate climates are not directly applicable in milder climates, where prolonged torpor can be interspersed with more frequent arousals and occasional foraging.