From the Nightly Rest to the Years-Long Wait
The first essay in this series established sleep as a state of adaptive inactivity — withdrawal from metabolic expenditure when wakefulness offers no survival advantage (Siegel, 2009). If that argument holds, a prediction follows: organisms facing more extreme or prolonged environmental conditions than those that trigger ordinary sleep should exhibit more extreme or prolonged versions of the same withdrawal. This essay tests that prediction across the biological spectrum, from the daily torpor of small mammals to the seasonal dormancy of hibernating bears, from the metabolic suspension of the aestivating African lungfish to the architectural ingenuity of the unihemispherically sleeping dolphin. Along the way, it encounters the only hibernating primate — a discovery that challenges long-held assumptions about the relationship between cognitive complexity and metabolic flexibility. The evidence confirms the prediction — and does something more. It reveals that sleep, torpor, hibernation, and aestivation are not categorically separate states with distinct biological logics. They are points on a single continuum of adaptive metabolic withdrawal, calibrated by species, season, and environmental condition. The implications for how human sleep variation is understood — and pathologised — are substantial.
If sleep is, at its foundation, an energy conservation strategy — the organism withdrawing from metabolic expenditure when activity offers no survival advantage — then a clear biological prediction follows. The strategy should be scalable. Organisms facing environments more severe, more prolonged, or more energetically punishing than those that trigger ordinary overnight sleep should exhibit more severe, more prolonged, or more energetically suppressive versions of the same withdrawal. The continuum should exist because the principle that generates it is continuous.
This is not a novel prediction. It is the logical extension of Jerome Siegel's adaptive inactivity hypothesis (Siegel, 2009), and it is one that the natural world has been testing, and confirming, across millions of years of evolutionary pressure. What is surprising is not that the continuum exists, but how far it extends — and how much it illuminates about the nature of sleep when examined from its far end, looking back toward the familiar.
The journey across that spectrum begins close to home: with the small mammals that enter daily torpor, and works outward — through the seasonal dormancy of bears, the suspended animation of lungfish, the metabolic extremity of aestivating amphibians — toward the edges of what biology can sustain. At each point, the same underlying logic is operating. The form it takes changes dramatically. The principle does not.
The hummingbird is among the smallest endotherms on Earth. Maintaining a body temperature of approximately 40°C while weighing less than five grams requires a metabolic rate so high that a hummingbird feeding continuously through the day may consume up to half its body weight in nectar. At night, when feeding is impossible, the arithmetic of staying warm becomes unsustainable. The solution is torpor: a controlled reduction in body temperature and metabolic rate during the inactive period, allowing the bird to survive the night on a fraction of the energy that full thermoregulation would require. In the morning, the bird rewarms itself — a process that takes minutes — and resumes full activity (Geiser, 2004).
This is daily torpor: a shallow, readily reversible suppression of metabolic activity tied to the animal's natural inactive period. It differs from sleep in degree rather than in kind. The body temperature drops further, the metabolic suppression is deeper, and the arousal threshold is higher than in ordinary sleep — but the underlying logic is identical. The organism is withdrawing from the cost of wakefulness during a period when wakefulness offers no advantage, and doing so with whatever depth of suppression its energy budget demands.
Daily torpor is widespread across small mammals and birds, including bats, small rodents, dunnarts, and multiple avian species (Geiser, 2004). The capacity to enter it is not a quirk of a few specialised lineages — it is a broadly distributed biological tool, available to organisms whose size makes the metabolic arithmetic of full thermoregulation during inactivity prohibitively expensive. Larger animals, with their more favourable surface-area-to-volume ratios, can afford to maintain body temperature through the night without suppression. Smaller animals frequently cannot.
The significance of this for the continuum argument is straightforward: daily torpor is sleep's immediate neighbour on the spectrum. The behavioural and physiological boundary between deep sleep and shallow torpor is not a cliff edge. It is a gradient, and many small animals move along it fluidly in response to ambient temperature, food availability, and season.
The assumption that primates are cognitively and metabolically too complex for true hibernation held largely unchallenged for most of the twentieth century. Primates, the reasoning went, depend on sophisticated neural function and continuous social coordination that cannot be suspended for months at a time. Then research on the fat-tailed dwarf lemur (Cheirogaleus medius) of Madagascar dismantled that assumption entirely.
The fat-tailed dwarf lemur is the only primate in the world known to hibernate for an extended period of time. Hibernation can last up to seven months and is defined by periods of torpor — severely decreased metabolism, heart rate, and body temperature — interspersed with metabolically active periods of rewarming called interbout arousals. This is not a marginal or atypical hibernation. It is full seasonal dormancy: the animal spends the Malagasy dry season from approximately April or May to October in a tree hole, dormant, sustaining itself entirely on fat stored in its distinctively swollen tail. By the beginning of torpor, the tail can account for up to 40% of the animal's body weight.
The physiological figures are striking. During torpor, a fat-tailed dwarf lemur's heart rate drops from about 200–300 beats per minute to as low as six beats per minute. Its body temperature drops too, and becomes driven by the ambient temperature of the environment. Unlike the polar bear, which maintains near-normal body temperature during denning, the dwarf lemur is a genuine deep hibernator: body temperature tracks the surroundings, metabolic suppression is profound, and arousal from torpor requires active rewarming.
What makes the fat-tailed dwarf lemur especially significant for the argument being developed here is what happens during the interbout arousals. During an interbout arousal, a dwarf lemur's body temperature rises, its heart rate increases, its breathing becomes more regular, and its brain shows massive bouts of REM sleep. Afterward, the animal drops back into torpor. The animal is not merely waking to perform maintenance tasks and returning to dormancy. It is sleeping — properly, with full REM architecture — during the arousal period. This observation has profound implications for the relationship between hibernation and sleep, discussed further in section 2.3 below.
The dwarf lemur's existence as a hibernating primate challenges a comfortable assumption: that the capacity for deep metabolic withdrawal is an evolutionary primitive, available only to cognitively simpler organisms. A primate — an order that includes the most neurologically complex animals on Earth — can enter and emerge from seven months of metabolic suspension. The adaptive withdrawal principle is not constrained by neural sophistication. It is available wherever the environmental logic demands it.
As the days shorten and food availability declines in temperate and arctic environments, some species extend their torpor from a daily event into a seasonal one. This is the transition from daily torpor to hibernation proper — and it is a transition in duration and depth rather than a change in fundamental mechanism.
The metabolic suppression of true hibernation is substantially more profound than daily torpor. In small hibernators such as ground squirrels and dormice, body temperature may fall to within a degree or two of ambient temperature, heart rate drops to a few beats per minute, breathing becomes intermittent, and metabolic rate may be suppressed to as little as 1–5% of normal basal metabolic rate (Storey, 2001). These animals are not simply sleeping deeply. They are operating on biological minimum — a metabolic rate so low that weeks or months of survival become possible on stored fat reserves alone.
Critically, hibernation is not simply one long torpor bout. It is a series of torpor episodes, each lasting days to weeks, interrupted by brief arousals to euthermic (normal body temperature) states. These arousals were long treated as something of a paradox: they are metabolically expensive, consuming a significant proportion of the winter's total energy budget, yet the animal returns promptly to torpor without feeding or drinking. Why would a hibernating animal periodically interrupt its energy-saving state to warm itself up, only to cool back down again?
Electroencephalographic research has provided a compelling answer. Studies of hibernating animals — including, significantly, the fat-tailed dwarf lemur — have demonstrated that sleep as conventionally defined, with its characteristic NREM and REM architecture, cannot occur during deep torpor. Body temperature during torpor is too low to sustain the neural processes that generate normal sleep (Krystal et al., 2013). The consequence is that the hibernating animal accumulates what functions as a sleep debt during each torpor bout — the neural processes associated with sleep are suspended, not accomplished. The interbout arousal is, in significant part, the animal surfacing to sleep: to discharge the accumulated neural debt through proper NREM and REM sleep cycles before returning to metabolic suppression (Frontiers in Neuroanatomy, 2019). Interbout arousals in the fat-tailed dwarf lemur tend to occur every 6–12 days during hibernation, and the massive REM bouts observed during those arousals support the sleep discharge hypothesis directly.
This is a discovery of considerable conceptual importance. It reveals that hibernation and sleep are not merely analogous — they are functionally interdependent. The hibernating animal is not exempt from the need to sleep; it has found a way to defer that sleep in blocks, discharging it periodically during brief euthermic windows. The two states are coupled. Hibernation is not sleep's alternative; it is a strategy that incorporates sleep within a broader cycle of adaptive withdrawal.
The scaling of metabolic suppression across the torpor-hibernation continuum is striking in its own right. Storey (2001) documents metabolic rate reductions ranging from approximately five-fold in shallow daily torpor to one-hundred-fold or more in the deepest hibernation states. This is the same biological lever being pulled by degrees, not a different mechanism being engaged.
The polar bear is perhaps the most commonly invoked example of animal hibernation in popular accounts — and one of the most misrepresented. The reality is considerably more interesting than the popular image, and considerably more useful for the argument being developed here.
Most polar bears do not hibernate at all. Male polar bears and non-pregnant females remain active throughout the Arctic winter, hunting seals on sea ice and managing the energetic demands of one of the harshest environments on Earth through fat reserves and metabolic efficiency rather than dormancy (Polar Bears International, 2024). The popular image of the hibernating polar bear is, for most of the population, simply wrong.
The exception — and it is a biologically extraordinary one — is the pregnant female. Between October and November, pregnant polar bears excavate maternity dens, typically in snowdrifts on land or sea ice, and enter a state that is best described not as hibernation but as maternity denning: a physiologically distinct form of adaptive withdrawal calibrated not to resource scarcity alone but to the simultaneous demands of reproduction (USGS, 2024).
During maternity denning, the polar bear female's body temperature falls modestly — from a normal 37–38°C to approximately 35°C (Whiteman et al., 2015). Heart rate, by contrast, can fall to as low as 8–10 beats per minute during sleep (Folk et al., 1980). Metabolic rate under denning conditions is reduced by approximately 25–50% from expected values (Robbins et al., 2012). She does not eat, drink, urinate, or defecate throughout the denning period — which may last up to eight months in total when the pre-denning fast is included — with metabolic wastes recycled biochemically rather than excreted (Polar Bear Range States, 2024).
This is already remarkable. But the most biologically striking feature of polar bear maternity denning is not what the female stops doing. It is what she continues doing while in this state of reduced metabolic activity.
Within the maternity den, the female gives birth — typically to one or two cubs in December or January — and immediately begins nursing them. The cubs are born blind, toothless, and weighing barely half a kilogram. They are among the most altricial neonates of any large mammal: entirely dependent, incapable of thermoregulation, requiring continuous warmth and frequent nursing to survive. The mother's milk is exceptionally rich, containing approximately 31% fat (Polar Bears International, 2024), and she nurses actively through the denning period, sustaining the rapid growth of cubs from 500 grams at birth to approximately 10 kilograms by emergence in March or April.
The den itself is a feat of passive engineering: a small snow chamber whose structure traps the mother's body heat, maintaining an internal temperature up to 25°C warmer than the Arctic exterior (Polar Bears International, 2024). The mother regulates this microclimate through her own metabolic heat output — which must be calibrated precisely, since too little heat risks the cubs, and too much wastes irreplaceable energy reserves.
The polar bear maternity den is not a shutdown. It is a reconfiguration. The organism has withdrawn from one class of energetically expensive activity — foraging, locomotion, thermal regulation against an extreme external environment — while simultaneously maintaining and indeed intensifying another: the reproductive and nurturing demands of parturition, lactation, and cub thermoregulation. The bear is doing some of the most physiologically demanding work of her life while fasting, while motionless, and while operating at a significantly reduced overall metabolic rate.
This is the adaptive inactivity principle in its most complex expression. The withdrawal is selective. What is suspended is the activity that offers no survival advantage in current conditions. What continues — indeed, what is prioritised — is the activity that represents the organism's most critical reproductive investment. The bear is not dormant in any simple sense. She is sleeping strategically, conserving what can be conserved in order to devote the freed resources to what cannot wait.
This challenges the active/inactive binary that underlies most popular accounts of sleep and dormancy. The polar bear maternity den demonstrates that an organism can be simultaneously in a state of metabolic withdrawal and in a state of intensive biological productivity. The two are not opposites. The withdrawal enables the productivity. And that observation reaches back, with considerable force, toward the human parent at three in the morning — not dormant, not fully resting, but engaged in a form of biological work — caregiving, feeding, responding — that the body treats as the highest-priority task available, suspending full alertness to a degree compatible with that task, and no further.
Hibernation is a response to cold and scarcity — the organism withdrawing when low temperatures and absent food make sustained activity impossible or prohibitively expensive. But the adaptive withdrawal principle is not directionally constrained. An equivalent logic operates in response to heat and drought. The state it produces is called aestivation: seasonal dormancy in response to high temperatures or water shortage, which is physiologically distinct from hibernation but governed by the same underlying principle (Storey, 2016; Navas et al., 2023).
Aestivation is widespread across invertebrates, fish, amphibians, and some reptiles. Its metabolic logic mirrors hibernation's: the organism suppresses activity, reduces metabolic rate, and enters a state of suspended or near-suspended animation until conditions become viable for normal life again. The environmental trigger is different — heat and desiccation rather than cold and scarcity — but the adaptive response is structurally identical. Here again, the biology is not selecting a different survival strategy. It is running the same strategy against a different environmental signal.
No single organism makes the continuum argument more forcefully than the African lungfish (Protopterus sp.). The lungfish's aestivation represents the biological extreme of adaptive metabolic withdrawal — and its specific physiological details are remarkable enough to require careful description.
During the dry season, as rivers and wetlands of sub-Saharan Africa desiccate, the lungfish burrows into the mud of the drying riverbed. It secretes a mucus cocoon around its body — leaving only a small opening at the mouth for gas exchange — and enters aestivation. In this state, the animal survives without food or water for the duration of the dry season, and in extreme cases for up to four years (Oxford Scientist, 2023). Heart rate falls from approximately 25 beats per minute when active in water to as low as two beats per minute during deep aestivation (Oxford Scientist, 2023). Oxygen intake falls to approximately half that of an active fish in water.
Research on the lungfish has identified three distinct phases of aestivation, each with its own transcriptional and physiological profile: induction, maintenance, and arousal. The brain of the African lungfish is able to coordinate a whole-body response to induce aestivation and to arouse from aestivation. This is not passive shutdown. The organism is actively managing a complex sequence of physiological transitions, each phase characterised by different patterns of gene expression and metabolic adjustment. The induction phase is characterised by reduction in glycolytic capacity and metabolic activity, suppression of protein synthesis and degradation, and an increase in defence against ammonia toxicity. The maintenance phase involves an increase in oxidative defence capacity, and upregulation of transcription, translation, and glycolytic capacities in preparation for arousal.
The selectivity of what continues during aestivation is as striking as the depth of what is suppressed. The lungfish immune system remains active throughout, protecting the skin from pathogens — the animal's most direct interface with the surrounding mud (Navas et al., 2023). The mucus cocoon itself functions not merely as a physical barrier but as a living tissue: research has shown that the cocoon actively traps bacteria, providing an outer layer of immunological defence during a period when the animal could not mount an active response to infection (Heimroth, 2018). Additionally, increased expression of myostatin in aestivating lungfish is effective at preventing muscle atrophy — a precisely targeted upregulation of a muscle-preserving gene during a period of total immobility. The animal is not simply shut down. It has made a series of precise molecular allocations: the minimum necessary processes continue; everything else is suspended.
One further detail is of particular significance for the argument running across this essay series. The mRNA expression levels of growth hormone and prolactin were upregulated in the brain of Protopterus annectens during the induction phase of aestivation. This is directly relevant to the growth hormone discussion in Sleep as Biology, which examines the claim that sleep is necessary for growth hormone release and finds the causal evidence weaker than commonly presented. If growth hormone expression is upregulated in a lungfish brain during the induction of profound metabolic suppression — during a state that is, if anything, the antithesis of what popular science imagines as optimal growth-promoting sleep — the hormone's relationship with reduced-activity states appears to be far more ancient, far more biologically flexible, and far less specifically tied to any particular sleep architecture than the simple association with slow-wave sleep in humans implies. The hormone appears to be part of a broader adaptive metabolic signalling system, one that evolution was running in fish long before the first mammal evolved the specific sleep stage with which it became associated in humans.
The African lungfish is the most extreme example, but it is not isolated. African clawed frogs (Xenopus laevis) have a similar strategy, digging into and remaining immobile in the mud during the seasonal drying of their lakebed habitat. Sea cucumbers (Apostichopus japonicus) aestivate for up to 100 days, a period marked by cessation of feeding, degenerative atrophy of the digestive tract, metabolic inhibition, and weight loss. Nevertheless, normal structure and function can be restored upon release from aestivation.
In each case, the pattern is identical: withdraw when conditions make activity non-viable; conserve; resume fully when conditions permit. The metabolic suppression across the hibernation and aestivation spectrum is not modest. Storey (2001) documents that metabolic rate reductions of five- to one-hundred-fold or more support months or even years of dormancy across different species. This is the same biological principle — adaptive withdrawal — operating across an extraordinary range of intensity, duration, and environmental context.
The bottlenose dolphin faces a constraint that most terrestrial animals do not: it is a mammal that must breathe air while living in water. Unconsciousness in that environment is not a matter of vulnerability to predators alone — it is a question of survival from one breath to the next. The biological solution is one of the most architecturally ingenious adaptations in the animal kingdom: unihemispheric slow-wave sleep, in which one cerebral hemisphere sleeps while the other remains active, enabling the animal to surface and breathe, maintain motor coordination, and continue monitoring its environment (Lyamin et al., 2008; Rattenborg, Amlaner, and Lima, 2000).
The hemispheres alternate. After a period, the sleeping hemisphere wakes and the waking hemisphere sleeps. The dolphin is never fully unconscious. It achieves what the body requires from sleep — neural restoration, the cycling of slow-wave states — through a radically different architecture than the consolidated bilateral sleep of most terrestrial mammals. The constraint has shaped the solution.
Unihemispheric sleep is not unique to dolphins. It has been documented across a range of aquatic and semi-aquatic mammals including fur seals, manatees, and beluga whales — each facing a version of the same constraint — and is widespread in birds, where it serves a different primary function: predator detection (Rattenborg, Amlaner, and Lima, 2000). The solution is similar; the problem being solved is different. The same architectural flexibility has been recruited by evolution for more than one purpose.
In mallard ducks sleeping in groups, individuals positioned at the exposed edge of the group — where predation risk is highest — show dramatically more unihemispheric sleep than those in the protected centre, and they direct the open eye away from the group toward the direction from which a threat is most likely to approach (Rattenborg, Lima, and Amlaner, 1999). The system is not merely capable of unihemispheric sleep; it is capable of adjusting the proportion and orientation of that sleep in real time, in response to perceived risk. The architecture of sleep is dynamically calibrated to the environment, not fixed by biology alone.
The frigatebird takes this further still. Electrophysiological recordings of great frigatebirds (Fregata minor) during non-stop oceanic flights lasting up to ten days demonstrate that these birds sleep in flight — with either one hemisphere at a time or both hemispheres simultaneously — for an average of 0.69 hours per day (Rattenborg et al., 2016). Back on land, the same birds sleep for 12.8 hours per day. The ecological demands of the flight determine the allocation. The biology provides the tool; the environment determines how it is deployed.
Unihemispheric sleep is the continuum argument's most architecturally striking example, because it demonstrates that what varies across the spectrum is not only the depth and duration of adaptive withdrawal but its very structure. The brain does not simply suppress more or less. It can partition itself, running different states simultaneously in different regions, allocating rest and wakefulness in spatial rather than purely temporal terms.
This is a long way from the binary of asleep/awake that underlies most popular and clinical accounts of sleep. It is a biological system of remarkable flexibility and precision — not a switch but a continuum of states, managed dynamically, in response to a continuously shifting balance of demands. The brain that can direct one hemisphere into slow-wave sleep while the other steers a course through a rising air current over a moonlit ocean is not a brain that treats sleep as a fixed requirement. It treats sleep as a resource to be allocated with precision, in the form and at the time that circumstances permit.
The evidence examined in this essay supports a single conclusion: sleep, daily torpor, seasonal torpor, hibernation, and aestivation are not categorically distinct biological states. They are points on a continuum of adaptive metabolic withdrawal, unified by a common principle and differentiated by degree, duration, and the specific constraints of the organism's ecology.
The common principle is the one established in Sleep as Biology: the organism withdraws from the metabolic cost of wakefulness when the environment offers no advantage sufficient to justify that cost. What changes along the continuum is the severity of the environmental signal and the depth of the withdrawal it triggers.
The hummingbird in daily torpor and the lungfish in four-year aestivation are running the same adaptive programme. The difference between them is the duration and severity of the environmental condition that makes wakefulness non-viable. The hummingbird's night lasts hours. The lungfish's dry season can last years. The biology scales accordingly. The dwarf lemur, sleeping for seven months in a Malagasy tree hole and surfacing every few days to dream, sits between them — a primate running the same ancient programme, disproving the assumption that evolutionary sophistication exempts an organism from its reach.
The terminology in this area has historically been inconsistent, and precise usage matters for accurately representing the research. Drawing on Storey (2001; 2016) and Navas et al. (2023), the following distinctions hold:
A shallow, readily reversible suppression of metabolic rate during the inactive period, typically lasting hours. Common in small endotherms. Body temperature drops modestly; arousal is quick. The boundary with deep sleep is a gradient rather than a threshold.
Seasonal; a response to cold temperatures and food scarcity. Multiple torpor bouts separated by brief arousals to euthermy, during which sleep — including REM sleep — occurs. Metabolic rate may be suppressed to 1–5% of basal rate in deep hibernators. The arousals serve, in significant part, to discharge accumulated sleep debt.
Seasonal; a response to high temperatures and water shortage. Physiologically distinct from hibernation in its triggers and some of its mechanisms, but governed by the same adaptive logic of metabolic suppression during conditions that make activity non-viable. Proceeds through distinct induction, maintenance, and arousal phases, each with its own molecular signature.
A selective reconfiguration of metabolic activity rather than uniform suppression; withdrawal from energetically costly locomotion and foraging while maintaining the reproductive and nurturing demands of parturition and lactation. Metabolic suppression is moderate (25–50%) rather than profound. Body temperature remains near-normal.
A structural rather than metabolic adaptation; the spatial partitioning of sleep across cerebral hemispheres to enable partial wakefulness during the rest period. Primarily documented in aquatic mammals and birds; serves respiratory and predator-detection functions respectively.
These are distinct expressions of a shared principle, not fundamentally different biological phenomena. Each represents evolution's solution to a specific version of the same problem: how to withdraw from the cost of wakefulness when conditions demand it, in a form compatible with the organism's particular survival constraints.
The continuum has a direct implication for how human sleep variation is understood. If sleep is one point on a spectrum of adaptive withdrawal — and if the defining feature of all points on that spectrum is their calibration to the organism's circumstances rather than adherence to a fixed template — then the expectation that human sleep should conform to a specific pattern, duration, and architecture is biologically unwarranted.
The human infant waking three times in the night is not violating a sleep norm. It is expressing a sleep pattern calibrated to infant metabolic needs, feeding requirements, and developmental demands — a pattern that sits at the polyphasic end of the human sleep range, appropriate to its place in the life course. The adult who sleeps six hours rather than eight, who wakes briefly in the night, who finds their alertness distributed unevenly across the day, is not malfunctioning. They are a member of a species whose sleep system was never designed to conform to a fixed industrial template.
A dwarf lemur in a tree hole in Madagascar emerges from months of torpor, sleeps through a storm of REM dreams, and drops back into silence. A polar bear nurses newborns in a snowdrift while barely eating for months. A frigatebird sleeps for forty minutes a day above an ocean that has no landing. None of these is a failure of the sleep system. Each is the sleep system doing precisely what it evolved to do: allocating adaptive withdrawal to the precise form and duration that this organism, in this environment, at this moment in its life, requires.
The cultural construction of anxiety around human deviation from a historically recent sleep norm is examined in the third essay in this series (see: Sleep as Culture). The biology, viewed from across the full spectrum, does not support that anxiety. It supports something more generous, and more accurate.
The spectrum described in this essay extends, in evolutionary time, further than any individual lifetime. There is a compelling scientific hypothesis — explored in the palaeontological literature though difficult to prove from the fossil record alone — that the capacity for metabolic withdrawal was a decisive survival advantage across at least one of Earth's great mass extinction events.
The end-Cretaceous impact event, approximately 66 million years ago, produced an impact winter: sunlight blocked by ejecta, temperatures collapsing, photosynthesis failing, food chains disrupting from the bottom up. For animals dependent on continuous feeding and active thermoregulation, the months or years of cold and darkness that followed were lethal. The non-avian dinosaurs — almost certainly metabolically active, food-dependent animals — did not survive. The mammals that did survive into the Palaeocene were predominantly small, burrowing, and, by the evidence of their descendants, likely capable of torpor or hibernation. The surviving bird lineages were similarly small-bodied and metabolically flexible.
The adaptive withdrawal principle, traced across the pages of this essay, may have done more than help individual animals survive a dry season or a cold winter. It may have allowed entire lineages to outlast the end of the Mesozoic — to sleep, in effect, through a planetary catastrophe, and wake into a world the large specialists had vacated.
The continuum, viewed at that scale, runs from the hummingbird's eight hours of nightly torpor to the ancestors of everything alive today.
To look across the full spectrum of adaptive metabolic withdrawal — from the hummingbird's nightly torpor to the lungfish's years-long suspension, from the dolphin's half-sleeping mind to the polar bear mother nursing cubs in her snow den, from the dwarf lemur surfacing from seven months of torpor to discharge a night's worth of accumulated dreams — is to see something that is obscured when sleep is studied only in the narrow register of human overnight rest.
Sleep is not a fixed requirement. It is a strategy. The organism takes whatever form of that strategy its circumstances demand — however deep, however long, however architecturally unusual — and deploys it with a precision that millions of years of evolutionary pressure have refined to extraordinary calibration. The lungfish does not choose to aestivate. It does not calculate that four years underground is the optimal response to a drying riverbed. The strategy is built in, triggered by environmental signals, and implemented through biological mechanisms of remarkable sophistication.
The human infant does not choose to wake at 2am. The biology does. The question is not whether this is a problem. The question is what problem the biology is solving — and the answer, viewed across the full spectrum, is clear. It is the same problem that every organism on the spectrum is solving. How to allocate the finite resources of a living body across the competing demands of survival, reproduction, and development, in the conditions that actually exist, rather than the conditions that cultural norms have decided ought to exist.
The spectrum does not make sleep less important. It makes it more clearly understood. And understanding it clearly is the beginning of treating the parents who are lost in it — at 3am, with a waking infant and a head full of anxiety — with the generosity that the biology actually supports.
Blanco, M.B., Dausmann, K.H., Faherty, S.L. and Yoder, A.D. (2024). 'Primate hibernation: the past, present, and promise of captive dwarf lemurs.' Annals of the New York Academy of Sciences. doi: 10.1111/nyas.15206.
Duke Lemur Center (2024). 'Fat-tailed dwarf lemur.' Available at: Duke Lemur Center
Folk, G.E., Larson, A. and Folk, M.A. (1980). 'Physiology of hibernating bears.' Bears: Their Biology and Management, 4, 373–380.
Frontiers in Neuroanatomy (2019). 'Sleep, torpor, and hibernation: the relationship between sleep and metabolic suppression.' Frontiers in Neuroanatomy, 13, 87.
Geiser, F. (2004). 'Metabolic rate and body temperature reduction during hibernation and daily torpor.' Annual Review of Physiology, 66, 239–274.
Heimroth, R.D. (2018). 'Mucosal immune system of African lungfish.' Doctoral dissertation, University of New Mexico.
Krystal, A.D., Schopler, B., Kobbe, S., Williams, C., Rakatondrabe, Y., Yoder, A.D. and Bhaskaran-Nair, K. (2013). 'The relationship of sleep with temperature and metabolic rate in a hibernating primate.' PLoS ONE, 8(9), e69914.
Lyamin, O.I., Manger, P.R., Ridgway, S.H., Mukhametov, L.M. and Siegel, J.M. (2008). 'Cetacean sleep: an unusual form of mammalian sleep.' Neuroscience and Biobehavioral Reviews, 32(8), 1451–1484.
Navas, C.A., Gomes, F.R. and Carvalho, J.E. (2023). 'Aestivation in nature: physiological strategies and evolutionary adaptations in hypometabolic states.' International Journal of Molecular Sciences, 24(18), 14093.
Ong, J.L.Y. et al. (2017). [Myostatin expression during lungfish aestivation; prevention of muscle atrophy.] Cited in Frontiers in Genetics, PMC9886888.
Oxford Scientist (2023). 'A fish out of water: How African lungfish sleep through the heat.' Oxford Scientist.
PMC4977174. Chew, S.F. et al. (2013/2014). 'Aestivation and brain of the African lungfish Protopterus annectens.' Temperature, 1(2), 82–83.
PMC9886888. 'Aestivation induces widespread transcriptional changes in the African lungfish.' Frontiers in Genetics, 2023.
Polar Bear Range States (2024). 'Reproduction.' Available at: Polar Bear Range States
Polar Bears International (2024). 'The basics of polar bear denning.' Available at: Polar Bears International
Rattenborg, N.C., Amlaner, C.J. and Lima, S.L. (2000). 'Behavioral, neurophysiological and evolutionary perspectives on unihemispheric sleep.' Neuroscience and Biobehavioral Reviews, 24(8), 817–842.
Rattenborg, N.C., Lima, S.L. and Amlaner, C.J. (1999). 'Facultative control of avian unihemispheric sleep under the risk of predation.' Behavioural Brain Research, 105(2), 163–172.
Rattenborg, N.C., Voirin, B., Cruz, S.M., Tisdale, R., Dell'Omo, G., Lipp, H.P., Wikelski, M. and Vyssotski, A.L. (2016). 'Evidence that birds sleep in mid-flight.' Nature Communications, 7, 12468.
Robbins, C.T., Lopez-Alfaro, C., Rode, K.D., Tøien, Ø. and Nelson, O.L. (2012). 'Hibernation and seasonal fasting in bears: the energetic costs and consequences for polar bears.' Journal of Mammalogy, 93(6), 1493–1503.
Siegel, J.M. (2009). 'Sleep viewed as a state of adaptive inactivity.' Nature Reviews Neuroscience, 10(10), 747–753.
Storey, K.B. (2001). 'Turning down the fires of life: metabolic regulation of hibernation and estivation.' In K.B. Storey (ed.), Protein Adaptations and Signal Transduction. Amsterdam: Elsevier.
Storey, K.B. (2016). 'Metabolic flexibility: hibernation, torpor, and estivation.' Comprehensive Physiology, 6(2), 737–771.
United States Geological Survey (2024). 'Polar bear maternal denning.' Available at: USGS
Whiteman, J.P., Harlow, H.J., Durner, G.M., Anderson-Sprecher, R., Albeke, S.E., Regehr, E.V., Rode, K.D. and Ben-David, M. (2015). 'Summer declines in activity and body temperature offer polar bears limited energy savings.' Science, 349(6245), 295–298.
Topics: #sleepbiology #adaptiveinactivity #torpor #hibernation #aestivation #lungfish #polarbear #dwarf lemur #unihemisphericsleep #dolphin #frigatebird #sleepspectrum #sleepscience #infantsleep #YoungFamilyLife #psychology
© 2026 Steve Young and YoungFamilyLife Ltd. All rights reserved.
This essay was developed collaboratively using AI assistance to research academic sources and refine content structure, while maintaining the author's original voice, insights, and "Information Without Instruction" philosophy. No part of this essay may be reproduced, distributed, or transmitted in any form or by any means, including photocopying, recording, or other electronic or mechanical methods, without the prior written permission of the copyright holders, except in the case of brief quotations embodied in critical reviews and certain other noncommercial uses permitted by copyright law.
For permission requests, contact: info@youngfamilylife.com