Most conversations about menopause, to the extent they happen in a clinical setting at all, start and end at the same set of symptoms. Hot flashes. Night sweats. Sleep disruption. Mood changes. These are real, they are common, and for many women they are severe enough to significantly affect quality of life. But they are also, in an important sense, the surface of a much deeper physiological story — one that involves the brain directly, in structural and functional terms, and one that most women are not hearing from the people who are supposed to be helping them navigate this transition.

The cognitive dimension of menopause — the memory changes, the concentration difficulties, the particular kind of mental fatigue that many women in their late forties and fifties describe — has been systematically underrepresented in clinical guidance and research funding for decades. That is beginning to change, but the change is arriving slowly, and the practical consequence is that women are frequently left to interpret their own symptoms without context, without a framework, and without information about interventions whose effectiveness is, at this point, reasonably well supported by evidence — provided the timing is right. The timing, it turns out, is everything.

What the brain actually goes through

A 2026 review published in The Lancet titled “Advances in understanding of cognitive symptoms during menopause” brought together the current state of evidence on what happens neurologically during this transition, and the picture it presents is more specific and more structural than the popular understanding of menopause typically includes. Estrogen is not merely a reproductive hormone. It has well-documented neuroprotective effects — it supports synaptic plasticity, promotes the production of acetylcholine (a neurotransmitter central to memory and attention), and appears to modulate the brain’s inflammatory response. When estrogen levels decline during the menopausal transition, the brain is not simply losing a hormone. It is losing a system of support it has relied on throughout adulthood.

The structural consequences are measurable. Research cited by the Menopause Society has documented reductions in gray matter volume in the frontal and temporal cortex and in the hippocampus — precisely the regions involved in memory formation, executive function, and the ability to hold and manipulate information in working memory. These reductions are not subtle on a population level. They are consistent enough across studies to be considered a feature of the menopausal transition rather than an incidental variation. What this means, practically, is that the brain fog many women report during perimenopause is not psychosomatic, not a side effect of stress or poor sleep alone, and not a symptom that politely awaits acknowledgment before making itself felt in daily life.

The cognitive symptoms women are experiencing but not naming

There is a particular kind of suffering that comes from experiencing symptoms you cannot name, in a domain where your reports have historically been met with skepticism or normalization. Many women going through perimenopause describe a cognitive texture that is difficult to articulate precisely because it is diffuse — not a single dramatic deficit but a constellation of subtle difficulties that compound over time. Forgetfulness that feels qualitatively different from ordinary absentmindedness. Difficulty holding a thread of thought through a complex task. A kind of mental friction that wasn’t there before, an extra effort required to do things that previously felt automatic.

The research vocabulary for this cluster of experiences covers attention, working memory, verbal memory, and executive function — all the cognitive capacities associated with the prefrontal and hippocampal regions where gray matter reductions have been documented. The SWAN (Study of Women’s Health Across the Nation) cohort, which has followed women longitudinally through the menopausal transition for over two decades, found that cognitive performance declines measurably during perimenopause. Crucially, the SWAN data also suggests that this decline may not be permanent — there is evidence of possible reversal, or at least stabilization, in the postmenopausal phase as the brain adapts to its new hormonal environment.

That potential reversal is important information. It means that what women experience during perimenopause is not necessarily a preview of permanent cognitive decline but a transition period with its own arc — one that the brain navigates, imperfectly and with varying degrees of difficulty, toward a new equilibrium. The problem is that understanding this arc, and making informed decisions about whether and how to intervene, requires a conversation that is not yet happening routinely in clinical settings.

The timing problem with hormone therapy

The most consequential piece of information in the current evidence base — and the one most likely to remain unshared in a routine clinical visit — is that the effectiveness of hormone therapy for cognitive outcomes is not uniform across time. It depends critically on when treatment is initiated, and the window during which initiation appears most beneficial is the same window during which most women are still actively navigating the transition and most actively need support.

An observational study published in Neurology found that estrogen therapy initiated in midlife — during or shortly after the menopausal transition — was associated with improved verbal memory. The same intervention initiated later in life showed no such association. This is not a minor calibration note. It is a fundamental characteristic of how the intervention works, and it means that a woman who waits until her sixties to discuss hormone therapy with a doctor, perhaps because the cognitive conversation never happened in her fifties, may have missed the window during which that therapy could have meaningfully supported brain health.

This timing dependence is sometimes described as the “critical window hypothesis” — the idea that the neuroprotective effects of estrogen are most available when the brain’s estrogen receptors are still responsive and the menopausal transition is still underway. The research supporting this hypothesis is actively contested. A 2025 meta-analysis in The Lancet Healthy Longevity, applying stricter risk-of-bias criteria, found no evidence for a cognitive benefit tied to the timing of hormone therapy. Other analyses, including a Weill Cornell meta-analysis of 34 randomised trials, found timing-dependent effects on verbal memory for certain formulations. The broad signal is present in parts of the literature, but it is not yet settled science. Individual variation, hormonal formulation, and interaction with other risk factors all affect outcomes in ways the research has not fully resolved.

But the broad signal — that earlier intervention is more effective than later intervention for cognitive outcomes — is consistent enough that leading researchers have begun calling explicitly for earlier, more routine discussion of these options with patients.

The UK Royal College of Obstetricians and Gynaecologists identified the cognitive effects of menopause as one of its top ten research priorities — a designation that reflects both the seriousness of the issue and the relative thinness of the clinical infrastructure currently built around it.

Why the conversation isn’t happening

The reasons the conversation isn’t happening are multiple, and none of them are particularly flattering to the systems involved. Menopause has historically been undertreated and under-researched relative to its prevalence and impact. The WHI study of the early 2000s, which raised concerns about hormone therapy and was widely interpreted as a broad warning against it, cast a long shadow over the field — even though subsequent analysis substantially revised that picture, particularly for younger women and for the specific question of cognitive outcomes. That shadow has been slow to lift from clinical practice.

There is also the matter of consultation time. A standard appointment is not well structured for a conversation that requires explaining neurological mechanisms, walking through evidence about timing and formulation, discussing individual risk factors, and arriving at a genuinely informed decision. Many women do not bring the cognitive symptoms up, partly because they are uncertain whether they are real or significant, partly because they have absorbed the cultural message that menopause is something to be endured rather than managed. And many clinicians, even those who are receptive, do not ask — either because it falls outside their training, because they are uncertain of the evidence, or simply because the appointment ends before the topic arises.

What changes if the conversation does happen — earlier, more routinely, and with better information on both sides — is that women can make decisions about their own brain health during the window in which those decisions carry the most weight. Not all women will want or be appropriate candidates for hormone therapy. There are legitimate individual differences in risk profile, personal preference, and clinical judgment that should shape those decisions. But the decision cannot be made well if the information never arrives. The current situation, in which timing matters enormously and most women are not told that timing matters, is not an acceptable equilibrium — and the evidence base is strong enough that calling for more routine clinical discussion is not premature.

The post The conversation women aren’t having with their doctors about menopause and memory loss isn’t just overdue — it may be one of the most important health decisions of their fifties appeared first on Space Daily.

On a clear, moonless autumn night, away from city lights, the patch of sky between the constellations Andromeda and Cassiopeia contains a faint, fuzzy smudge that looks like a slightly out-of-focus star. The smudge is not a star. It is the Andromeda Galaxy, designated M31, the nearest large galaxy to our own and the most distant object that an unaided human eye can see. The light reaching your retina from Andromeda has been travelling for approximately 2.5 million years. It started its journey across intergalactic space at a moment when, on Earth, the first members of the genus Homo had recently appeared in East Africa and were beginning to chip the earliest stone tools out of pebbles. The light is older than our genus’s mastery of the rock.

According to NASA’s Hubble Messier Catalogue reference on M31, Andromeda lies at a distance of 2.5 million light-years, contains an estimated trillion stars, and spans roughly 260,000 light-years across — about twice the diameter of the Milky Way’s main disk, depending on how each galaxy’s outer halo is measured. It is the largest member of our Local Group of galaxies, a small cluster of about 80 galaxies bound together by gravity that includes the Milky Way, the Triangulum Galaxy, the Magellanic Clouds and a few dozen smaller satellite galaxies. Andromeda and the Milky Way are the two heavyweights of the group, separated by 2.5 million light-years of nearly empty space and approaching each other at roughly 110 kilometres per second. In about 4.5 billion years, the two galaxies are predicted to merge.

What “2.5 million light-years” means

A light-year is the distance light travels in one year, moving at 299,792 kilometres per second through a vacuum. It works out to roughly 9.46 trillion kilometres. The “year” in “light-year” therefore refers to the time the light takes, not to the date when the light was emitted. A galaxy “2.5 million light-years away” is 2.5 million light-years’ worth of empty space distant, meaning that the light from it requires 2.5 million years to cover the gap.

The implication is the one that makes astronomical observation a form of time travel in reverse. Any observation of a distant object is necessarily an observation of that object’s past, by the amount of time the light has taken to reach the observer. When you look at the Sun, you are seeing it as it was about eight minutes ago. When you look at Proxima Centauri, the nearest star outside the Solar System, you are seeing it as it was about four years ago. When you look at the brightest stars of the constellation Orion, you are seeing them as they were a few hundred to a few thousand years ago, depending on the specific star. When you look at Andromeda, you are seeing it as it was 2.5 million years ago. The galaxy has continued to evolve in the meantime. Whatever has happened in M31 since the light left, no observer on Earth can see yet.

The poetic version of this fact is sometimes used by NASA’s own science-communication team. NASA’s “Catch Andromeda Rising” guide for night-sky observers phrases the framing succinctly: M31 is “so far away that the light you see left M31’s stars when our earliest ancestors figured out stone tools.”

What was happening on Earth then

The corresponding moment in Earth’s history is well documented in the palaeoanthropological record. According to the Smithsonian Institution’s Human Origins Program, the genus Homo — the group containing our species and our closest extinct relatives — appears in the African fossil record from about 2.8 million years ago. The earliest known specimen, called LD 350-1, was discovered at the Ledi-Geraru site in Ethiopia’s Afar region in 2013 and dated to roughly 2.75-2.8 million years ago. According to the Becoming Human project at Arizona State University’s Institute of Human Origins, the LD 350-1 mandible is “one of the best fossil representatives from this poorly understood period of human evolution,” giving researchers their clearest single window into the emergence of our genus. By the time the light from Andromeda that is now reaching Earth was just beginning its journey, the genus Homo had been in existence for perhaps 200,000 to 300,000 years.

The same period saw the appearance of the first deliberately produced stone tools in the archaeological record. The earliest stone tools, the Oldowan industry, date to about 2.6 million years ago, with cut-marked animal bones from the same period showing that early humans were using sharp stone flakes to butcher carcasses. Homo habilis, the species whose Latin name literally translates as “handy man,” lived from about 2.4 million to 1.65 million years ago, and was given the name specifically for its association with these early stone tools.

The light that reaches your eye from Andromeda tonight began travelling toward you during the early phase of this period. While the first photons were crossing the gap between M31 and the Milky Way, a small number of early hominins in East Africa were taking the first reliably documented steps toward the technology that would, two and a half million years later, build the telescopes capable of imaging the galaxy the light came from.

Why Andromeda is the visible limit

The reason Andromeda is the most distant naked-eye object, despite being roughly the size of six full Moons in angular extent across the sky, is that it is also extremely faint. Most of the galaxy’s surface brightness is spread thinly across a large area, with only the bright core visible to the unaided eye. Under truly dark skies, observers report being able to see fainter portions of the disk extending well beyond the bright nucleus, but in practice most casual observers see only a small, slightly elongated smudge a few times brighter than the surrounding sky background.

Other galaxies are technically visible to keen observers under exceptional conditions. The Triangulum Galaxy, M33, lies at about 2.7 million light-years and is sometimes claimed as a naked-eye object by experienced amateurs at very dark sites. M81, in Ursa Major, is much further at 12 million light-years and has been reported by a small number of naked-eye observers under perfect conditions. These observations are at the absolute limit of human visual capability and require dark adaptation, perfect transparency, and skilled averted vision. For practical purposes, Andromeda is the conventional and widely-accepted “most distant naked-eye object.” Anyone who can see it without a telescope is, in the strict sense, seeing further than any other unaided human observation.

The galaxy looks unremarkable from Earth because of the distance. Up close, M31 contains roughly a trillion stars, dozens of satellite galaxies, supermassive black holes, supernovae, and the same vast structural complexity that the Milky Way contains. Hubble’s most recent high-resolution survey, released in 2025, assembled the largest photomosaic ever made by the space telescope: 2.5 billion pixels stitched together from 600 overlapping snapshots taken over a decade of observations, capturing the glow of about 200 million individual stars — roughly one fiftieth of one percent of M31’s total stellar population. From 2.5 million light-years away, the entire galaxy appears as a faint patch of haze, smaller in subjective impression than a moonbeam. The light has been on its way since before any human ancestor knew what a tool was.

The post The light from the Andromeda Galaxy — the most distant object visible to the naked human eye — left its source about 2.5 million years ago, which means when you look at it on a clear night, you are seeing light that began its journey to Earth around the time the first members of our genus, Homo, were learning to use stone tools appeared first on Space Daily.

We tend to think of laughter as a social performance — the audible signal that something is funny, the punctuation on a joke well received. Even people who study emotion professionally can drift into treating laughter as essentially expressive, as the outward visible surface of an inner state. But a growing body of research is pushing back against that framing, and the pushback is coming from neuroscience rather than philosophy.

Laughter, it turns out, is a biological event with measurable consequences for the hormonal environment, the neural reward system, and — in the case of children — the actual architecture of the developing brain. It is not merely the sign of a good mood. It is, in important respects, a driver of one.

A 2026 book, “The Brain Loves to Laugh” by Dr. Jacqueline Harding, an early childhood researcher at Middlesex University, published by Routledge, brought a degree of biological specificity to this question that has rarely been attempted at the developmental level. Harding’s analysis synthesized research across neuroscience, developmental psychology, and endocrinology to ask what laughter does to the brain — not in the abstract, but in the physiological and structural sense, and particularly during the period when the brain is most susceptible to experience-dependent shaping. The findings complicate the idea that laughter is something that happens to children. They suggest it is something that happens inside them, at a level that shapes who they become.

The reward circuit connection

One of the more striking findings in Harding’s analysis is the mapping of laughter onto the brain’s mesolimbic reward system — the same distributed network activated by food, sex, social bonding, and other stimuli that evolution has decided are worth pursuing. This is not a metaphor about how laughter feels good. It is a description of neural architecture. The experience of genuine laughter recruits the ventral tegmental area, the nucleus accumbens, and the prefrontal cortex in patterns that overlap substantially with other primary rewards. Dopamine is released. So are serotonin, endorphins, and oxytocin.

What this means, from a developmental standpoint, is that laughter is not a secondary or incidental feature of a child’s emotional life. It is wired into the same motivational circuitry that drives learning, attachment, and the pursuit of pleasure more broadly. The child who laughs is not simply reacting — their brain is generating the same neurochemical conditions associated with reward and approach behavior that are foundational to motivated engagement with the world.

This helps explain something that developmental researchers have noted for decades but struggled to fully account for: the surprising intensity with which young children seek out the experiences and people that make them laugh, long before they have language to explain why.

It also reframes laughter’s developmental timeline. Laughter precedes speech — children laugh reliably before they produce words, and the emergence of shared laughter between caregiver and infant is one of the earliest markers of social bonding. The fact that this emerges so early, and that it maps onto the same reward circuitry as other primary biological drives, is not coincidental. It appears to be how the social brain bootstraps itself into function before language is available to do the same work.

What the cortisol data shows

Beyond the reward system, Harding’s analysis is specific about what laughter does to the hormonal environment — and the finding that has attracted the most attention is the effect on cortisol. Cortisol is the primary stress hormone in humans, produced by the adrenal glands in response to perceived threat or demand. It is not inherently harmful — cortisol plays important roles in metabolism, immune function, and alertness — but chronically elevated cortisol is associated with a wide range of negative outcomes, and in developing children, sustained cortisol elevation has particular consequences for neural development that research has tracked with increasing precision.

Laughter, Harding’s analysis found, physically lowers circulating cortisol. This is not a claim about mood or subjective wellbeing. It is a measurable change in the hormonal environment, and it comes paired with a reduction in epinephrine — the other major stress-response neurochemical — while simultaneously raising the neurochemicals associated with positive affect and social connection. A systematic review and meta-analysis of interventional studies on spontaneous laughter and cortisol levels provides convergent evidence for this effect across populations, and a 2025 meta-analysis of laughter interventions in children found large effect sizes for anxiety reduction in pediatric patients — specifically in hospital settings using structured clown-therapy interventions. This suggests the hormonal mechanism has meaningful real-world consequences, not just lab-based correlates.

The phrase “physically lowers cortisol” is worth pausing on. It is not unusual, in popular writing about emotional states, to describe psychological experiences in language that implies biological reality without committing to it. The research here does commit. When a person laughs — genuinely laughs, not a performed social laugh but the involuntary kind — the body produces less of the hormone associated with threat-response and more of the hormones associated with approach, bonding, and reward. That is a biological event. Its consequences are biological consequences.

How this restructures the developing brain

The most significant dimension of Harding’s analysis, from a developmental perspective, is the argument about what repeated emotional experiences do to the architecture of a young brain. Early emotional states, she argues, do not merely pass through a child — they become embedded in its neural structure. The brain develops in the context of its dominant emotional environment, and the circuits that are most frequently activated during early childhood are the circuits that develop most robustly. This is a version of the Hebbian principle — neurons that fire together wire together — applied to affective experience at scale.

The implication is that a child who experiences frequent shared laughter is not simply having more pleasant moments than a child who does not. They are developing, gradually and through repetition, a brain that has built stronger infrastructure around the states associated with those moments: reward, safety, approach, connection, the resolution of playful tension. The prefrontal network that laughter activates — and that humor, as a cognitively demanding activity requiring the resolution of conflicting ideas, exercises with particular intensity — is the same network involved in executive function, emotional regulation, and the management of stress.

This last point about humor as cognitive work is underappreciated. Harding’s analysis notes that humor is genuinely demanding — understanding a joke requires holding two incompatible frameworks simultaneously and resolving the incongruity between them. That is not a trivial cognitive task, and doing it repeatedly appears to exercise the neural machinery of flexible thinking in ways that have downstream effects on cognitive and emotional resilience. The child who laughs a lot is, in this account, also a child whose brain is being worked in particular ways that matter for development.

The co-regulation dimension of this is equally important. When an adult and child share laughter — when the adult’s face and voice and body communicate delight, and the child’s nervous system responds to that signal — what is happening is not merely bonding in the social sense. Research into parent-child co-regulation during positive shared experiences — including play and laughter — has found measurable physiological coordination between caregiver and child, including heart rate alignment and coordinated brain activity, suggesting their nervous systems are actively attuned during these moments.

The child’s limbic system is, through that alignment, acquiring a working model for what regulated emotional states look like and feel like — a model it can eventually deploy independently. Co-regulation through shared joy is, in this sense, a form of instruction in self-regulation that requires no words and no deliberate teaching.

What remains when the laughter fades

There is a temptation, when encountering research like this, to reach immediately for prescriptions — to convert findings about laughter and neural development into a program, a set of recommendations, a checklist of things parents should do more often. That is probably not the most useful response to what the science is showing. The research does not describe a deficit to be corrected. It describes a mechanism that is already operating in most children’s lives, in the ordinary texture of play and silliness and shared delight that tends to happen naturally when adults and children spend time together without too much pressure on either side.

What the neuroscience adds is a more accurate description of what is actually happening during those moments. The child who collapses in giggles is not simply expressing happiness. Their hypothalamic-pituitary-adrenal axis is producing less cortisol. Their reward network is receiving a signal that the present moment is safe and worth approaching. Their prefrontal circuitry is being exercised in ways that contribute to cognitive flexibility and emotional regulation.

Their nervous system is synchronizing with the nervous system of the person laughing with them, and that synchrony is building a model they will carry forward. None of this requires anything more complicated than what most adults, at their best, already bring to the children in their lives. The science is not an instruction manual. It is an explanation for something that was already working.

The post Laughter activates many of the same brain reward circuits as food and sex, and a 2025 study finds it measurably lowers cortisol and may restructure how the developing brain builds resilience to stress appeared first on Space Daily.

The tomato is generally believed to have arrived in Europe in the early 16th century, brought by Spanish explorers — most likely Hernán Cortés after the conquest of Tenochtitlán in 1521. For the next two and a half centuries, Europeans grew the plant in their gardens and refused to eat it. Tomatoes appeared in herbals, in botanical illustrations, in ornamental displays, and in private correspondence between curious naturalists. They did not appear on European dinner plates in any meaningful way until the late 1700s. The reason was not pewter plates, lead poisoning, or any specific incident. The reason was botanical resemblance. The tomato is a member of the nightshade family, Solanaceae, and Europeans correctly identified that the plant was related to species they already knew to be dangerous.

What Europeans actually saw

The earliest documented European reference to the tomato as food comes from the Italian herbalist Pietro Andrea Mattioli, who described the fruit in 1544 and classified it as a type of mandrake. According to a 2025 Smithsonian Magazine update on the cultural history of the tomato, originally published in 2013 by K. Annabelle Smith and updated by Kayla Randall, Mattioli’s classification of the tomato as a mandrake “had later ramifications.” The classification was not arbitrary. Mandrake (Mandragora officinarum) is a real plant, used in European folk medicine and witchcraft traditions for centuries, and it is genuinely poisonous. It is also a member of the same family, Solanaceae, as the tomato. So is deadly nightshade (Atropa belladonna), the bittersweet nightshade (Solanum dulcamara), the black nightshade (Solanum nigrum), henbane (Hyoscyamus niger), and a number of other plants well known in early modern Europe to be variously poisonous, narcotic, or hallucinogenic.

From a 16th-century botanist’s perspective, the tomato fitted neatly into a known family of dangerous plants. The leaves were similar in shape to those of black nightshade. The flowers were similar to those of belladonna. The fruit was small and round, in the early imported varieties yellow or pale red rather than the large red tomatoes of modern cultivation. Mattioli’s classification was reasonable. The tomato was related to plants that killed people. The natural assumption, in the absence of any direct experimental evidence to the contrary, was that the tomato also killed people.

The leaves and stems are actually toxic

The assumption was not entirely wrong. The leaves and stems of the tomato plant do contain alkaloids that would, in sufficient quantity, make a person sick. The principal compound is tomatine, a glycoalkaloid related to solanine, the better-known toxic compound in green potatoes. According to a peer-reviewed review on tomato glycoalkaloid toxicology published in Food Chemistry in 2024, tomatine concentrations are highest in the stems, roots and leaves of young tomato plants, and progressively diminish as the fruits ripen. The compound can cause vomiting, diarrhoea, abdominal pain and lethargy in mammals at sufficient dose, and acts as a natural fungicide and insecticide protecting the plant from microbial and herbivore attack.

An adult would have to consume a substantial quantity of tomato leaves to experience symptoms, but the toxic compounds are real. Livestock occasionally poisons itself by overgrazing on tomato plants. Children who chew on the leaves can become mildly unwell. The plant defends itself against insects and grazers with the same chemistry that makes the rest of the family a poor choice for casual experimentation. Europeans observing the tomato as a new species, examining its leaves, smelling its pungent foliage, and noting its membership in a recognised family of poisonous plants were not making a foolish judgement. They were making a botanically defensible one. The error was specifically in extending that judgement to the ripe fruit.

Two hundred years of hesitation

The tomato’s reputation in Europe was therefore a botanical reputation rather than an experiential one. The fruit was not actually killing anyone. No physician of the 1600s could point to a chain of confirmed tomato-related deaths. The plant was simply assumed dangerous on the grounds that its relatives were dangerous, and the assumption persisted in the absence of widespread tests of the contrary.

In the warmer climates of southern Europe, where tomatoes grew more readily, this reluctance eroded first. Italian cookbooks began including tomato recipes in the late 17th century. The earliest known recipe for tomato sauce was published in Naples in 1692, in Antonio Latini’s two-volume cookbook Lo Scalco alla Moderna (“The Modern Steward”). Latini, a former orphan who had risen to become steward to the first minister of the Spanish viceroy of Naples, included a recipe for “Salsa di Pomodoro alla Spagnuola” — tomato sauce in the Spanish style — calling for roasted and peeled tomatoes mixed with chopped onion, chilli, thyme, salt, oil and vinegar. The sauce was intended for meat and fish rather than pasta, but it marked the formal entry of the tomato into recorded European cooking. By the 18th century, tomatoes were a regular feature of Mediterranean cuisine, particularly in southern Italy and Spain. In the colder regions of northern Europe — England, the Netherlands, Germany, the Scandinavian countries — the suspicion lasted longer. English gardeners grew tomatoes ornamentally well into the 1800s, and the fruit only entered British and American kitchens widely in the second half of the 19th century.

The popular story that the tomato’s bad reputation was driven by lead leaching from pewter plates has been widely repeated in food-history journalism, but the explanation has been challenged by chemists. As cited in the Smithsonian Magazine update, Dr. Joe Schwarcz, director of the Office for Science and Society at McGill University in Montreal, has dismissed the lead-leaching story directly, noting in a 2023 video for the Montreal Gazette that “the amount that would be leached out would be trivial, and you’d never get sick from it.” Wine and vinegar — both more acidic than tomatoes — were extensively consumed from pewter vessels throughout the medieval period without producing any analogous panic. The pewter story, in the form that has circulated in recent decades, appears to be a colourful retrospective explanation rather than a documented historical cause.

How the fruit eventually won

The rehabilitation of the tomato in northern Europe and North America happened slowly through the 18th and 19th centuries, driven by gradually accumulating evidence that the fruit was, in fact, safe to eat. According to the Thomas Jefferson Foundation’s Monticello research, Jefferson was an early American adopter of the tomato. His butler Étienne Lemaire purchased tomatoes for Presidential dinners during Jefferson’s time in office, and Jefferson’s garden book records the planting of tomatoes at Monticello from 1809 until 1824. Jefferson had referred to tomatoes as a common Virginia garden plant in his 1781 Notes on the State of Virginia, suggesting they were already familiar in some American gardens before the wider rehabilitation got underway. In Salem, New Jersey, in 1820, a local gentleman named Robert Gibbon Johnson is said to have eaten a basket of tomatoes on the steps of the courthouse in front of a curious crowd, with no ill effects. The story is colourful and probably embellished, but it captures the social process accurately. The tomato was a fruit Europeans and Americans had to decide, collectively and gradually, to start eating.

By the end of the 19th century, the rehabilitation was complete. The development of Pizza Margherita in Naples in 1889, the spread of Italian immigration to the United States in the late 19th and early 20th centuries, and the rise of tin canning all helped the tomato establish itself as a staple ingredient in Western cuisine. The original botanical worry — that the tomato belonged to a family of poisonous plants — remained literally true. It was just that the fruit itself, of this one member of that family, turned out to be harmless. Two centuries of European caution had been based on a reasonable inference that happened to be wrong about which parts of the plant were dangerous.

The post Tomatoes were considered poisonous in much of Europe for nearly two centuries after they were introduced from the Americas — not because of anything dangerous in the fruit itself, but because the plant belongs to the same botanical family as deadly nightshade, with leaves and stems that really do contain toxic alkaloids, and the misunderstanding kept the fruit out of European cuisine until the late 1700s appeared first on Space Daily.

In 2023, Saudi Arabia paid approximately $140,000 to import construction-grade sand from Australia. The figure is small in dollar terms, but the trade flow it represents is the visible part of a much larger phenomenon. Saudi Arabia, a country dominated by some of the largest sand deserts on Earth, cannot use the sand in its own deserts for most construction work. The grains are the wrong shape. They have been polished smooth by thousands of years of wind erosion, and smooth round grains do not bond with cement in the way that concrete requires. Construction-grade sand has to come from somewhere else — from rivers, lakes, seabeds, or quarries, where the grains have been broken sharp by water rather than rounded by air. For a country building cities the size of NEOM and the Red Sea Project, the result is that even abundant sand is the wrong sand, and imports become a structural feature of the economy.

Why desert sand fails

The materials science is straightforward and well understood. Sand grains come in a range of shapes, depending on how they were produced. Grains that have been carried by rivers, tumbled in streams, ground by glaciers, or pulverised in quarries tend to have rough, angular surfaces. They lock together when packed, like irregular puzzle pieces, and they bond mechanically with the cement paste in concrete to produce a strong composite material. Grains that have been carried by wind, in contrast, behave very differently. Each collision between airborne grains, repeated countless times across the dunes of a desert, slightly rounds off the corners and edges. After thousands of years of this process, the grains end up smooth, spherical, and uniform in size. They are beautiful under a microscope. They are nearly useless in concrete.

The behaviour of desert sand in a wet concrete mix is sometimes described in industry literature as resembling ball bearings. Smooth round grains slide past one another rather than interlocking. They fail to engage mechanically with the cement paste, leaving microscopic voids and weak interfaces throughout the cured material. Concrete made primarily with desert sand cracks more easily, weighs more for the same strength, and ages worse than concrete made with angular sand. For a small structure, the difference might be tolerable. For a 200-storey skyscraper or a kilometre-long bridge, it is not. The structural engineers responsible for Saudi Arabia’s mega-projects need sand whose grains can do their structural job, and the grains in the Empty Quarter cannot.

What gets imported, and from where

The construction sand reaching Saudi Arabia comes from a small number of countries with abundant water-eroded sand reserves and the export infrastructure to ship it economically. Australia is among the most important suppliers. According to the Observatory of Economic Complexity, Australia exported approximately $273 million worth of sand in 2023, making it the second-largest sand exporter in the world. Saudi Arabia, the United Arab Emirates, and other Gulf states are among the regular destinations. Australia’s geological history, with its rivers, quarries, and glacial deposits, has produced the sharp-grained sand that concrete production requires. Australia ships it. The Gulf buys it.

The asymmetry is sometimes treated as an absurdity, the kind of fact that seems to contradict common sense, but it is the logical outcome of how sand actually forms. According to a 2026 analysis by Gulf Good News referencing UN Environment Programme research, Dubai’s Burj Khalifa required approximately 330,000 cubic metres of concrete, with most of the sand component imported from overseas because local desert sand could not provide adequate structural strength. The Palm Jumeirah artificial island in the UAE consumed 94 million cubic metres of marine sand, dredged from specific locations in the Persian Gulf where the grain size was suitable, and even that supply could not be drawn from the surrounding desert. The pattern repeats across the Gulf. Mega-construction demands angular sand. Local deserts cannot supply it. Foreign rivers, quarries, and seabeds do.

The wider problem

Saudi Arabia’s situation is the most counterintuitive example of a global pattern. According to the May 2026 United Nations Environment Programme report Sand and Sustainability: An Essential Resource for Nature and Development, the world extracts approximately 50 billion tonnes of sand and gravel every year — a fivefold increase since 1970, when annual extraction stood at 9.6 billion tonnes. Sand is now the second most consumed natural resource on Earth, after fresh water. Demand has grown at an average annual rate of 3.2 percent over the past half-century, and UNEP projects that demand for sand used in buildings alone could rise by up to 45 percent by 2060. The volume of sand humanity already uses each year is enough to construct a wall 27 metres high and 27 metres thick around planet Earth.

The UNEP report emphasises that most of this sand cannot come from deserts. The reserves usable in construction are concentrated in river systems, coastal areas, and continental shelves, and they are being extracted faster than geological processes can replace them. The result is what UNEP describes as the “sand gap,” in which unregulated sand mining is causing riverbed erosion, the destruction of marine habitats, the collapse of beach ecosystems, and the disappearance of small islands. The countries that supply construction sand are paying environmental costs to do so. The countries that import it are insulated from those costs only because their geography happens to produce the wrong kind of sand.

The longer-term response is shifting toward alternatives. Manufactured sand, produced by mechanically crushing rocks into the angular grain shapes that concrete requires, is becoming an increasing share of construction supply in countries that have started to take the problem seriously. Recycled concrete, in which old buildings are crushed and re-incorporated into new ones, is another partial solution. Saudi Arabia itself is investing in both approaches, and is considering domestic manufactured-sand production as part of its Vision 2030 infrastructure plans. The total amount of imported sand the country actually requires for any given year is therefore a moving figure, dependent on how quickly alternatives scale. What stays constant is the underlying physics. The grains in the Arabian deserts have been rounded by the wind for thousands of years. They will not bond with cement. The country that has more sand than almost any other still has to buy the sand it can actually use.

The post Saudi Arabia imports sand — despite being a country dominated by desert, the sand of the Arabian Peninsula is too smooth and round for use in construction concrete, because thousands of years of wind erosion have polished the grains to a shape that does not bond well with cement, and most construction sand must be imported from countries like Australia appeared first on Space Daily.

In Sequoia National Park in California, a giant sequoia named the President stands 247 feet tall, carries roughly 2 billion leaves, and has been alive for about 3,200 years. It germinated as a small seed in the late twelfth century BC. At the time, the New Kingdom of Egypt was building additions to the temples at Karnak. The Trojan War, if it actually happened, was still within living memory. The first Olympic Games would not be held for another four centuries. The founding of Rome, in the traditional reckoning, would not occur for another 425 years. The Parthenon would not be built for another seven centuries. Most of recorded human history is younger than the President tree.

The President is not even the oldest known giant sequoia. According to a National Park Service history of sequoia age estimates, the Grizzly Giant in Yosemite’s Mariposa Grove was estimated at 3,800 years of age by early 20th-century researchers, and was considered by many to be the oldest of all living giant sequoias. Other historic estimates pushed individual trees as old as 4,000 years. The most carefully verified figures, based on ring counts of stumps in the early 20th century, identified a small number of giant sequoias more than 3,000 years old, with the oldest at about 3,200 years.

Why sequoias live so long

The biological mechanism behind sequoia longevity is a combination of features that together produce an organism unusually resistant to the ordinary causes of tree death. The bark of a mature giant sequoia is up to three feet thick, fibrous, and rich in tannins that deter insects and fungi. The wood itself contains compounds that resist decay, so even fallen sequoia logs remain intact for centuries. The trees are notably fire-adapted: their thick bark insulates the living cambium beneath, and surface fires that would kill most species sweep through a sequoia grove without ending the trees’ lives. In fact, periodic fires are required for sequoia reproduction. Sequoia cones release their seeds in response to the heat of a passing fire, exposing the freshly cleared mineral soil that seedlings need.

The biggest threat to a mature giant sequoia is not age but mechanical failure. The trees grow so large that root systems eventually become unable to anchor them against high winds, particularly in saturated soil after heavy storms. Most sequoias that die in old age do so by falling over rather than by senescence in any biological sense. There is no indication that giant sequoias have a maximum natural lifespan in the way most species do. They simply continue to grow, slowly, for as long as they remain upright. According to Atlas Obscura’s profile of the General Sherman tree, the largest sequoia by volume is “adding volume faster than ever, overturning previous theories that trees grow more slowly as they get bigger.”

Two species, often confused

The popular phrase “sequoia and redwood” conflates two distinct species. The giant sequoia (Sequoiadendron giganteum), confined to about 75 groves on the western slope of the Sierra Nevada in California, is the species producing the famous 3,000-year-old specimens. The coast redwood (Sequoia sempervirens), found along the Pacific coast from southern Oregon to central California, is a closely related but separate species. Coast redwoods are the tallest trees on Earth, with the current record-holder, Hyperion, measuring 380 feet, but they typically do not live as long as giant sequoias. The oldest documented coast redwoods are around 2,200 years old. The world’s oldest individual tree of any kind is neither a sequoia nor a redwood, but a bristlecone pine (Pinus longaeva) named Methuselah, growing in California’s White Mountains, with a verified age of approximately 4,855 years.

What makes the giant sequoias famous is the combination of their longevity with their sheer mass. The General Sherman tree, the largest single-stem tree in the world by volume, contains roughly 1,487 cubic metres of wood and weighs over 2,000 tons. The National Park Service’s current age estimate places it at about 2,200 years, with other published figures ranging from 2,200 to 2,700 years depending on the dating method. Earlier 20th-century estimates were considerably higher, reaching 3,500 years or more, but successive scientific revisions have brought the figure down. The uncertainty reflects the practical difficulty of dating a living tree without coring it deeply enough to count its innermost rings, which for the oldest sequoias would require boring through metres of dense, often partially-rotten heartwood.

What 3,200 years actually looks like

The historical context becomes more striking the more carefully you spell it out. A tree that germinated in 1,175 BC was already a sapling when the Phoenician alphabet was being developed. It was approximately 425 years old when Rome was founded in 753 BC by the traditional reckoning. It was approximately 728 years old when work began on the Parthenon in 447 BC. It was approximately 1,075 years old when Julius Caesar was born in 100 BC. It was approximately 1,200 years old when the Roman Empire was founded under Augustus in 27 BC. It was approximately 1,290 years old when the Roman Empire reached its territorial peak under Trajan around 117 AD.

The tree was already ancient by every meaningful measure when the Western Roman Empire fell in 476 AD. It had been alive for 2,460 years when the Norman Conquest of England occurred in 1066. It had been alive for 2,975 years when the United States declared independence in 1776. It has now lived through the entire span of recorded human history of California’s indigenous Yokuts, Tubatulabal, and Mono peoples, who lived in proximity to the giant sequoia groves for the last several thousand years and who, by oral tradition, regarded the trees with religious significance. The first widely-documented European sighting of a giant sequoia, by Augustus T. Dowd, occurred only in 1852. From the tree’s perspective, that encounter happened less than 6 percent of its life ago.

The General Sherman, the President, the Grizzly Giant, and the other named individuals of the giant sequoia groves are quietly older than nearly every cultural reference humans use to anchor the deep past. They were there when those reference points happened. They are still there now.

The post Sequoia and redwood trees alive today were already mature when the Roman Empire was at its peak — the oldest living giant sequoias are over 3,000 years old, which means they were standing in California before the Parthenon was built in Athens, before Julius Caesar was born, and before the Roman Empire even existed appeared first on Space Daily.

A lost continent lies in pieces beneath southern Europe. Geologists call it Greater Adria: a landmass roughly the size of Greenland that broke off the supercontinent Gondwana, drifted north for tens of millions of years, then collided with Europe and was driven down into the mantle. What remains was scraped off the top during that collision and now sits folded into the mountains of Italy, Greece and the Balkans.

The detailed picture comes mainly from one large piece of work: a 2019 reconstruction published in Gondwana Research by Douwe van Hinsbergen of Utrecht University, with colleagues in Oslo and Zürich. The existence of a buried Adriatic continent was not the new part. Geologists had understood the broad outline for decades. What the paper added was detail: a time-lapse reconstruction of the whole Mediterranean, assembled slice by slice.

What the study actually did

The reconstruction took about ten years. Van Hinsbergen has described the task as a jigsaw whose pieces had been broken, curved and stacked on top of one another, then scattered across more than thirty countries from Spain to Iran, each with its own geological surveys and its own maps. He has called the region, plainly, “a geological mess.”

The team used plate-reconstruction software together with field data and the magnetic orientation locked into rocks at the moment they formed. Those tiny palaeomagnetic signals act as a record of where a rock sat, and which way it faced, in the distant past. Reading them across thousands of samples let the group wind the deformation backwards and watch the region reassemble.

What Greater Adria was

Here the “lost continent” label can mislead. Greater Adria was not a green, populated land that sank like the Atlantis of legend. For much of its life it was largely underwater: a shallow continental shelf in a warm, tropical sea, where sediments settled and slowly hardened into limestone. The closest living comparison is Zealandia, the mostly submerged continent whose few high points are New Zealand and New Caledonia.

It broke from Gondwana about 240 million years ago, in the Triassic, and drifted north. By around 140 million years ago it had reached roughly the size and shape of Greenland.

Then, between about 100 and 120 million years ago, it ran into Europe.

How a continent disappears

Continental crust is generally too buoyant to sink. That is the usual rule, and it is part of why this story is unusual. When Greater Adria met Europe, most of the continent was forced down beneath it, into the mantle, where it remains. Only the uppermost layers, the lighter sedimentary rock, were too light to follow. They were peeled off in the collision, the way a sleeve crumples when an arm is pushed under a table, and left piled into the mountain belts that run through the region today.

Some of that scraped-off rock is familiar. Limestone from Greater Adria, cooked under heat and pressure, became the marble that the Greeks and Romans quarried for their temples. The continent did not vanish without trace. It became scenery, and then it became building stone.

What is solid, and what is a model

The reconstruction is detailed, but it is a reconstruction. Laurent Jolivet, a geologist at Sorbonne University who was not part of the work, told Science that the broad tectonic history had been known for some time, and that the achievement here was the unprecedented level of detail in the systematic time-lapse model.

That distinction matters for reading the result. The northward drift, the collision and the burial rest on a large body of structural and palaeomagnetic data, and are not seriously in dispute. The precise outlines, the exact timing of each stage, and the shape of the continent at any single moment are model outputs, built from scattered and deformed evidence, and they carry the uncertainty that comes with that.

This is one detailed study rather than a closed case, and the fine print of the reconstruction will be revised as more data comes in.

What the reconstruction is good for

For van Hinsbergen’s group, the model is also a tool. Once the Mediterranean is reassembled, other features can be returned to where they began, including extinct volcanoes and ore deposits, which has a bearing on where those resources are found now. A buried continent, read backwards, turns out to be one way of working out how the surface above it was put together, and what got locked inside the mountains on the way.

The post A lost continent lies in pieces beneath southern Europe. Geologists call it Greater Adria: a Greenland-sized landmass that was crushed under Europe, leaving its scraped-off remains scattered through the mountains of Italy, Greece and the Balkans. appeared first on Space Daily.

The planet’s name is HD 189733b. It is one of the closest extrasolar planets to Earth that can be studied in detail, and one of the most thoroughly characterised exoplanets in the astronomical literature. Its star, HD 189733, sits in the constellation Vulpecula, the Little Fox, north of the celestial equator. The star itself is faintly visible to a small telescope from any dark backyard in the northern hemisphere on a clear summer night. The planet is not. The planet has never been directly photographed and probably never will be by any current generation of telescope.

Everything that is known about it has been deduced from the way its star’s light changes as the planet passes in front of, alongside, and behind it. The deductions, after twenty years of accumulated work, describe one of the most hostile environments in the catalogued universe.

It was discovered on 5 October 2005 by François Bouchy and colleagues at the Haute-Provence Observatory in southern France, using the Doppler-spectroscopy method to detect the small gravitational tug of the planet on its host star. The planet is approximately the mass and size of Jupiter. It sits roughly 4.6 million kilometres from its star, which is about one-thirtieth of the distance between the Sun and Mercury. It completes one orbit every 2.2 Earth days. At that distance, it is tidally locked. The same hemisphere faces the star at all times.

On the side facing the star, it is approximately 1,000 degrees Celsius.

What it would be like to be there

The astronomers who have studied HD 189733b in detail describe an atmosphere that has no analogue in the solar system. The temperature differential between the planet’s permanently lit day side and its permanently dark night side, measured by NASA’s Spitzer Space Telescope in 2007, is approximately 260 degrees Celsius. That differential drives atmospheric winds at speeds that, in the upper atmosphere on the day side, reach approximately 7,000 kilometres per hour. As the European Space Agency set out in its 2013 announcement of the planet’s confirmed colour, the wind speeds are roughly seven times the speed of sound. On Earth, by comparison, the strongest sustained surface winds ever recorded reached approximately 410 kilometres per hour during a tropical cyclone. HD 189733b’s winds are approximately seventeen times faster.

The atmosphere is composed primarily of hydrogen and helium, like Jupiter’s, but it also contains a high concentration of silicate particles. Silicates are the family of minerals that make up most of the Earth’s crust, including sand, quartz, and the basaltic rocks that form ocean floors. On HD 189733b, at atmospheric temperatures exceeding 1,000 degrees Celsius, silicate particles condense in the atmosphere from gaseous form into small molten droplets of glass.

The droplets do not fall straight down. They are driven sideways by the 7,000 km/h winds, at velocities at which a single droplet impacting a surface would carry the energy of a small artillery shell. The planet, on the side facing its star, is therefore experiencing continuous horizontal precipitation of molten glass at hurricane speeds, at temperatures that would melt aluminium.

The rain is the weather. There is no break in it.

How they figured out it was blue

The visual colour of an exoplanet 63 light-years away cannot be observed in the conventional sense. The planet is far too faint and far too close to its star to be photographed directly. The team that established HD 189733b’s true colour, led by Tom Evans at the University of Oxford, used a technique called secondary eclipse spectroscopy. Their paper, published in Astrophysical Journal Letters on 1 August 2013, describes the method in detail.

The Hubble Space Telescope’s Space Telescope Imaging Spectrograph observed the HD 189733 system continuously through several full orbital cycles of the planet. During each orbit, the planet passes behind the star from the telescope’s perspective. In the moments before and after the planet disappears behind the star, the telescope is receiving light from both the star and the planet. In the moments when the planet is hidden, the telescope is receiving light from the star alone. By subtracting the second measurement from the first, the team could isolate the light reflected by the planet alone.

The drop in brightness as the planet vanished behind its star was measurable specifically in the blue part of the spectrum, between 290 and 450 nanometres. The drop in the red and near-infrared was much smaller. The published interpretation is that the planet reflects blue light at approximately three to four times the rate it reflects red light, which makes it, in the visual range, a deep cobalt blue.

If a human observer could be positioned in space within the HD 189733 system, at a safe distance, the planet would appear to them as a small, deep blue point of light, almost indistinguishable in colour from the way the Earth appears to astronauts looking back from the International Space Station.

The mechanism that produces the blue colour, however, is completely different.

What the blue actually is

Earth appears blue from space for two reasons. The most obvious is the reflection of light from the oceans, which cover approximately 71 per cent of the planet’s surface. The second, less commonly understood, is Rayleigh scattering in the atmosphere. Short-wavelength light, including blue, scatters more efficiently off the molecules of nitrogen and oxygen in the air than long-wavelength light does. This is why the sky is blue. The same effect contributes to the planet’s blue appearance from orbit.

HD 189733b has no oceans and probably no liquid water of any kind. The temperatures are too high for water to exist as a liquid anywhere in the atmosphere. The blue colour comes entirely from the silicate particles. The droplets of molten glass suspended and condensing in the atmosphere scatter blue light preferentially, in much the same way that nitrogen and oxygen molecules in Earth’s atmosphere scatter blue light. The mechanism is a different kind of Rayleigh-like scattering, off particles that are themselves molten and being driven sideways by hurricane-speed winds, but the optical outcome is similar.

HD 189733b is, by the resemblance of one colour to another, a kind of cosmic mimicry. A planet that looks, from sixty-three light-years away, like Earth. A planet that, on inspection, has nothing in common with Earth except the wavelength of the light it reflects.

Why it matters

HD 189733b belongs to a class of exoplanets called hot Jupiters: gas giants similar in mass and composition to Jupiter, orbiting their stars at distances much closer than Mercury orbits the Sun. The first hot Jupiter was discovered in 1995. Since then, several hundred have been confirmed in the published exoplanet catalogue maintained by NASA’s Exoplanet Exploration Program. They are, on the data so far, surprisingly common in the galaxy. They are also, on the same data, completely absent from our own solar system.

The reasons hot Jupiters form, and the processes that drive them inward to such close orbits around their stars, are still subjects of active investigation. HD 189733b, because of its relative closeness to Earth and its bright host star, has become one of the most-studied hot Jupiters in the astronomical literature. The 2013 confirmation of its blue colour was the first time the visible-light colour of any exoplanet had been measured directly. The same observational programme, and follow-ups using the James Webb Space Telescope, have detected water vapour, carbon dioxide, methane, and atmospheric haze in the planet’s upper layers, building up a picture of an atmosphere chemically rich and physically violent at scales no observation of a solar-system planet has matched.

The exoplanet has, in the years since its discovery, been informally referred to in the astronomical literature as the planet where it rains glass. The wind speeds, the temperatures, and the silicate atmospheric chemistry are now well established. The geological details — whether the molten glass droplets reach the planet’s deeper layers as glass or evaporate back into vapour, whether the planet has a coherent solid core or whether its interior is a continuous fluid down to whatever pressure ultimately produces metallic hydrogen — remain open questions.

What 63 light-years actually means

The light that the Hubble Space Telescope captured in 2013 had been travelling toward Earth since approximately 1950. The light Hubble might capture from HD 189733b today began its journey toward us during the early 1960s. The planet itself, in real time, is doing whatever it has continued to do for the four billion years it has existed. The astronomers who study it are studying its past.

If a human civilisation around HD 189733 were, at this moment, observing Earth through the same kind of telescope Hubble represents, they would be looking at images of Earth as it was in 1962. They would be receiving the light Earth was reflecting during the Cuban Missile Crisis, the early Mercury space programme, and the year before the death of John F. Kennedy.

Earth, from sixty-three light-years away, also looks like a deep blue dot.

The difference is that ours, on closer inspection, has oceans.

The post There is a planet 63 light-years from Earth where the rain is made of molten glass, the winds blow at 7,000 kilometres per hour, the daytime temperature is over 1,000 degrees Celsius, and the planet itself, viewed from space, is the same deep blue as Earth. appeared first on Space Daily.

Freezing solid is supposed to be the end. Ice forming inside a body generally stops the chemistry of life. For almost every animal on Earth, that is exactly how it works. The wood frog did not get the memo.

Each winter, Rana sylvatica lets a large share of the water in its body turn to ice, loses the mains signs of life we measure, and then, months later, restarts. The freezing is the famous part. The thawing is the part that should keep biologists awake.

What “frozen” actually means here

The common assumption is that freezing must mean slowing down, a deep chill that lowers the heart rate and breathing to a crawl. The wood frog does not slow down. It stops. When frozen, the frog shows no heartbeat, no breathing, no blood circulation, no muscle movement, and no detectable brain activity.

The proportion of ice is hard to believe. In a study published in Physiological Reviews, Kenneth and Janet Storey noted that around 65% of water in their bodies can be frozen as extracellular ice, with no physiological vital signs, before returning to normal life within hours of thawing. The ice forms outside the cells, in the spaces between them and around the organs, not inside the cells themselves. That distinction is most of the story.

Janet Storey, a research associate at the Institute of Biochemistry at Carleton University, has described the effect plainly. “They look like they’re totally dead, and then they’re not,” she told the Up Here. The frog looks dead because, by the ordinary measures, it is indistinguishable from dead. It is not.

It is tempting to reach for the phrase “clinically dead,” and the wording almost fits. “Clinical death refers to the medical state involving the complete and irreversible cessation of all body functions,” the cryobiologist Jon Costanzo of Miami University has explained. The word doing the work there is “irreversible.” A frozen wood frog reverses, spontaneously and completely, which is precisely why Costanzo has been careful that the frog only loosely qualifies, and why what its brain is doing during the freeze remains an open question.

How it survives

The trick is sugar. As explained by the folks at National Park Service, glucose keeps the the frogs blood from freezing. As they noted “Hibernating wood frogs can tolerate blood sugar levels 100 times higher than normal without the damage suffered by human diabetics when their blood sugar is only 2 to 10 times above normal”. 

Wild frogs even appear to rehearse the whole performance before committing to it. In Alaska, wood frogs go through repeated freeze-thaw cycles in early autumn before settling into the long freeze of winter, and those cycles seem to prime the system.

The reversal 

Restarting a stopped heart, rebooting a brain with no recorded activity, and clearing months of accumulated metabolic waste, all without permanent injury, is harder to account for than the freezing itself.

The field evidence is striking. Working with wild Alaskan frogs in their natural winter burrows, Larson and colleagues tracked 18 animals that stayed frozen for months at a stretch. The frogs survived being frozen for up to 218 days at minimum temperatures below minus 18 degrees Celsius, with every frog surviving. 

Recovery is fast once it begins. On thawing, the heart and brain restart spontaneously as the soil warms in spring, the contractions resuming on their own after months of silence. Why the restart works at all, after every vital sign has been absent for so long, is the question researchers find hardest to answer.

What the frog is quietly telling us

The wood frog is perhaps a working model for the long-running effort to freeze and bank human organs. A review by Al-Attar and Storey treats the frog’s natural freeze tolerance as a template for cryopreservation and biobanking. 

The gap that research is trying to close is large. As of her 2018 comments, Janet Storey noted that “so far there’s nobody that’s been able to freeze an entire organ and get it to survive and function when it comes back.” 

The stakes are concrete. Tens of thousands of people sit on the U.S. organ transplant waiting list at any given time, far more than the number of transplants performed each year. A way to bank organs for longer would likely change those numbers. 

The post More than 60% of the water in a wood frog’s body can freeze solid each winter: its heart stops, it stops breathing, and for more than 7 months it can lie essentially a frogsicle, before it thaws out in spring and simply hops away appeared first on Space Daily.

In October 1994, a German archaeologist named Klaus Schmidt visited a hilltop near the city of Urfa in southeastern Turkey, in a region of rolling limestone uplands close to the Syrian border. Schmidt had been working at a nearby Neolithic site called Nevalı Çori and was searching the surrounding country for related sites that an earlier survey, conducted in 1963 by a joint team from the University of Chicago and the University of Istanbul, had identified but dismissed. The hilltop the 1963 survey had passed over was called Göbekli Tepe, which translates from Turkish as “potbelly hill.” A local family, the Yıldız family, who owned and farmed the land, had been reporting odd stones turned up by their ploughs for years.

Schmidt arrived at the hill and recognised, almost immediately, that the smooth flat-topped stones the Yıldız family had been ploughing around were the upper surfaces of T-shaped limestone pillars of the kind he had been excavating at Nevalı Çori, buried up to their tops in the hillside. He began systematic excavation the following year, in 1995, under the auspices of the Şanlıurfa Museum and the German Archaeological Institute. The DAI’s own published account of the project sets out the early discovery in detail. Schmidt directed the excavation continuously until his death in 2014. The work has continued under his successors at the German Archaeological Institute and the Şanlıurfa Museum and continues today.

What Schmidt uncovered, and what subsequent excavations have continued to uncover, is a site that, by every prior model of human prehistoric development, should not exist.

What is at the site

Göbekli Tepe consists, in the parts that have been excavated so far, of approximately twenty circular and oval enclosures cut into the bedrock of the hilltop. Each enclosure is bounded by a low limestone wall and contains a ring of T-shaped limestone pillars, with two larger pillars standing at the centre of the ring. The UNESCO World Heritage Centre, which inscribed Göbekli Tepe on the World Heritage List in 2018, dates the construction to between 9,600 and 8,200 BCE on the basis of the radiocarbon analyses conducted during the German Archaeological Institute’s excavations.

The pillars range in height from approximately 3 metres for the smaller examples to 5.5 metres for the largest, and they weigh between 10 and 50 tonnes each. The pillars are carved from limestone quarried from outcrops within several hundred metres of the site, shaped using only chipped stone tools, and set vertically into sockets cut directly into the bedrock. The largest known pillar at the site, still partially embedded in the limestone bedrock of a nearby unfinished quarry, would have stood 7 metres tall and weighed approximately 50 tonnes if the builders had ever extracted it.

The surfaces of the pillars carry detailed relief carvings of wild animals. Foxes appear on multiple pillars. So do lions, boars, gazelles, vultures, scorpions, and snakes. Some pillars carry abstract symbols whose meaning has not been determined. Several pillars carry stylised representations of human arms and hands carved along their sides and fronts, suggesting that the pillars themselves were understood as anthropomorphic figures rather than as architectural elements. A few carry images of human heads alongside the animals, including one widely discussed example showing a human head in the wings of a vulture.

The site was, on the available evidence, used for somewhere between one and two thousand years, and then deliberately buried. The Pre-Pottery Neolithic builders filled in their own enclosures with stone, debris, and animal bones, raising the level of the surrounding ground until the pillars were entirely concealed. The hilltop was then abandoned. The reasons for the burial are not understood. The preservation of the site by its own builders is the reason any of it has survived for archaeologists to find.

Why it should not exist

The standard model of how human civilisation developed, established across roughly a century of archaeological work between the late nineteenth century and the late twentieth, held that monumental architecture was a product of agricultural societies. The reasoning was straightforward. Building enclosures with limestone pillars weighing tens of tonnes requires sustained, coordinated labour by large numbers of people over extended periods. Sustained coordinated labour at that scale requires a reliable food supply that does not depend on each individual hunting and gathering daily. A reliable food supply at that scale requires agriculture. Therefore, the model held, monumental architecture appears only after agriculture, and agriculture appears only in settled communities, and settled communities appear only after the Neolithic Revolution.

The earliest dated layers at Göbekli Tepe were laid down in approximately 9,600 BCE, which is approximately 11,600 years ago. That is several centuries before the earliest archaeological evidence of agriculture anywhere on Earth. The excavated layers contain no domesticated plants. They contain no domesticated animals. As Schmidt set out in his 2000 paper on the first five years of excavation, published in the journal Paléorient, the animal bones recovered from the site, in quantities reaching the tens of thousands, are all wild species: gazelles, wild boars, wild aurochs, wild sheep, and various deer. The cereal grains recovered are wild varieties of einkorn wheat and barley. The Göbekli Tepe builders were, by the unambiguous evidence of their refuse, hunter-gatherers.

They were also building one of the largest monumental complexes anywhere in the world for the next four thousand years.

The site predates the construction of Stonehenge by approximately 6,000 years. It predates the construction of the Egyptian pyramids by approximately 7,000 years. It predates writing by approximately 6,000 years. It predates the wheel by approximately 6,000 years. It predates pottery in the region by approximately 1,500 years. It was built by people who had not yet domesticated any plant or animal, did not yet live in permanent settlements year-round, did not have any form of metal, and were using exclusively chipped stone tools of the kind found at hunter-gatherer sites elsewhere in Eurasia and the Near East at the same date.

What it implies about agriculture

The most consequential implication of Göbekli Tepe, in the published interpretations of Schmidt and the subsequent excavation teams, is not the bare temporal fact that monumental architecture preceded agriculture. It is the suggestion that the conventional causal direction may be reversed.

The standard model held that agriculture created the food surpluses that allowed complex society, which in turn allowed monumental ritual architecture. Göbekli Tepe inverts the sequence. The site appears to have been a regional gathering place to which hunter-gatherer groups travelled from significant distances, possibly for ritual purposes connected with the wild animal imagery on the pillars. Sustaining such gatherings, on the scale the construction work would have required, would have placed substantial pressure on the wild food resources of the surrounding landscape. The earliest archaeological evidence for the domestication of einkorn wheat, set out by Heun and colleagues in Science in 1997, comes from a region called Karaca Dağ, located approximately 30 kilometres from Göbekli Tepe.

The inference some archaeologists have drawn from this geographic and temporal proximity is that the demands of sustaining the Göbekli Tepe gatherings may have driven the experiments in selective cultivation that produced the first domesticated cereals. On that interpretation, agriculture is the consequence of the ritual gathering rather than its prerequisite. Humans did not invent farming and then build temples. They built temples and then invented farming to keep the gatherings fed.

The interpretation is contested. Other archaeologists have argued that the connection is correlational rather than causal, that domestication may have been underway elsewhere in the Fertile Crescent independently, and that Göbekli Tepe’s relationship to early agriculture is one of contemporaneity rather than causation. A 2020 paper in the Cambridge Archaeological Journal by Gil Haklay and Avi Gopher used computer modelling to argue that the three main enclosures at Göbekli Tepe were planned as a single geometric whole rather than built piecemeal over generations, which adds a further layer of complication to the model of how the work was coordinated. The dispute remains live in the literature. What is not contested is the temporal sequence itself. The monumental construction came first. The agricultural revolution followed it.

What has not yet been found

The excavation Schmidt began in 1995 has been continuous for thirty years. Approximately twenty enclosures have been investigated in some detail. The ground-penetrating radar surveys that have been conducted across the rest of the hilltop indicate that the buried complex is substantially larger than what has been excavated so far. The current estimates suggest that fewer than 10 per cent of the site’s structures have been uncovered, and that the total number of T-shaped pillars at the site may eventually exceed 200.

The Yıldız family’s ploughs have been working ground that lies, at most, two or three metres above limestone pillars 11,500 years old. The 1963 survey team walked across the same hilltop and saw the same stones the family had been turning up. They did not recognise them. The pillars Schmidt identified in 1994 had been visible at the surface of the hill for at least three decades, and probably much longer, before any archaeologist looked at them with the right eye.

The site that has now produced the most consequential revision of the established model of how human civilisation developed was, until thirty years ago, an unploughed corner of farmland that local people had been reporting to the museum in the nearest city for years, and that the international archaeological community had walked past.

What else is buried in the rest of the hilltop has not been excavated yet.

What else is buried in other hilltops that the surveys have walked past, or never visited, is a question the published literature does not address.

The post Stonehenge is widely known as one of the oldest monumental stone structures in the world, but hunter-gatherer societies in southeastern Turkey built circles of T-shaped limestone pillars 6,000 years earlier, weighing up to 50 tonnes each and predating the human invention of agriculture by approximately 4,000 years appeared first on Space Daily.

Drifting through the Milky Way may be billions, and on some estimates trillions, of rogue planets: worlds bound to no star, some flung out of the systems where they formed, others perhaps never attached to a star at all. They are also called free-floating, nomad, or starless planets. The reason the figure spans such a wide range is simple. We have confirmed very few of them, and counting things you cannot see is hard.

Microlensing searches have produced fewer than ten likely low-mass free-floating planet detections, according to a survey of the field by IEEE Spectrum. Direct-imaging surveys of young star-forming regions have turned up many more planetary-mass candidates, though their formation history is harder to classify. Almost everything beyond that handful is inference and modelling, and the gap between a small number of firm detections and a projected population in the billions is the real state of play.

How you find a planet that gives off no light

A planet with no sun emits almost nothing a telescope can catch directly. The main way these objects are found instead is gravitational microlensing, which does not see the planet at all. It sees what the planet’s gravity does to the light of a star behind it.

When a rogue planet passes almost exactly between Earth and a distant background star, its gravity bends and briefly magnifies that star’s light. The star appears to brighten, then fade, and for a planet-sized lens the whole event is short. The first credible detection of an Earth-mass candidate came this way: an event designated OGLE-2016-BLG-1928, reported in 2020 by Przemek Mróz of the University of Warsaw and colleagues, with a brightening timescale of about 41.5 minutes, the shortest then recorded and only caught because the data were taken at high cadence. The object was estimated at roughly 0.3 Earth masses. It remains a candidate rather than a confirmed rogue, because the data cannot rule out that it sits very far from a host star rather than being truly unbound.

The strength of the method is that it can find low-mass objects nothing else can reach. The weakness is built into it. Each event happens once and never repeats, which makes the mass of any single object hard to pin down, and makes the whole population something you reconstruct statistically rather than observe one by one.

Where the large numbers come from

The headline estimates rest mainly on this statistical reconstruction. A 2023 analysis of a nine-year survey by the Microlensing Observations in Astrophysics collaboration, led by Takahiro Sumi at Osaka University, concluded that free-floating planets are far more common than earlier work assumed, and supported the idea that the galaxy holds more such objects than it has stars.

It is worth being clear about what that is. It is not a tally. It is an extrapolation from a small number of brief lensing events, folded through models of how often such events should occur. One NASA-backed estimate puts the Milky Way’s rogue planets at roughly twenty per star, or trillions of worlds in total. Other modelling work lands much lower, closer to one free-floating planet per star over the mass range considered. The honest conclusion is not that anyone has counted them. It is that the population could be enormous, and the uncertainty is still enormous too.

Where they come from, which is also unsettled

There are two broad accounts of how a planet ends up with no star, and they are not exclusive.

One is ejection. A planet forms in orbit around a star, then gets thrown out, usually by a gravitational shove from a larger planet or a passing star in a crowded cluster. Simulations of dense star-forming regions, such as work modelling the Orion Trapezium cluster at Leiden, produce rogue planets this way in large numbers.

The other is that some of these objects never had a star. They may have formed directly from collapsing gas, the way stars do, but with too little mass to ignite. The International Astronomical Union has suggested the term sub-brown dwarf for objects formed like stars but below the mass needed for fusion, which blurs the line between a small failed star and a large free-floating planet.

A recent wrinkle came from the James Webb Space Telescope. Observing the Orion Nebula, Samuel Pearson and Mark McCaughrean reported around 540 planetary-mass candidates, of which about nine per cent appeared to be in wide pairs, which they nicknamed JuMBOs, for Jupiter-mass binary objects. Pairs are awkward for the ejection story, since it is hard to throw two objects out of a system together and keep them bound to each other. The result is strange enough that it remains contested, with later analyses questioning how many of the pairs hold up and whether the objects should be called planets at all.

What the next survey is built to settle

The instrument expected to move this from extrapolation toward a census is NASA’s Nancy Grace Roman Space Telescope, now targeted for launch no earlier than September 2026, with a commitment to launch no later than May 2027. Roman will run a dedicated microlensing survey from space, staring at a strip of sky toward the galactic centre for months at a time, above the atmospheric blurring that limits ground-based work.

The expectations have grown as the modelling has improved. An earlier estimate put Roman’s likely haul at around 50 Earth-mass rogue planets. A 2023 study led by Naoki Koshimoto at Osaka University raised that to roughly 400. Japan’s PRIME telescope in South Africa is intended to make simultaneous ground-based observations, which would help measure masses rather than just count events. A 2025 paper by William DeRocco and colleagues, posted to the arXiv, works through how Roman’s data could be used to reconstruct the free-floating planet mass function, the distribution of how many of these worlds exist at each mass.

There is also the prospect of pairing Roman with the European Space Agency’s Euclid, already in orbit. A joint survey, according to BBC Science Focus, could turn up more than 100 rogue planets in its first year.

What to watch

The useful thing about Roman’s survey is that it has a clear way to be wrong. The models predict hundreds of detections. If Roman finds far fewer, or none, the population estimates and the detection methods both come back under review.

For now the honest position is that rogue planets are real, that a handful have been firmly detected, and that whether they number in the billions or the trillions, and how they mostly form, are open questions a single mission is now built to narrow. The figure to watch is not the trillion. It is the first few hundred, and whether they show up.

The post Drifting through the Milky Way may be billions — perhaps even trillions — of rogue planets: worlds with no sun of their own, some flung from the systems where they formed, now wandering the galaxy in darkness. appeared first on Space Daily.

In May 2026, Microsoft began cancelling most internal Claude Code licences for its Experiences + Devices division, the group responsible for Windows, Microsoft 365, Outlook, Teams, and Surface. Engineers were instructed to transition their workflows to GitHub Copilot CLI by 30 June, which is also the last day of Microsoft’s fiscal year.

The instruction came from Rajesh Jha, Microsoft’s Executive Vice President for the division. Microsoft had opened Claude Code access to Experiences + Devices in December 2025, inviting thousands of developers, programme managers, and designers to use it in real engineering workflows. According to reporting by The Verge, the tool became popular quickly, popular enough that it started displacing GitHub Copilot CLI in daily use. That created a problem that was both financial and strategic: Microsoft sells Copilot to the rest of the world and cannot credibly do that while its own engineers migrate away from it at scale.

Jha’s internal message to staff, quoted by The Verge, was careful not to frame the decision as a reversal. “Claude Code was an important part of that learning,” he wrote. “At the same time, Copilot CLI has given us something especially important: a product we can help shape directly with GitHub for Microsoft’s repos, workflows, security expectations, and engineering needs.” The official framing is toolchain unification. The timing, the fiscal year boundary, and the cost pressure in the wider market suggest cost reduction was also a factor, though Microsoft has not said so directly.

The Uber figure, and what it does and does not mean

The Microsoft decision did not happen in isolation. It followed a disclosure from Uber that has been circulating in enterprise technology circles since April: Uber’s CTO for Mobility and Delivery, Praveen Neppalli Naga, confirmed to The Information that the company had exhausted its entire 2026 AI coding tools budget by April, four months into the fiscal year.

The mechanism was not a single large contract. It was adoption velocity. Uber rolled out Claude Code and Cursor to its engineering organisation in December 2025. The company introduced an internal leaderboard ranking teams by total AI tool usage volume, which accelerated uptake of both tools sharply. By March, 84 per cent of Uber’s roughly 5,000 engineers were classified as agentic coding users. Monthly per-engineer costs reached between $500 and $2,000 for heavy users, against an average of $150 to $250 across the organisation. At that rate of consumption, a full-year budget evaporated before summer.

COO Andrew Macdonald, speaking on the Rapid Response podcast and reported by Fortune on 26 May, gave the most direct public account of the resulting discomfort. “That link is not there yet,” he said, referring to the connection between rising AI spend and consumer feature output. “Maybe implicitly there’s more that is getting shipped, but it’s very hard to draw a line between one of those stats and ‘Okay now we’re actually producing like 25% more useful consumer features.'” Uber had told engineers that AI tool usage was good, incentivised it with a visibility mechanism, and then discovered that good incentives produce exactly the behaviour they reward, regardless of whether the underlying ROI case has been established.

The $3.4 billion figure that has appeared in several secondary reports refers to Uber’s total research and development budget for 2025, not its AI coding tools budget specifically. The figure that was exhausted was the AI tools allocation within that envelope. The distinction matters: the Uber situation is a story about AI tool spend outpacing its own budget line, not about Uber spending $3.4 billion on Claude Code.

Why token-based billing changes the forecasting problem

Both the Microsoft and Uber cases reflect a structural difficulty that a number of enterprises are encountering for the first time. Traditional software licences are predictable: a fixed number of seats at a fixed price. Agentic AI tools priced on token consumption are not. Usage scales with the ambition of the task. An engineer running a multi-step refactor across a large codebase generates vastly more tokens than one asking for a function suggestion. When agentic workflows become standard and developers run tasks in parallel, the per-user cost envelope widens in ways that quarterly budget models built in late 2025 could not easily have anticipated.

Anthropic moved Claude Code from flat-fee pricing to usage-based billing for autonomous agents earlier this year, a shift that reflects the same economic reality from the other side: agents use far more compute per task than standard chat interactions, and flat fees priced for chat do not sustain the infrastructure cost of agentic use at enterprise scale. Fortune cited a Gartner projection in its 26 May report that while per-token inference costs will fall roughly 90 per cent by 2030, enterprise AI bills will not fall proportionally because agentic workflows require far more tokens per task, and because AI providers will not fully pass cost reductions through to customers.

The result is a market where usage-based pricing is rational for vendors, adoption incentives inside organisations have a documented tendency to outrun the budgets attached to them, and the tools are genuinely useful enough that restricting access after a blowout feels punitive to the engineers affected.

Where Opus 4.8 enters the picture

Anthropic launched Claude Opus 4.8 on 28 May 2026, three days after Fortune published the Uber story and roughly two weeks into the Microsoft transition. The timing is coincidental rather than causal, but the proximity is not irrelevant to how the launch has been received.

The release made Opus 4.8 available immediately via the Claude API, Amazon Bedrock, Google Cloud Vertex AI, and GitHub Copilot. That last integration is directly relevant to the Microsoft story: engineers in Experiences + Devices who are being moved from Claude Code to GitHub Copilot CLI now have access to Opus 4.8 through the tool they are being directed toward. The model is available to Copilot Pro+, Business, and Enterprise users, with a 15-times premium request multiplier applied until Copilot’s usage-based billing launches on 1 June.

On pricing, Anthropic held standard API rates flat at $5 per million input tokens and $25 per million output tokens, matching Opus 4.7. The new Fast mode, available to organisations in research preview via their account manager, is priced at $10/$50 per million, roughly three times cheaper than the Fast mode pricing on Opus 4.7. Anthropic described the release as a “modest but tangible improvement” over Opus 4.7, with benchmark gains in agentic coding, reasoning, and knowledge work. The most specific capability claim in Anthropic’s announcement is that Opus 4.8 is approximately four times less likely than Opus 4.7 to let flaws in code it has written pass without flagging them, a reliability characteristic that matters specifically in agentic workflows where the model is reviewing its own output.

It would be a misreading to treat the flat pricing as a direct response to the Uber and Microsoft situations. Anthropic has held Opus pricing steady across several recent releases. But the context in which enterprise buyers are evaluating Opus 4.8 is one where the cost structure of agentic Claude deployments has become a live CFO-level concern at two of the most prominent adopters on record. The launch landed into that conversation whether Anthropic intended it to or not.

What the Microsoft move actually settles, and what it does not

The clearest reading of the Microsoft decision is that it is about platform control as much as cost. GitHub has deep integration with code review, pull requests, CI pipelines, and repository metadata. Microsoft owns GitHub. Copilot CLI built on top of that integration has structural advantages for internal Microsoft workflows that Claude Code, sitting outside that ecosystem, cannot easily replicate. Jha’s reference to Microsoft’s “repos, workflows, security expectations, and engineering needs” is an accurate description of why a vertically integrated tool wins on governance grounds even when it loses on developer preference.

What the decision does not settle is whether Claude Code is less capable than Copilot CLI for the engineering tasks Microsoft’s teams actually perform. The internal signals point the other way: engineers adopted Claude Code at a rate that required an active policy decision to reverse. Copilot CLI now inherits a comparison it did not ask for, and engineers who preferred Claude Code will be evaluating it against a tool they were using by choice.

Uber’s situation is different in kind. Uber is not pulling back on Claude Code for platform reasons. It ran out of budget because adoption worked precisely as intended, and because the ROI case for agentic coding spend has not yet been made in terms that connect to consumer product output. CTO Neppalli Naga described the company as “back to the drawing board” on AI budgeting. Uber has also said it plans to test OpenAI’s Codex alongside Claude Code as it explores broader agentic deployment, which suggests the budget problem is being treated as a forecasting failure rather than a signal to stop.

The broader pattern here is not that enterprise AI coding tools are failing. It is that organisations incentivised adoption without building the financial and governance infrastructure to manage what successful adoption looks like at scale. That is a solvable problem, and both Uber and Microsoft appear to be solving it, in different ways and for different reasons.

The primary sources for this piece are: The Verge’s reporting on the Microsoft internal decision, Fortune’s 26 May 2026 article by Jake Angelo on Uber’s COO comments, The Information’s earlier reporting on the Uber budget blowout cited by Fortune, and GitHub’s 28 May changelog entry on Opus 4.8 availability in Copilot.

The post Microsoft is canceling Claude Code licenses across its Experiences + Devices division by June 30, steering thousands of engineers toward GitHub Copilot, while Uber burned through its entire 2026 AI tools budget on Claude Code and Cursor in just four months — and Opus 4.8 launched into that exact crisis appeared first on Space Daily.

A typical fair-weather cumulus cloud, the kind that drifts across a summer sky looking like a fluffy white pillow, contains roughly half a million kilograms of water suspended in the air. Translated into more familiar units, that is about 1.1 million pounds, or 550 short tons (about 500 metric tonnes). Comparable weights include 100 adult African elephants, five adult blue whales, or rather more than a fully loaded Boeing 747. The cloud weighs all of this, and yet it does not fall. The reasons it does not fall combine three distinct physical mechanisms, working together rather than in order of importance.

The calculation is straightforward, and the United States Geological Survey’s Water Science School publishes the standard version of it. A typical cumulus cloud occupies roughly one cubic kilometre of atmosphere, or one billion cubic metres. Atmospheric scientists estimate the average liquid-water density of a cumulus cloud at about half a gram per cubic metre. Multiplying these two figures together gives 500 million grams, or 500,000 kilograms. The cloud is mostly empty air, with a small mass of water droplets distributed through a very large volume. The water is the heavy part. The air is the support.

Why a cloud floats: three contributing factors

The popular explanation for why clouds float treats them as a kind of mist of tiny droplets held aloft by air resistance against gravity. This is part of the story, but not the whole story. The full picture combines three physical effects, which together account for the cloud’s quiet suspension above the ground.

One factor, and the one most often missed in popular explanations, is buoyancy. The reason a cloud sits on top of the air below it is in some ways the same reason oil floats on water: the cloud-laden air is, on average, less dense than its surroundings, and the denser surrounding medium supports it from below. This sounds counterintuitive, because the cloud contains water and the air around it does not. The resolution is that water vapour, the gaseous form of water, is actually lighter than air. A water molecule (H₂O, molecular weight 18) is less massive than a nitrogen molecule (N₂, weight 28) or an oxygen molecule (O₂, weight 32), and a parcel of warm, moisture-laden air is therefore less dense than a parcel of cool, dry air at the same pressure. Once water has condensed into liquid droplets, the situation is more complicated, but the parent air parcel that produced the cloud is still typically warmer and less dense than the dry air around it.

A second factor is the active one. Cumulus clouds do not just hover; they are also held up by rising columns of warm air, called updrafts. According to NASA’s reference on convective cloud formation, cumulus clouds form when surface air warmed by the sun-heated ground rises, expands and cools as it ascends, and reaches the dew point at which water vapour begins to condense around airborne aerosol particles such as dust, sea salt or pollen. The condensation releases latent heat, which further warms the parcel and accelerates its rise. The result is a continuous updraft beneath and within the cloud, typically moving upward at metres per second. The water droplets are held aloft not by stillness but by motion: they are riding an active column of rising air.

The third factor is the one most often cited in popular accounts. The droplets themselves are extraordinarily small, with typical cumulus cloud droplets measuring roughly 20 micrometres in diameter, or 0.02 millimetres. A droplet of this size, falling through still air, has a terminal velocity of around 1 to 2 centimetres per second. Updrafts in even a modest cumulus cloud move air upward an order of magnitude faster than this. A droplet trying to fall at one centimetre per second through air rising at, say, two metres per second simply does not fall. It is suspended in a fluid whose net motion is upward, with only a small downward component contributed by its own weight. By comparison, a typical raindrop is about 2 millimetres across, a hundred times the diameter of a cloud droplet, and falls at roughly 9 metres per second, fast enough to fall through even strong updrafts.

What changes when it rains

The transition from a floating cloud to a raining one is a transition in droplet size. Cloud droplets grow primarily by collision and coalescence: small droplets, jostled by turbulence within the cloud, bump into and merge with their neighbours, gradually accumulating mass. A droplet that grows from 20 micrometres to 200 micrometres has increased its terminal velocity from about 1 centimetre per second to about 70 centimetres per second. A droplet that grows to 2 millimetres falls at 9 metres per second. At some point in this size progression, the falling speed exceeds the updraft speed, and the droplet begins to fall through the cloud rather than being carried along by it. The droplets that emerge from the bottom of the cloud as rain have grown roughly a thousand times in mass during their time within the cloud.

A thunderstorm cloud, technically a cumulonimbus rather than a cumulus, contains far more water than a fair-weather cumulus. A typical thunderstorm cloud can have a mass on the order of a million tons of water rather than five hundred tons, with a corresponding requirement for far more vigorous updrafts to keep it aloft. Updrafts in severe thunderstorms can exceed 30 metres per second, enough to hold hailstones the size of golf balls suspended for many minutes while they grow by accreting layer after layer of supercooled water. When the updraft finally weakens, all of that mass falls at once, which is why thunderstorms produce heavy precipitation in short bursts.

A fair-weather cumulus cloud, by contrast, is in a quiet equilibrium. It is roughly a kilometre across, contains roughly half a million kilograms of suspended water, and is held aloft by a combination of buoyancy from being warmer and less dense than its surroundings, an active updraft from the convection that produced it, and the very small terminal velocity of its constituent droplets. Take any of these three away and the cloud changes character: cool the parcel and it begins to sink; remove the updraft and it begins to settle and dissipate; let the droplets grow and it starts to rain. The cloud as it appears, suspended quietly above a summer field, is the product of three physical processes in balance.

The post A single cumulus cloud — the kind that looks like a fluffy white pillow drifting across a summer sky — typically contains hundreds of tons of water by weight, suspended in the air because the water droplets are small enough that air resistance keeps them aloft against the pull of gravity appeared first on Space Daily.

At the centre of most large galaxies sits a supermassive black hole. When that black hole is actively consuming surrounding material, it becomes what astronomers call an active galactic nucleus. The infalling matter forms a disc, heats to extreme temperatures, and generates powerful jets of plasma that fire outward from the poles at close to the speed of light. When those jets happen to point toward Earth, the object is classified as a blazar. The orientation changes nothing about the physics; it changes what we see. A jet aimed at us delivers an intensity of radiation and particle flux that a sideways-on jet does not, making blazars among the most energetically extreme objects observable in the sky.

This matters for the story of KM3-230213A, the neutrino detected at 01:16 UTC on 13 February 2023 by the KM3NeT/ARCA detector at the bottom of the Mediterranean, roughly 3,450 metres below the surface off the coast of Sicily. With an estimated energy of approximately 220 petaelectronvolts, it remains the most energetic neutrino ever recorded. The previous record, from IceCube’s dataset, sat near 10 PeV. The gap between them is roughly a factor of twenty.

Where that neutrino came from has not been settled. Several candidate explanations have been put forward over the past year. A paper published in the Physical Review Letters in February 2026 argues that a population of blazars is a plausible origin. It is, as the authors are explicit in stating, one hypothesis among several. It has not been confirmed.

How neutrinos are produced in blazars

Inside a blazar jet, protons can be accelerated to extreme energies. When those protons interact with photons or other matter inside the jet, they produce pions, and those pions decay into neutrinos and gamma rays. This mechanism, broadly called hadronic production, is why high-energy neutrino detections and high-energy gamma-ray observations are connected: both are products of the same underlying process.

The connection also gives the hypothesis a testable constraint. Any model that invokes blazars as neutrino sources must not produce more gamma rays than have actually been observed. The extragalactic gamma-ray background has been measured carefully by the Fermi Gamma-ray Space Telescope, and a proposed blazar population cannot exceed it. This is one of the key tests the Bendahman paper applies.

The team used an open-source modelling tool called AM3 to simulate a realistic population of blazars, fixing parameters like magnetic field strength and emission region size to values established by independent observations. They then varied two quantities: the baryonic loading, which governs how much energy is carried by protons relative to electrons and therefore how many neutrinos can be produced; and the proton spectral index, which determines how the energy is distributed across the proton population. For each combination of these parameters, they calculated the expected diffuse neutrino flux and the corresponding gamma-ray output, then compared both against actual measurements from KM3NeT, IceCube, and Fermi.

They found a region of parameter space in which the blazar population could account for an event like KM3-230213A while remaining consistent with the gamma-ray constraints. The result positions blazars as physically viable. It does not identify a specific blazar source for this specific neutrino. The paper’s conclusion is that the scenario is plausible and merits further investigation, not that the question has been answered.

Why the absence of a counterpart complicates things

When a high-energy neutrino is detected, the standard follow-up procedure involves searching for an electromagnetic counterpart, a signal in radio, optical, X-ray, or gamma-ray light arriving from the same direction at approximately the same time. For KM3-230213A, no such counterpart was found. The KM3NeT collaboration conducted searches for correlations with known Galactic and extragalactic sources in the direction of the event (right ascension 94.3 degrees, declination minus 7.8 degrees) and found nothing significant.

The absence of a counterpart rules out some source scenarios more cleanly than others. A single dramatic event, a flare or an outburst from one identified object, would generally be expected to produce an accompanying electromagnetic signal. Its absence is one reason Bendahman and colleagues lean toward a diffuse origin: if the neutrino comes not from one spectacular burst but from the accumulated flux of many blazars integrated across large distances, there may be no single object to point to and no associated flare to find.

As Bendahman noted in the EurekAlert press release accompanying the paper, this reasoning does not completely exclude a point-like source. It does shift the prior toward a diffuse population explanation, which the blazar model provides.

The IceCube constraint and what it requires of any model

The IceCube Neutrino Observatory at the South Pole has been collecting data since 2010 with a larger effective detection volume than KM3NeT had at the time of the event. It has not observed any neutrino comparable to KM3-230213A. This non-detection is not a minor footnote: it imposes a real constraint on the expected rate of ultra-high-energy neutrino events, and any proposed source population must be consistent with it.

The tension between KM3NeT’s detection and IceCube’s non-detection has been estimated at between two and three-and-a-half sigma across several analyses, depending on assumptions about the source spectrum and angular region. That sits below the conventional threshold for claiming a significant discrepancy, but it is not easily waved away either.

The Bendahman paper addresses this directly. Their blazar population model is tested not just against the KM3NeT observation but against IceCube’s upper limits as well. They find a scenario in which blazars can produce a neutrino flux consistent with the KM3NeT detection while the IceCube non-detection remains statistically unremarkable. The model threads the needle, but only within a specific parameter range, and only as a statistical argument about expected rates, not as a demonstrated resolution of the IceCube tension.

What other explanations are on the table

The blazar population hypothesis is the most recent well-developed proposal from the KM3NeT collaboration itself, but it is not the only one circulating in the literature.

The cosmogenic neutrino scenario holds that KM3-230213A was produced not at an astrophysical source but in transit, when an ultra-high-energy cosmic ray collided with a photon from the cosmic microwave background. This process, expected to produce neutrinos in a broadly similar energy range, was analysed in a companion paper to the original Nature publication. The cosmogenic explanation has the advantage of not requiring a specific identified source, but the IceCube non-detection makes any steady, isotropic source harder to sustain without careful tuning.

Separate analyses have examined whether specific known objects could be the source. One paper by researchers at Peking University and Chongqing University investigated associations with gamma-ray bursts, searching a broader region around the event’s coordinates while allowing for possible Lorentz invariance violations that might have delayed the neutrino relative to any accompanying photons. No definitive association was found. Another paper pointed to a specific blazar, PKS 0605-085, as a candidate point source, based on its proximity to the reconstructed direction. The angular uncertainty of KM3-230213A is currently around 1.5 degrees, which leaves a sizeable search cone, and PKS 0605-085 has not been confirmed as the source.

A paper published in Physical Review Letters in March 2026, by Vedran Brdar and Dibya S. Chattopadhyay, takes a different approach entirely, focusing not on where the neutrino came from but on what may have happened during its journey. Their proposal involves sterile neutrinos, hypothetical particles that do not interact via the standard weak force, oscillating into active neutrinos over the 147-kilometre path through rock and seawater to the KM3NeT detector. The same transformation would be far less likely over the 14-kilometre path to IceCube from the same sky position, potentially explaining the discrepancy between detectors. This scenario requires physics beyond the Standard Model and remains speculative. The authors describe it as a possible resolution, not a demonstrated one.

A more exotic proposal, published separately in Physical Review Letters, suggested the event could have originated in the final evaporation of a primordial black hole. This hypothesis is not supported by independent evidence and has not been taken up broadly in the follow-up literature.

What the next data should resolve

KM3NeT/ARCA was operating with 21 detection strings at the time of the event, roughly ten per cent of its planned final configuration. Construction has continued since. The completed detector will cover approximately one cubic kilometre of deep water with around 200,000 photomultiplier tubes, and an online alert system is being developed to notify other observatories within seconds of a candidate high-energy event, enabling the kind of rapid multi-wavelength follow-up that might finally attach a counterpart to the next event of this kind.

The collaboration also expects a positioning system upgrade to tighten the directional reconstruction from the current 1.5 degrees to a target of around 0.1 degrees. That improvement, applied retroactively to KM3-230213A as well as to future detections, would substantially shrink the search cone and either implicate or exclude several of the current candidate sources.

The post Supermassive black holes are pointing jets of plasma directly at Earth — and a population of them may have produced the highest-energy neutrino ever recorded appeared first on Space Daily.

On a clear afternoon in Paris, the very top of the Eiffel Tower, 330 metres above the ground, is not exactly where it was at sunrise. It has moved a few centimetres to the west. By midday it has moved a few more centimetres to the south. By late afternoon it is several centimetres east of where it began. By the time the sun sets and the iron cools back to the surrounding air temperature, the top of the tower returns to its starting position. Over the course of a sunny day, the summit traces a slow, irregular curve roughly 15 centimetres in diameter, or just under six inches. The cause is not wind. Wind makes the tower sway and shudder on its own faster timescale. The cause of the slow daily circle is the sun.

According to the official Eiffel Tower website maintained by the Société d’Exploitation de la Tour Eiffel, “the sun only hits one of the 4 sides of the Tower creating an imbalance with the other 3 sides, that remain stable, thus causing the Eiffel Tower to lean. In this way, the sun’s movement over the course of a clear day can cause the top of the Tower to move in a more or less circular curve measuring approximately 15 centimetres in diameter.”

The physics, in three lines of arithmetic

The mechanism is the simplest thermal physics there is. Solids expand when they get warmer, because the atoms in them vibrate more vigorously and on average sit slightly further apart. The amount of expansion is described by the material’s coefficient of linear thermal expansion. According to a 2025 explainer in The Conversation by architecture professor Federico de Isidro Gordejuela, the puddled iron and steel components used in the Eiffel Tower have a coefficient of approximately 12 × 10⁻⁶ per degree Celsius. That figure means a one-metre iron bar grows by 12 micrometres for each degree Celsius of warming, which is roughly the width of a human hair.

The Eiffel Tower is not a one-metre bar. It is 330 metres tall after the installation of a new digital radio antenna in March 2022, which added 6 metres to the previous 324-metre height. The top of the original 1889 iron lattice structure, before any antennas, sits at 300 metres. Multiply 300 metres by 12 micrometres per metre per degree, and you get 3.6 millimetres of expansion per degree Celsius along the tower’s vertical axis. A 40-degree temperature change between a cold Paris winter and a sun-heated summer surface gives 14 centimetres of vertical expansion. The Eiffel Tower’s seasonal height range, as monitored by engineers with continuous strain-gauge readings, is between 12 and 15 centimetres, comfortably within what the simple calculation predicts.

Why the tower leans

The seasonal vertical expansion is uniform across the whole tower, because cold winters and hot summers heat all four faces about equally. The daily circular drift at the top is different. On a sunny day, the sun is in the east in the morning, the south at noon, the west in the afternoon. Each of these positions illuminates a different face of the tower’s four-sided lattice, while the other three faces remain in shade. The illuminated face heats up several degrees above the shaded faces, and that face expands more than the others. The result is that the tower bends, a few millimetres per metre of height, away from the sun.

Because the sun moves across the sky over the course of the day, the warmest face changes. The lean changes with it. In the morning, when the east face is illuminated, the top of the tower leans west. By noon, with the south face illuminated, the top leans north. In the afternoon, with the west face illuminated, the top leans east. After sunset, with no differential heating, the top returns to its starting position above the centre of the base. The trace of the top’s position over a full day is a roughly circular curve, with the irregularities reflecting variations in cloud cover and wind. The amplitude of the daily drift is about 7 centimetres in each horizontal direction, for a circle approximately 15 centimetres in diameter.

What Gustave Eiffel knew

The behaviour is not a surprise to the engineers who built the tower. Gustave Eiffel and his team, including the chief engineers Maurice Koechlin and Émile Nouguier, were well aware of thermal expansion when they designed the structure for the 1889 Exposition Universelle. Late-nineteenth-century iron-and-steel construction was already routine for railway viaducts and large bridges, and any competent engineer of the period understood that 300-metre iron structures would change shape with temperature. The tower’s design includes the lattice geometry and riveted joints that allow thermal movement to be distributed across thousands of small connections rather than concentrated in a few stressed points. The seasonal 15-centimetre rise and the daily lean are deliberately accommodated by the structure, not problems to be solved.

The same physics governs nearly every large engineered structure on Earth. The Garabit Viaduct, also designed by Eiffel and completed in 1884, is 565 metres long; the Forth Bridge in Scotland is 2.5 kilometres long; modern long-span bridges and tall buildings are all engineered with expansion joints, sliding bearings, or flexible connections to accommodate thermal movement. Railway tracks have expansion gaps welded into them at calculated intervals. The Eiffel Tower is the most visible of these examples because it is tall, slender, and isolated against the Paris sky, with its movement directly observable to anyone with a precise enough measurement system.

What you would see if you could see it

The drift is too slow to perceive in real time. A 15-centimetre circle traced over the course of a 10-hour day means the top is moving at an average speed of about half a centimetre per hour, far below the threshold of human visual detection. The Eiffel Tower has been continuously instrumented for structural monitoring since 2021, with GPS sensors, accelerometers, inclinometers, and strain gauges tracking the spire’s inclination, the deformations of its four pillars, and ambient temperature and humidity across the structure. The data shows the tower’s response to temperature, sunlight, wind, and the cycle of seasons with millimetre precision.

From the perspective of a tourist standing at the base, the tower looks as still as any building. From the perspective of the iron itself, the tower is in continuous quiet motion. The sun warms one face, that face stretches, the tower leans. The sun moves, the next face warms, the lean shifts. The structure that has dominated the Paris skyline since 1889 is, on any clear day, also a 7,300-tonne sundial, with the position of its summit telling you, within a few centimetres, where the sun is.

The post On a sunny day, the top of the Eiffel Tower slowly drifts in a small circle about six inches wide — it isn’t the wind, it’s the sun, heating one side of the iron at a time and making the whole tower lean a little away from whichever side is warmest appeared first on Space Daily.

Sound in air travels at about 343 metres per second, the figure that fighter pilots cross when they break the sound barrier. Sound in seawater travels at about 1,500 metres per second. The ratio is roughly 4.3 to 1. The reason is straightforward physics: sound waves propagate through a medium by transferring vibrational energy between adjacent molecules, and the denser and stiffer the medium, the faster that transfer happens. Water is roughly 800 times denser than air, with much smaller intermolecular distances, and it transmits acoustic energy with much less loss per metre than air does. The same whale call that would fade to nothing within a kilometre in air can carry for hundreds of kilometres through the deep ocean.

According to NOAA’s Ocean Service reference on underwater sound, the speed of sound in seawater varies between roughly 1,450 and 1,540 metres per second depending on temperature, salinity and pressure. The first reasonably accurate measurement was made in 1826 on Lake Geneva by the Swiss physicist Jean-Daniel Colladon and the French mathematician Charles-François Sturm, who used an underwater bell, a flash of gunpowder for synchronisation, and two boats positioned ten miles apart. Their figure was within a few percent of the modern measurement, despite the relatively crude instruments available.

Why whale songs cross oceans

The speed of sound in water is only part of what makes long-distance whale communication possible. The other part is the way sound bends in the ocean. Sound speed in seawater depends on temperature and pressure, and these vary with depth. Near the surface, the water is warm and the sound speed is high. Below the warm surface layer, the temperature drops sharply through the thermocline, and the sound speed drops with it. Below the thermocline, the temperature is nearly constant but pressure continues to rise, and the sound speed begins to climb again.

The result is a minimum in the sound speed at roughly 1,000 metres depth in mid-latitudes, with faster speeds both above and below. According to the Discovery of Sound in the Sea reference, an academic resource maintained by oceanographers at the University of Rhode Island and elsewhere, this minimum creates a natural waveguide. A sound wave emitted at or near 1,000 metres depth that strays upward into faster water gets bent back down. A sound wave that strays downward into faster water gets bent back up. The wave is trapped, channelled by the gradients of temperature and pressure, and propagates horizontally with almost no energy loss to absorption.

This is the SOFAR channel — Sound Fixing and Ranging — discovered toward the end of the Second World War by Maurice Ewing and Joseph Worzel at the Woods Hole Oceanographic Institution. A low-frequency sound emitted into the SOFAR channel can travel thousands of kilometres before its energy dissipates. Blue whales, fin whales and other large baleen whales produce calls below 20 hertz, on the edge of human hearing or below it, and these low-frequency calls couple efficiently into the SOFAR channel. Researchers led by Christopher Clark at Cornell University have tracked individual whales across entire ocean basins using the long-distance propagation that the SOFAR channel makes possible.

The “Jezebel Monster”

The Cold War history of underwater sound is where the popular framing of “sonar operators filtering out whale noise” comes from. Beginning in 1950, the US Navy developed a global underwater listening network called the Sound Surveillance System, or SOSUS, designed to detect Soviet submarines by their low-frequency acoustic signatures travelling through the SOFAR channel. According to the Discovery of Sound in the Sea account of SOSUS history, the system was very successful at detecting noisy diesel and nuclear Soviet submarines. It was also picking up sounds that no one initially knew how to classify.

One particularly persistent unknown source was labelled the “Jezebel Monster” by SOSUS analysts. The sounds were low-frequency, came from no known submarine, and could be heard from enormous distances. They turned out to be the calls of blue and fin whales, propagating through the SOFAR channel exactly as Soviet submarines would have been, and detected by SOSUS arrays exactly as Soviet submarines would have been. The whales were not being filtered out in the everyday sense of being a routine nuisance. They were being identified as a category of acoustic source that the Navy had not previously known to look for. Once identified, the calls became valuable rather than confusing: they confirmed that the SOFAR channel was working as predicted, and they later became the basis of the most comprehensive long-distance whale-tracking research ever conducted.

At the end of the Cold War, the Navy declassified the technical operation of SOSUS sufficiently to allow civilian researchers with security clearances to use the system. The result has been a generation of marine biology research that would have been impossible without the Cold War infrastructure. Whales communicating across thousands of kilometres of ocean are now routinely tracked by hydrophone arrays that were originally listening for Soviet ballistic-missile submarines. The same physics, the same channel, the same long-distance propagation.

How loud is loud, under the ocean

Whale calls below 20 hertz are produced at source levels of up to 188 decibels relative to one microPascal at one metre — a figure that does not directly map onto the decibel scale used for air-based sound, because the reference pressures are different, but that corresponds to one of the loudest sustained biological sounds on Earth. A blue whale’s call at source can be heard at a level above background ocean noise for hundreds of kilometres in unmodified ocean conditions. The 2012 paper by Denise Risch and colleagues in PLOS One documented the inverse case: humpback whales in the Stellwagen Bank National Marine Sanctuary off Cape Cod measurably reduced their singing during an Ocean Acoustic Waveguide Remote Sensing experiment 200 kilometres away. The whales could detect the experiment, and modify their behaviour in response, from a distance comparable to the route between London and Paris.

The modern problem is anthropogenic noise. Shipping traffic, oil-and-gas exploration, military sonar, and underwater construction now produce so much low-frequency noise that the effective range over which whales can communicate with each other has shrunk substantially. Christopher Clark has estimated that the acoustic environment in which whales operate has been reduced to a small fraction of what it was a century ago, because human noise occupies the same frequency band as whale calls. The physics of underwater sound has not changed. What has changed is the level of competing noise sharing the channel. The whales are still singing. The ocean is just no longer quiet enough for the songs to travel as far as they once did.

The post Sound travels about four times faster underwater than it does through air — which is why whale songs can travel hundreds of miles across the ocean, and why early submarine sonar operators sometimes picked up the calls of distant whales communicating from hundreds of miles away appeared first on Space Daily.

In 1096, scholars were already teaching in the streets of Oxford, in the kingdom of William II, the son of William the Conqueror. England had been under Norman rule for thirty years, the Domesday Book had been compiled ten years earlier, and the First Crusade had just been declared by Pope Urban II at Clermont. Across the Atlantic, in the central highlands of what is now Mexico, the Aztec people would not arrive in the Valley of Mexico for another two hundred years. They would not see the omen of the eagle, the cactus and the snake that, according to tradition, told them where to build their capital, until 1325. By the time the first stones of Tenochtitlán were laid on a small island in Lake Texcoco, Oxford had been a centre of teaching for 229 years.

The comparison is one of those facts that everyone half-remembers as plausibly true and few people stop to verify. A 2024 Smithsonian Magazine piece by Meilan Solly documented the contrast in some detail. The dates check out. The cultural intuition about which institution feels older does not.

Oxford does not actually have a founding date

The 1096 date for Oxford is the earliest documented evidence of teaching activity in the town, not the formal foundation of the university. According to Oxford’s own history page, “there is no clear date of foundation but teaching existed at Oxford in some form in 1096.” The university grew rapidly from 1167, when Henry II banned English students from attending the University of Paris during his ongoing feud with Thomas Becket, sending a generation of English scholars home to study in Oxford instead. By 1188, the historian and royal clerk Gerald of Wales was giving public readings to assembled Oxford dons. By around 1190, the first known overseas student, Emo of Friesland, had arrived. By 1248, Henry III had issued the university a royal charter. By 1264, three original colleges — University, Balliol and Merton — were operating as residential institutions for students. The institution emerged over the course of a century and a half, not in any single founding moment.

The other claimants for “oldest university” are older still. The University of Bologna, generally regarded as the oldest in continuous operation, dates to 1088. The University of Paris emerged around 1150. These three institutions, together with a small handful of others, laid the foundations for the European university system that the rest of the world eventually adopted. Oxford is the second-oldest university in continuous operation in the world, and the oldest in the English-speaking world.

Tenochtitlán: the city founded on an omen

The Aztec people, who called themselves the Mexica, arrived in the Valley of Mexico in the early 14th century after a long migration from a homeland in the northwest of Mexico called Aztlán. According to Britannica’s account of the city, the founding of Tenochtitlán in 1325 followed a long pilgrimage during which the Mexica’s patron god Huitzilopochtli instructed them to settle wherever they saw an eagle perched on a prickly pear cactus, eating a snake. They saw the omen on a small island in Lake Texcoco, in the Valley of Mexico. The image is preserved on the modern Mexican flag.

The city the Mexica built on that island grew rapidly. Originally confined to two small islands, Tenochtitlán was extended through the construction of artificial islands, called chinampas, until the city covered more than five square miles of the lake. It was connected to the mainland by causeways, supplied with fresh water by aqueducts, and traversed by a network of canals. Estimates of its peak population vary considerably. The Britannica entry gives a figure of about 400,000 people in 1519, “the largest residential concentration in Mesoamerican history” and larger than any contemporary European city, including Paris, London, or Rome. Other scholarly sources give more conservative figures in the range of 200,000 to 300,000. By any measure, it was among the most populous cities in the world at the time of Spanish contact.

The Aztec Empire as a political entity dates not from the founding of the city but from the formation of the Triple Alliance in 1428, when Tenochtitlán joined with the neighbouring cities of Texcoco and Tlacopán to dominate central Mexico. By that point, Oxford had been operating as a university for over three centuries, had survived multiple plague outbreaks, had produced Roger Bacon and John Wycliffe, and had been teaching the Latin classics to clerics, lawyers and merchants’ sons for so long that the institution was already considered ancient by its own English contemporaries.

What the comparison actually changes

The intuition that gets disrupted is not about Oxford. Most people are aware that the university is old. The intuition that gets disrupted is about the Aztecs. Aztec civilization tends to feel anchored in the distant past, in the same general mental category as Egyptian pharaohs, Mesopotamian temples, or ancient Greece. The actual historical position of Tenochtitlán is much closer to the modern present. The city was founded 229 years after Oxford began teaching, conquered by Cortés in 1521, and razed within the lifetime of people who were born in the same year as William Shakespeare’s grandparents. Aztec civilization, in the sense of Tenochtitlán-centered Mexica society, existed for less than two hundred years before its destruction. The empire proper, dating from the Triple Alliance, existed for less than a century.

Oxford has, by contrast, existed continuously through every European event from the Crusades through the Reformation, the Industrial Revolution, two world wars, and the digital age. The same town, the same general institution, the same broad teaching tradition has been operating without interruption for over 900 years. The Aztecs were latecomers to history. Oxford was already producing graduates when the first Mexica arrived in the Valley of Mexico, and was already a chartered university when the first stones of the Templo Mayor were laid. The mental image of which civilization belongs to the deeper past, in this comparison, is almost exactly reversed.

The post Oxford University is older than the Aztec Empire — teaching at Oxford began in 1096, while the Aztec capital Tenochtitlán was founded in 1325, which means Oxford was already more than two centuries old by the time the civilization that built one of the most sophisticated cities of the medieval world had even begun appeared first on Space Daily.

Parts of Canada are quietly short on gravity. The familiar explanation blames an ice sheet that pressed the crust down and then melted away thousands of years ago. That story is partly true, but satellite measurements suggest it explains less than half of the Hudson Bay gravity low, with much of the rest likely tied to deeper mantle structure and slow motion inside the solid Earth.

The region in question sits around Hudson Bay, a broad area where the local pull reads slightly below the global average. Geophysicists have known about the low since gravity surveys mapped it in the 1960s. The deficit is far too small to feel: you would not weigh noticeably less standing on the shore, and nothing floats. But it is real, it covers a large area, and for decades it had no settled cause.

What GRACE actually measured

The modern picture comes from GRACE, the Gravity Recovery and Climate Experiment, a pair of satellites flown jointly by NASA and the German Aerospace Center from 2002. The two spacecraft followed each other around the same orbit, and the tiny changes in the distance between them tracked variations in the gravity field below. NASA’s overview of the GRACE mission sets out the basic method.

What made GRACE useful for Hudson Bay was not a single snapshot but the trend over time. Some of the gravity field over Canada is changing year on year. Some of it is effectively static on human timescales. Separating those two parts is what made the difference.

The ice-sheet explanation, and its limit

The leading account is glacial isostatic adjustment, often shortened to GIA. The Laurentide Ice Sheet covered most of Canada at the last glacial maximum, in places several kilometres thick. Its weight pushed the crust down into the softer mantle beneath. When the ice melted, roughly ten thousand years ago, the crust began rising back, and it is still rising today, by something under a centimetre a year. Until that rebound finishes, the region holds a mass deficit, and less mass means a weaker local pull.

The trouble is the size of the gap. As early as 1992, a study in Geophysical Research Letters by Thomas James found that the best deglaciation models of the day could reproduce only 15 to 30 per cent of the observed Hudson Bay low.

The rest had to come from somewhere else.

The 2007 result

The clearest separation came from a 2007 paper in Science by Mark Tamisiea, Jerry Mitrovica and James Davis, GRACE Gravity Data Constrain Ancient Ice Geometries and Continental Dynamics over Laurentia. Using GRACE data from April 2002 to April 2006, the authors isolated the part of the gravity field changing in step with the ongoing rebound, and set it apart from the part that does not move on that timescale.

Two findings came out of that separation.

The first concerned the ice itself. The pattern of present-day uplift implied that the Laurentide complex was not one dome but two, sitting to the west and east of Hudson Bay, which matched one of the two competing reconstructions of the ice sheet’s shape.

The second concerned the gravity low. According to the paper, rebound models that match the measured uplift rates account for about 25 to 45 per cent of the static gravity anomaly. On the authors’ reading, most of the low is not a leftover from the ice age at all.

This is one study, and the exact split should be read as a result from this dataset and this set of models rather than a fixed constant. But it points the same way as the earlier work, and the direction of the answer has held.

What “the rest” is, and is not

Here the language needs care. The remainder is usually attributed to mantle convection, the slow movement of rock deep inside the Earth. The popular retelling often describes the mantle as a sea of molten magma. It is not. The mantle is mostly solid rock that deforms and flows over very long timescales, closer to extremely stiff putty than to liquid.

In the Hudson Bay case, the idea is that cold, dense material sinking in the mantle draws the surface down and pulls mass away, lowering the gravity field above it. The Tamisiea paper frames its result in those terms, treating the non-rebound portion as a constraint on the buoyancy of the deep continental root beneath Laurentia, some of the oldest crust on the planet.

What GRACE does not do is photograph that flow. The convection contribution is inferred as the part left over once the rebound signal is removed, so the strength of the inference depends on how well the rebound itself is modelled. That is the honest limit of the result.

What to watch

GRACE itself ended in 2017. Its successor, GRACE Follow-On, launched in 2018 and continues to track the same field, which means the rebound signal over Canada is still being measured rather than frozen at the figure of two decades ago. The longer the record, the more tightly the changing part can be told apart from the static part.

For now the Hudson Bay low sits at the meeting point of two very different timescales: an ice age that ended in the last ten thousand years, and a circulation in the mantle that runs over tens of millions. The satellites are good at separating the two. Putting a precise number on how much each one owes is still open.

The post Parts of Canada are quietly short on gravity. The standard story blames an ice sheet that pressed the crust down and vanished thousands of years ago, but satellites suggest that explains less than half of it. The rest comes from something churning far deeper in the mantle. appeared first on Space Daily.

The Richat Structure, the concentric ring formation in the Adrar plateau of Mauritania often called the Eye of the Sahara, was photographed from orbit by the crew of Gemini IV in June 1965, at a time when the leading hypothesis on the ground was that it had been formed by a meteor impact. The orbital photograph reached the public before the geological consensus did.

The structure is about 40 kilometres across, depending on whether you count the outer rings, with some descriptions stretching that to 50. It sits near the town of Ouadane in northern Mauritania, on a plateau dry and bare enough that the rings of differing rock types stand out against the surrounding desert with the kind of clarity that mostly only works from high altitude.

It does not work from ground level. A traveller walking through the Richat Structure would see ridges, slopes, and gullies, and would not necessarily recognise that the ground beneath them is part of a near-perfect circle the size of a medium European city.

What the Gemini IV crew actually saw

James McDivitt and Ed White launched aboard Gemini IV from Cape Kennedy on 3 June 1965. The mission ran four days. White’s twenty-minute spacewalk on the second orbit became the photograph history remembered. The crew also took a large number of Earth-observation images as the spacecraft passed over Africa, the Middle East, and the Americas, sometimes opportunistically, sometimes on the flight plan.

One of those images, taken on 4 June 1965, captured the Richat Structure from orbit, showing the whole formation at a scale that helped turn it into a global curiosity. NASA’s Earth Observatory notes that the image helped bring wider global attention to a feature French geographers had already named and partly described from aerial photographs in the 1930s and 1940s, but which had not been seen at this kind of scale before.

This is the part of the story that gets compressed in popular retellings. The Eye was not discovered from space. It had been on French colonial maps for decades. What space did was give the world a single image of the whole thing at once, in a way that no aerial survey at lower altitude had managed.

The impact crater hypothesis and why it held

For most of the period between the structure’s first scientific description in 1948 by the French geographer Jacques Richard-Molard and the studies of the 1960s and 1970s, two ideas competed. One was that the Richat was a deeply eroded dome of uplifted rock. The other was that it was an astrobleme, the geological signature of an ancient meteor impact.

The impact theory had real reasons behind it. The Richat looks like a crater. It has a raised outer perimeter and a sunken centre. A 1952 expedition led by the French naturalist Théodore Monod found three nearby smaller features that were genuine impact craters: Aouelloul, Temimichat-Ghallaman, and Tenoumer. If the Sahara had been hit, repeatedly, by smaller bodies, the case for a larger neighbour seemed reasonable.

What undid the impact hypothesis was the absence of the things impacts leave behind. There was no central peak. There was no clear evidence of shock metamorphism: no melted rock, no shocked quartz with the diagnostic deformation lamellae produced by hypervelocity impacts. Early reports of coesite, a mineral that forms under impact pressures, were later identified as misidentifications. The pattern of features pointed somewhere else.

The somewhere else was an uplifted geological dome shaped by igneous intrusions, hydrothermal alteration, collapse, and long erosion, rather than a single neat crater-making event. Different rock types erode at different rates. The harder igneous rocks form the high ridges. The softer sedimentary rocks form the lower troughs between them. The bullseye is differential erosion, not impact geometry.

What the recent work shows

The hypothesis has been refined considerably in the last two decades. A 2005 paper by Guillaume Matton and colleagues in Geology, titled “Resolving the Richat enigma,” proposed a model in which doming above an alkaline igneous complex was followed by hydrothermal karstification, with the centre of the structure dissolved and collapsed by hot fluids working through the carbonate rock. A 2014 paper by Matton and Michel Jébrak filled in the alkaline-hydrothermal complex picture in more detail.

More recent work by Abdeina and colleagues, published in 2024 in Lithos under the title “How old is the Eye of Africa?”, attempts to date the igneous phases of the Richat more precisely. The picture that has emerged is of an isolated Cretaceous alkaline complex with a multi-phase history of intrusion, doming, and erosion. The International Commission on Geoheritage recognised the Richat Structure as a site of global geoheritage significance in 2022.

The impact hypothesis has not been retained by anyone working on the structure for a long time.

What the story is really about

The interesting thing about the Eye of the Sahara, in our reading, is not the structure itself, which is well understood by the people who study it. It is the lag between what could be seen from orbit and what was understood on the ground.

Earth-observation imagery from crewed and uncrewed orbital platforms has, since the 1960s, produced a steady stream of features that look like one thing from above and turn out to be another when investigated at the surface. The Richat is the famous example. There are others. A circular feature in the desert is not, by itself, a crater. A bright spot on a coastline is not, by itself, an algal bloom. The image suggests a hypothesis. The hypothesis is then tested by people with hammers, drill cores, and a willingness to be wrong.

That is the working relationship between orbital remote sensing and field geology, and it is not new. The Gemini IV photograph of the Eye of the Sahara is one of the first popular illustrations of the pattern. It is also a reminder that an image, by itself, does not constitute an explanation. It constitutes a question, well framed, from an unusually good vantage point.

The latest high-resolution images of the Richat, captured by Landsat 8 and 9 in March 2026, appeared as NASA Earth Observatory’s Image of the Day on 16 April. The structure looked the same as it had from Gemini IV. What had changed, in the intervening sixty-one years, was the explanation underneath it.

The post The Eye of the Sahara is a giant bullseye in the Mauritanian desert, up to fifty kilometres across, and astronauts were already photographing it from orbit while geologists still believed it was a meteorite impact crater, long before anyone worked out it was something else entirely. appeared first on Space Daily.

The river runs through the Mayantuyacu region of central Peru, in the Province of Puerto Inca in Huánuco State, on the eastern flank of the Andes where the foothills give way to the Amazon basin. The Asháninka people who have lived along its banks for centuries call it Shanay-timpishka, which translates from their language as “boiled with the heat of the sun.” The river is approximately 9 kilometres long, up to 25 metres wide, and up to 6.1 metres deep at its hottest sections. It is a tributary of the Pachitea River, which itself feeds the Ucayali, which itself feeds the Amazon, which itself empties into the Atlantic Ocean approximately 5,000 kilometres to the east.

The river’s source is marked by a boulder shaped like the head of a snake.

The Asháninka believe the river is the work of Yacumama, the Mother of Waters, a giant serpent spirit who gives birth to hot and cold water from her body. The boulder at the headwaters is, in the local cosmology, Yacumama’s head. The shaman of Mayantuyacu, who has authority over access to the river, has historically permitted only a small number of outsiders to study it.

How hot it actually is

The river’s temperatures have been measured by a small number of scientific expeditions, beginning systematically in 2011. The hottest single temperature recorded in the river is 99.1 degrees Celsius (210.4 degrees Fahrenheit), measured in a hot spring at the river’s hottest pool. The hottest average river temperature recorded along the heated stretch is approximately 95 degrees Celsius. The water cools as it flows downstream, dropping from near-boiling at its hottest pool to ambient Amazon river temperature within a few kilometres of its emergence.

The 4-kilometre heated stretch is consistently hot enough to be lethal to anything that enters it. The geothermal scientist Andrés Ruzo, who has spent more than a decade studying the river, has described what he has personally witnessed happening to animals that fall in. The eyes go first. Eyes, he has noted, cook very quickly, turning a milky-white colour within seconds. The flesh follows. The animal, in the time it takes a human observer to register what is happening, is essentially being cooked alive in the water it has fallen into.

The local people have used the river’s water, harvested from the cooler downstream sections, for cooking, brewing tea, and cleaning, for at least as long as the oral history of the area records. The water itself, despite its temperature, is remarkably safe to drink once cooled. Chemical analysis has shown that the water is, by the standards of Amazon waterways, unusually free of the parasites and pathogens that contaminate most other local water sources. The Asháninka have long understood the river as a place of healing as well as danger.

Why no one believed it existed

The river was, until very recently, considered by the international scientific community to be a legend. Boiling rivers, the standard model held, exist only in association with active volcanoes. The geothermal heat that sustains them comes from magma chambers within a few kilometres of the surface. Yellowstone, Iceland, the geothermal features of New Zealand and Japan and Chile, all sit directly above active volcanic systems. The Peruvian Amazon does not.

The nearest active volcanic centre to Shanay-timpishka is more than 700 kilometres away. The interior of the Amazon basin is, geologically, one of the most tectonically quiet regions on Earth. There is no magma close to the surface. There is no obvious heat source. By every prior model of how geothermal water comes to be at the temperatures Shanay-timpishka achieves, the river should not exist where it does.

Ruzo, who grew up in Peru and trained as a geothermal scientist at Southern Methodist University in Texas, first heard about the river from his grandfather as a child. He spent the next twelve years assuming it was a legend. When he raised the possibility with his PhD colleagues during his work on a thermal map of Peru, the response was unanimous. Boiling rivers exist, his colleagues said, but they are always near volcanoes. There are no volcanoes in the central Peruvian Amazon. Therefore the boiling river does not exist.

It was Ruzo’s aunt, on a family visit to Peru, who corrected him. She had been to the river herself. She had swum in its cooler downstream sections. Her friend was married to the shaman who guarded it. In November 2011, Ruzo travelled to Mayantuyacu with his aunt as a guide, became the first geoscientist granted shamanic permission to study the river, and set out the scientific account that would eventually become his 2014 TED Talk and his 2016 book on the project. Until that point, the river had been known to local people and to a handful of curious travellers, but had not been the subject of any sustained scientific investigation. It had also been doubted by the Peruvian government, by international academic institutions, and by the fossil fuel industry, all of whom had reason to want to know whether the geothermal claim was real but had not committed resources to confirming it.

Related: In 1964, IBM risked its entire corporate empire on the System/360, a chaotic gamble to make all of its future machines compatible with the same software — and the architecture proved so robust that modern enterprise mainframes today are still running sections of binary code written more than sixty years ago

What is heating the water

The current best understanding of the river’s mechanism, set out in Ruzo’s published and conference work, is that it is heated by a deep hydrothermal circulation system rather than by any local volcanic source. Rainwater falling in the surrounding mountains, including possibly as far away as the Andes, percolates through fault systems into the deep crust. The water descends several kilometres into the bedrock, where the natural geothermal gradient heats it to temperatures approaching the boiling point. The heated water then rises back to the surface through other fault systems in the Mayantuyacu region, emerging at a small number of specific hot springs that feed the river. The heat is therefore not magmatic. It is the ordinary background heat of the Earth’s crust, concentrated by a fault geometry that allows water to descend, heat, and return to the surface in the same location.

The mechanism is not, in itself, unusual. Deep hydrothermal circulation is a well-documented process. What is unusual at Shanay-timpishka is the scale. The volume of water reaching the surface, and the temperature at which it emerges, exceeds anything else documented at a non-volcanic geothermal site on the planet. The river is, by current measurement, the largest documented non-volcanic geothermal feature on Earth.

The full scientific characterisation of the system remains incomplete. Ruzo’s geothermal analysis has been published in conference proceedings, in a 2016 TED Books volume, and in the National Geographic Society’s Young Explorer reports, but the detailed geochemistry and hydrology of the river’s source have not yet appeared in a peer-reviewed scientific journal. The work continues.

What it tells us about everything else

In 2021, a team of biologists from the University of Miami’s Jungle Biology lab, led by the graduate student Riley Fortier and his colleague Alyssa Kullberg, recognised that the Shanay-timpishka river offered something that no laboratory could replicate. The river produces a permanent thermal gradient along its course, with cool forest at one end and forest growing in soils heated by the river at the other. The rainfall, the humidity, the soil chemistry, and the species pool are all approximately the same at every point along the gradient. Only the temperature changes.

The site is, in effect, a natural warming experiment running across a 9-kilometre stretch of Amazon rainforest.

In November 2024, Fortier, Kullberg, Ruzo, and their colleagues published their first major findings in the peer-reviewed journal Global Change Biology. They had measured the canopy and understory temperatures along the river at thirteen monitoring stations over a full year. The annual average air temperature in the coolest forest plots was 24 to 25 degrees Celsius. In the hottest plots near the river, the annual average was 28 to 29 degrees Celsius, with maximum air temperatures approaching 45 degrees Celsius. The hotter plots, the team found, supported approximately 11 per cent fewer tree species per degree of warming. The composition of the surviving species shifted toward heat-tolerant pioneer trees. The forest, in the hottest sections, was visibly simpler, drier, and more sparsely vegetated than the cooler sections only a few hundred metres away.

As Fortier set out in coverage by Mongabay, the implication is direct. The temperature gradient at Shanay-timpishka is approximately what the broader Amazon basin is projected to experience over the next several decades under continued warming. The species lost from the hot end of the gradient are species that, on the published trajectory, may be lost across much wider areas of the Amazon if global temperatures continue to rise. The river that the Asháninka understood as the work of Yacumama, and that the Spanish conquistadors reported as part of the legend they assumed was a fantasy, has turned out to be one of the most useful natural laboratories on the planet for understanding what the tropical rainforests of the next century will look like.

What it means for the river itself

Mayantuyacu and the surrounding region face the same pressures as the rest of the western Amazon. Illegal logging, cattle ranching, road construction, and oil and gas exploration are gradually encroaching on the forest within which the river flows. The Boiling River Project, founded by Ruzo as a conservation nonprofit, has worked since 2016 to advocate for legal protection of the river and its surrounding watershed. The Peruvian government has not yet granted the site formal protected status. The shaman of Mayantuyacu and his successors continue to control direct access to the headwaters.

The river has flowed at its current temperatures for, by the best available estimates, at least several thousand years and possibly far longer. The Asháninka knowledge of the river’s behaviour, recorded in oral tradition rather than scientific literature, contains observations of its seasonal flow rates and temperature variations that the published scientific literature has not yet matched in detail. The local people, in this case, have been the long-term observers of a geothermal system that science is still in the process of catching up with.

What other rivers run hot in other parts of the world’s tropical rainforests, in places that international science has not yet visited and that local people have not yet been asked, is a question the published literature does not address.

The post Anything that falls into a four-kilometre stretch of a river in the central Peruvian Amazon dies within seconds, because the water reaches temperatures of up to 100 degrees Celsius, despite the river sitting more than 700 kilometres from the nearest active volcano and in a region of the planet with no known magmatic activity. appeared first on Space Daily.