Losing one variety of gut bacteria may lead to type 2 diabetes as people age.

Old mice have less Akkermansia muciniphila bacteria than young mice do, researchers report November 14 in Science Translational Medicine. That loss triggers inflammation, which eventually leads cells to ignore signals from the hormone insulin. Such disregard for insulin’s message to take in glucose is known as insulin resistance and is a hallmark of type 2 diabetes.

Researchers have suspected that bacteria and other microbes in the gut are involved in aging, but how the microbes influence the process hasn’t been clear. Monica Bodogai of the U.S. National Institute on Aging in Baltimore and colleagues examined what happens to mice’s gut bacteria as the rodents age. The mice lose A. muciniphila, also called Akk, and other friendly microbes that help break down dietary fiber into short-chain fatty acids, such as butyrate and acetate. Those fatty acids signal bacteria and human cells to perform certain functions.
Losing Akk led to less butyrate production, Bodogai’s team found. In turn, loss of butyrate triggered a chain reaction of immune cell dysfunction that ended with mice’s cells ignoring the insulin.

Treating old mice and elderly rhesus macaques with an antibiotic called enrofloxacin increased the abundance of Akk in the animals’ guts and made cells respond to insulin again. Giving old animals butyrate had the same effect, suggesting that there may be multiple ways to head off insulin resistance in older people in the future.

For the first time, humans have built a set of pushy, destructive genes that infiltrated small populations of mosquitoes and drove them to extinction.

But before dancing sleeveless in the streets, let’s be clear. This extermination occurred in a lab in mosquito populations with less of the crazy genetic diversity that an extinction scheme would face in the wild. The new gene drive, constructed to speed the spread of a damaging genetic tweak to virtually all offspring, is a long way from practical use. Yet this test and other news from 2018 feed one of humankind’s most persistent dreams: wiping mosquitoes off the face of the Earth.

For the lab-based annihilation, medical geneticist Andrea Crisanti and colleagues at Imperial College London focused on one of the main malaria-spreading mosquitoes, Anopheles gambiae. The mosquitoes thrive in much of sub-Saharan Africa, where more than 400,000 people a year die from malaria, about 90 percent of the global total of malaria deaths.

To crash the lab population, the researchers put together genes for a molecular copy-and-paste tool called a CRISPR/Cas9 gene drive. The gene drive, which in this case targeted a mosquito gene called doublesex, is a pushy cheat. It copies itself into any normal doublesex gene it encounters, so that all eggs and sperm will carry the gene drive into the next generations. Female progeny with two altered doublesex genes develop more like males and, to people’s delight, can’t bite or reproduce.

In the test, researchers set up two enclosures, each mixing 150 males carrying the saboteur genes into a group of 450 normal mosquitoes, males and females. Extinction occurred in eight generations in one of the enclosures and in 12 in the other (SN: 10/27/18, p. 6).

This is the first time that a gene drive has forced a mosquito population to breed itself down to zero, says Omar Akbari of the University of California, San Diego, who has worked on other gene drives. However, he warns, “I believe resistance will be an issue in larger, diverse populations.” More variety in mosquito genes means more chances of some genetic quirk arising that counters the attacking gene drive.

But what if a gene drive could monkey-wrench a wild population, or maybe a whole species, all the way to extinction? Should people release such a thing? To make sense of this question, we humans will have to stop talking about “mosquitoes” as if they’re all alike. The more than 3,000 species vary considerably in what they bite and what ecosystem chores they do.

The big, iridescent adults of Toxorhynchites rutilus, for instance, can’t even drink blood. And snowmelt mosquitoes (Ochlerotatus communis) are pollinators of the blunt-leaved orchid (Platanthera obtusata), ecologist Ryo Okubo of the University of Washington in Seattle said at the 2018 meeting of the Society for Integrative and Comparative Biology.
Estimating what difference it would make ecologically if a whole mosquito species disappeared has stirred up plenty of speculation but not much data. “I got pretty fed up with the hand-waving,” says insect ecologist Tilly Collins of Imperial College London. So she and colleagues dug through existing literature to see what eats An. gambiae and whether other mosquitoes would flourish should their competitor vanish.

So far, extermination of this particular mosquito doesn’t look like an ecological catastrophe, Collins says. Prey information is far from perfect, but diets suggest that other kinds of mosquitoes could compensate for the loss. The species doesn’t seem to be any great prize anyway. “As adults, they are small, not juicy, and hard to catch,” she says. The little larvae, built like aquatic caterpillars with bulging “shoulders” just behind their heads, live mostly in small, temporary spots of water.
The closest the researchers came to finding a predator that might depend heavily on this particular mosquito was the little East African jumping spider Evarcha culicivora. It catches An. gambiae for about a third of its diet and likes the females fattened with a human blood meal. Yet even this connoisseur “will readily consume” an alternative mosquito species, the researchers noted in July in Medical and Veterinary Entomology.

Collins also thinks about the alternatives to using genetically engineered pests as pest controls. Her personal hunch is that saddling mosquitoes with gene drives to take down their own species is “likely to have fewer ecological risks than broad-spectrum use of pesticides that also kill other species and the beneficial insects.”

Gene drives aren’t the only choice for weaponizing live mosquitoes against their own kind. To pick just one example, a test this year using drones to spread radiation-sterilized male mosquitoes in Brazil improved the chances that the old radiation approach will be turned against an Aedes mosquito that can spread Zika, yellow fever and chikungunya.

Old ideas, oddly enough, may turn out to be an advantage for antimosquito technologies in this era of white-hot genetic innovation. Coaxing the various kinds of gene drives to work is hard enough, but getting citizens to sign off on their use may be even harder.

The monarch butterfly isn’t the only insect flying up and down North America in a mind-boggling annual migration. Tests show a big, shimmering dragonfly takes at least three generations to make one year’s migratory loop.

Ecologist Michael Hallworth and his colleagues pieced together the migration of the common green darner, described December 19 in Biology Letters, using data on forms of hydrogen in the insects’ wings, plus records of first arrivals spotted by citizen scientists.
The study reveals that a first generation of insects emerges in the southern United States, Mexico and the Caribbean from about February to May and migrates north. Some of those Anax junius reach New England and the upper Midwest as early as March, says Hallworth, of the Smithsonian Migratory Bird Center headquartered in Washington, D.C.

Those spring migrant darners lay eggs in ponds and other quiet waters in the north and eventually die in the region. This new generation migrates south from about July until late October, though they have never seen where they’re heading. Some of these darners fly south in the same year their parents arrived and some the next year, after overwintering as nymphs.

A third generation emerges around November and lives entirely in the south during winter. It’s their offspring that start the cycle again by swarming northward as temperatures warm in the spring. With a wingspan as wide as a hand, they devote their whole lives to flying hundreds of kilometers to repeat a journey their great-grandparents made.
Scientists knew that these dragonflies migrated. Dragonfly enthusiasts have spotted swarms of the green darners in spring and in fall. But which generations were doing what has been tricky to demonstrate. “Going in, we didn’t know what to expect,” Hallworth says.
Tracking devices that let researchers record animals’ movements for more than a week or two haven’t been miniaturized enough to help. The smallest still weigh about 0.3 grams, which would just about double a darner’s weight, Hallworth says. So researchers turned to chemical clues in darner tissues. Conservation biologist and study coauthor Kent McFarland succeeded at the delicate diplomacy of persuading museums to break off a pinhead-sized wing tip fragment from specimens spanning 140 years.

Researchers checked 800 museum and live-caught specimens for the proportion of a rare heavy form of hydrogen that occurs naturally. Dragonfly wings pick up their particular mix of hydrogen forms from the water where the aquatic youngsters grow up. Scientists have noticed that a form called hydrogen-2 grows rarer along a gradient from south to north in North America. Looking at a particular wing in the analysis, “I can’t give you a zip code” for a darner, Hallworth says. But he can tell the native southerners from Yankees.

An adult darner, regardless of where it was born, is “a green piece of lightning,” says McFarland, of the Vermont Center for Ecostudies in White River Junction. Darners maneuver fast enough to snap insect prey out of the air around ponds across North America. The front of an adult’s large head is “all eye,” he says, and trying to catch samples for the study was “like hitting a knuckleball.”

Although the darners’ north-south migration story is similar to that of monarchs (Danaus plexippus), there are differences, says evolutionary biologist Hugh Dingle of the University of California, Davis, who has long studied these butterflies. Monarchs move northward in the spring in stepwise generations, instead of one generation sweeping all the way to the top of its range.

Also, Dingle says, pockets of monarchs can buck the overall scheme. Research suggests that some of the monarchs in the upper Midwest do a whole round trip migration in a single generation. As researchers discover more details about green darners, he predicts, the current basic migration scheme will turn out to have its quirky exceptions, too.

The adolescent saber-toothed cat on a summertime hunt realized too late that she had made a terrible miscalculation.

Already the size of a modern-day tiger, with huge canine teeth, she had crept across grassy terrain to ambush a giant ground sloth bellowing in distress. Ready to pounce, the cat’s front paw sank into sticky ground. Pressing down with her other three paws to free herself, then struggling in what has been called “tar pit aerobics,” she became irrevocably mired alongside her prey.

Scenarios much like this played out repeatedly over at least the last 35,000 years at California’s Rancho La Brea tar pits. Entrapped herbivores, such as the sloth, attracted scavengers and predators — including dire wolves, vultures and saber-toothed Smilodon cats — to what looked like an easy meal. Eventually the animals would disappear into the muck, until paleontologists plucked their fossils from the ground in huge numbers over the last century.

Five million or so fossils have been found at the site. But “it’s not like there was this orgy of death going on,” says Christopher Shaw, a paleontologist and former collections manager at the La Brea Tar Pits and Museum in Los Angeles. He calculates that such an entrapment scenario, dooming 10 or so large mammals and birds, would have needed to occur only once per decade over 35,000 years to account for that bounty of fossils.

At La Brea, the collection of Smilodon fatalis fossils alone includes more than 166,000 bones, from an estimated 3,000 of the ill-fated prehistoric cats. Famed for their fearsome canines, which grew up to 18 centimeters long, S. fatalis weighed as much as 280 kilograms, bigger than most of today’s largest lions and tigers.
Fossils of S. fatalis, the second largest of three Smilodon species that roamed the Americas during the Pleistocene Epoch, have been found across the United States and in South America, west of the Andes as far south as Chile. And a recent study put S. fatalis in Alberta, Canada, about 1,000 kilometers north of its previously known range.

But the La Brea fossil site, unique in offering up so many specimens, is the source of the vast majority of knowledge about the species. There, fossils of dire wolves and saber-toothed cats together outnumber herbivores about 9-to-1, leading scientists to speculate that both predators may have formed prides or packs, similar to modern lions and wolves. Yet a small number of experts argue against cooperative behavior for Smilodon, reasoning that pack-living animals would have been too intelligent to get mired en masse.
New studies may help settle the debate about Smilodon’s sociality, and answer questions about how the cat lived and why it died out 10,000 to 12,000 years ago.

“We have an innate curiosity to understand what it was doing and why it went extinct,” says Larisa DeSantis, a vertebrate paleontologist at Vanderbilt University in Nashville. Now, she says, “we can answer these questions.”

DeSantis is studying microscopic wear on fossil teeth and chemical signatures in the enamel to reveal Smilodon’s diet. Other scientists are doing biomechanical studies of the skull, fangs and limbs to understand how the powerful cat captured and killed its prey. Some researchers are extracting DNA from fossils, while others are gathering data on the paleoclimate to try to piece together why Smilodon died out.

“It’s the T. rex of mammals … a big, scary predator,” says Ashley Reynolds, a paleontology Ph.D. student and fossil cat researcher at the University of Toronto. She presented the Alberta fossil find in October in Albuquerque at the Society of Vertebrate Paleontology conference. Explaining why Smilodon cats continue to excite researchers, she says, “They’re probably the baddest of all the cats that have ever existed.”
Safety in numbers
Whether Smilodon was a pack hunter has long been debated (SN: 10/28/17, p. 5) because living in groups is rare among large cats today. But an unusual number of healed injuries in the Smilodon bones at La Brea makes it unlikely that these cats were solitary, DeSantis and Shaw reported in November in Indianapolis at a meeting of the Geological Society of America.

More than 5,000 of the Smilodon bones at La Brea have marks of injury or illness: tooth decay, heavily worn arthritic joints, broken legs and dislocated elbows that would have occurred before the animals’ tar burial. Dramatic examples include crushed chests and spinal injuries, which the cats somehow survived. “You would actually wince to see these horribly, traumatically injured specimens,” says Shaw, who is also coeditor of the 2018 book Smilodon: The Iconic Sabertooth.

One particularly debilitating injury was a crippled pelvis, but evidence of new bone growth shows that the animal lived long enough for healing to occur. “There was a lot of infection, pain and smelly stuff, and just a really awful situation for this animal, but it survived well over a year,” Shaw says. “To me that indicates [the injured cat] was part of a group that helped it survive by letting it feed at kills and protecting it.”

Shaw and DeSantis looked at a series of specimens with what were probably agonizing maladies in the teeth and jaws, including fractured canines and massive infections that left animals with misshapen skulls.

“These animals probably couldn’t have gone out … to kill anything,” Shaw says. “You know how it is when you have a toothache. This is like that times 100.”
DeSantis compared microscopic pits and scratches on the surface of the teeth of injured animals with microwear on the teeth of seemingly healthy Smilodon cats. The injured cats’ dental surfaces indicated that the animals were eating softer foods, which would have been less painful to chew, “likely a higher proportion of flesh, fat and organs, as opposed to bone,” she says.

The findings are consistent with the interpretation that Smilodon was a group-living animal, she says, and that the cats “allowed each other access to food when [injured pack members] couldn’t necessarily take down their own prey.”

Reynolds agrees that the healed injuries are persuasive evidence that Smilodon lived in groups. “When you see an animal with really nasty injuries that healed somehow, it does make you wonder if they were cared for.”

Not everyone is convinced, however. Ecologist Christian Kiffner of the Center for Wildlife Management Studies in Karatu, Tanzania, has studied modern carnivores such as African lions and spotted hyenas. “Relatively long survival of Smilodon fatalis individuals after dental injuries had occurred does not necessarily provide airtight evidence for a specific social system in this species,” he says. “It is very, very difficult to use patterns in Pleistocene carnivore [fossil] assemblages to make inferences about behavior of an extinct species.”

Even if the saber-toothed cats did live in groups, the animals’ exact social structure remains an open question, Reynolds says. Modern lion prides have numerous females and several younger males led by an alpha male, with intense competition between male lions. As a result, males are much bigger than females, as the males must work hard to defend their positions.

Despite searching, scientists have not found obvious evidence of a size difference between the sexes in Smilodon; researchers can’t even tell which La Brea fossils are male or female. Size differences between the sexes, if they existed, may have been small.

“That lack of sexual dimorphism is odd,” says Blaire Van Valkenburgh, a UCLA paleontologist who studies fossil carnivores. Sex-related size differences are seen in many big cats today, most particularly lions. She thinks the lack of sexual dimorphism in Smilodon might hint at a different social structure. Perhaps males weren’t competing quite so intensely for access to females. Maybe there was no single alpha male preventing the majority of males from making a move.

Family affair
Perhaps Smilodon groups had an alpha female rather than an alpha male, or an alpha pair. Such is the case in modern wolves and coyotes, which have less pronounced size differences between sexes than lions do. The prehistoric cats “could have had extended family structures [similar to wolves] where uncles and aunts hung around, because it probably took a while to raise the young saber-toothed cats,” Van Valkenburgh suspects.

Kittens may have taken a long time, as long as 22 months, to get most of their adult teeth, she says. The upper canines took even longer, as much as three years or more, to reach their massive size, researchers reported in PLOS ONE in 2015. Modern lions, in contrast, typically have all of their adult teeth by 17 months, Van Valkenburgh says.

Smilodon kittens also probably went through a substantial learning curve before attempting to take down large prey. “It took longer for them to learn how to safely kill something without breaking their teeth or biting in the wrong place and hurting themselves,” Van Valkenburgh speculates.

Pack living would enable this slower development: “If you’re a social species, you can afford to grow at a slower rate than a nonsocial species because you have a family safety net,” Reynolds says. She is studying Smilodon fossils from Peru’s Talara tar pits for evidence of slow bone development using bone histology, examining thin cross sections under a microscope to determine such things as age and growth rate.
To understand how saber-toothed cats eventually took down prey, Van Valkenburgh joined paleobiologist Borja Figueirido of the University of Málaga in Spain and others. The group studied the biomechanics of Smilodon’s killing bite and how the animal used its sabers. That work, published in the October 22, 2018 Current Biology, adds to a consensus that the cat used its powerful forelimbs, which existed even in the youngsters (SN Online: 9/27/17), to pin prey before applying a lethal bite to the neck.

“The specialization of being a saber-toothed appears to have been partly to effectively take prey larger than yourself and to do that very quickly,” Van Valkenburgh says. With the prey tightly gripped, a Smilodon cat would position itself so that one or two really strong canine bites would rip open the pinned animal’s throat.

In contrast, lions suffocate prey — one lion may clamp its jaws around the neck, crushing the windpipe, while another uses its mouth to cover the victim’s nose and mouth. Using this slower method would have increased Smilodon’s chances of injuring or damaging those precious canine teeth.

Diverging senses
Smilodon and its extinct saber-toothed relatives are on a branch of the cat family tree that is far from today’s cats. Scientists think Smilodon’s branch diverged from the ancestors of all living cats about 20 million years ago. Given the evolutionary distance, researchers are still trying to determine how similar — or different — Smilodon was from its living feline cousins. A recent focus has been the cat’s sounds and senses.

At the October vertebrate paleontology conference, Shaw presented evidence that Smilodon may have roared, as do lions, tigers, leopards and their close relatives. The clues come from 150 La Brea fossils that were once part of the hyoid arch, or larynx, in the Smilodon throat. (Tar pits stand out for preserving tiny bones rarely found elsewhere.) The small fossils are very similar in shape and style to those of roaring cats. House cats and others that purr have a different arrangement of bones.
Smilodon may have “used this type of communication as an integral part of social behavior,” Shaw says. Roaring, however, is not a sure sign of pack living, Reynolds notes; most roaring cats today do not live in large groups.

How Smilodon’s sense of smell compared with living cats’ is something else researchers wonder about. To probe this part of the extinct animal’s biology, a team lead by Van Valkenburgh looked at Smilodon’s cribriform plate — a small, perforated bone inside the skull. Smell-sensing nerve cells pass through holes in the plate from the olfactory receptors in the nose to the brain. The size and number of holes are thought to correlate with the number of receptors and, therefore, the extent of an animal’s sense of smell.

To confirm this link, Van Valkenburgh’s team combined CT scans and 3-D images of skulls from 27 species of living mammals with information on the number of olfactory receptor genes. A CT scan of a skull revealed that Smilodon may have had slightly fewer olfactory receptor nerve cells than a domestic cat, the researchers reported at the paleontology conference. Smilodon’s sense of smell was perhaps 10 to 20 percent less keen than a modern lion’s, says Van Valkenburgh, whose team reported the findings in the March 14, 2018 Proceedings of the Royal Society B.

Smilodon “might have relied more heavily on their eyes and their ears,” she says. Perhaps, in an ancient evolutionary divergence, Smilodon’s level of reliance on smell went in a slightly different direction than in modern big cats.

Saber-toothed swan song
As the pieces of the Smilodon puzzle fall into place, perhaps the biggest remaining mystery is why the animal disappeared 10,000 to 12,000 years ago. Debate about the extinction of some of North America’s large mammal species swings between blaming humans and climate change (SN: 11/10/18, p. 28). While humans, who probably arrived on the continent more than 15,000 years ago, and Smilodon certainly knew one another in the Americas, they may not have overlapped at La Brea, Shaw says. The earliest evidence of people in the Los Angeles Basin is about 11,000 years ago, by which time Smilodon may or may not already have gone. Nevertheless, human hunting of large prey elsewhere in the Americas could have led to a scarcity of food for the big cats, he says.

One theory holds that Smilodon went through tough times at La Brea when lack of prey forced the saber-toothed cats to consume entire carcasses including bones. This has been posited as the reason for all those broken teeth among the La Brea fossils. But DeSantis isn’t convinced; she thinks breakages happened during scuffles with prey. She says dental microwear suggests that Smilodon was not eating great quantities of bone.
Some opportunistic carnivores, such as cougars, did eat bone and managed to survive to the modern day. Perhaps Smilodon couldn’t adapt to hunting smaller prey when larger herbivores disappeared, also around 10,000 to 12,000 years ago (SN: 11/24/18, p. 22).

“A lot of the large prey on the landscape go extinct,” DeSantis says. “You lose out on the horses, camels, giant ground sloths, mammoths and mastodon. That’s got to have had an impact.”

The challenge of dating fossils from the tar pits has been one hurdle to understanding exactly what was going on with Smilodon over time. Bones deposited over many thousands of years get jumbled by movement in the tar, for reasons experts don’t fully understand. Plus, the tar itself becomes embedded in each specimen, complicating carbon dating.

However, new methods of chemically pretreating fossils to remove the tar have made carbon dating much easier and cheaper — and a multi-institutional project is now dating hundreds of Smilodon and other bones. Researchers will soon be able to track changes in Smilodon over the 35,000 years of prehistory recorded at La Brea and correlate fossil changes to known changes in climate over that time.

“We’re going to have a much better handle,” Van Valkenburgh says, “on what was going on towards the end of their existence.”

This article appears in the March 30, 2019 issue of Science News with the headline, “The Baddest Cat of All: Fresh details say saber-toothed Smilodon helped injured pack members.”

An ancient four-legged whale walked across land on hooved toes and swam in the sea like an otter.

The newly discovered species turned up in 2011 in a cache of fossilized bones in Playa Media Luna, a dry coastal area of Peru. Jawbones and teeth pegged it as an ancient cetacean, a member of the whale family. And more bones followed.

“We were definitely surprised to find this type of whale in these layers, but the best surprise was its degree of completeness,” says Olivier Lambert, a paleontologist at the Royal Belgian Institute of Natural Sciences in Brussels.
Jaw, tooth and spine features, described April 4 in Current Biology, don’t quite match anything else in the fossil record, setting the skeleton apart as a new species, dubbed Peregocetus pacificus (meaning “the traveling whale that reached the Pacific Ocean”). At 42.6 million years old, it’s the oldest whale skeleton found in the New World, though some fossilized whale teeth from North America may be even older.

Big, possibly webbed feet and long toes would have allowed P. pacificus to dog-paddle or swim freestyle. And like modern otters and beavers, this whale’s vertebrae suggest that its tail also functioned as a paddle. With tiny hooves and strong legs and hips, the animal could walk on land. But “it was definitely a better swimmer than walker,” Lambert says.

Whales got their start on land and gradually adapted to a water-dwelling lifestyle. The first amphibious whales emerged more than 50 million years ago near what’s now India and Pakistan. The new species shares some similar features with Maiacetus and Rodhocetus, two early whales from that area. P. pacificus’ age supports the idea that whales migrated across the South Atlantic and around South America to the Pacific Ocean in their first 10 million years of existence.

Multiplying 2 x 2 is easy. But multiplying two numbers with more than a billion digits each — that takes some serious computation.

The multiplication technique taught in grade school may be simple, but for really big numbers, it’s too slow to be useful. Now, two mathematicians say that they’ve found the fastest way yet to multiply extremely large figures.

The duo claim to have achieved an ultimate speed limit for multiplication, first suggested nearly 50 years ago. That feat, described online March 18 at the document archive HAL, has not yet passed the gauntlet of peer review. But if the technique holds up to scrutiny, it could prove to be the fastest possible way of multiplying whole numbers, or integers.
If you ask an average person what mathematicians do, “they say, ‘Oh, they sit in their office multiplying big numbers together,’” jokes study coauthor David Harvey of the University of New South Wales in Sydney. “For me, it’s actually true.”

When making calculations with exorbitantly large numbers, the most important measure of speed is how quickly the number of operations needed — and hence the time required to do the calculation — grows as you multiply longer and longer strings of digits.

That growth is expressed in terms of n, defined as the number of digits in the numbers being multiplied. For the new technique, the number of operations required is proportional to n times the logarithm of n, expressed as O(n log n) in mathematical lingo. That means that, if you double the number of digits, the number of operations required will increase a bit faster, more than doubling the time the calculation takes.
But, unlike simpler methods of multiplication, the time needed doesn’t quadruple, or otherwise rapidly blow up, as the number of digits creeps up, report Harvey and Joris van der Hoeven of the French national research agency CNRS and École Polytechnique in Palaiseau. That slower growth rate makes products of bigger numbers more manageable to calculate.

The previously predicted max speed for multiplication was O(n log n), meaning the new result meets that expected limit. Although it’s possible an even speedier technique might one day be found, most mathematicians think this is as fast as multiplication can get.

“I was very much astonished that it had been done,” says theoretical computer scientist Martin Fürer of Penn State. He discovered another multiplication speedup in 2007, but gave up on making further improvements. “It seemed quite hopeless to me.”

The new technique comes with a caveat: It won’t be faster than competing methods unless you’re multiplying outrageously huge numbers. But it’s unclear exactly how big those numbers have to be for the technique to win out — or if it’s even possible to multiply such big numbers in the real world.

In the new study, the researchers considered only numbers with more than roughly 10214857091104455251940635045059417341952 digits when written in binary, in which numbers are encoded with a sequence of 0s and 1s. But the scientists didn’t actually perform any of these massive multiplications, because that’s vastly more digits than the number of atoms in the universe. That means there’s no way to do calculations like that on a computer, because there aren’t enough atoms to even represent such huge numbers, much less multiply them together. Instead, the mathematicians came up with a technique that they could prove theoretically would be speedier than other methods, at least for these large quantities.

There’s still a possibility that the method could be shown to work for smaller, but still large, numbers. That could possibly lead to practical uses, Fürer says. Multiplication of these colossal numbers is useful for certain detailed calculations, such as finding new prime numbers with millions of digits (SN Online: 1/5/18) or calculating pi to extreme precision (SN Online: 12/10/02).

Even if the method is not widely useful, making headway on a problem as fundamental as multiplication is still a mighty achievement. “Multiplying numbers is something people have been working on for a while,” says mathematical physicist John Baez of the University of California, Riverside. “It’s a big deal, just because of that.”

A nebulous void in Egypt’s Great Pyramid of Giza has been unveiled thanks to strange subatomic particles called muons.

Scientists first identified the void in 2016 using muons, heavy relatives of electrons that can penetrate through solid materials. Thought to be a corridor-shaped hole, the void was located near a chevron-shaped structure visible on the pyramid’s north face. Further muon measurements revealed new details of the void’s size and shape, scientists from the ScanPyramids team report March 2 in Nature Communications.
The new muon measurements indicate that the void is a 9-meter-long corridor about 2 meters wide by 2 meters tall, close to the pyramid’s north face. ScanPyramids researchers made additional measurements with ground-penetrating radar and ultrasonic testing, they reported March 2 in NDT & E International. The detailed measurements allowed the scientists to use an endoscope to take images inside the chamber, the team announced. The images reveal a corridor with a vaulted ceiling, presumably one that was hasn’t been seen by humans since the pyramid was built more than 4,500 years ago. The corridor’s purpose is still unclear.
Muons are created when high-energy particles from space called cosmic rays crash into the Earth’s atmosphere. Muons are partially absorbed as they rain down onto structures such as the pyramids. Using detectors placed inside the pyramid, scientists from ScanPyramids zeroed in on regions where more muons made it through, indicating they’d traversed less material, which let them map out the location of the void.

Scientists also recently used muons to probe an ancient Chinese wall (SN: 1/30/23), a nuclear reactor and various volcanoes (SN: 4/22/22).

Artificial intelligence has passed the last major milestone in mastering poker: six-player no-limit Texas Hold’em.

Games like poker, with hidden cards and players who bluff, present a greater challenge to AI than games where every player can see the whole board. Over the last few years, computers have become aces at increasingly complicated forms of one-on-one poker, but multiplayer games take that complexity to the next level (SN Online: 5/13/15).

Now, a card shark AI dubbed Pluribus has outplayed more than a dozen elite professionals at six-player Texas Hold’em, researchers report online July 11 in Science. Algorithms that can plot against several adversaries using such spotty information could make savvy business negotiators, political strategists or cybersecurity watchdogs.
Pluribus honed its initial strategy by playing against copies of itself, starting from scratch and gradually learning which actions helped to win. Then, the AI used that intuition for when to hold and when to fold during the first betting round of each hand against five human players.

During subsequent betting rounds, Pluribus fine-tuned its strategy by imagining how the game might play out if it took different actions. Unlike artificial intelligence trained for two-player poker, Pluribus didn’t speculate all the way to the end of the game — which would require too many computations when dealing with so many players (SN: 4/1/17, p. 12). Instead, the AI imagined several moves ahead and decided what to do based on those hypothetical futures and different strategies that players could adopt.

In 10,000 hands of Texas Hold’em, Pluribus competed against five contestants from a pool of 13 professionals, all of whom had won more than $1 million playing poker. Every 100 hands, Pluribus raked in, on average, about $480 from its human competitors.
“This is roughly the amount that elite human professionals aspire to beat weaker players by,” implying that Pluribus was a savvier player than its human opponents, says Noam Brown of Facebook AI Research in New York City. Brown, along with Tuomas Sandholm of Carnegie Mellon University in Pittsburgh, created Pluribus.

Now that AI has poker in the bag, algorithms could test their strategic reasoning in games with more complex hidden information, says computer scientist Viliam Lisý of the Czech Technical University in Prague, who was not involved in the work. In games like Kriegspiel — a chess spin-off where players can’t see each other’s pieces — the unknowns can become far more complicated than a few cards held close to opponents’ chests, Lisý says.

Video games like StarCraft, which allow many more types of moves and free players from rigid, turn-based play, could also serve as new tests of AI cleverness (SN: 5/11/19, p. 34).

Homo sapiens who reached Europe around 54,000 years ago introduced bows and arrows to that continent, a new study suggests.

Researchers examined tiny triangular stone points and other artifacts excavated at a rock-shelter in southern France called Grotte Mandrin. H. sapiens on the move probably brought archery techniques from Africa to Europe, archaeologist Laure Metz of Aix-Marseille University in France and colleagues report February 22 in Science Advances.

“Metz and colleagues demonstrate bow hunting [at Grotte Mandrin] as convincingly as possible without being caught bow-in-hand,” says archaeologist Marlize Lombard of the University of Johannesburg, who did not participate in the new study.
No bows were found at the site. Wooden items such as bows preserve poorly. The oldest intact bows, found in northern European bogs, date to around 11,000 years ago, Metz says.

Previous stone and bone point discoveries suggest that bow-and-arrow hunting originated in Africa between about 80,000 and 60,000 years ago. And previously recovered fossil teeth indicate that H. sapiens visited Grotte Mandrin as early as 56,800 years ago, well before Neandertals’ demise around 40,000 years ago and much earlier than researchers had thought that H. sapiens first reached Europe (SN: 2/9/22).

“We’ve shown that the earliest known Homo sapiens to migrate into Neandertal territories had mastered the use of the bow,” Metz says.

No evidence suggests that Neandertals already present in Europe at that time launched arrows at prey. It’s also unclear whether archery provided any substantial hunting advantages to H. sapiens relative to spears that were thrust or thrown by Neandertals.
Among 852 stone artifacts excavated in a H. sapiens sediment layer at Grotte Mandrin dated to about 54,000 years ago, 196 triangular stone points displayed high-impact damage. Another 15 stone points showed signs of both high-impact damage and alterations caused by butchery activities, such as cutting.

Comparisons of those finds were made to damage on stone replicas of the artifacts that the researchers used as arrowheads shot from bows and as the tips of spears inserted in handheld throwing devices. Additional comparative evidence came from stone and bone arrowheads used by recent and present-day hunting groups.

Impact damage along the edges of stone points from the French site indicated that these implements had been attached at the bottom to shafts.

The smallest Grotte Mandrin points, many with a maximum width of no more than 10 millimeters, could have pierced animals’ hides only when shot from bows as the business ends of arrows, the researchers say. Experiments they conducted with replicas of the ancient stone points found that stone points less than 10 millimeters wide reach effective hunting speeds only when attached to arrow shafts propelled by a bow.

Larger stone points, some of them several times the size of the smaller points, could have been arrowheads or might have tipped spears that were thrown or thrust by hand or launched from handheld spear throwers, the researchers conclude.

Lombard, the University of Johannesburg archaeologist, suspects that the first H. sapiens at the French rock-shelter hunted with bows and arrows as well as with spears, depending on where and what they were hunting. Earlier studies directed by Lombard indicated that sub-Saharan Africans similarly alternated between these two types of hunting weapons starting between about 70,000 and 58,000 years ago.

H. sapiens newcomers to Europe may have learned from Neandertals that spear hunting in large groups takes precedence on frigid landscapes, where bow strings can easily snap and long-distance pursuit of prey is not energy efficient, Lombard says.

But learning about archery from H. sapiens may not have been in the cards for Neandertals. Based on prior analyses of brain impressions on the inside surfaces of fossil skulls, Lombard suspects that Neandertals’ brains did not enable the enhanced visual and spatial abilities that H. sapiens exploited to hunt with bows and arrows.

That’s a possibility, though other controversial evidence suggests that Neandertals behaved no differently from Stone Age H. sapiens (SN: 3/26/20).If Grotte Mandrin Neandertals never hunted with bows and arrows but still survived just fine alongside H. sapiens archers for roughly 14,000 years, reasons for Neandertals’ ultimate demise remain as mysterious as ever.

Whether it’s Katy Perry poaching dancers from once-BFF Taylor Swift or Clytemnestra orchestrating the murder of her husband Agamemnon, betrayal is a dark, persistent part of the human condition. Unlike garden-variety deception, betrayal happens in established relationships, destroying trust that has developed over time. It’s usually unexpected, and it yields a unique, often irreparable, wound. In fact, betrayers have a special place in hell, literarily: In Dante’s Inferno, they occupy the ninth and final circle; mere fraudsters dwell in the eighth.

While most of us are familiar with betrayal, investigating it is really hard. (Consider all the complications of a study that asks people in trusted relationships to betray each other.) Case studies of real betrayals can provide insight after-the-fact, but without a time machine, finding studies that reveal big picture patterns about the lead-up to treachery are scarce.

“We all know betrayal exists,” says Cristian Danescu-Niculescu-Mizil, a computer scientist at Cornell University who spends a lot of time thinking about what language reveals about relationships. “But finding relevant data is really hard.”

So when Danescu-Niculescu-Mizil heard about a Diplomacy, a strategy game rife with betrayal, he figured it might serve as a good proxy for real life treachery. And he was right: Studying the patterns of communication between the players revealed that betrayal is sometimes foreseeable. But like many relationships that collapse in betrayal, teasing out what goes wrong and who is at fault isn’t so easy.
Unlike Risk and other war games, Diplomacy is all about, well, diplomacy (John F. Kennedy and Henry Kissinger reportedly were fans). Set in Europe before World War I, the nations/players have to form alliances to win. But chance is removed from the equation; players don’t roll dice or take turns. There’s only diplomacy: a negotiation phase where players converse, form alliances and gather intelligence (these days, typically online), and a movement phase where everyone’s decisions are revealed and executed all at once. Betrayal is so integral to Diplomacy that, as noted on a “This American Life” episode, stabbing an ally in the back is referred to by the shorthand “stabbing.”

Danescu-Niculescu-Mizil, colleague and fan-of-the-game Jordan Boyd-Graber, and colleagues examined 249 games of Diplomacy with a total of 145,000 messages among players. When they used a computer program to compare exchanges between players whose relationships ended in betrayal with those whose relationships lasted, the computer discerned subtle signals of impending betrayal.

One harbinger was a shift in politeness. Players who were excessively polite in general were more likely to betray, and people who were suddenly more polite were more likely to become victims of betrayal, study coauthor and Cornell graduate student Vlad Niculae reportedJuly 29 at the Annual Meeting of the Association for Computational Linguistics in Beijing. Consider this exchange from one round:

Germany: Can I suggest you move your armies east and then I will support you? Then next year you move [there] and dismantle Turkey. I will deal with England and France, you take out Italy.

Austria: Sounds like a perfect plan! Happy to follow through. And—thank you Bruder!

Austria’s next move was invading German territory. Bam! Betrayal.

An increase planning-related language by the soon-to-be victim also indicated impending betrayal, a signal that emerges a few rounds before the treachery ensues. And correspondence of soon-to-be betrayers had an uptick in positive sentiment in the lead-up to their breach.
Working from these linguistic cues, a computer program could peg future betrayal 57 percent of the time. That might not sound like much, but it was better than the accuracy of the human players, who never saw it coming. And remember that by definition, a betrayer conceals the intention to betray; the breach is unexpected (that whole trust thing). Given that inherent deceit, 57 percent isn’t so bad.

When I spoke to Danescu-Niculescu-Mizil, he said that more important than the clues themselves is the shift in the balance of behavior in the relationship. Positive or negative sentiment of one player isn’t what matters, it’s the asymmetry of the behavior of the two people in the relationship. He likens the linguistic tells to body language: While you wouldn’t use it as a sole basis for decision-making, if you know how to interpret it, it might give you an advantage.

More work is needed to explore whether these patterns exist in real life. And while the research did reveal some patterns, it can’t say anything about cause and effect or who is at fault. Perhaps, for example, the extensive planning of the eventual victims came off as super bossy and frustrating to the eventual betrayer. After all, Clytemnestra’s betrayal of Agamemnon came after he killed their daughter Iphigenia. That kind of bad blood may be unforgivable.