Alan Antczak / DVIDS

A trained Atlantic bottlenose dolphin leaps out of the water during a photo session with the Space and Naval Warfare Systems Center Pacific Marine Mammal Team in San Diego.

The U.S. Navy doesn't yet exactly know how a 130-year-old brass torpedo got to the bottom of the Pacific off the coast of San Diego, but they have a couple of dolphins to thank for rediscovering the rare weapon.

The find was so unexpected that the humans didn't believe the dolphins at first.

The marine mammals have been trained by the Navy's Space and Navy Warfare Systems Center Pacific, or SSC Pacific, to hunt for underwater mines and mark their locations. Divers place mine-shaped objects on the sea bottom, and then they teach the dolphins to find them. "It's all part of training to show the dolphins what they're going to be exposed to when they're on real-world missions," SSC Pacific spokesman Jim Fallin told NBC News on Monday.

During an exercise in March, conducted not far from California's historic Hotel del Coronado, the trainers sent a dolphin down to look for the pre-positioned target objects. The dolphin dove down, came back up — and gave the trainers a signal they didn't expect. "It had found something where we knew something shouldn't be," Fallin said.

The training team dismissed that first signal as a false positive. But when the same team went back to the same place with a different dolphin, the location was flagged again, Fallin said. That's when the trainers started taking the animals seriously.

A piece of naval history
SSC Pacific worked with recovery divers and bomb disposal experts to check out what the dolphins had found. At first, they thought the object was merely an old tail section from an aerial drop mine. They quickly changed their minds.

"It was apparent in the first 15 minutes that this was something that was significant and really old," Christian Harris, operations supervisor for the SSC Pacific Biosciences Division, said in a news release. It turned out to be the tail section from one of the first self-propelled torpedoes developed and used by the U.S. Navy, known as the Howell torpedo.

U.S Navy / SSC Pacific

The fins of a Howell torpedo can be seen preserved in water after the object was recovered with the aid of dolphins.

U.S. Navy

The only other Howell torpedo known to exist today is at the Naval Undersea Museum in Keyport, Wash.

More sections were brought up and submerged in water for preservation. Eventually the torpedo will be flown to the Naval History and Heritage Command at the Washington Navy Yard for more thorough study. "What's missing at this point is the nose, and we're not sure where that is," Fallin said.

The 11-foot-long (3.4-meter-long) torpedo was developed by Lt. Cmdr. John A. Howell between 1870 and 1889. The Navy says it was driven by a 132-pound (60-kilogram) flywheel that was spun up to 10,000 rpm prior to launch. It had a range of 400 yards, a speed of 25 knots, and a warhead filled with 100 pounds of gun cotton.

"It was the first torpedo that could be released into the ocean and follow a track," Harris said. "Considering that it was made before electricity was provided to U.S. households, it was pretty sophisticated for its time."

Howell torpedoes were used on Navy battleships and torpedo boats until 1898, when they were replaced by Whitehead torpedoes. Only 50 of the Howells were ever were built. The only other Howell that exists today is sitting inert in the Naval Undersea Museum in Keyport, Wash.

How did the torpedo get there? Fallin said he "can't add any information other than that it was there," he said.

Day of the dolphins
This isn't the first unexpected object located by the Navy's mine-hunting dolphins: Previously, the mammals have detected sunken items including a submerged car and a lobster trap in a place "where a lobster trap wasn't supposed to be," Fallin said. But the Howell torpedo could well rank as the most significant archaeological find for a finny troop that's trained for war.

The dolphins' finest hour came during the Persian Gulf conflicts, when they spotted underwater hazards and served as sentries for the U.S-led coalition's vessels.

"Dolphins remain the pre-eminent capability for the Navy in counter-mine identification," Fallin said. "There's no technology that the Navy has today that replicates the dolphins' natural ability to identify mines ... although our lab is working on those futuristic technologies. We're designing those technologies around the sonar capabilities that are inherent in dolphins. Unmanned autonomous robots have been proven to be pretty capable at this point in shallow water. The technology holds promise."

It's all in a day's work for the dolphins — and for SSC Pacific, an arm of the Navy's Space and Naval Warfare Systems Command that focuses on command and control, communications, computers, intelligence, surveillance and reconnaissance — a group of technologies known as C4ISR. "We represent the nation's only full-spectrum C4ISR laboratory," Fallin said.

More about dolphins intelligence:

• Dolphins appear to do nonlinear mathematics
• Are dolphins the world's second-smartest animals?
• Dolphins sought to protect against terrorists

Alan Boyle is NBCNews.com's science editor. Connect with the Cosmic Log community by "liking" the NBC News Science Facebook page, following @b0yle on Twitter and adding the Cosmic Log page to your Google+ presence. To keep up with NBCNews.com's stories about science and space, sign up for the Tech & Science newsletter, delivered to your email in-box every weekday. You can also check out "The Case for Pluto," my book about the controversial dwarf planet and the search for new worlds.

Dognition

Duke neuroscientist Brian Hare tests a dog's cognition using a simple set of toys.

Is your dog an Einstein or a Charmer? For $60 (woof!!), a new business venture called Dognition will help you put your pooch through a series of fun playtime activities to find out how your dog thinks. The metrics generated by those experiments … I mean, fun playtime activities … are being fed into a research project that could for the first time determine how the cognitive traits of various breeds differ.

"Dognition.com is ultimately about people's dogs, and finding out about your dog," Duke University neuroscientist Brian Hare, one of the venture's co-founders, told NBC News. "That's what you're paying for. I buy fancy dog food for my dog, and just like I want to take care of his stomach, I want to take care of his mind, too. Skip the next couple of chew toys, and your dog and you will really enjoy doing something a little different."

The business venture builds on Hare's work as the director of the Duke Canine Cognition Center and an associate professor in evolutionary anthropology at Duke's Center for Cognitive Neuroscience. It also meshes with a newly published book by Hare and his wife, Vanessa Woods, titled "The Genius of Dogs."

How smart are dogs?
Don't expect Dognition's cognitive assessment to measure your pet's IQ: Hare says a dog's intelligence can't be described with a single number. (Come to think of it, the same caveat should apply to humans.)

"Because we use standardized testing in all walks of life, it leads you to believe that there's just one measure of intelligence, and there's a number, and that's it," Hare said. "But when you start studying cognitive science, and look at other species, that all starts to crumble."

It's also fruitless to try calculating whether dogs are smarter than cats, or chimps, or bonobos. "That's like trying to answer the question, 'Is a hammer better than a screwdriver?'" Hare said. Instead, he and his colleagues look at how dogs and other species address problem-solving challenges and communicate with humans. It turns out that dogs are geniuses when it comes to figuring out what humans are trying to tell them — which suggests that our world is truly going to the dogs.

"Dogs are bizarrely successful," Hare said. "They have more jobs than ever in this age of the Internet and the International Space Station."

Researchers have even argued that humans and dogs are locked in a co-evolutionary embrace that began tens of thousands of years ago. Last month, for example, one research team determined that canine digestive systems have adapted to the relatively starchy diet served up by modern humans.

How it works
Hare's research into dog cognition began back in 1995, with studies of how dogs looked for hidden treats when humans tried giving them hints. Those experiments, which are done using simple household items such as plastic cups (plus tons of treats), are laid out in Dognition's Canine Assessment Toolkit.

After you plunk your money down, Dognition's website takes you through a personality questionnaire about your dog: For example, how excited does your dog get around other dogs, grown-ups, children? Do fireworks scare your pup? Then, Dognition guides you through a battery of tests that are as fun as playing fetch, or hide-and-seek. The results are uploaded to Dognition HQ, and you get back a detailed profile of your dog's mental habits, based on where Fido's performance ends up on a chart of independent vs. social problem-solving skills.

Different areas of the chart are associated with nine different canine archetypes: Ace, Stargazer, Maverick, Charmer, Socialite, Protodog, Einstein, Expert or Renaissance Dog. That can give you something to brag about on Dognition's Facebook page, but it also can shed new light on why dogs do the things they do, or how you can get through to them better. "We've got a bunch of really fantastic trainers who have signed up to help," Hare said.

Researchers get a reward as well: The data from hundreds of Canine Assessment Tests can be correlated with breed, age and other factors. "To collect the amount of data we've taken in during our month-long beta program would have taken us a couple of years," Hare said.

Eventually, Hare and his colleagues hope to map out the substantive cognitive differences between dog breeds — differences that have not yet been studied scientifically. "The reason we don't know anything about breed differences is that we currently don't have the tools available to look at the number of dogs that would allow us to answer the interesting questions," Hare said.

Dognition could fix that. And it also could open up new possibilities for some of humanity's best friends.

"One of the things we're hoping to do is, suppose there's that dog that may not be the most attractive dog physically, but the dog is wonderfully behaved," Hare said. "What Dognition.com can do is help people understand more about what's inside that dog, and not just its physical appearance — and see that, wow, this dog is amazing."

More about dog intelligence:

• Dog's vocabulary makes her a star
• Gallery: Cat vs. dog intelligence
• The world's 10 most intelligent animals
• How smart is your dog? Give him an IQ test

Dognition offers a $59.95 Canine Assessment Toolkit as well as a $129.95 annual membership bundle that includes enhanced games and other goodies.

Alan Boyle is NBCNews.com's science editor. Connect with the Cosmic Log community by "liking" the log's Facebook page, following @b0yle on Twitter and adding the Cosmic Log page to your Google+ presence. To keep up with Cosmic Log as well as NBCNews.com's other stories about science and space, sign up for the Tech & Science newsletter, delivered to your email in-box every weekday. You can also check out "The Case for Pluto," my book about the controversial dwarf planet and the search for new worlds.

TEAM Network via AFP / Getty Images

A jaguar prowls through the Central Suriname Nature Reserve at night.

The individual scenes tell wonderful wildlife stories: A jaguar goes on the prowl during the night. An elephant takes a stroll through the jungle. A mountain gorilla strikes a pose with a baby on her back. But when you put together 52,000 of those hidden-camera scenes, you can see the bigger picture: The more territory you set aside for wild mammals, the more diversity you'll find. And soon you may be able to contribute to that bigger picture as well.

The tens of thousands of "Candid Camera" moments come from the first worldwide camera-trap mammal study, conducted by a consortium known as the Tropical Ecology Assessment and Monitoring Network, or TEAM Network. In a study published by the Philosophical Transactions of the Royal Society B, scientists documented 105 species that showed up in nearly 52,000 images from seven protected areas in the Americas, Africa and Asia.

The camouflaged cameras were designed to snap pictures when they detect the heat signature of an animal nearby — whether it's a mouse opossum weighing just an ounce, or an African elephant weighing more than four tons. Wild creatures weren't the only things that were caught on camera: The traps also picked up pictures of tourists or poachers who happened to walk by.

WCS / TEAM Network via AFP - Getty Images

A mountain gorilla (Gorilla berengei berengei) lies in front of a hidden camera with a baby on its back in Uganda's Bwindi Impenetrable Forest.

At each of the sites, 60 camera traps were set up for a month, with each camera covering a 2-square-kilometer area. Images for the TEAM project were collected between 2008 and 2010. The pictures showed plenty of cute animals, including giant anteaters, a curious chimpanzee, tapirs, jaguars and cougars. But the study's lead author, Jorge Ahumada of Conservation International, said the cuteness of the pictures was not the focus of the project.

"We're really concerned not about a particular species, but about the health of an ecosystem," he told me today.

Camera traps have been used before for a wide range of species-specific projects, including studies of shimmying bears, seldom-seen leopards and extremely rare rhinos. But the TEAM scientists are more interested in conducting a wider census of species in protected areas around the world. The results could suggest particular locales where conservationists and policymakers should concentrate their efforts.

"Sometimes we operate in the dark, because we don't have a systematic global effort," Ahumada said.

Trento Museum of Science / TEAM Network via AP

This image provided by the Trento Museum of Science shows an African elephant in Tanzania's Udzungwa Mountains on Oct. 11, 2009. It's one of almost 52,000 photos of 105 mammal species taken as part of the first global camera trap mammal study by the TEAM Network.

The scientists have already found that mammalian species diversity is related to how much land is set aside for a protected area, as well as how fragmented the area is. The Central Suriname Nature Reserve in South America had the most species show up on camera (28), while Laos' Nam Kading National Protected Area in Southeast Asia had the least (13). In the Laotian reserve, not a single insect-eating mammal was seen.

"The results of the study are important in that they confirm what we suspected: Habitat destruction is slowly but surely killing our planet’s mammal diversity," Ahumada said in a news release. "We take away two key findings from this research. First, protected areas matter: The bigger the forest they live in, the higher the number and diversity of species, body sizes and diet types. Second, some mammals seem more vulnerable to habitat loss than others: Insect-eating mammals — like anteaters, armadillos and some primates, are the first to disappear — while other groups, like herbivores, seem to be less sensitive."

TEAM Network via AFP - Getty Images

An ocelot walks by a camouflaged camera in Brazil's Manaus nature reserve. The picture is one of nearly 52,000 images collected by the TEAM Network.

Ahumada told me that reduced biodiversity could have a big impact on other environmental issues as well, including climate change and Earth's carbon balance.

"Some scientists suggested not too long ago that if you remove mammals from forests ... you will shift the forest community toward trees that will have much less wood, much less density of carbon per unit weight. That's an unintended consequence of not paying attention to these animals and the whole ecosystem," he said.

He said the findings reported today were just "an initial snapshot" of species diversity in protected areas. "Now, for some of these sites, we have four or five years of data," he added. The TEAM Network has already expanded from seven to 17 sites, and it's aiming to have camera traps in 40 protected areas around the world by 2013. That will provide a regularly updated census of mammalian species in the world's hot spots for biodiversity.

And who knows? Maybe someday there'll be a hot spot near you. Ahumada said he and his colleagues are planning to make their image-analysis software available to the public, so that anyone with a camera trap can gather data about the creatures that pass by.

"Involving citizens in science is great," he said. "The more information we have, the more we know is happening."

TEAM Network via AFP - Getty Images

A South American tapir (Tapirus terrestris) makes an appearance in a camera-trap photo taken in the Central Suriname Nature Reserve.

More about biodiversity:

• More pictures from the camera-trap mammal study
• Tiger poachers caught on camera ... their own
• Biological gems found in the Philippines
• Search for 'biodiversity' on msnbc.com

The TEAM Network is a partnership involving Conservation International, the Missouri Botanical Garden, the Smithsonian Institution and the Wildlife Conservation Society. The effort is partially funded by those institutions and the Gordon and Betty Moore Foundation.

In addition to Ahumada, the authors of "Community Structure and Diversity of Tropical Forest Mammals: Data from a Global Camera Trap Network" include Carlos E.F. Silva, Krisna Gajapersad, Chris Hallam, Johanna Hurtado, Emanuel Martin, Alex McWilliam, Badru Mugerwa, Tim O'Brien, Francesco Rovero, Douglas Sheil, Wilson R. Spironello, Nurul Winarni and Sandy J. Andelman.

TEAM's local partners in the study are Instituto Nacional de Pesquisas da Amazonia (INPA), Conservation International Suriname, Organization for Tropical Studies, Museo Tridentino di Scienze Naturali, and Institute of Tropical Forest Conservation. The seven protected areas monitored for the study were Bwindi Impenetrable Forest (Uganda), Udzungwa Mountains National Park (Tanzania), Bukit Barisan Selatan National Park (Indonesia), Nam Kading National Protected Area (Laos), Central Suriname Nature Reserve (Suriname), Manaus (Brazil) and Volcan Barva Transect (Costa Rica). 

Connect with the Cosmic Log community by "liking" the log's Facebook page or following @b0yle on Twitter. You can also add me to your Google+ circle, and check out "The Case for Pluto," my book about the controversial dwarf planet and the search for new worlds. 

Dreamstime.com

Researchers find that males spend less time in elaborate courtship displays such this peacock's feathers when males far out number females. Instead, they get sneaky to get a mate.

When a woman walks into a male-crowded bar she's unlikely to be showered with courtly attention — that is if findings about mating in the animal kingdom translate to the human realm.

"She might just be watching them fight it out and then have one particularly possessive one making sure others aren't getting access to her," Laura Weir, a postdoctoral fellow at Simon Fraser University in British Columbia, Canada, told me today. 

In other words, as the dudes duke it out with each other, one little weasel will sneak over and trap her in a corner and try to keep her all to himself?

"Exactly," she said, although she stressed her reluctance to take the analogy too far. The data, she noted, is compiled from the mating behaviors of the birds and bees … and alligators, fish, frogs, lizards and lobsters, too. 

Operational sex ratio
Her research focuses on the influence of the so-called operational sex ratio on competition for mates. Operational sex ratio is the ratio of males and females ready to get it on at any one time and place. 

In the animal kingdom, like at a bar, the ratio can often be heavily male biased. For example, in a lot of species, males are often first to arrive to the mating ground so they can establish territories.

The females will come afterward, and depending on the pace of female arrivals over  time, "you can have very biased sex ratios during some times of the mating season," Weir said.

When the first female arrives, the general thought is the males will get aggressive toward each other, fighting with each other to take out the competition. 

"We found that there is this increase in aggression to a point, but then they stop using aggression as a tactic to get females and they change to other tactics like sneaking in or scrambling around looking for females," Weir said.

Dying courtship
All this aggression and sneaking around comes at the expense of courtship, which is a costly, time-consuming investment. Think birds such as peacocks with their fabulous displays of feathers or sparrows constantly updating their playlists.  

Instead of investing in the displays and songs to attract the very best mate, the males put their energy into just trying to find a mate, any mate, which involves more covert sneaking around.

On the flip side, males in these situations tend to be more guarded of the mates they secure.

"If they've mated with her, they want to ensure they are the only one who's done it," Weir said. "And so, rather than go off and fight with other males or try to court another female, they'll just cling to the female that they've already mated with."

Overall, Weir and her colleagues note, the findings illustrate a considerable flexibility in the mating structure within species, which is likely related to the end goal of life: reproduction. 

At least, that's the next question they hope to tackle in their research. 

"If the goal of the biological world is to leave more offspring, are these changes in behavior beneficial to the males when they are actually competing for mates and competing to fertilize eggs?" asked Weir.

Weir is a lead author of a paper describing this research in the February issue of The American Naturalist. 

More stories on the science of animal sex:

• Same sex behavior nearly universal in animals
• Hook-ups in the wild: Do animals enjoy sex?
• Urine spray signals sex, violence to crayfish
• New tunes, not oldies, lure the feathered ladies
• Of mice and men: Tens sex lessons from the wild
• Despite flash, males are simple creatures
• Polygamous mice are better breeders

BW Quist and R Faruqi / Northwestern University

This is a view of the model whisker array built to explore how sensory and motor data are combined in the brain to create a perception.

A virtual model of rat whiskers may help scientists unlock the mystery of how our brains turn the mechanics of touch into perceptions.

"Our sense of touch is very mysterious. You can reach into your pocket or your purse and without even looking, you can identify your keys, a coin, or a paperclip," Mirta Hartmann, who studies sensory and neural systems engineering at Northwestern University, explained to me today.

Her lab is using rat whiskers to understand how the brain goes from the mechanics of touch to a perception. "In the same way that we use our hands to go out and actively explore different objects, rats use their whiskers," Hartmann said.

Whiskers are less complicated to study than the human hand, which has sensors all over. The response of the sensors depend on the viscoelasticity of the skin.

Rat whiskers, by contrast, have senors only at the base. In addition, "rats cannot grasp with their whiskers, they can only explore. Our hand movements are complicated because we can grasp and manipulate objects, as well as tactually explore," Hartmann  noted.

Whisker model
She can colleagues studied the structure of the rat head and whisker array — 30 on each side of the face arranged in a regular pattern — to create their virtual model.

Rats use these whiskers to whisk objects 5 to 25 times per second. This is different than cats or dogs, which also have whiskers but aren't able to "move them back and forth that much," Hartmann noted.

The model allows the researchers to simulate the rat whisking against different objects and predict the full pattern of inputs into the whisker system as a rat encounters an object. These simulations can then be compared against real rat behavior.

"It allows us to start to simulate what's going to happen as the rat comes up to an object and explores it with its whiskers," she said.

Human touch
This information, in turn, should lead to insights to what's going on in the human brain as the hand fishes around a pocket or purse.

"There's just electricity in your brain and there's just mechanical signals on your hand. And somehow your brain is able to turn that contact pattern into electricity that generates a perception," Hartmann said. "That whole process is very mysterious. We need basic research to try and figure out how that happens."

In addition, the research is being used to create robots with whiskers, which can use the motion of the whiskers to generate three-dimensional spatial representations of the environment. The technology could be used, for example, on robots designed to explore dark places.

A paper describing the research was published Thursday online in Public Library of Science Computational Biology.

More stories on whiskers and sense of touch:

• Tomorrow's robots could have whiskers
• How whiskers help rats find their way
• Stretchy solar cells to power e-skin

John Roach is a contributing writer for msnbc.com. Connect with the Cosmic Log community by hitting the "like" button on the Cosmic Log Facebook page or following msnbc.com's science editor, Alan Boyle, on Twitter (@b0yle).

University of Michigan Museum of Paleontology

The ocean in Devonian times: Is the past prologue when it comes to biodiversity collapse?

A long, long time ago — between 378 million and 375 million years ago — about half of all species on the planet vanished. The trigger for this mass extinction, one of five known in Earth's history, was a lethal combination of sea level rise and invasive species, according to a new study.

"The basic processes that normally result in new species forming were blocked," study author Alycia Stigall, a paleobiologist at Ohio University, told me today.

In normal times, species are always going extinct, but as they die off, new species arise. That keeps the planet's number of species relatively constant. "When you take some species away but don't replace them, the overall result is a collapse in global biodiversity," she said.

The findings suggest that the planet's current ecosystems, which are experiencing loss of biodiversity, could meet a similar fate.

How species rise and fall
One path to the rise of new species starts when a population is split in half due to a geological event, such as the rise of a mountain chain that prevents the two halves of the population from interbreeding. Over time, the two groups develop into new species.

 During the Late Devonian period that Stigall studied, most of Earth's creatures lived in shallow sea basins. As sea levels rose due to warming temperatures and shifting land masses, these basins were connected — that is, the barriers that kept them separate disappeared. New species stopped forming.

"The Devonian has normal extinction rates," Stigall noted. In other words, the number of species dying off wasn't abnormally high, as it was when a space rock slammed into Mexico's Yucatan Peninsula and killed off the dinosaurs about 65 million years ago. "So it is really the stopping of speciation that results in a major collapse.

"The other thing that happens when you raise sea level is some species move into that basin that didn't previously live there, and that's the invasive species," she added.

Some of these invasive species fail to carve out a niche for themselves and die off, but others — typically generalists that can eat almost anything and survive in a range of environments — outcompete the natives. "They basically take over and stop the ability of new species from forming," Stigall said.

Lessons for today
Many scientists say that we are in the throes of a sixth great mass-extinction event. The two main reasons are habitat loss — land converted to human use is less available to other critters — and the fact that we are moving species around the planet.

"What we can expect in the long term is that because we have this global movement of invasions, we can expect speciation rates to be very low, but we also know that because of habitat loss, the extinction rate is very high. So we are really looking at a very bad combined effect," Stigall said.

Knowing this, conservationists may be smartest to focus their efforts on generalist-type native species, she added.

"Things that are very narrowly adapted, specialist species, are unlikely to survive. They are unlikely to speciate in the future, and they are also unlikely to survive the habitat loss," she said. "So things like polar bears that are really cute — there's just not much we are going to be able to do for them."

Stigall's findings were published today in the journal PLoS ONE.

More stories about mass extinctions:

• Earth's Timeline: Where does the Devonian era fit in? 
• Scientists: Haiti's wildlife faces mass extinction
• Is algae secret ingredient in mass extinctions?
• Mass extinctions may follow one-two punch
• Study sees mass extinction via warming
• Did Great Salt Lakes trigger mass extinction?
• Meteors not only culprit in mass extinction
• Mass extinction threat: Earth on verge of huge reset button?

John Roach is a contributing writer for msnbc.com. Connect with the Cosmic Log community by hitting the "like" button on the Cosmic Log Facebook page or following msnbc.com's science editor, Alan Boyle, on Twitter (@b0yle).

Scientists have long debated whether climate change or human hunters doomed woolly mammoths to extinction about 10,000 years ago. A new study suggests that delayed weaning due to the prolonged winter darkness north of the Arctic Circle may also be to blame.

The finding, based on an analysis of woolly mammoth teeth from Old Crow in the Yukon, shows that woolly mammoths didn't begin eating plants and other solid foods before the age of 2 and, in some cases, 3. This is much later than customary for modern-day elephants in Africa, according to Jessica Metcalfe, a doctoral student in earth sciences at the University of Western Ontario.

"In modern Africa, lions can hunt baby elephants but not adults. They can’t kill adults. But they can kill babies, and by and large, they tend to be successful when they hunt at night because they have adapted night vision,” Metcalfe explained in a media statement. "In Old Crow, where you have long, long hours of darkness, the infants are going to be more vulnerable, so the mothers nursed longer to keep them close."

She believes this prolonged weaning may have added to the pressures that doomed the woolly mammoths to extinction.

"Today, a leading cause of infant elephant deaths in Myanmar is insufficient maternal milk production," she said. "Woolly mammoths may have been more vulnerable to the effects of climate change and human hunting than modern elephants not only because of their harsher environment, but also because of the metabolic demands of lactation and prolonged nursing, especially during the longer winter months."

Metcalfe, who explains her tooth analysis research in the video above, published the findings in the journal Palaeogeography, Palaeoclimatology, Palaeoecology. Check out the stories below for more information on the woolly mammoth.

• Mammoth DNA could spark resurrection
• Woolly mammoth hair yields fantastic DNA
• Humans drove mammoths to their doom
• Climate change wiped out mammoths

 John Roach is a contributing writer for msnbc.com. Connect with the Cosmic Log community by hitting the "like" button on the Cosmic Log Facebook page or following msnbc.com's science editor, Alan Boyle, on Twitter (@b0yle).

Josh Wickham / Univ. of Florida / IFAS file

Researchers found that a high-mercury diet had an effect on the mating behavior of white ibises confined in a net-covered aviary at the University of Florida. They said the degree of homosexual pairing increased along with the birds' mercury exposure.

A years-long study at the University of Florida suggests that mercury pollution can alter the hormones of white ibises to make males more likely to mate with other males.

"We knew that mercury can disrupt hormones -- what is most disturbing about this study is the low levels of mercury at which we saw effects on hormones and mating behavior," Peter Frederick, a wildlife ecology and conservation professor who led the study, said in a news release this week. "This suggests that wildlife may be commonly affected."

The study was published online on Wednesday in the Proceedings of the Royal Society B.

Frederick and his co-author, Nilmini Jayasena, were hasty to warn against drawing any inferences about the roots of human homosexuality. They didn't set up the experiment to find out what makes birds gay -- rather, they were trying to figure out why ibises in the Everglades went through a stretch of poor breeding in the early 1990s, followed by a baby boom in the late 1990s.

Scientists knew that improvements in the birds' watery habitat was one factor behind the increased breeding, but they suspected that mercury concentrations played a role as well. During the downswing in breeding, low-level mercury contamination made its way increasingly into the Everglades via municipal and medical waste incineration -- but that waste became more regulated at about the same time as the start of the baby boom.

To find out if there was a connection between mercury contamination and a low birth rate, the university set up a 13,000-square foot net-covered aviary in 2005. They brought in 160 ibises, and divided the birds into four groups with equal numbers of males and females. Each group was fed a different diet -- low, medium or high mercury, or no mercury at all. The highest level of mercury was no higher than what the birds would have consumed if they had been in the wild during the early 1990s.

In 2006, about 55 percent of the high-mercury-diet males were nesting with other males. Frederick said the degree of male-male pairing was proportional to the degree of mercury in the diet. That played a role in breeding differences, as well: In comparison with the control group, high-mercury males were less likely to be approached by females during courtship. All of the mercury-consuming males were less prone to perform the ritual head bows and bobs that are part of the ibises' mating ritual.

Frederick and Jayasena, who was Frederick's doctoral student and is now based at the University of Peradeniya in Sri Lanka, reported that the high-mercury-diet females produced 35 percent fewer fledglings than the females in the control group.

The study lasted for three breeding seasons. After the experiment, the birds were put on a cleansing, mercury-free diet for several months and then released back into the wild.

Does all this mean homosexuality is linked to mercury pollution? For the ibises, maybe. For humans, almost certainly not. Frederick pointed out that there have been a number of long-term studies on the effects of mercury on humans, and none of those studies has noted a change in sexual behavior. Speaking more generally, the researchers noted that sexual preference is a much more complex phenomenon for humans than it is for birds.

In a report on Nature's website, a German expert on animal physiology cautioned that the results might not even be applicable to other bird species. Heinz Köhler of the University of Tübingen told Nature that this might be something that's just between ibises. "Their behavior may be less fragile and more robust to methylmercury," he said.

More perspectives on the research:

• Nature: Mercury causes homosexuality in male ibises
• BBC: Mercury 'turns' wetland birds homosexual
• New Scientist: Mercury poisoning makes male birds homosexual
• Miami Herald: Mercury makes ibises gay, study says

Connect with the Cosmic Log community by "liking" the log's Facebook page or following @b0yle on Twitter.