Smithsonian Science » predator

Tiny, new brains prove just as adept as large, mature brains among tropical orb-web spiders

John Barrat — Wed, 02 Jun 2010 19:18:24 +0000

When it comes to brains, is bigger better? For instance, can the tiny brain of a newly hatched spiderling handle problems as adeptly as the brain of a larger adult spider? One theory, known as the size limitation hypothesis, predicts that the miniscule brains of tiny creatures contain fewer neurons and thus have less capacity, making these creatures more prone to mistakes.

Photo right: An adult Nephila clavipes on its web with an insect it has captured.

To test this theory, Thomas Hesselberg of the Smithsonian Tropical Research Institute in Panama, recently examined and compared the designs of webs woven by newly hatched, mid-size and adult orb-web spiders of the species Nephila clavipes and Eustala illicita. (Nephila grow explosively and an adult can weigh up to 500 times more than it did when newly hatched.) He collected a number of spiders from each age group, induced them to weave webs on square frames in a laboratory and then took careful measurements of each web. He also conducted observations of the webs of both species in the wild. A paper on Hesselberg’s work was published recently in the scientific journal “Ethology.”

Photo left: An N. clavipes resting at the hub of its laboratory web.

Orb-web spiders have a fixed reserve of sticky capture spiral silk inside their bodies prior to web building. This reserve is thought to be entirely used up as the web is completed. “As orb spiders build the capture spiral from the outer periphery towards the hub [or center of the web], they need to match web size to silk reserves,” Hesselberg writes. Younger spiders that haven’t mastered this behavior, he reasoned, should have a larger area free of silk at their web’s center hub. Or, to cover for a miscalculation, a young spider may increase the distance between the spirals of its capture silk spun out nearer to the web’s hub.

In his observations of the different webs, he found no evidence that adult Nephila or Eustala spiders have any more brain power or build webs any more effectively than newly-hatched spiderlings. “Neither species showed clear signs of being behaviorally limited or more prone to committing errors as spiderlings than were older juveniles or adults,” Hesselberg concluded.

Photo right: The web of an E. illicita spider in the wild. (Photos by Thomas Hesselberg).

One miscalculation may lie, Hesselberg adds, in the human assumption that “the orb web is the result of a demanding computational behavioral process. Theoretical models suggest that the construction of the orb web might be achieved by following a few simple rules of thumb and thus not pose any significant computational challenge for the spider.”

–John Barrat

Researchers discover treefrog embryos can evaluate different features of vibrations

John Barrat — Wed, 16 Dec 2009 16:16:58 +0000

Stuck to a leaf like a blob of jam, the gelatinous egg clutch of the red-eyed treefrog (Agalychnis calidryas) of Central America is a food favored by arboreal snakes. Five minutes is all it takes for a snake to gobble down a clutch of a few dozen frog eggs. Yet, as their unhatched siblings are being devoured, some of the red-eyed treefrog embryos can sense the vibrations of the snake’s feasting and hatch prematurely. Squirming and wriggling the tadpoles escape by dropping from the leaf and into a pond below.

Photo: A female red-eyed treefrog deposits a clutch of eggs on a leaf overhanging a pond.

Recently, researchers from Boston University and the Smithsonian Tropical Research Institute (STRI) in Panama have been taking a closer look at the vibrations that red-eyed treefrog embryos use as cues to trigger early hatching. The researchers—Michael Caldwell and J. Gregory McDaniel of Boston University and Karen Warkentin of both Boston University and STRI—embedded tiny recording devices into red-eyed treefrog egg clutches and recorded the low-frequency vibrations caused by snakes as they ate the eggs and also by tropical rain storms. They played back these vibrations to egg clutches in a laboratory and found the embryos hatched in response to the snake-generated vibrations and not to the rain vibrations.

This infrared video recording show a parrot snake (Leptophis ahaetulla) attacking a red-eyed treefrog egg clutch. Some of the treefrog embryos escape the snake by hatching early.

This experiment showed that the embryos were responding to the vibrations and didn’t need chemical or visual cues from snakes, Warkentin says, and that the embryos could differentiate between snake vibrations and rainstorm vibrations.

What puzzled the researchers was that the low-frequency vibrations that triggered the embryo hatchings were in some ways very similar to vibrations caused by many benign forest stimuli—such as rainfall, wind or non-predatory animals. “Hatching early in response to benign stimuli would be a serious error” Warkentin explains, “since premature hatchlings are vulnerable to predators in the pond. We reasoned that if a defense—such as hatching early—is costly for prey, then mechanisms should be in place to avoid false alarms, just as mechanisms exist to recognize and defend against predators.”

Photo: The device in this egg clutch is an accelerometer, used to record vibrations. The vibration recording from a snake attack on an egg clutch is shown below.

In recent laboratory work the scientists improved their understanding of how these false alarm mechanisms work in regard to rainfall. The low-frequency vibrations generated by tropical rainstorms are accompanied by two elements that snake-feeding vibrations do not have: high frequency vibrations and an initial buildup of intensity. When these two features were removed from rainfall recordings and the edited recordings were played back to embryos in the lab, many of the embryos hatched. The intensity buildup and high frequency vibrations played simultaneously with low frequency vibrations indicative of danger suppressed the hatching response.

“Although we don’t know the precise mechanism—whether, for instance, the high-frequency vibrations mask perception of the low-frequency vibrations that indicate danger, or whether the embryos perceive high frequency vibrations as indicative of safety—we do know that the embryo behavior is affected by these two features of rain vibrations in a way that benefits them,” Warkentin says.

A paper on this research: “Is it safe? Red-eyed treefrog embryos assessing predation risk use two features of rain vibrations to avoid false alarms,” was published in the journal Animal Behavior at the Web address: www.elsevier.com/locate/anbehav

Photo: A male red-eyed tree frog (Photos and video by Karen Warkentin)

The Web site for the Warkentin Lab at Boston University is: people.bu.edu/kwarken/

—John Barrat

Appalachian Trail survey aims hidden cameras at large predators

John Barrat — Wed, 20 May 2009 16:56:18 +0000

Describing his project of counting bears, bobcats and other predatory mammals along the Appalachian Trail, National Zoological Park wildlife ecologist William McShea looks to American literature for a comparison. “This is a big Tom Sawyer-type thing, where I want all this work done, and I want to get volunteers to do it for me,” he says with a chuckle. “So far, it’s working.”

Like Mark Twain’s fictional Tom Sawyer —who, when faced with the burden of whitewashing 30 yards of board fence, persuaded other boys to paint it for him—McShea has rounded up 100 eager volunteers (“who are totally jazzed,” according to the scientist) to collect data for him in the woods along the famous footpath that runs from Georgia to Maine.

For now, McShea’s project covers just a portion of the trail, a 570-mile stretch from the southern border of Virginia to the northern boundary of Maryland. His volunteers, most of them recruited from hiking clubs that maintain the trail, are responsible for setting up some 50 cameras at predetermined points along the route, leaving them in place for a month, and then moving each camera to a new spot.

Infrared sensors allow the cameras to take a photograph whenever an animal strays within range of the lens. The project’s first phase ran from April through November, during which time McShea’s volunteers set up cameras to take animal snapshots at about 350 locations along the Appalachian Trail.

With luck, the photo-shoot project should do more than tally wild carnivores across a ribbon of eastern woodland; it also should tell McShea something about the condition of the landscape those animals roam.

“The Appalachian Trail really doesn’t change that much from Georgia to Maine,” explains McShea, who works at the National Zoo’s Conservation and Research Center, in Front Royal, Va. “It’s mature oak forest up on the top of a ridge, and it just keeps going and going. But what changes is the surrounding landscape. The trail goes through suburbia, through national forest; it passes major highway systems.”

Because predators, especially larger ones, need to cover lots of territory to get an adequate diet, recording their presence or absence as the trail snakes across the varied habitats of the eastern United States measures the fitness of those habitats for wildlife, providing what McShea calls “an index of wildness along the trail.”

The project came about when McShea attended a 2006 conference at which officials from the National Park Service were seeking ways to use volunteers to gather environmental data about the trail. McShea suggested the predator survey, modeled after ones he has done in China and Malaysia, where he trained staff at wildlife reserves to set up and use cameras to take a census of animal populations.

By early 2007, McShea was training volunteers in this country in the fine points of automated wildlife photography. He taught them to strap cameras to trees at about knee height. “Most animals are shorter than you think they are,” he explains, and the camera’s infrared sensor needs to be low enough to be triggered, not just by bears but by foxes and raccoons.

To avoid creating a gallery of hiker portraits, cameras were set up away from the Appalachian Trail itself but along nearby animal trails. Volunteers also were given directional coordinates within a predetermined segment of the Appalachian Trail and instructed to set up their camera within 100 meters of that location.

In addition to cameras, “the stink” was provided to aid McShea’s helpers. “That’s what we call it,” says Ricki Ashcraft, an education specialist at the Conservation and Research Center, who as a volunteer is helping to manage three cameras along a stretch of the Appalachian Trail in Shenandoah National Park. This scent lure, extracted from animal musk glands, is obtained from trapping-supply companies. “It smells to high heaven,” she says. But a drop left on a stick or stone in front of a camera will make a passing predator pause just long enough for the project’s digital cameras to wake up from “sleep mode” and get the picture.

McShea says his volunteers are enthusiastic about the work because “they maintain sections of the trail and they’re always wondering, ‘What’s here? Do they have bobcats on their section of trail, or bears or weasels?’ Some are even doing it because they hope they may photograph a mountain lion.”

For the record, McShea does not believe wild cougars still roam the Appalachians. But perhaps with a nod to Tom Sawyer, he says he’s promised “a bonus” to the first volunteer who records one.

What the project’s cameras have captured is an abundance of bears, bobcats and coyotes; a handful of startled hikers; both red and gray foxes; and a surprisingly small number of raccoons, opossums and skunks.
After shutting down for the winter, McShea intends to resume his experiment in what he calls “citizen science” for another seven months beginning in the spring. Then, having honed procedures for using trail-club volunteers to set up cameras, record habitat information and sprinkle stink, he hopes “to talk someone into letting me do the entire trail.”

That “megatransect” along the entire 2,175-mile Appalachian Trail would, like the current project, seek to determine how the East’s wild carnivores are coping within the variety of natural and highly developed landscapes that bracket the trail through 14 states. Like much of McShea’s research, whether studying giant pandas in China or deer in Virginia, the work has a practical bent.

“I’m trying to be helpful to land managers,” McShea says, “trying to give the guys who work in the Park Service and the Forest Service and the state game agencies information that helps them do their jobs better. And I think this distribution-of-predators study is something they can use to say whether they’re doing a good job or not.”