Robert Barlow sees the world through a horseshoe crab's eyes -- at least, through two of the creature's 10 light-sensing organs. And what he has seen is surprising.

The lowly limulus -- not a crab at all, actually, but an arachnid (a spider cousin) -- is certainly one of nature's weirder children. It looks like a refugee from the set of a bad science fiction movie -- say, "Planet of the Cheap Special Effects." So primitive that scientists commonly refer to it as a "living fossil," it bears a striking resemblance to 350-million-year-old impressions left in rock by its forebears.

But this animal oddity has long been a favorite object of scientific study and has yielded numerous insights about fundamental biological processes. Each of the horseshoe crab's lateral eyes contains about 1,000 photoreceptors, known as ommatidia, and each one is about 100 times bigger than the cones and rods found in the human eye, making the limulus eye an enticing model for vision researchers. Nature is ever thrifty: Systems that work well in a lower life form often continue to be used by creatures further up the evolutionary chain. And so the eyes of the horseshoe crab have, through the years, given us a peek into our own.

Researcher H. Keffer Hartline did pioneering work on the horseshoe crab's eyes beginning in the 1920s, measuring impulses from the nerves and showing how the individual ommatidia influenced each other, working together to accentuate contrast and enhance the animal's perception of the edges of things. This phenomenon, lateral inhibition, turned out to be a highly "conserved" trait throughout evolutionary development and is important in human senses such as vision, touch and hearing. Hartline's work earned him the Nobel prize for physiology or medicine in 1967.

But what does the horseshoe crab use its eyes for? Hartline often joked that he had spent his life "studying vision in a blind animal." Barlow, who worked with Hartline at the Marine Biological Laboratory here and has spent summers at the lab studying horseshoe crabs since 1962, set out to determine what the animal uses its eyes for -- and to move from the study of how individual neurons behave to the more complex task of constructing a realistic model of how animals see. Zeroing In on a Mate Barlow and his fellow researchers feel they solved the first question a few years ago, when they were able to show that the horseshoe crab is very, very good at seeing other horseshoe crabs, the better to mate with them. To prove that the animals are not guided to each other by scent, Barlow put cement casts of the animals into the water of shallow coves where they would normally come to mate. Even on dark, cloudy nights, male crabs approached the casts and attempted to mate with them. That finding should not surprise anyone familiar with basic concepts of evolution, Barlow explains: "You have to do three things in order to survive in this world -- you've got to eat, avoid being eaten and procreate."

Several systems have evolved to help the horseshoe crab find its mate. The creature's eyesight is closely regulated by its biological day/night cycle, or circadian rhythm. Unlike most other animals, whose eyes contain both rods and cones to adapt to changes in light levels, the horseshoe crab's lateral eye contains only one kind of receptor, the ommatidium -- a collection of cells that adapts to day and night by an internal clock that changes the way the receptor operates. As night falls, the brain sends signals to the animal's lateral eyes, which stimulate the release of a chemical, octopamine. (Similar brain signals, or "efferent inputs," have been found by researchers in other sensory systems and animals.) The octopamine cranks up the receptor's sensitivity a millionfold to pick up the faintest glimmers of light. This system, along with a host of other nightly performance-improving changes, allows males to spot females equally well by day or night. Cracking the Code Now, Barlow, a professor of neuroscience and bioengineering at Syracuse University, is working with researchers Frederick Dodge of Syracuse and Maureen Powers of Vanderbilt on the second task: cracking the horseshoe crab's neural code for vision. The horseshoe crab's eye is "complex enough to be interesting, yet simple enough to be understood," he said. "Our goal is to look and see if we can understand how this information is used by the brain."

The researchers, with students Erik Herzog, Christopher Passaglia and Darrell Porcello, have been re-creating the crab's visual network on a computer. But not just any computer: The scientists use a "massively parallel" supercomputer, which employs many processors linked together to solve problems. Researchers believe the system can be programmed to resemble the network of neurons in living systems. They fed in equations that decades of research have shown to describe the actions and interactions of retinal cells, and then they directly transmitted signals from the animal's optic nerve into the computer. The computer processed this input, using the equations, and converted it into an image on a video screen. They compared the image produced by the computer with a video feed from a tiny camera attached to the animal's shell. Surprising Sophistication The researchers hoped to see two similar images, but what showed up on their screens last May startled them. When they showed the animal a low-contrast underwater scene of a gray cylinder against a gray background, the camera barely picked up the image -- but the computer-generated image showed the cylinder clearly. Somehow, the neurons in the crab's visual system were working together to produce far more contrast and definition than the researchers expected to see.

The computer program thus revealed an emergent property of the crab's visual system -- an unexpected aspect that the simulation helped make apparent. Barlow now believes the limulus eye is specially tuned to make shapes that are important to it pop out from the background, operating as a "global feature detector" that makes a nearby horseshoe crab all but impossible to miss.

"It looks out into the world and enhances {the image} -- plucks it out from the physical world and sends it on to the brain. . . . We had no idea that the eye was as sophisticated as this."

The research, funded by two institutes of the National Institutes of Health and by the National Science Foundation, is continuing. Next, the researchers will analyze their model to get a better understanding of how the eye generates its signals. That research will dovetail with other work in their lab measuring signals within the animal's brain.

The success of the recent experiment, not yet published, gives Barlow even greater respect for the nervous system of this supposedly primitive animal. Simulating just one of its functions requires a computer that fills a room and costs millions of dollars.

"This little eye of a horseshoe crab does what it does in less time and in less space," said Barlow. "One appreciates how complex nervous systems like the one between our ears have to be." JUMP INFORMATION APPENDED FROM FILE N:\NEP\DAYS\071095\07100158.NVP Page: 00

A DIFFERENT POINT OF VIEW For six decades, the 10-eyed horseshoe crab has served as a powerful model for scientists seeking to understand vision -- particularly in studies on the exchange of messages between the animal's brain and its two lateral eyes. SOURCE: Scientific American CAPTION: Researcher Robert Barlow has tapped into the optic nerve of horseshoe crabs and compared their signals with the images from the "crab cam" attached to the animal's carapace, or shell. Above, a fully loaded test subject heads into the water, attached to monitoring equipment by a coaxial cable. A close-up of the horseshoe crab's lateral eye (left) shows the array of hundreds of ommatidia, or light receptors.