It was nothing more than the letter "H," yellowish in color and the size of the head of a match, but to Harold Churchey it looked a lot like a miracle.

Blind for decades, Churchey saw the letter in his dysfunctional right eye while serving as a test subject in an intricate and intensive effort at Johns Hopkins University to artificially replace damaged eyes. A team of researchers there, and scientists at five other sites worldwide, are attempting to do something usually left to faith healers: bring sight back to the blind.

"I've never looked at the face of my grandson," said Churchey, a Sharpsburg, Md., resident and for many years the manager of a snack bar in the Hagerstown courthouse. "But I do believe I will before I die."

In place of the damaged human workings of Churchey's 72-year-old eyes, the researchers are trying to produce an artificial vision system that can electronically capture and transmit images to the brain to create sight. The system would consist of a computerized miniature video camera mounted on a pair of glasses, and a sophisticated computer chip that would be surgically implanted in the eye. The camera would transmit its image to the computer chip, which would be connected to tissue in the back of the eye called the retina so it could transmit images to the brain.

The Hopkins researchers have been working on the idea for a decade, and they remain a few years away from their goal of permanently implanting a chip in a blind person. So far, they have tested only a crude version of the system. Nonetheless, and despite some daunting obstacles that remain, they are convinced they will some day bring vision to people who now have none. And if they succeed, 2 million Americans who are blind because their retinas are damaged will have an option most never imagined possible.

"When we first proposed the idea of a retinal microchip, most if not all ophthalmologists believed that no one would ever see using [one]," said Mark S. Humayan, leader of the Hopkins team at the Wilmer Ophthalmological Institute. "This belief was obviously based on incomplete information. . . . We're convinced that achieving some level of vision--enough to allow a person to walk through a room and maybe recognize a face--is a realizable goal."

Humayan and his efforts attracted widespread notice last week when information leaked out that singer Stevie Wonder had talked with the doctor recently about whether he might be helped by a retinal microchip implant.

Wonder, who has been blind since soon after his birth, does not have the kind of blindness most likely to be helped by Humayan's chip. Researchers believe the people most likely to benefit are those suffering from retinitis pigmentosa, a rare inherited disorder that causes blindness by destroying the retina, and age-related macular degeneration, another condition that causes a deterioration of the central part of the retina in the elderly.

But Humayan said during a news conference yesterday that he would not rule out working with Wonder until he had fully tested his vision. Humayan also said that regardless of whether he might be helped someday, Wonder had voiced interest in conducting a telethon or some other effort to "raise consciousness about what can be done in this field."

Vision experts said the retinal implant project is one of several lines of research showing promise for restoring sight to the blind, although the other, less invasive approaches are probably further away from therapeutic use.

Peter Dudley, program director for retinal diseases at the National Eye Institute, said researchers have identified the genetic abnormalities that cause many types of blindness and are experimenting with ways to repair those genes. In addition, he said that growth factors that promote regeneration of damaged cells in the eye also have been identified and hold promise. These biological approaches, he said, offer more of an actual cure than do the implants.

Dudley's institute is one of several sponsors of Humayan's work--including the National Science Foundation--and the institute will fund his upcoming animal trials of the retinal chip implant. The animal studies, which Humayan said will begin soon, will be conducted on a group of blind dogs provided by the National Institutes of Health.

During the dog experiments, the full artificial vision system will be tested for the first time. If successful, Humayan said, they will lead to human studies and an application to the Food and Drug Administration for approval. The Hopkins team, along with two other groups working on related projects at Harvard University and in Illinois, has applied for $12 million from the NIH.

According to Humayan, Churchey and his twin brother are two of 17 blind people who have been successfully tested in the initial phases of the research. Those tests involved the brief implanting of tiny electrodes into the eye. The electrodes were connected to a computer chip outside the eye, which sent out impulses representing various shapes. The patients, most of whom had never seen light with the eye that was being tested, did see the light and were able to make out some shapes and sizes--thereby proving that remaining neurons in the eye can be artificially activated.

The computer chip being used was designed by Wentai Liu, a professor of electrical engineering at North Carolina State University. Liu described the chip as similar to microprocessors in general use today, but with a very specialized design.

"Our biggest challenge is to learn how to encase the chip so it can be implanted in the eye," Liu said. Fluids in the eye have to be kept outside the chip, and the chip has to kept from further damaging the retina, which has been likened in its fineness to wet tissue paper. It particular, the chip has to operate without so much power that its heat causes damage.

The implant requires complicated technology, and Humayan has been predicting imminent success for quite a few years. In fact, Humayan acknowledged yesterday that he told Churchey back in the mid-1990s that he probably wouldn't have to wait much longer for an implant.

Nonetheless, Churchey said he was eager to take part in any more tests Humayan might require.

"From the very first time they put that probe in my eye, I knew they were on the right track," he said. "People will do anything to get sight."

Simulated Sight

The "eye chip" would simulate eyesight by bypassing defunct neurons and sending electrical signals directly to the optic nerve. Here is how it would work.

1. Cameras attached to eyeglasses transmit images as radio waves.

2. A computer chip embedded in the back of the eye picks up the signal.

3. The chip converts the signal to electrical impulses, and sends them through the optic nerve. Vision centers in the brain interpret the impulses.