You can't see the tiny injury between the first and second vertebrae of Christopher Reeve's neck, and even if you could, it wouldn't look like much. But Reeve is always aware of the little wound. Ever since he sustained it in a 1995 riding accident, the actor best known for playing Superman has had virtually no movement or sensation below the neck and has been largely dependent on a ventilator to breathe.

Reeve, however, doesn't plan to stay that way. On Sept. 25, 2002, his 50th birthday, he hopes to rise to his feet, lift a glass and toast the people who have helped him through the past few years. "I wouldn't bet the farm on it," he says. "But there's a chance it might happen."

Remarkably, there are other people--sober, scientific people--who agree. For centuries, doctors have considered the spinal cord an impossible thing to heal. Choked by proteins that block regeneration, denied other proteins that foster growth, dammed up by scar tissue at the site of an injury, a spinal cord that gets hurt tends to stay hurt. But for more than a decade, researchers have been learning to overcome these problems, figuring out ways to heal damaged cords and switch the power back on in spines long since gone dead. Even if Reeve and others don't walk by 2002, there is no limit to what may happen in the decades that follow. Says clinical neurologist Ira Black of the Robert Wood Johnson Medical School in Piscataway, N.J.: "There's been a revolution in our view of the spinal cord and its potential for recovery."

Much of what is behind the new hope is a better understanding of why the cord doesn't heal itself. In 1988 neuroscientist Martin Schwab of the University of Zurich isolated substances in the central nervous system whose sole purpose appears to be to block growth. In a healthy spine, the chemicals establish boundaries that regulate cell growth. After an injury, they do little but harm. In recent years, however, Schwab has developed antibodies that neutralize the growth blockers, allowing regeneration to occur. MORE>>

PAGE 1 | 2