Larry Curtiss (above) and his faculty advisors pieced together a contraption that could do what no one thought was possible and ultimately led to modern fiber optics—at a cost of about $250 for their earliest machines.


This is an article from the fall 2018 issue of LSA MagazineRead more stories from the magazine.


For their lunchtime games of bridge, a handful of physics faculty relaxed with playing cards, food, and coffee in the Randall Laboratory basement on central campus. For a young physics student in the 1950s like Larry Curtiss (B.S. 1958), he says all these years later, “It was a wonderful place for an undergraduate to go to ask an occasional question and get quick advice.” Tapping the group at the card game, a student could get several different and useful answers about how to solve an issue in the lab.

Curtiss was trying to build a long, flexible instrument that could snake down through the mouth, throat, and esophagus and project images of the stomach: a gastroscope. Since his sophomore year, Curtiss had worked on a team with his physics advisor, Professor Wilbur Peters, and a resident in the U-M Medical School, Basil Hirschowitz. Their most stubborn problem was that test images looked faded and washed out: Light was leaking out of the hair-thin glass fibers in the fiber-optic bundle that formed the important snaking part of the instrument.

The card-playing professors suggested that Curtiss should insulate each glass fiber with a thin layer of plastic. That way, he could limit the light’s path through each individual fiber. When Curtiss wondered aloud whether glass might work as a better material to insulate the strands, the erudite professors scoffed at his silly idea. 

But after getting laughed out of the room and putting the professors’ advice into practice, Curtiss says that his attempts at plastic coatings “began three months of futile effort.”

Then, Curtiss quietly started trying his own solution in the lab while his peers and supervisors were out of town at a conference. “So I wouldn’t be embarrassed if it turned out to be a fiasco.”

He bent over the contraption that he’d built with help from the old Physics Machine Shop. Many of its parts came from odds and ends they’d found around Randall Lab. Curtiss included an empty can of oatmeal to carefully wind thin glass strands around as they cooled coming out of the small, cylindrical furnace that had melted the glass.

He placed a glass tube into the top of his furnace and slipped a glass rod through the tube’s center. Both dripped from the heat, and Curtiss slowly pulled a thread through the bottom of the furnace, thin as a hair, with an outer glass coating over an inner glass strand.

He was giddy with the great result: “I was walking backward down the hall, and at 40 feet away, I could still see the glow of the furnace through the fiber,” he says.

“That was my moment of joy.”


Curtiss and the team patented their invention in 1956 as a “flexible light transmitting tube.” After first testing it on himself, Hirschowitz used the new gastroscope to examine a patient’s stomach in 1957.

Advancing at Light Speed

A century before Curtiss, early experimenters with stomach scopes had to practice with inflexible instruments on cadavers and sword swallowers. 

The procedure was dangerous. Early scopes often tore holes in the esophagus, knocked teeth out, and burned organ linings with hot lamps—the only technology available to light inner cavities enough to see. Doctors gave up on trying to make good gastroscopes, likely due to horrible accidents with patients.

When Curtiss landed on his solution, the discovery was profound. Physicians now could look inside the human body without cutting it open and without traumatizing the patient. And Curtiss had addressed a tough technical problem and made fiber optics possible: People now could bend light around corners with cheap, flexible glass, and without losing much brightness. In some ways, his insight led to major advancements in telecommunications, helping replace bulky copper wires with sleek fiber optics.

In retrospect, the idea came from simple reasoning in physics.

“If you ever try to move a wagon by pushing it with your leg to get it away from the grass and onto the sidewalk, the wagon always wants to turn and go back on the grass,” says Curtiss. “That’s because it goes a little bit faster on the sidewalk than it does on the grass.

“Light’s the same way,” he says. “It wants to turn away from where it goes faster and move into the space where it goes slower.”

Curtiss had coated his inner strand of high-refractive glass, where light moves more slowly, with a thin layer of low-refractive glass, where light moves more quickly. Light entering one end of the glass fiber would travel through the inner glass, constantly reflecting away from the outer glass that surrounded the inner fiber. With nowhere to go but bounce straight through the inner glass strand, the light could follow that path all the way through to the other end of the fiber without leaking out at all. Physics calls it “total internal reflection.”

With a bundle of these glass fibers, Curtiss, Peters, and Hirschowitz could transmit light, and thus an image, through bends and coils in the flexible bundle—from the stomach into their view through an eyepiece outside the body. They’d been working on the project for just two years. After testing their new fiber-optic gastroscope on Hirschowitz, they soon used the scope on patients with great success. 

What made the gastroscope work was physics, of course, but also the collaboration between physicists who understood how light behaves and a doctor who knew how to poke around inside a patient. 

Taking off for the Summer

The instrument’s use spread so quickly that in ten years, thousands of fiber-optic gastroscopes had sold. By the 1970s, the technology advanced so rapidly that new models quickly became obsolete. The fiber-optic scope became the standard among doctors examining stomachs and soon extended to other organs, and dentists, and microscopes. Even Rolls-Royce could peer inside their airplane engines without dismantling them, thanks to the invention.

Curtiss finished his physics degree at U-M and headed to grad school, working on the side with the company that made the first commercial fiber-optic gastroscope. When the company asked Curtiss to spend the summer helping with the product, he assumed that they’d finish the project in time for fall classes. 

“I went down for the summer to New York City,” he says, “and that summer was 22 years long.” He never did get his Ph.D.

Now 82 years old, Curtiss worked in the medical instrument industry for his whole career, “pretty much having a ball, doing a lot of the things I found fun to do,” he says.

He’s modest about his glass insulation technique. But Curtiss has gotten patents for his work, and an early gastroscope he built has been held at the Smithsonian Institution since 1989.


Top photo: Courtesy of U-M Bentley Historical Library. Inline photo courtesy of Smithsonian Institution Archives