“Most people left the field,” Thomas told an interviewer. “They felt that this couldn’t ever be done.” Some physicians felt strongly that the transplants “shouldn’t go on as an experimental thing,” he said.

Thomas found that dogs given whole-body radiation and marrow from their litter mates had the same problems that humans did with infection, rejection and graft-versus-host disease, in which immune cells in the graft attack the patient’s organs as foreign. But occasionally, a dog survived the procedure and remained healthy. Thomas surmised that matching donors to patients was the key.

In 1963, he moved to the University of Washington and developed a system for matching the tissue types of dogs. By the mid-1960s, he showed that most irradiated dogs that received marrow from matched donors survived long term. Around the same time, researchers elsewhere had worked out methods for matching human tissue types. In 1969, Thomas began transplanting marrow from matched siblings to patients with advanced leukemia.

Patients were irradiated in an underground bunker at a former military facility, the only place in the Seattle area with the required radiation sources, then rushed by ambulance to a sterile hospital ward.

“We went back and forth like Ping-Pong balls,” Thomas said.

He inoculated horses to obtain a serum to treat graft-versus-host disease, sometimes getting kicked in the process. Because patients were highly susceptible to infection until the transplanted marrow took hold, Thomas had their food sterilized in an autoclave normally used for germ-proofing surgical tools. Most of his initial 54 patients died of transplant complications or leukemia despite the precautions, but six experienced complete remissions.

These small successes led him to conduct transplants in patients with less advanced leukemia. In 1979, he reported curing half the leukemia patients who underwent transplantation when their disease was in chemotherapy-induced remission. Bone marrow transplants now can cure 70 to 80 percent of the healthiest children and teenagers with leukemia.

Thomas joined the Fred Hutchinson Cancer Research Center, an affiliate of the University of Washington, in 1974, serving as its first director of medical oncology and later as director of the clinical research division. The center has become a world leader in bone marrow transplantation.

Thomas was named professor emeritus of medicine at the University of Washington in 1990. He became a member of the National Academy of Sciences in 1982 and received the National Medal of Science in 1990.

Besides his wife, his survivors include two sons, E. Donnall Jr. and Jeffrey, and one daughter, Elaine Thomas.