The world’s first atomic bomb was detonated on July 16, 1945, in the New Mexican desert — a result of a highly secretive effort code-named the Manhattan Project, whose nerve center lay nearby in Los Alamos. Just 49 months later, the Soviets detonated a nearly identical device in Central Asia, and Washington’s monopoly on nuclear arms abruptly ended.

How Moscow managed to make such quick progress has long fascinated scientists, federal agents and historians. The work of three spies eventually came to light. Now atomic sleuths have found a fourth. Oscar Seborer, like the other spies, worked at wartime Los Alamos, a remote site ringed by tall fences and armed guards. Mr. Seborer nonetheless managed to pass sensitive information about the design of the American weapon to Soviet agents.

The spy fled to the Soviet Union some years later; the F.B.I. eventually learned of his defection and the espionage but kept the information secret.

His role “has remained hidden for 70 years,” write Harvey Klehr and John Earl Haynes in the current issue of Studies in Intelligence, the C.I.A.’s in-house journal; their article is titled “On the Trail of a Fourth Soviet Spy at Los Alamos.” In separate interviews, the sleuths said they were still gathering clues regarding the exact character of Mr. Seborer’s atomic thefts.

Mr. Klehr is an emeritus professor of politics and history at Emory University, and Mr. Haynes is a former historian for the Library of Congress. Both have written books on Soviet spies and American communism, often together. Their tale has an eerie resonance at a time when Russian intelligence agencies are again at the center of American life.