MOSCOW, OCT. 3 -- A week after American scientists exploded the world's first atom bomb in the desert of New Mexico, President Harry S. Truman had a meeting in the German town of Potsdam with Joseph Stalin. Attempting to strike as casual a note as possible, he told the Soviet leader that the United States had discovered a "new weapon of unusual destructive force."

After mentally preparing himself for a barrage of questions from his suspicious wartime ally, Truman was amazed that Stalin showed no particular interest in his revelation. It now turns out that there was a very good explanation for the Soviet dictator's cool response: Stalin could well have been better informed about the making of the U.S. atom bomb than Truman himself.

Until he succeeded Franklin Roosevelt as president in April 1945, Vice President Truman knew very little about the top secret Manhattan Project. Recently declassified Soviet espionage documents and interviews with former Soviet intelligence agents make clear that copious information about the development and construction of the bomb had been available to the master of the Kremlin since the fall of 1941.

While the bare outlines of one of the espio-nage coups of the century have been known to Western governments since the trial of Julius and Ethel Rosenberg in 1951, the full extent of Soviet penetration of the U.S. atom bomb project is only now becoming apparent. In an interview with The Washington Post, former Soviet spy master Anatoly Yatskov claimed that the FBI had succeeded in uncovering "only half, perhaps less than half" of his network of agents in the United States.

"No other intelligence service in the world succeeded in penetrating the wall of secrecy around the Manhattan Project," boasted Yatskov, who is better known in the United States as Anatoly Yakovlev, the Soviet diplomat who was indicted in the Rosenberg case but, having returned to Moscow in 1946, was never brought to trial. "We {intelligence agents} performed an important service. We helped Russia achieve strategic parity with the United States much sooner than would otherwise have been the case."

While acknowledging that some of his information came from the British atom spy Klaus Fuchs, Yatskov insists that another scientist inside the experimental nuclear station at Los Alamos, N.M., also was passing secret information to Moscow. Soviet intelligence documents indicate that the agent, who Yatskov said is still alive, joined the Manhattan Project in 1942, at least 18 months before Fuchs arrived in the United States. This agent was code-named "Perseus," or simply "Mr. X."

Perseus evidently played such an important part in the network that -- if his existence is confirmed -- it would relegate the Rosenbergs' role in supplying atomic secrets to Moscow to the periphery.

A testimonial to the extraordinary achievements of the Soviet spy network in Europe and the United States came from Igor Kurchatov, the father of the Soviet A-bomb. In recently published notes to his Kremlin masters, written in March 1943, Kurchatov said information received from intelligence sources would enable Soviet physicists to resolve all problems connected with the splitting of the atom in a "significantly shorter period."

"The receipt of these materials has a huge, invaluable significance for our state and science," enthused Kurchatov, the only Soviet scientist with direct access to the Western intelligence material. "Now we have important guidelines for subsequent research that enable us to bypass many laborious phases involved in tackling the uranium problem and reveal new scientific and technical ways of solving it."

Almost half a century after the launching of the Manhattan Project, some of the information obtained by the Kremlin's atom spies is still extremely sensitive. The planned publication of detailed intelligence documents obtained from Los Alamos in an obscure Russian scientific journal was held up this week after U.S. and Russian officials complained that they could enable a "third party" to build an atom bomb.

This account of how the Kremlin stole the West's atom secrets has been pieced together from interviews with Russian scientists and intelligence agents and a flood of material that has appeared in the Russian press over the past year. The final piece in the jigsaw puzzle was provided by the scientific article, which will appear in the journal Questions of the History of Natural Sciences and Technology and was made available to The Washington Post ahead of publication.

By invading the Soviet Union on June 22, 1941, Adolf Hitler made politics a good deal simpler for left-wing scientists and intellectuals in the West. Up until that moment, many of them had found their sympathies toward the Soviet Union shaken by the 1938 Nazi-Soviet nonaggression pact. But as the German armies swept through Ukraine, their doubts were swept away. Their duty was now clear: to do everything in their power to help the Soviets bear the brunt of Nazi aggression.

Although the Nazi invasion was a military disaster for the Soviet Union, it produced an intelligence gold mine. Within months of the German attack, recently declassified Russian espionage documents make clear, information about the most secret Western weapons systems began pouring into Moscow.

On Sept. 25, 1941, the head of Soviet intelligence in London, Anatoly Gorsky, sent a memorandum to Moscow about a session of the nuclear advisory committee to the British war cabinet that had been held nine days earlier. British scientists, he reported, were confident that a uranium bomb could be produced within two years.

The source of the top secret information was an agent called "Leaf," since identified as the British diplomat Donald Maclean, a member of the famous Cambridge spy ring. In addition to technical details about the construction of an atom bomb, Maclean revealed that the British were devoting top priority to the building of a uranium factory.

Confirmation that the Western allies were on their way to developing the world's first atom bomb soon came from a Soviet agent in New York, Morris Cohen. As an idealistic young Communist, Cohen had volunteered to fight on the republican side in the Spanish Civil War. While recuperating from a leg injury in Barcelona in 1938, Cohen had been recruited to work for Soviet intelligence. His network of contacts with other left-wing sympathizers was to prove invaluable to Moscow in the years ahead.

Yatskov claims that Cohen was approached in New York by an "acquaintance," a physicist who said he had been invited to take part in top secret work on the building of an atom bomb. Soviet intelligence agents in New York passed this information on to Moscow, along with a recommendation to recruit the physicist, who was to become Perseus.

In an apparent attempt to conceal the physicist's true identity, Soviet intelligence officials are vague about when Perseus began working for Moscow. But circumstantial evidence suggests that this must have taken place between September 1941 and July 1942, when Cohen left New York and joined the Army.

An article in the weekly journal New Times in April 1991 by Col. Vladimir Chikov, a public relations officer for the KGB, maintains that Perseus acted out of purely ideological considerations and refused an offer of money. According to Chikov, Perseus was convinced that the Americans intended to use the bomb not against Nazi Germany but against the Soviet Union.

"The Pentagon is of the opinion . . . that it will take the Soviet Union decades to harness atomic energy. In the meantime, America will destroy socialism by means of the uranium bomb," Perseus is quoted by Chikov as saying.

There are some contradictions and inconsistencies in Chikov's version of how Perseus was recruited, which make his New Times article a somewhat unreliable source. But unless the Russian intelligence agency is playing a gigantic hoax, there seems little doubt that the Kremlin had such an agent. The latest batch of intelligence documents contains information that could only have come from a scientist with direct access to the innermost secrets of the Manhattan Project.

The recruitment of Perseus in the United States appears to have roughly coincided with the recruitment of Fuchs in Britain. A refugee from Nazi Germany, Fuchs had been working on a top secret atom project code-named Tube Alloys. In late 1941, as German troops were sweeping toward Moscow, he volunteered his services to the Soviet Union. His first contact, he later confessed, was the London station chief of the GRU military intelligence agency.

In Moscow, meanwhile, Stalin and his aides were too preoccupied with the German invasion to pay much attention to the information flowing in from their intelligence agents. It was not until March 1942 that the commissar for internal affairs, Lavrenti Beria, got around to briefing Stalin fully about the Western plan to develop an A-bomb.

At first, both Stalin and Beria were skeptical about the intelligence reports. Beria feared that the Western allies might be attempting to trick the Kremlin into vast expenditures of money and manpower. "If this is disinformation, I'll throw you all into jail," the secret police chief roared, according to Yatskov.

The paranoia of Stalin and Beria was somewhat allayed by the enthusiastic response of Kurchatov, the Soviet Union's top nuclear scientist. In a note dated March 7, 1943, Kurchatov reported that information from Britain (apparently from Fuchs) would force Soviet scientists to revise many of their previous theories about nuclear reaction and the production of uranium-235.

Kurchatov conceded it was possible that the material could be a hoax, designed to distract Soviet scientists from other more important tasks. On balance, however, he felt the information was probably genuine. "There are some conclusions that seem to me doubtful, but the responsibility for them probably lies with English scientists rather than the quality of the {intelligence} information," he wrote.

Kurchatov's remaining doubts were soon swept away by important new intelligence information arriving from the United States (presumably from Perseus). On March 22, 1943, he reported that the American research suggested "an entirely new approach to the entire problem of uranium" that was "exceptionally attractive."

The new American approach described by Kurchatov centered on the use of plutonium rather than uranium-235 as the basic element in the atom bomb. Plutonium is easier to produce than uranium-235 because it does not require the same elaborate separation process. It was a plutonium bomb that the Soviets eventually tested in August 1949, four years after the first American test.

Excited by the new developments, Kurchatov asked that Soviet intelligence agents in the United States be instructed to gather more information from their American sources. He dictated a series of technical questions about the properties of plutonium that needed to be clarified.

Stalin had meanwhile authorized the establishment of a special research facility in Moscow, Laboratory Number Two, now known as the Kurchatov Institute, to develop a Soviet A-bomb. A team of leading scientists was assembled to work under Kurchatov's direction, but nobody else was told where he was getting his ideas. Many attributed the exciting new discoveries to the scientific genius of their director.

At some point in early 1943, Moscow received the first detailed description of the top secret Los Alamos facility in New Mexico, the main development site for the American A-bomb. A message from Vasily Zarubin, the head of Soviet intelligence in New York, reported that security arrangements at the camp were exceptionally tight. Nevertheless, he boasted, "we are taking further measures to plant our reliable sources in Los Alamos."

A 28-year-old intelligence officer, Anatoly Yatskov took up his low-ranking cover post at the Soviet consulate in New York in early 1941. His Soviet passport had been issued in the name of Anatoly Yakovlev -- and this is the name by which he was identified at the Rosenberg trial.

By the time Yatskov was assigned to full-time work on the bomb project in early 1943, Morris Cohen had left New York and joined the Army. But his wife Lona agreed to cooperate with Soviet intelligence. According to Yatskov, Lona Cohen undertook two courier missions to Albuquerque to meet Perseus on his behalf.

One of the two missions, Yatskov recalls, almost ended in disaster. After picking up the intelligence material, which was stuffed in the bottom of a box of tissues, Lona Cohen was startled to find plainclothes policemen checking the bags of passengers boarding the Chicago-bound train at the Albuquerque station. She decided to board the train at the last moment. She asked the plainclothes man to hold the box for her while she searched for a ticket and ID. After searching her bags, the agent handed back the box.

"When she got back to New York and gave me the materials, she joked that they had been in the hands of the police. It was a really dangerous moment," said Yatskov, who retired in 1985 after stepping down as the head of a faculty at the KGB's spy training college in Moscow.

Yatskov also acted as the Soviet controller of Harry Gold, the Swiss-born chemist who served as a courier for Fuchs after he joined the Los Alamos project in mid-1944. But he denies knowing the Rosenbergs, who were named as Soviet spies by Gold in 1951 and later executed. It is impossible to tell whether his denial is genuine or motivated by a wish to support the repeated claims of innocence by the Rosenbergs.

Unlike Gold, Morris and Lona Cohen never attracted the attention of law enforcement agencies while they were in the United States. They later reappeared in Britain under the names Peter and Helen Kroger. Using the cover of antiquarian booksellers, they provided technical support for the celebrated KGB sleeper agent Gordon Lonsdale until their arrest in 1961. They eventually were exchanged for a British agent in the Soviet Union and now live in Moscow. Their activities in the United States did not become public knowledge until the appearance of the New Times article last year.

Russian intelligence officials responded to a request for an interview with the Cohens by saying that Lona was dying of cancer and Morris was at her bedside. But in a recent telephone conversation with the American espionage historian Walter Schneir, Lona Cohen acknowledged that she had made at least one trip to Albuquerque.

Asked whom she had met in Albuquerque, Cohen replied that she could not remember. "A scientist?" Schneir asked. "A physicist," she replied.

According to Yatskov, Perseus supplied Lona Cohen with information about the construction and testing of the plutonium bomb that exploded in the desert of New Mexico on July 16, 1945. The declassified Russian intelligence archives contain an exceptionally detailed technical description of the bomb, but it is not clear whether this came from Perseus or Fuchs.

"The information we acquired enabled us to gain time in building our own bomb," said Yatskov, who was awarded the Red Banner of Labor medal for his services to Soviet science. "How much time we gained is hard to tell, but a few years at least. If it had not been for the information that other people were developing a bomb, we would not have started our own effort until the end of the war."

Although most of the recently released documents deal with the atom bomb, one dated Dec. 31, 1946, suggests that Western scientists also supplied information to Moscow about the more sophisticated hydrogen bomb. The document is a note from Kurchatov to KGB chiefs acknowledging the receipt of information about "American work on the superbomb," as the H-bomb was known.

Yatskov, who left the United States in December 1946, said that information from the agents in Los Alamos dried up after the Soviet Union tested its own plutonium bomb in September 1949. "When we had our own bomb, they gradually stopped working with us. They had done all they could for the Soviet Union," he said.