Policy makers at the State and Defense departments are sharply divided about the meaning and import of a nuclear-testing proposal by President Reagan to Soviet Foreign Minister Andrei A. Gromyko and the U.N. General Assembly.

The renewed bureaucratic combat flares up as Reagan is declaring that arms-control agreements with the Soviets will have a high priority in his second term and aides are emphasizing his readiness for far-reaching negotiations.

In a post-election news conference in Los Angeles last week, Reagan insisted that his administration is "united in the idea of arms control" and said various devices, such as a special arms-control "envoy," are under consideration to facilitate progress with Moscow.

Secretary of State George P. Shultz called on the Soviet Union to undertake "concrete deeds" to get the ball rolling toward major agreements, and a senior presidential aide reported from Los Angeles that Moscow has expressed interest in Reagan's concept of broad "umbrella" talks covering most of the outstanding problems.

The optimistic tenor of headlined statements is in stark contrast to the persistent internal struggle over the leadership, direction and substance of arms-control policy. As portrayed by many officials and unofficial observers, the semi-public argument over whether there should be an arms control "czar" in the second Reagan administration, and who that person should be and should report to, suggests that the maneuvering among senior policy makers has only just begun.

Interagency gridlock among the contending figures and forces is widely reported to have been one of the two fundamental impediments to successful arms-control negotiations in Reagan's first four years. The other major impediment, nearly everyone agreed, was the uncertain state of decision-making power in the Kremlin.

The dispute over Reagan's proposal to send U.S. and Soviet experts to each other's nuclear test sites indicates the depth of discord even on peripheral issues of arms control that seem from the outside to be settled by presidential pronouncement.

Reagan's proposal was made publicly in his address to the U.N. Sept. 24 and repeated privately to Gromyko at the White House four days later. It called for the exchange of visits by experts to nuclear test sites "to measure directly the yields of tests of nuclear weapons."

There is no dispute in the government about the desirability of such an exchange. But officials of the State and Defense departments are sharply at odds about why this is being proposed, what it would accomplish and what the next step would be if the Soviets agree to discuss it.

At the heart of the controversy is whether it is in the U.S. interest to ratify two treaties on limiting underground nuclear explosions which were signed by Presidents Richard M. Nixon and Gerald R. Ford in 1974 and 1976, respectively, but have been not ratified by the Senate. Reagan refused to ask for ratification on grounds that the means for verifying compliance were insufficient.

The treaties took on additional prominence last month when Soviet President Konstantin U. Chernenko, in an interview with The Washington Post, listed their ratification as one of the steps the United States might take to improve relations between the nuclear superpowers. The Soviets have repeatedly asked for the treaties to be ratified and have refused even to discuss earlier Reagan administration proposals for changes until that step is taken.

State Department officials, who have been working behind the scenes for more than a year to resurrect the idea of ratifying the test treaties, are promoting the exchange of experts as a major step toward improving verification and, in this way, setting the stage for ratification.

But Defense Department officials reject this view. They insist that the limited on-site monitoring proposed by Reagan would not solve the verification problem, and they continue to oppose ratification of the nuclear testing treaties.

To the public, it might seem self-evident that the president's idea of exchanging experts to monitor the size and nature of nuclear tests would be related to the question of ratifying the disputed treaties. Otherwise there would appear to be little point in the exercise.

Surprisingly, officials in both departments agreed in interviews that ratification was deliberately not addressed in the interagency decision-making that led to Reagan's September proposal. "If we tried to hammer out where it would lead, we would never have gotten it into his speech," a State Department source said.

A Defense Department official said that "the president was purposely vague in his speech" when discussing the exchange of experts and that an interagency committee is at work to discuss where to go from here.

Shultz, in an interview last week, challenged the Soviets to agree to the exchange of experts. He noted that nuclear testing was one of the areas listed by Chernenko as offering promise for improvement in U.S.-Soviet relations.

In what evidently was a veiled reference to the bureaucratic controversy, Shultz said that Reagan's proposal for the exchange of experts "could be, but is not necessarily, related to those test ban treaties."

In addition to the disagreement over the meaning of any exchange of experts, there also is discord about how many such exchanges the president is seeking.

State Department officials have let it be known that, in their view, two or three visits by American experts to Soviet nuclear test sites probably would be sufficient to "calibrate" sensitive monitoring equipment in U.S. installations, so that future tests could be measured from outside the Soviet Union with much greater accuracy and assurance.

A senior Defense Department official called that assessment "technical nonsense." Because of geological differences in Soviet testing areas, this official said, on-site monitoring at every Soviet test site would be essential before the United States could be confident of its measurements.

The 1974 Threshold Test Ban treaty, negotiated by Nixon, prohibited nuclear weapons tests in excess of 150 kilotons (the equivalent of 150,000 tons of TNT, more than 10 times the power of the U.S. bomb that destroyed Hiroshima in 1945).

The 1976 Peaceful Nuclear Explosions treaty, negotiated by Ford, placed restrictions on nuclear explosions for nonweapons purposes such as changing the course of a river.

The Carter administration did not push for ratification of the treaties, largely because it was seeking to negotiate a much broader treaty, known as the comprehensive test ban treaty, prohibiting nuclear weapons tests of all sizes.

The Reagan administration refused to pursue the comprehensive test ban negotiations and initially said nothing about ratifying the 1974 and 1976 treaties. Both the United States and the Soviet Union had agreed to adhere to the limitations of the two treaties even though they had not been ratified, but the Reagan administration complained that Moscow had tested weapons above the 150-kiloton limit. The Soviets denied it.

Reagan decided in July 1982 to seek "improved verification procedures" that could lead to ratification of the two treaties. It took until the following year to hammer out interagency agreement on a specific U.S. proposal, which called for on-site inspection.

The Soviets refused even to meet to discuss it, saying that the treaties had been negotiated and signed in good faith. Moscow insisted that they be ratified by the Senate before changes were discussed. The Soviets pointed out that the treaties themselves provided for post-ratification exchanges of technical data, which could aid the calibration of U.S. measuring devices.

After being turned down three times in seeking to change the treaties, the State Department more than a year ago initiated interagency efforts to forge a new approach: the on-site monitoring of a limited number of tests in order to calibrate the U.S. verification instruments. However, the department was never able to obtain interagency agreement or presidential approval for this approach.

"It was typical of the State Department reaction to modify your proposal before it was even tabled with the Russians," a Defense Department official said caustically.

This fall, when the White House was canvassing the government for ideas for Reagan to present to the United Nations, the State Department resurrected the proposal for a limited amount of on-site monitoring. It was quickly accepted by the White House as an item in Reagan's speech and subsequently for presentation to Gromyko.

While conceding this point, the Defense Department never agreed that Reagan's proposal should be intended to advance ratification of the two treaties. Nor has the president or his national security staff intervened to decide the context or significance of the proposal.

The Soviets so far have shown no sign of accepting Reagan's proposal. Chernenko, in his Post interview, said the United States should prove its sincerity "precisely by ratifying" the threaties "and not by inviting observers, as suggested by the American side, who would merely dispassionately ascertain the fact of explosion."

State Department officials said they did not take Chernenko's statement to be a definitive rejection of Reagan's proposal. Shultz, asked to characterize where the proposal stands with the Soviets, said, "I think it's lying there on the table basically."