Universities across the United States say they are facing increasing government pressure not to publish some research because of claims it might be used by terrorists.

University officials say the U.S. Department of Justice and the Department of Defense, in a move that runs contrary to the principles of academic freedom and the free exchange of ideas, are insisting that the government check research before papers are published and approve participation by foreign nationals in university research.

Concerns about these strings now being tied to research funding were heightened earlier this month when the Massachusetts Institute of Technology revealed it turned down a $400,000 research grant because the government demanded foreign students not be allowed to participate in the project. Nearly half of the graduate students in the natural and physical sciences at MIT are from abroad.

University researchers worry the trend toward greater restrictions over their research could stifle the tradition of open science through which academics publish and debate their findings in the advancement of knowledge.

"When the Soviet Union tried to keep its research secret during the Cold War, their whole scientific apparatus atrophied," former Air Force Secretary Sheila Widnall, now an aeronautics professor at MIT, told Science.

The controversy at MIT is not unique. Last fall, the University of California at Berkeley turned down funding from the Army Corps of Engineers after being asked to provide the names of all foreign nationals who would be involved in a research project.

In addition to demanding that foreign students be barred from some research, the government is increasingly insisting that research it funds be kept secret or submitted for approval before publication.

Traditionally, university research in the U.S. has fallen into two broad categories - classified or open. But since the Sept. 11 attacks, the Bush administration has been considering new controls for research that is "sensitive but unclassified" including the authority to control publication.

White House officials say such controls are necessary in order to prevent potentially dangerous information being used by terrorists, but many in the academic community are questioning restrictions that are being imposed on projects that would seem to be of little interest to terrorists.

The Justice Department, for instance, recently demanded the right to approve, prior to publication, a Cornell University study on physical abuse of female students. Robert Richardson, Cornell's vice provost for research, said the university turned down the grant.

Nevertheless, in a trend that is alarming many in the scientific community, some institutions are agreeing to the restrictions. The California Institute of Technology recently allowed the Army Research Laboratory the right to review before publication a professor's work on computer simulation.

The National Academies, a body chartered by Congress to advise the government on science policy, has urged Washington to maintain the principle that any university research not explicitly classified as secret should be published freely.