THE NATIONAL Institutes of Health in recent years has suffered from a set of lax ethics rules, laxly enforced. Freed in 1995 from regulations that had limited their outside income, some senior NIH scientists pocketed fat consulting fees, at times for work that overlapped, or even conflicted, with their government jobs.

In one instance recently uncovered by congressional investigators, a researcher at the National Institute of Mental Health, one of the NIH's 27 components, was paid $517,000 by the drug company Pfizer over five years; the money wasn't cleared with agency officials or reported on his financial disclosure form. In another, a National Cancer Institute lab chief, along with a colleague from the Food and Drug Administration, was assigned to collaborate with a private company to develop cancer testing; the pair then signed a consulting deal with one of the firm's competitors. The skin diseases branch chief at the National Institute of Arthritis was retained to testify as an expert witness on the acne drug Accutane, at a reported hourly rate of $600; among his points was the alleged inadequacy of government-approved warnings on the drug.

"I've reached the conclusion that drastic changes are needed," NIH director Elias Zerhouni told a House committee. Dr. Zerhouni's prescription, which goes beyond his original recommendations, may turn out not to be strong enough medicine. It's a legitimate question whether any outside consulting at all should be allowed. But given the unusual role of the NIH -- its scientists are more comparable to academic researchers than to government regulators like those at the FDA -- the Zerhouni approach is worth a try.

Senior employees wouldn't be allowed to do outside consulting. Those allowed to engage in consulting work would be limited to 400 hours annually; their pay couldn't amount to more than a quarter of their government salary (and no more than half could come from any one source). None of the consulting work could involve their government duties. All consulting would have to be publicly disclosed. NIH employees couldn't serve on corporate boards or accept stock or stock options as compensation; they couldn't hold more than $5,000 worth of stock in any pharmaceutical or biotechnology firm. And no consulting deals would be allowed with universities that get NIH funding.

One of the arguments in favor of allowing continued outside consulting work is not to inhibit the NIH's ability to recruit and retain top-tier scientists, who could command higher pay in the private sector. Another sneaky way around the pay disparity has been for the NIH to exploit a loophole designed to let the government hire "special consultants" above the usual federal pay rates. This exception has been used widely at the NIH, with 21 of 27 institute and center directors paid as special consultants, at salaries of up to $235,000. Nearly one-third of NIH employees hired last year were brought on using this inartful dodge.

As with the Securities and Exchange Commission, which has been granted exemptions from regular government pay scales because of difficulties in competing with the private sector to attract lawyers and accountants, we don't begrudge giving the NIH some flexibility in setting salaries. But this can't be taken too far: There are rewards to working at the NIH that alleviate the pay differential. And salaries ought to be set in an above-board manner. As Rep. James C. Greenwood (R-Pa.), who has overseen the congressional probe of the NIH as chair of the Energy and Commerce subcommittee on oversight and investigations, put it, "The gaming must end."