Scanning the journals of research, you can get the impression that venerable values of truthfulness and openness are crumbling in the get-rich economy that swiftly turns scientific knowledge into wealth. It is a correct impression, and, despite proclamations of concern from the chieftains of science and occasional correctives, the known episodes of dubious dealing continue to increase.

A new case of an unwholesome old practice is reported in the Journal of the American Medical Association, where a review of a series of drug tests concluded that favorable results were puffed up through repetition in various publications, while negative information was played down or ignored. This pharmacological deception has repeatedly been used by drug manufacturers and the academic accomplices who perform the tests and write the reports, according to JAMA.

Independent researchers who sought to examine the original test data ran into a stone wall--"despite the fact," JAMA ingenuously asserted, "that a scientific duty assumed by a researcher who accepts the credit that goes with publication is to answer the questions and criticisms of fellow scientists and readers." Maybe. But if the researcher balks, there's no redress.

Ironically, the American Medical Association, which publishes JAMA, ran off the ethical high road two years ago when it entered into a lucrative royalty deal to allow use of the AMA logo on Sunbeam Corp. appliances, without testing the goods. Under protest from AMA members, the deal was rescinded and several senior executives exited the AMA.

At the nation's other great medical publication, the New England Journal of Medicine, a search is underway for a replacement for editor Jerome Kassirer. Though honored for his medical knowledge and editorial judgment, he was fired by the publisher, the Massachusetts Medical Society, for resisting plans to exploit the prestigious name of the Journal in various new publications and commercial ventures.

The idealists of science have traditionally opposed secrecy in research, arguing that it impedes progress. But with the recognition that laboratories are the gold mines of our time, secrecy has become a fact of working life in modern science. In the biomedical sciences and in anything related to computer technology and software development, university laboratories are enmeshed in countless contractual deals that guarantee industrial partners first look at their laboratory findings, and exclusive rights in some cases.

"Secrecy in Science" was the topic of a conference held at MIT last spring by the American Association for the Advancement of Science. The gathering heard tales of litigation threatened and carried out by industrial firms in disputes with university researchers over the meaning of non-disclosure agreements. In many of these cases, even when the professors win, they personally lose by impairing the university's attractiveness as an industrial partner.

Even if academic-industrial deals are bad for science, they're good for the economy, according to the new breed of entrepreneurs bridging the two sectors. Maybe so, though it's doubtful that the overall economy benefits from deals that restrict the dispersion of scientific knowledge. What is certain, however, is that the melding of academe and commercial interests can deprive the public realm of independent, disinterested expertise.

"University entrepreneurship shifts the ethos of academic scientists toward a private orientation and away from the public interest role that has largely dominated the scientific culture since the middle of the century," according to Sheldon Krimsky, of Tufts University. "An independent reservoir of scientific experts who are not tied to special interests is critical for realizing the potential of a democratic society," he argues. "The benefits to academia of knowledge entrepreneurship pale against this loss to society," Krimsky warns.

Daniel S. Greenberg is a Washington journalist specializing in science, medicine and politics.