IT IS ALWAYS true that some fields of scientific research are more promising than others, and the pattern incessantly shifts. For the great patrons of the sciences--that is, national governments--the constant question is where to increase the funding most productively. Every five years, a group of eminent American scientists offers Congress its advice in the form of a short book of essays on those fields that seem most imminently hopeful. It represents a loose consensus within the various national academies of science, engineering and medicine; the most recent of those books, under the title "Frontiers in Science and Technology," has just appeared.

The striking thing about it is the strong emphasis on biology, at the level of the cell and the molecule. Of the eight topics the book addresses, four are drawn from that field. Two others are the mandatory ones: robots and lasers. Two are less obvious subjects lying in the broad region where physics and engineering overlap-- the study of surfaces, and turbulence in fluids. Turbulence theory has applications to the design of more efficient turbines, but also to the medical care of stroke victims. Surface science is relevant to the design of both integrated circuits and corrosion-resistant bridges.

In a period of massive increases in defense spending, you can safely assume that support for science-- or, more accurately, support for some sciences--is going to rise quite significantly. Although much of the money will go into applications, some of it will trickle into basic work in the purest sense. But it won't be spread equally among all the sciences. Defense money tends to favor some areas of physics and chemistry.

Against that background, the academicians are reminding Congress that, in terms of fundamental understanding, there are very large gains not far ahead in fields that the defense money is least likely to touch. The infectious diseases that were the killers through most of history have largely been brought under control. They have been replaced in the mortality statistics by others--the diseases of the heart and circulatory system, cancers, metabolic diseases--very much more complex in their causes. But "we now have a window into many of these disorders," as Floyd Bloom of the Salk Institute observes in his introduction to the book. "This may indeed be the era of biology. . . ."

Hardly anything is less predictable than the progress of scientific discovery. But we are coming into a time when the changes in the federal budget, if uncorrected, are likely to favor the hardware aspects of science. Congress is getting good advice here--to maintain a balance between the hardware and basic biology, where things are now moving very fast.