A wise man once noted that it is very difficult to make predictions, especially about the future. Fortunately, it occasionally happens that someone like physicist Gerald Feinberg, author of "Solid Clues: Quantum Physics, Molecular Biology, and the Future of Science," decides to ignore this warning and think deeply about the present state of human endeavor and the paths it is likely to follow in the future.

Feinberg has chosen to think about the nature of science -- in particular, the natures of modern physics and biology. He feels that these are the sciences "with the future in their bones," and who can argue otherwise? For each science, he examines the present state of the art, picking out the crucial questions being asked by researchers today.

In physics the questions concern a problem of ancient and honorable lineage -- namely, what are the building blocks from which the matter of the universe is formed? The concepts of quarks and electrons that dominate modern physics are laid out in overview, and anyone interested in picking up the current thinking of physicists working at the frontiers of knowledge could do much worse than to read these sections.

In biology the discussion becomes somewhat less rigorous, reflecting the present state of knowledge in that field. The basic problem here is to explain the enormous diversity among living beings on the earth, and this problem, in turn, leads to the fascinating question of how a single fertilized egg can develop, without outside instruction, into a full-grown individual.

Having laid the groundwork for his inquiry in this way, Feinberg proceeds to identify the key topics that he believes will form the core of future science. Some of these are familiar -- the origin and fate of the universe, black holes, the origin of life on earth. Some were, to me at least, unexpected. Among the questions he asks in physics, for example, is why nature should be arranged symmetrically, which is another way of asking why snowflakes are so beautiful. This is an extraordinarily profound question, and, while I hope we will be able to answer it someday, I'm not as hopeful as Feinberg that we'll find the answers soon.

This section has some of the book's best exposition. In reading about the current research on aging (one of Feinberg's fundamental biological problems), I had always been struck by the apparent chaos that seems to hold sway in the field. It seemed that virtually every researcher had his own theory, and none of these theories seemed to have anything to do with any other. Feinberg, in a logical tour de force, actually manages to make sense of this whole picture. He breaks down the theories of aging into two categories -- programmed theories (in which aging is built into our genes) and wear-and-tear theories (in which aging is the result of accumulated insults and genetic accidents) and then examines the successes and failures of each.

Feinberg then examines what he calls "engines of change" in science -- new experimental techniques that allow us to see new realms of nature, new calculative techniques (including large computers) that allow us to deal with ever more complex knowledge in a rational way, and the various societal attitudes that support (and sometimes hinder) those who do science. Finally, he turns to the question of the shape of science in the 21st century.

I will confess to a slight feeling of letdown when the book finally got down to specifics. The list of future technologies (unlike the list of future basic research) seemed strangely prosaic. For example, Feinberg mentions "nanoengineering" (nano is the prefix used to designate a billionth -- thus, a nanosecond is a billionth of a second). He talks of acquiring the ability to build new materials, atom by atom -- materials so strong that they could be strung between the earth and the moon to carry freight up and down. That the Silicon Valley boys can already come close to "nanoengineering" surfaces of material takes some of the gloss off this sort of prediction. This isn't a criticism of the book, simply an illustration of how difficult it is for even the most imaginative minds to see the possible future effects of technology and science.

The book closes on an upbeat, with Feinberg examining a number of social influences that might affect science and coming to the cautiously optimistic conclusion that things will probably go well for the scientific enterprise for the foreseeable future.

If I had any problems with "Solid Clues" (and I had very few), they centered on the amount of space that the author has chosen to devote to what are at best speculations by theoretical scientists. The literature is always full of these wild ideas, and they are almost always ephemeral. Where, for example, are the parallel universes of yesteryear? I suspect that many of the things Feinberg mentions in an offhand way will be gone by the time this book is available, and the reader is going to be hard-put to know what to take seriously and what to ignore. Another point that bothered me was his argument (in a section on the science of the future) that watching "Star Trek" will cause children to turn away from scientific careers. I would have thought that a half-hour's conversation at any meeting of the Society of Physics Students would have been enough to convince the author otherwise.