If the general public believes that scientists know more about the natural world than it does, that is understandable. Science, Nature, and other top-tier journals with effective public relations departments issue regular press releases on upcoming articles. Journalists duly write these up after interviewing obliging authors and one or two scientists from other labs to obtain an outsider's perspective. This weekly routine has made these "one-paper" science and medical stories so familiar that news readers might be forgiven for thinking that scientists have pretty much conquered their fields. John Horgan's The End of Science may have had working scientists rolling their eyes when it was published last year, but interested lay readers are still turning its pages.

Such media coverage might explain why biomedical research may seem predictable to casual observers. Pick a disease, find its gene, predict its usefulness for screening and eventually treating affected individuals, and move onto the next one. Or study the habits of a subpopulation before declaring that nutrient x protects against or brings on disease y. The media repetition of these themes reinforces the view that we are chipping away at the mountain of biological knowledge and that we are cutting it down to hillock size. Perhaps that is why a publisher would release a title like The End of Science while most researchers know that the often undescribed mountain still presents a challenging climb.

The scope of this challenge may explain why scientists rarely, if ever, discuss what they don't know with the public. Understandably, they prefer to discuss the implications of their latest contributions. It makes sense for them to stress what they have just made known. That is where the grant money is and where their career fortunes lie.

Fortunately, some authors see the benefit, value, and fun in reviewing the limits of our knowledge. Twenty years ago, The Encylcopaedia of Ignorance edited by Ronald Duncan and Miranda Weston-Smith (Pocket Books, 1977) was published. Now out of print, it included 50 chapters written by leading scientists on unanswered questions and problems in nearly as many areas of science. The discussions, however, were detailed and definitely not directed at the general public. Last year, James S. Trefil's Edge of the Unknown: 101 Things You Don't Know About Science and No One Else Does Either (Houghton Mifflin, 1996), directed specifically at general readers, appeared.

The latest contribution to this small genre is Why Aren't Black Holes Black by Robert M. Hazen with Maxine Singer. Like Trefil's book, this one is intended for the general public, but any scientist reading the chapters discussing issues outside his or her field of specialization might be briefly entertained. The style is conversational and, while many questions cannot be included in only 300 pages, the book's scope is broad. Topics range widely in scale, from a discussion of how atoms combine to a brief overview of "missing" matter in the universe. The life sciences are the subject of six of the fourteen chapters and cover the origin of life, the genetic code, evolution, development and aging, memory, and behavioral genetics.

This diversity reflects many of the research topics studied at the authors' place of employment. Hazen is a scientist at the Carnegie Institution of Washington's Geophysical Laboratory and a professor of earth science at George Mason University. His "written-with" coauthor, Maxine Singer, was chief of the Laboratory of Biochemistry at the National Cancer Institute. She is now president of the Carnegie Institution.

The volume is not helped by the introduction by Stephen Jay Gould. It is the type of undisciplined, somewhat random essay that popular writers sometimes offer after achieving bestseller status. Its message seems to be that successful writers need to be edited more, if anything, after they have gained fame than before.

The book also unfortunately lacks an index, but it does offer a useful list of references for additional reading.

Fortunately the authors express their opinions in the text, and this makes it even more interesting than it might have been otherwise. They note, for instance, that "President George Bush pooh-poohed the environmental crisis. In spite of his skepticism, or perhaps in light of it, in 1989, he instituted the multibillion-dollar United States Global Change Research Program. . . ."

In another example, they reveal what could be interpreted as arrogance typical of pedestrian administrators when they write, "Any scientist who does not include his or her research field in a list of the top unanswered questions should consider switching fields." This attitude is too common and provincial for such talented authors. It ignores the fact that no one knows what lies in every unexplored grotto of nature. It ignores the belief still held by some that a discovery in one obscure field, cross-referenced with another finding in a seemingly unrelated field, may produce fresh insights and unexpected progress. It also promotes the boorish conviction that progress is predictable. It is, of course, but only if we choose to limit ourselves. Fortunately they fill the remaining text with much more worthwhile and interesting insights.

Nonscientists reading this book will come away with a more than a fair overview of what we know and what we don't about science today. That is enough to recommended it. Scientist readers may well be entertained by discussions outside their areas of expertise. A few researchers, bogged down by grant applications and teaching, may even be reminded why they choose to do what they do.

"What We Don't Know" - review from the New York Times Book Review of September 28, 1997. "[The book] fails to address adequately the issue of limitations on what science can achieve in principle."

The Carnegie Institution - the wide-ranging Washington institution's Web site is attractive and information-packed. It is a private, nonprofit organization conducting "basic research and advanced education in biology, astronomy, and the earth sciences."

Joseph F. Traub - a professor of computer science in the Columbia University Computer Department of Science, Traub explores turning the limits of scientific knowledge into part of the study of science.