If Richard Lewontin is right, we're in for a big letdown when the Human Genome Project is completed. The promise used to justify the $3 billion effort is that the 3 billion base pairs in human DNA make up a "Book of Life" that will reveal the secrets of development, disease and everything in between. But as Lewontin points out in It Ain't Necessarily So-a collection of nine extended pieces from The New York Review of Books-there are a few troublesome details scientists "forgot" to mention when they were lobbying for the project more than a decade ago.

For one thing, we have only the foggiest picture of how our 100,000 genes interact to regulate one another's expression and to direct protein production. Extrapolating from the genome to the whole organism is therefore akin to writing a history of New York City based on the phone directory. Another problem is that DNA, by itself, doesn't produce or explain anything. Genes specify which amino acids should be used to build a protein, but proteins themselves "are made by other proteins, and without that protein-forming machinery nothing can be made," Lewontin observes. Life, in other words, is a co-production between DNA and its cellular surroundings, the choreography of which cannot be recorded solely in the genome.