It hasn't been a good week for biology. I've just learned that Francis Crick has died, and while I was at SDB I heard news that grieved me even more: Ed Lewis has died.

Lewis was one of the giants of Drosophila genetics. He was a student of Thomas Hunt Morgan, and one of the rare practitioners of pure genetics at the time molecular biology was swallowing up the field. He's mentioned several times in Jonathan Weiner's book, Time, Love, Memory (you've all read it, right? If not, get it—it's a wonderful view of neurobiology and genetics and molecular biology, from the perspective of Seymour Benzer). I'll quote a few passages from that book that give a taste of what Lewis was like.

Seymour Benzer was not an unpleasant human being, but he was one of the revolutionaries. In Church Hall he and Delbrück and sometimes Watson (when he came to visit) strode through the corridors talking about events "in the days of genetics" as if those days were ancient history—even though Ed Lewis was still sitting in his lab crossing mutant flies, and the man who had made the first map of the genes [Sturtevant] was kneeling just outside the building, weeding his irises. Someone in the public relations department at CalTech once interviewed Benzer as part of an in-house oral history project. She asked Benzer if Lewis had been held in contempt by the new breed of molecular biologists in the 1960s. "No. He was a nice guy," Benzer replied. "He was very good with flies. But at the same time it seemed sort of like having a Greek mythology scholar; it's nice to have one around for the university at large. He taught the genetics class, and kids counted flies. Of course I'm giving you the jaundiced point of view. He was the true inheritor of the Morgan-Sturtevant tradition, and that was just fine."

Lewis himself felt his isolation painfully. "Drosophila went into almost total eclipse," he says. "Delbrück would pound the table: 'Genetics is dead! Genetics is dead! Genetics is dead!'" Over and over, Delbrück said it in so many words: Molecular biology is the only biology.

(Many years later, around the corner from the campus, sitting in Max Delbrück's favorite old rocking chair, Manny Delbrück would giggle when she remembered the apocalyptic speeches her husband used to make. "You see," she said, "Max didn't know any other biology.")

As it turned out, what Lewis was learning in his Fly Room would someday thrill a new generation of molecular biologists and would win him a share of a Nobel Prize. But at that time, Lewis saw no point in trying to convince Delbrück that what he was doing was interesting. Lewis was smaller, gentler, and quieter than most of the bullyboys. He liked to keep tanks of rare tropical fish and anemones in his laboratory, and he raised a generation of octopi, which Benzer, who had never seen an octopus embryo, found supremely beautiful. Lewis had long, owlish eyebrows like nerve endings looking for a home. And Lewis was an owl, like Benzer himself. Sometimes before dawn, when Benzer roamed the corridors thinking about genes, nerves, and behavior, he would hear, through Lewis's closed door, the sounds of a flute.

Lewis was persistent, and that work he was doing quietly, while the molecular biologists were oblivious, would pay off in the end.

In a normal fly, the thorax has three segments. The first segment has a pair of legs. The second segment has a pair of legs and a pair of wings. The third segment has a pair of legs and a pair of balancers, called halteres. Lewis discovered that bithorax's problem is not a single mutation, but a cluster of mutations on the third chromosome. In a fly embryo, these mutations confuse the identity of one segment with the identity of the next. When Lewis mapped them patiently decade after decade, he found that the genes in the bithorax complex are arranged in the same order along the chromosome as the parts of the body they affect. That is, if one goes down the chromosome from top to bottom, one comes to genes that affect the growth of the fly from the top of the head to the tip of the abdomen. The genes that control the development of the head and antennae are at one end of the complex, and the genes that control the development of the tip of the abdomen and the anus are at the other end. What is more, these genes turn on one after the other as the fly embryo is growing, in anatomical order, beginning with the head and working down towards the anus; and if they switch on in a different order or if one of them misfires, the body plan of the fly is disarranged. In this sense, the fly itself is a map of its genes. As one drosophilist writes, "It is as if the insect's entire body is the expression of a giant chromosome made visible to the naked eye."

The complex of genes would eventually lead molecular biologists inside the problem of development the way Benzer's mutants would lead them inside the problem of behavior. What Lewis found in flies would turn out to be fundamental throughout the tree of life. New tools of molecular biology would augment the old tools of genetics and mapping to produce breakthrough after breakthrough; and the same tools would work equal wonders on Benzer's mutants. Clock mutants and the savoir-faire mutants would provide the first set of picture windows into the workings of genes and behavior at the level of the anatomies of atoms.

In the last years of the century, Lewis, standing by his old teacher Sturtevant's iris bed or besieged by reporters in the faculty parking lot, would say with a grin, "It was pure genetics. It was pure genetics." Nothing molecular about it. He would remember how Delbrück had pounded the table and denounced the fly, and he would murmur, so softly that the reporters had to ask him to raise his voice, "I'm so glad I stayed with it."

I didn't know Ed Lewis personally, but his work was my first terrifying exposure to the rigors of science. As an undergraduate, I worked in a fly lab at the University of Washington that was much concerned with homeotic mutants, and I vividly recall taking home some of Lewis's papers on the subject, and trying to understand them...and let me tell you, Lewis's work was an amazing exercise in logical puzzle-solving. This wasn't just "mutant gene X causes phenotype Y" stuff; this was working through a tangle of gain-of-function and loss-of-function and regulatory mutants, and deciphering how a whole network of genes cooperated to yield a specific pattern of gene expression. I could barely understand a word of it. I think I've been spending the last twenty-five years working to get the background to understand what Lewis, and others like him, do.

For a taste of what Lewis did, here's a review he wrote for the Nobel Foundation (pdf). It starts out gently enough, but unless you know a bit of genetics, by the time you hit the sections on half-tetrad analysis and cis and trans regulation of the bithorax complex, you'll know how I felt back in 1978. This is glorious, brilliant science.