Time, Love , Memory

by Jonathan Weiner

  For years a fly man at the University of Pennsylvania, Vincent Dethier, had suspected that even flies could learn. Where most people see flies as “little machines in a deep sleep,” Dethier once wrote, he looked through the microscope at their fantastically intricate armored bodies, “their staring eyes, and their mute performances,” and could not help wondering if there might be “someone inside.” Dethier tried to prove that flies can remember but he never could, and after eighteen years he gave up. Today when Benzer lectures about genes and memory, he often flashes on the projection screen an old headline from the Washington Post about Dethier’s conclusions. The Post ran a rather unflattering close-up photograph of the face of a fly with a caption that passed the finger-shaking judgment “Can’t learn anything.”

  In those years Benzer taught an undergraduate course in behavior at Caltech. He always put the same question at the end of the final exam. The question was worth a case of beer and five hundred points (which meant an A-plus for a satisfactory answer): “Design an experimental situation in which you can show that Drosophila learn.” Many young techies rose to the challenge. One graduate student in Benzer’s laboratory arranged a tiny spotlight so as to cast a shadow of the fly on a sensor. By administering a series of little punishments in the form of heat, the student, Jeff Ramm, tried to train the fly to alter its posture. After a while he got discouraged and left the laboratory. Benzer used to complain that Ramm had quit just before the flies learned. “But I think that Jeff and Seymour had incompatible personalities,” says Chip Quinn. “Jeff Ramm was someone who didn’t have this tolerance for starting experiments when you don’t know what you are doing. And so he diffused off to Felix Strumwasser’s lab.” Strumwasser was a neurophysiologist, an expert in the workings of nerves. He had been one of the biggest naysayers when Benzer presented his genes-and-behavior project to the Sperry group.

  Another student of Benzer’s wrote a paper proposing to adapt a cockroach experiment called Horridge Leg Lifting. A student of invertebrate behavior, Adrian Horridge, had stuck a cockroach on a tiny diving board. If the cockroach’s legs dangled into the water, it would get a shock. Eventually the cockroach learned to lift its legs out of the water. Benzer’s student thought Horridge Leg Lifting might work with flies. But he quit the lab and went off into computer science. “Again,” says Quinn, “I think he quit just before they learned.”

  Benzer himself tried to teach flies in his countercurrent machine. He put an electric grid inside one of the test tubes and shocked the flies to teach them to stop going toward the light. If he shocked the flies over and over again, they would go toward the light less and less. When he turned off the current, the flies still did not run toward the light. For one brief happy period, Benzer thought he had taught his flies to avoid the light. But when he put them into a fresh test tube, they ran for the light as urgently as ever. Apparently the flies were not learning after all. They were laying down some sort of odor—perhaps the odor of panic, or the odor of singed fly hairs and feet—and avoiding a bad smell in a test tube is not the same thing as learning; it does not imply a remembrance of things past.

  When Quinn joined Benzer’s laboratory, he looked over the countercurrent machines and shock grids that Benzer had made. “And at some level, I had no idea what to do,” Quinn says. So he started simplemindedly repeating Benzer’s experiments and replicating his results, “just to see what was going on, because I didn’t know what to do.” He found that the flies did seem to be stinking up the test tubes. But he also found that if he took fresh flies—naive subjects, unshocked troops—and ran them through the countercurrent machine, these new flies would ignore the odor and head for the light as if the odor were not there. Quinn concluded that the odor of panic (or whatever it was) repelled the flies only if they had smelled it in conjunction with an electric shock. In other words, Benzer’s flies might have learned something after all: they might have learned to avoid the odor. “So that looked relatively encouraging,” Quinn says.

  Since the flies apparently paid some attention to odors, Quinn (guided by Benzer) decided to try perfuming one of Benzer’s countercurrent machines. “Caltech had a whole room full of old bottles of chemicals. So I went around and opened them and sniffed them.” Quinn had no way to know which odors to pick, because he did not know if the odors would smell the same to flies as they smelled to him. On the other hand, he had no better way to pick and choose than to please himself. He wanted a compound to be volatile enough that he could smell it, but not so volatile that it would go away immediately. So he went around the room opening vials at random. He chose a vial labeled “Octanol,” which smelled like licorice. He also chose a vial labeled “Methylcyclohexanol,” which smelled—as someone in his lab later put it—“a lot like tennis shoes in July.”

  Quinn lined the test tubes in the countercurrent machine with copper grids—very fine mesh, like tiny rolled-up window screens. He perfumed one test tube with octanol, another with methylcyclohexanol. The odors would linger powerfully on the grids for two hours or so. When he was ready to start the experiment, he laid this set of test tubes on his benchtop, just as Benzer had done, and lit Benzer’s fifteen-watt fluorescent bulb. He put about forty flies in the first tube and let them explore it for sixty seconds. Then he tapped the whole apparatus on a rubber mat to knock the flies down to the bottom of the tube (just like Benzer), as if he were making an official pronouncement in the dark room: “This meeting will now come to order.” Finally he slid the tubes so that the flies could walk, if they chose, into the tube ahead of them, which smelled of octanol.

  As in Benzer’s first experiments, almost all of the flies walked or ran toward the light—where they encountered a shock of seventy volts for fifteen seconds, which might kill a human being but only ruffled the fly’s bristles a bit. Then Quinn tapped the tubes on the benchtop again to return the flies to the first tube and he gave them sixty seconds to recover from the shock.

  Next he slid away the tube that smelled of octanol and replaced it with a tube that smelled of methylcyclohexanol. Again the flies ran for the light. This time Quinn did not give them a shock. After fifteen seconds he returned them to the first test tube.

  Quinn repeated this cycle: octanol and a shock, meth and no shock. In a way, he was talking to the flies: octanol, bad; methylcyclohexanol, good. (Of course, there was nothing intrinsically good or bad about either; he had tossed a coin before starting the experiment.)

  Finally, the test. Quinn slid into place a fresh test tube, one the flies had not encountered before. The tube was perfumed with octanol. He bent over the tubes by the fifteen watts of the fluorescent light and watched. More than half of the flies milled around and did not go into the tube.

  After returning all of the flies to the first test tube and giving them a minute to rest, he slid away the tube of octanol and slid into place a fresh tube of methylcyclohexanol. Most of the flies went into that tube.

  It was an eerie thing to watch. The flies were acting on their experience. Ever since Morgan’s Fly Room, geneticists had known that fruit flies have genes and chromosomes as we do and that they inherit their bodies and behavior as we do. But here the flies were doing something that even Quinn had not really expected to see a fly do: they were learning from their own experience. They have something inside that allows them to remember what has happened to them, as we do; and they can act on what they remember, as we do. It was a sight to give a human being pause, like William Blake’s visionary line: “Am not I a fly like thee / Or art not thou a man like me?”

  Quinn repeated the experiment with a second crowd of flies, and they learned the lesson, too. He tried teaching a third crowd of flies the reverse lesson: methylcyclohexanol bad, octanol good. The flies learned that lesson, too. Each time the change in the flies’ behavior was eerie to watch. It was as if something palpable had changed for them, as if the odors had turned into invisible doors. They acted as if one door was still wide open but the other door was almost closed and looke
d forbidding to most of the flies.

  To make sure he was not just kidding himself, Quinn asked a friend in the lab to set up the teaching machine for him, to arrange the perfumed test tubes. This way, when Quinn ran the experiment, he could not know which test tube held which odor and he could not know which odor led to a shock. He ran the experiment blind and recorded the flies’ behavior impartially: how many chose this tube and how many chose that one. Then he checked his friend’s notebook (in effect, taking off the blindfold) to see if the flies were really learning their lessons. They were.

  Not only were the flies learning, they were learning fast, as Benzer enjoyed pointing out chauvinistically. In one standard laboratory test of learning, an experimenter rings a bell and then blows a puff of air at a rabbit’s eye to make it blink. The rabbit learns to blink before the puff of air, but that takes about eighty lessons. Quinn’s flies learned in three.

  Once Quinn happened to notice that his flies avoided walking on a spill of dry powder—quinine sulfate. So instead of electrifying his grids, he tried dusting them with the powder, spreading some quinine sulfate on the copper with a fine artist’s brush. The flies learned to avoid the powder the same way they learned to avoid a shock. Benzer’s graduate student Bill Harris built a Y maze of black Plexiglas with small brass knobs. At the choice point of the Y maze the flies could go toward either a red light or a blue light. The flies learned those lessons too: red good, blue bad; or red bad, blue good.

  They seemed to be learning much the way human beings learn: by repetition. It is adaptive for living things’ memories to require repetition. A three-month-old baby will not remember something that happens once, but if the lesson is reinforced on a regular schedule, she will. “A single experience that is never to repeat itself is biologically irrelevant,” the physicist Schrödinger wrote. “Biological value lies only in learning the suitable reaction to a situation that offers itself again and again, in many cases periodically, and always requires the same response if the organism is to hold its ground.”

  Next, Quinn played with his flies to see how long they would remember their lesson. He taught a fresh group of conscripts to avoid octanol and let them rest for an hour. Then he put them through their paces again. One hour in the lives of fruit flies is like a few months in ours. After the break, most of the flies still remembered to avoid octanol, but some of them forgot. He gave another group of flies twenty-four hours (six years in human lives) to sit around in fly bottles and forget. Many of them still remembered; more of them forgot.

  “CHIP QUINN once described the ideal organism as one that has three large neurons, divides rapidly, and can learn to play the piano,” Benzer says. “Everybody wants a simple system.” Delbrück, Benzer, and Brenner had thought of the fungus, the fly, and the nematode worm as simple systems when they started: as something like gadgets in physics, atoms of behavior. “Sydney’s idea was, the nematode has a small number of neurons; therefore, it’s a simple system,” Benzer says. “I think he has changed his mind a little bit. He describes the lineage and the development of the nervous system of the nematode as ‘baroque.’ ” (This discovery was foreseeable even in the first days of the Enlightenment. “A worm is only a worm,” said Diderot. “But that only means that the marvelous complexity of his organization is hidden from us by his extreme smallness.”) Now Benzer and his students were beginning to realize that the fly is baroque too; they were delighted to discover that it can learn and act on what it learns.

  Because the fly has memory in its repertoire, Benzer and his students could begin to dissect this behavior too. They could use their tools of genetic dissection to take apart the act of remembering the same way they were taking apart the sense of time and the dance of love. A new student at the Benzer lab, Duncan Byers, dosed flies with the laboratory’s favorite mutagen, EMS, producing five hundred separate lines of mutants, and he began to test their memories, strain by strain, mutant by mutant, using the whole Quinnian rainbow of odors, including octanol, methylcyclohexanol, and quinine sulfate. Out of the five hundred lines of mutant flies, about twenty lines failed to learn. But almost all of those flies flunked other tests too. Some turned out to have problems seeing; others had problems smelling; some were sluggish or shaky walkers. Only one of the lines that flunked had normal instincts, normal senses, a healthy level of energy, but no talent for learning. That line of mutants turned up in Bottle 38. For them to appear so early in the search was auspicious. They fit Konopka’s Law: “If you don’t find it in the first two hundred, quit.” Byers examined them through the microscope. As eggs, larvae, and pupae, they looked normal, and they lived as long as normal flies. They ran to the light, they climbed walls, they flew, walked, courted, and copulated like normal flies. But for them every odor was an open door. They never seemed to learn.

  Benzer and his crew decided to name this new mutant after John Duns Scotus, the thirteenth-century philosopher. Duns Scotus’s disciples were known as Scotists, dunses, or dunces. In the sixteenth century, the dunces lampooned the new learning, and they were lampooned back by the natural philosophers who were the world’s first true scientists. The dunces lost the war, and the scientists made them eternal symbols of stupidity.

  Benzer’s raiders tried very hard to get dunce to learn something. They exposed the new mutants to a wide range of odors in all kinds of dilutions and combinations, and to a wide range of shocks, from 20 volts to 140 volts. But the flies kept their dunce caps. Benzer was delighted. With mutants like these, he and his students could now dissect the act of memory into a series of steps. They would search for mutants that could remember a little better than dunce—some for a few minutes, some for a few hours, some for a few days. If they could find them and figure out what made each one different from the next, they might be able to use dunce and the rest to trace the invisible steps through which an experience becomes a short-term memory and then a long-term memory. Benzer’s crew mapped dunce. The gene lies on the far left tip of the X chromosome, just a few map units away from white and from the mutants of Konopka, the mutants that lost the sense of time.

  BENZER’S PROJECT, the genetic dissection of behavior, was off to a strong start. He and his students had used genes to open three locked doors that had fascinated poets and philosophers from the beginning of Western thought.

  “The body is but a watch,” said Julien Offray de La Mettrie at the start of the Enlightenment, coining a slogan for a worldview that has lasted from his time to ours. The clockwork has been called the central metaphor of modern Western civilization. Certainly from the very beginning of modern science, we have seen the space around us as a clockwork of stars and the space inside us as a clockwork of organs, a clockwork of atoms. Benzer and his students, with one of their first stabs in the dark, had found a clockwork gene. They did not know yet if they had the key to the clock; but they hoped they had a piece of it, a point of entry into a type of behavior that is as potent for us as a symbol of science as it is basic to life itself.

  “The selfish gene” is another slogan for a worldview: a catchphrase for biologists—particularly biologists who study genes and behavior—since the 1970s. A body is a gene’s way of making more genes or an egg’s way of making more eggs, as in the quip that Benzer loves to quote from Butler. Mutants like fruitless were points of entry into the universal instinct that allows genes to move from one generation to the next and go wheeling across the longest geological reaches of time, from the beginning of life to the present, a span of four billion years.

  “A man is but what he knoweth,” wrote Sir Francis Bacon in a third slogan, a slogan that defines for us one of the traits we most value in being human. If we could not remember where we have been, we would not profit by our experience and we would not know who we are. The mutant dunce was a point of entry into the mechanisms that allow each of us to accumulate histories and apply the lessons we have learned at our choice points; mechanisms whose elaboration in our species has helped to set us apart from all the rest.
/>
  Darwin’s process has worked powerfully and unceasingly to shape all three of these classes of genes. If animals and plants did not have clock genes, they could not keep time with the world. They would drift in and out of day and night, living less efficiently than their competitors and sometimes falling into fatal encounters. If we did not have instincts for recognizing and winning the attentions of the opposite sex, we could not pass on our genes; we would die without issue. And if we did not have memories, we could not pass these other genes safely onward and most of us could not last a day without a great deal of help from our friends.

  Time, love, and memory are three bases of experience, three cornerstones of the pyramid of behavior. Benzer and his group had found a way inside all three in the first years of their Fly Room. Like Konopka’s discovery of a time-wounded fly in his two hundredth bottle, it was all surprisingly fast work. Konopka’s Law is broader than it sounds. Like so much that came out of the fly bottle, the message is universal. The Iliad and the Odyssey are the greatest epics ever written. Gutenberg’s Bible is the most beautiful book ever printed. Some of the world’s most memorable photographs were made during the first tests of the first photographic chemicals by Joseph Niepce and Louis Daguerre. Benzer’s sister’s husband, Harry Lapow (who gave Benzer his bar mitzvah microscope), spent years at Coney Island snapping pictures with a second-hand Ciroflex. The first day he brought the camera to the beach, he took a photograph that was included by Edward Steichen in the “Family of Man” exhibition at the Museum of Modern Art.