August 21, 1945.

RAEMER SCHREIBER
The blue flash was clearly visible in the room although it was well illuminated from the windows and possibly the overhead lights.

LOUIS SLOTIN
Well, that does it.

In the summer of 1945, the town on the mesa at Los Alamos had been running for two years on the same rhythm. It sat at 7,300 feet in the high desert of northern New Mexico, built from nothing on a former ranch school site in 1943. Most of the people who lived there were young. The average age was around twenty-five. Scientists and technicians worked shifts in the canyon laboratories six days a week, sometimes fourteen hours a day, and on Saturday nights they gathered in the lodge or someone's apartment and drank and tried to feel like people who lived in an ordinary place. They hiked the surrounding mesas on their days off. Their children played in the dirt streets between the rows of identical housing. The bombs had been dropped on Japan in August. The war was over. A third bomb, already in production, was cancelled. But the laboratories stayed open. The technicians and the physicists reported to work each morning the same as before. There was still material on hand, and there were still experiments to run with it.

One piece of that material was a sphere of plutonium-gallium alloy, 6.2 kilograms, eighty-nine millimeters across, small enough to cup in two hands. It had been manufactured for the weapon that was never dropped. Now it sat on a worktable in a laboratory at the Omega Site, a small cluster of buildings tucked into a canyon below the main mesa. Physicists used it for criticality experiments, edging the sphere toward the threshold of a sustained chain reaction and then pulling it back before the reaction could run. They called the work "tickling the dragon's tail." Within a year, two of them would be dead from handling it.

Harry Daghlian was twenty-four years old, from Waterbury, Connecticut. He had studied physics at Purdue University, where he joined a research group led by a physicist named Marshall Holloway. In 1943, when Holloway's group was invited to Los Alamos, Daghlian followed them in 1944. He was assigned to Otto Frisch's Critical Assembly Group at the Omega Site. In July of 1945, he had helped transport the first plutonium core south across the desert to the Trinity test site. He had watched the first atomic weapon detonate at dawn on the sixteenth. He knew what the material in front of him could do.

On the evening of August 21, Daghlian returned to the Omega Site after dinner. He was alone. The laboratory's safety rules did not yet require at least two scientists present for any criticality work, though Daghlian had no colleague with him. The only other person in the building was a security guard, Private Robert Hemmerly of the Special Engineer Detachment, who sat at a desk about twelve feet from the assembly table. Hemmerly was not a physicist. He was not part of the experiment. He was there to watch the door.

Daghlian began stacking tungsten carbide bricks around the plutonium core. The bricks reflected neutrons back into the sphere, and each layer pushed the assembly closer to critical. He added them one at a time, reading the monitoring instruments after each one, watching the count rate climb. When he held a final brick above the stack, the counters jumped. The assembly was too close. He pulled the brick back. It slipped from his hand. It fell onto the core.

The assembly went supercritical. A burst of radiation filled the space around the table. Daghlian reached in and knocked the brick away with his right hand. The reaction stopped. The exposure had lasted less than a second. Hemmerly, twelve feet away at his desk, had done nothing wrong. He had not touched anything. He had simply been in the room.

Daghlian's right hand began to blister within hours. Over the following days, the full course of acute radiation sickness set in. Nausea came first. Then burns spread from his hand across his body. Then the damage moved inward to his organs. He was taken to the base hospital. There was nothing the physicians could do to reverse what the radiation had done to his cells. His body failed by stages, one system after another. He died on September 15, 1945, twenty-five days after the accident. He was twenty-four years old.

Los Alamos convened a special committee to review what had happened. The committee wrote new rules. A minimum of two scientists would be present for all criticality work. At least two monitoring instruments with audible warnings would be running during every experiment. A written plan for operating methods and contingencies would be filed in advance. The rules were adopted. The plutonium sphere stayed in use. The rules did not address the specific technique that a senior physicist had already begun performing with it.

Louis Slotin was thirty-five, born in Winnipeg, Manitoba. He had earned a doctorate in physical chemistry at King's College London, then crossed the Atlantic and joined the Manhattan Project at the University of Chicago, where he worked under Enrico Fermi at the Metallurgical Laboratory. At Los Alamos, he had assembled the core of the bomb used at the Trinity test. He was one of the few men alive who had handled the internal components of a nuclear weapon with his bare hands. He had gone to the hospital to sit with Harry Daghlian in the weeks before Daghlian died. He had seen what radiation did to a man's body at close range and over time. By the spring of 1946, Slotin was preparing to leave Los Alamos. He had been offered a teaching position at the University of Chicago. His replacement, a physicist named Alvin Graves, needed to learn the laboratory's procedures before Slotin left.

The experiment Slotin performed most often involved lowering a beryllium hemisphere over the plutonium core to reflect neutrons back into it and measure how close the assembly came to sustaining a chain reaction. The approved method called for small shims between the two halves of the reflector, metal spacers that kept the hemispheres from closing completely around the sphere. Slotin did not use shims. He held the top hemisphere with one hand and kept it propped open with the flat blade of a screwdriver in the other. It gave him finer control, he said. Enrico Fermi is said to have told him he would be dead within a year if he kept working that way. Other physicists in the laboratory called the technique reckless. Slotin kept doing it.

On the afternoon of May 21, 1946, Slotin set up the experiment one more time. He was showing Graves the procedure before he left. Seven other people were in the room, including the physicist Raemer Schreiber. Slotin placed the plutonium core on the assembly stand. He lowered the beryllium reflector over it, holding the top half open with the screwdriver blade. He brought the two hemispheres closer together, watching the instruments climb. The screwdriver slipped.

The top reflector dropped into place around the core. The assembly went supercritical. The room filled with a blue flash and a wave of heat. Schreiber, standing a few feet away, saw the light clearly even though the room was bright from the windows. The burst lasted a fraction of a second. Slotin knocked the reflector off the core with his bare hand, breaking the reaction. He turned to the others and told them to mark exactly where they had been standing so their radiation doses could be calculated. He walked outside and vomited. He knew what was happening to him. He had watched it happen to Daghlian nine months before, in this same laboratory, from this same sphere of plutonium.

Slotin was taken to the base hospital. His left hand, the one that had knocked the reflector away, swelled and blistered within hours. Over the following days the damage spread through his body. The medical report noted: "From this day on, the patient failed rapidly." He died on May 30, 1946, nine days after the accident. He was thirty-five years old. Both accidents had fallen on a Tuesday. Both on the twenty-first of the month. Both men had been killed by the same sphere of plutonium. Both died in the same hospital room.

After Slotin's death, the physicists at Los Alamos gave the sphere a name. They called it the demon core. The laboratory ended all hands-on criticality experiments. Raemer Schreiber designed remote-control equipment that could perform the same work from a quarter of a mile away. No one would stand next to the assembly again.

No criminal charges were filed against anyone.

The core was melted down that summer and recast into a new weapon. Alvin Graves, the physicist who had been standing nearest to Slotin during the second accident, was hospitalized for weeks with severe radiation poisoning. He developed chronic neurological and vision problems. He died in 1965. Marion Cieslicki, another physicist who had been in the room, died of leukemia in 1965.

The guard who had been sitting at a desk twelve feet away on the night of the first accident died of leukemia thirty-three years later.