by Theo Emery
Mustard took longer to manufacture. Problems with synthesizing it continued to plague Conant and Organic Research Unit No. 1, slowing the drive toward large-scale production. As the weeks slipped by, the uncertainty over making mustard frustrated the Ordnance Department. George Burrell, the research chief, was frustrated, too. Burrell and Walker discussed the problem at one of the weekly meetings. Afterward, Burrell wrote a letter to Walker acknowledging the slow pace of progress with mustard, the most desirable offensive war gas in existence. But Burrell cautioned that it didn’t make sense to lurch forward with production without knowing the best method for making it. The wrong decision could mean a waste of time and money, he wrote.
There were so many problems with the British sulfur-monochloride method that the army moved ahead with a plan to produce mustard using the chlorohydrine method, signing a contract with the Commercial Research Company on Long Island, New York, on April 13 to manufacture five tons of mustard daily. Still, work continued night and day on the monochloride method. Eventually, Conant’s diligence paid off. He came up with a two-step process that maintained both the stability and toxicity of the mustard produced, with a result that was roughly 85 percent pure. Using this process, the Small-Scale Manufacturing Section produced a seven-hundred-pound batch of mustard. It soon became clear that the chlorohydrine method should be abandoned. The contract with the Long Island company was canceled, and work continued exclusively on the British method.
Finally, progress. From overseas, Fries’s chemists in the AEF Gas Service sent details of a new French reactor for making mustard, and the satellite labs worked on turning Conant’s small-scale production process into industrial-level production. By around May 1, the long-delayed plans for the mustard plant at Edgewood began to fall into place, and work on the plant began on May 18. It was to be a massive factory with a daily capacity of one hundred thousand pounds of mustard, consisting of four units radiating outward from a central ethylene compressor system, with each unit containing sixteen reactors. Mustard had taken so long to develop and taken so much nurturing that the Research Division chemists started calling it “the hill baby.”
Conant had made his mark. On April 8, he had received a commission as first lieutenant in the Chemical Service Section. A month later, the day before work began on the plant in Edgewood, he was promoted to captain. Soon, he would have a new assignment, one so shrouded in secrecy and thought to be so vital to winning the war that its details were barely committed to writing.
This new endeavor began with Winford Lee Lewis, now at work at Catholic University. Like Conant’s unit, his Organic Research Unit No. 3 had its own setbacks over the winter. Originally, the small unit of five to eight chemists had been charged with working on mustard gas. While Catholic’s facilities were better than American University’s in most ways, the chemical hoods that whisked toxic gases out of the lab there were inferior to those across town at the experiment station. And so the work on mustard remained at American, while Lewis moved on to other problems.
Lewis’s research unit had started experimenting in March with ricin, a poisonous protein derived from castor beans. About half of the men in the unit—now numbering about a dozen—were investigating large-scale production of ricin and whether bullets could be coated with the toxin. During the week of April 8, a member of the unit inhaled ricin powder and died. Despite this gruesome outcome, it didn’t hamper investigation of the poison. “Further tests of this method of administration are under way, designed to determine what promise ricin may have as a toxic dust cloud.” The other half of the men worked on experiments combining TNT and chlorine to create an explosive reaction that would produce phosgene.
But the experiments that proved to have the most promise, and would eclipse all the others in urgency and create an indelible legacy of the service, came almost by accident. Father Griffin, the lab chief who had gone to Johns Hopkins with Jim Norris, recalled a 1905 dissertation by a young priest named Julius Aloysius Nieuwland. For his doctoral work, Nieuwland had studied reactions between acetylene and other compounds. During one experiment, he had combined acetylene with arsenic trichloride, using aluminum chloride as a catalyst. The contents of the flask turned black, and when he cooled it in water, a thick, gummy substance resembling tar separated out, with a noxious odor. Nieuwland sniffed it and immediately fell ill, requiring hospitalization for several days.
Griffin pulled down the dissertation and pointed out the relevant section to Lewis. Organic Research Unit No. 3’s experiments began around April 13. The chemists first put one hundred grams of arsenic trichloride in a flask. Then, the chemists passed acetylene through the flask and put it in a water bath with little reaction. Then the chemists added a gram of aluminum chloride to the flask and heated it. Still nothing happened. Five more grams of aluminum chloride went in, and more acetylene was passed through the flask.
Now the chemists added more acetylene, but this time, they raised the temperature. A rapid reaction began in the flask. The liquid darkened, and a thin blue film developed on the top. The flask gave off a sickening odor that burned and irritated the eyes and noses of the chemists, who developed searing headaches that lasted for hours. They carefully distilled the substance, producing a few drops of oil that, like the undistilled liquid, had a nauseating smell and was extremely irritating to the eyes and nose. It was an auspicious result, Lewis’s report noted. “The headache resulting persists several hours and the material seems to be quite toxic,” he wrote.
The difficulty with the distillation process was the instability of the compound—when heated to separate it from water, it tended to explode and release arsenic trioxide. The solution to that problem came from Conant, who suggested using hydrochloric acid to remove the explosive catalyst. It worked. Lewis was able to distill and purify the mixture without danger of detonation.
About two weeks into the experiments, the scientists began testing the compounds in earnest. The work ended up producing three similar types of poisonous compounds, each one with slightly different properties. Two of them were extremely toxic blister agents and respiratory irritants, though one was far more potent than the other. The noxious fumes sickened the men and peeled the paint on the ceiling of the lab, so the chemists constructed a small-scale still on Maloney Laboratory roof, where they could produce larger batches in the open air. The compound proved to have powerful asphyxiating and lachrymal effects. The first reports on its properties appeared in mid-May with a series of skin tests. Another early report on research conducted at the University of Wisconsin in June described “very marked respiratory distress,” a rash that turned into blisters, edema, and scrotal burns “quite similar to those caused by mustard.” On May 24, a private named Lloyd M. Suthers arrived at the experiment station infirmary with burns on his foot—the first injury that the camp doctor attributed to the compound. The numbers grew rapidly—on June 5, the doctor saw six patients with such burns.
The research took a heavy toll on the chemists at Catholic. Organic Research Unit No. 3 suffered the greatest number of casualties of all the research units; at one point, half of the unit had suffered an injury connected to their experiments, and a third were sent away for ten days to two weeks to recover. Lewis himself was one of the casualties; after he tested the new substance on his own right hand, it swelled up painfully. Despite the injuries, reports claimed that the men’s spirits remained high, in part because the lab was run so well but also because of the chemists’ sense of self-sacrifice and loyalty to Lewis.
When Norris and Burrell saw Lewis’s reports, they must have been thrilled by what they read. Finding new war gases was no easy task, and this new compound met all of the criteria. It appeared lethal, and it was both a vesicant and a lachrymator, causing blisters and inducing tears. It could be manufactured without great difficulty and on a large scale. The precursor chemicals were plentiful and cheap. This new poison had all the promise of mustard gas and more, since its effects were felt immediately, while mustard
took time to cause its damage. Most important of all: the Germans didn’t have it.
On May 1, Norris ordered Organic Research Unit No. 3 to stop all other work and focus solely on this new compound. He increased the number of chemists at Catholic to nineteen and then to almost three dozen. The experiments became a singular priority. Secrecy, as ever, was crucial. Its chemical and technical name, chlorovinyldichloroarsine, couldn’t be used, so to confuse anyone who learned of the work on the hill, they adopted the same code used for mustard gas, G-34. They had another code name for it as well: “methyl,” an intentionally misleading description because the word had no connection to its constituent chemicals. To the men who worked at Catholic, however, it had a different name, derived from the name of the man who concocted it. They called it lewisite.
The shroud of secrecy that covered the work at both American and Catholic Universities also extended to the outpost on Jones Point. Walter Scheele’s new life as Dr. Smith had begun on April 8 in Peekskill, when he stepped through the lobby of the Eagle Hotel with his cane, accompanied by his Bureau of Investigation handler, Special Agent Warren Grimes, code-named Warren White. It was a half-mile drive to the boat landing. Levering’s launch was lashed to the quay where they had left it the previous night. The men boarded, and as the boat puttered out into the current, the river yawned open both upstream and down. Iona Island jutted out into the river to the north, and the Catskills rolled away to the west. The Hudson was slow moving and placid. It had less the look of a river than of a wide lake, with rocky shores and granite cliffs rising steeply into dense forest swaddling Bear Mountain.
The far shore grew close, the launch nosed up to the landing, and the men climbed out. It was a short walk to the American Potash Corporation plant. Grimes assessed the grounds, the buildings, the proximity to the river and the rail. It was a perfect location for Scheele’s work to proceed, since it would be difficult to get close to the plant unnoticed, and a secure fence would make approaching the buildings unseen impossible.
Inside the plant, Grimes and Scheele examined the facilities, took stock of the equipment, and determined what construction would be necessary. Grimes went to the office and asked for a list of the thirty or so workers on the plant payroll. He went down the list. Three Austrians worked for the company; they would have to go. He stopped at the name A. Kossovsky. A machinist and shift foreman who had brought a piano with him, Kossovsky was also a labor organizer with the radical IWW—“the Wobblies”—a union accused of pro-German leanings. Kossovsky had helped launch a strike the previous January. He had been trying to recruit employees at the American Potash Corporation to join the union. Fire him, Grimes told the company officials. Don’t even let him finish his shift—tell him to clear out immediately and to take his piano from the employee bunkhouse, too.
Grimes summoned the remaining employees, asking their full name, age, birthplace, marital condition, naturalization status, and connections to radicals. He explained to them that there would be a strict pass system from that point forward. No one would get into the plant without a stamped pass and signature.
He assigned four watchmen to duty—one at the gate, three patrolling the grounds. He warned them to keep an extra close eye on the brick building where explosives would be stored. Grimes ordered fencing material for delivery the next day and put men to work digging post holes.
From the moment Scheele arrived, Jones Point became a hive of activity, but his purpose remained a closely guarded secret. Though Grimes had purged Jones Point of potential troublemakers, there were still dozens of other employees ignorant of the new purpose of the laboratory, and it was crucial that they remain in the dark. Grimes made sure that everyone in the know always used the code names and that their true names were never uttered.
The house that was being outfitted for Scheele would also have sleeping quarters for Grimes or any federal agent who was there to keep an eye on Scheele. Thomas Edison’s chemist, Bruce Silver, had received permission to take a furlough from the inventor’s lab and work with Scheele at the laboratory. Grimes insisted that agents observe every experiment. As soon as each day’s tests finished, Scheele turned over the notes. Grimes told him nothing about the newspaper articles published after his arrest reporting that Scheele was in government custody. He also said nothing about the other figures in the bomb plot, some of whom had been in the headlines as well. He even contemplated locking up Scheele’s cabin at night, but decided that it would arouse more suspicion than it was worth.
The doctor spoke freely with Grimes. He relayed old gossip about the kaiser and his lovers in Germany. He rambled about the empire’s weaponry and claimed the German military attaché in Washington had stolen an American inventor’s plans for missiles propelled with liquid air. He talked about Germany’s plans to invade Canada and infiltrate Mexico and how Cuba was a nest of anti-American sentiment. Grimes listened carefully and jotted it all down.
Grimes reveled in his job as wrangler of the turncoat chemist. He had been at Scheele’s side almost every step of the way, and he wanted to stay there. But his time at Jones Point was coming to an end. In his report to the chief in Washington, he practically pleaded to remain on the job. “I would very much like to remain with him until the completion of the tests. It would be awkward to bring in a new man at this time,” he wrote. He signed it “Warren White.”
The appeal fell on deaf ears. At week’s end, the bureau’s new division superintendent, Charles DeWoody, ordered two other agents, Graham Rice and Francis X. O’Donnell, to Jones Point to replace Grimes. DeWoody didn’t specify why he dismissed Grimes, but Grimes was from Washington, and DeWoody was assigning men from his division in New York City. Their homes were closer, and they were familiar faces in the New York office. In his orders to Agent Rice—code-named “Riser”—DeWoody described the gravity of the situation. Not only was Rice guarding against the possibility of Scheele’s escape, he also had to protect the chemist. Accomplices in the ship-bombing ring were still at large, and they might want him dead. Grimes would brief Rice on the situation before returning to Washington. “All of his work must be maintained in the greatest secrecy,” DeWoody wrote.
Despite DeWoody’s insistence on secrecy, he had already proved himself less than reliable when he told the New York Times about Scheele’s arrest. He demonstrated his indiscretion again when he showed up in Peekskill a few days later for an inspection of Jones Point. Grimes was stunned to discover that the superintendent had brought his wife with him, as if they were on a weekend getaway to the countryside. To Grimes’s fury, the loose-lipped superintendent used Scheele’s and Grimes’s real names when introducing them to his wife, even though Grimes had insisted that code names must be used. When Grimes ushered DeWoody into the laboratory, he continued addressing Scheele by his real name, even though three plant employees were within earshot in an adjoining office with the connecting door open. DeWoody went on to tell Scheele facts that Grimes had carefully withheld, such as how the German spymaster Rintelen had said that Scheele would have to be “fixed”—that is, killed—and that it was common knowledge that Scheele was in the government’s custody. Alarmed, Grimes interrupted to tell DeWoody that he was mistaken. Irritated, the superintendent insisted that Scheele’s arrest was known all over the country. Scheele looked on as the two men argued, with the three employees still in the adjoining office.
DeWoody left after this spectacular bungle, but the damage was done. Grimes soon learned that at least two plant employees now knew that “White” wasn’t his real name. Worse, Scheele turned on Grimes and angrily told him he would stop work unless conditions changed at Jones Point, insisting that he would kill himself rather than continue under the restrictive conditions of his confinement.
The furtive activities of the men did not go unnoticed outside of the plant either. A week after DeWoody came up to Peekskill, O’Donnell brought Scheele to a drugstore in town. While they were there, an Office of Naval Intelligence agent spotted them and demanded t
hat the men accompany him to the police station, where the police chief interrogated them about why they were crossing the river each day to Jones Point. O’Donnell managed to sputter out an explanation without revealing Scheele’s identity or true purpose. Placated, the police sent the men on their way.
DeWoody, however, was irate. He felt the men should never have been in the drugstore in the first place, but he grudgingly acknowledged that no harm seemed to come from it. “I do not like the incident, but O’Donnell and Dr. Scheele seem well satisfied with the whole procedure and the outcome.”
There may have been another reason that DeWoody wanted men from the New York office alongside Scheele. It wasn’t just the plumbers and electricians at Jones Point who concerned DeWoody; it was Richmond Levering, whose business dealings in New York were getting new scrutiny. After the oilman’s departure under a cloud the previous fall, Levering’s reappearance had been no small worry for the department. Amazingly, in the short period since Levering had returned to the good graces of the department by scooping up Scheele in Cuba, whispers about the oilman’s transgressions had begun anew. Accusations of fraud that trailed his soured business deals gathered steam, with articles appearing in New York newspapers about angry shareholders in one of his companies. A new allegation had arisen against him: that Levering had attempted to bribe a stenographer in one of the cases against him. Then he had managed to create a diplomatic row by telling a Japanese diplomat that he was a target of Bolshevik assassins. Not only did the so-called plot prove unsubstantiated, Levering had revealed to the ambassador that he was involved with the Scheele case.