by Theo Emery
Despite Fries’s skepticism, the work in Washington was not breaking down. Just the opposite, in fact: through storms and cold, the chemists carried on their research. Van Manning’s January report on the offensive division’s progress summarized work on almost three-dozen projects, listing success after success, each one carefully bulleted and described in cautiously positive terms or as an outright achievement. He described new smoke devices used on the battlefield to obscure troops and hide trenches, and improvements to smoke boxes that belched chemical clouds to screen ships from submarines. Experimentation with flamethrowers had produced two models that spewed liquid flame one hundred feet. The station had tested two kinds of bombs to drop from airplanes and had been making progress on an American version of the Livens projectors, using smaller drums that could be shot farther than the British version.
Production of war gases was becoming more efficient. “The situation is satisfactory in regard to the manufacture of chlorine, chloropicrin and phosgene,” Manning wrote. “In fact the American Synthetic Color Company at Stamford expect to make 50,000 lbs. of chloropicrin per day from now on.” While that work was in the hands of the Ordnance Department, he boasted that “the preliminary research work was done by the Bureau of Mines.”
The chemists also made progress developing new gases: cyanogen chloride appeared to be a “very promising gas for use in gas shell,” Manning reported, while bromobenzyl cyanide proved to be the most powerful tear gas yet discovered. Magnesium arsenide, sodium arsenide, and calcium arsenide all had good results as well.
Over the months, the chemists had developed a system for investigating new war gases. First, the offensive research chemists analyzed a new gas’s physical and chemical properties, synthesizing small batches in the indoor laboratory. Samples went to the pharmacological section, where Yandell Henderson would study its properties and their effects on dogs, guinea pigs, monkeys: how lethal it was, whether it interfered with breathing, caused blisters, or had other effects. Then William McPherson, the chief of the Small-Scale Manufacturing Section, would work with his chemists on a process for synthesizing it in larger quantities, rigging up stills to make larger batches in the outdoor manufacturing shacks that dotted the grounds around the American University buildings. Fieldner’s Gas Mask Research Section ran tests with masks to determine whether the gases penetrated the different models of the American, French, and British masks. The Pyrotechnic and Dispersoid Sections tested ways of delivering it to the enemy, whether in artillery shell, Stokes mortars, Livens drums, as candles, or from cylinders. Throughout all of this testing, researchers investigated the availability of the precursor chemicals that were the raw ingredients. As gases moved through the research pipeline, the chemists who had become experts in their properties moved with them, graduating to each subsequent development stage. If the gas passed all of those hurdles and still proved satisfactory, then it was approved for mass production.
Mustard topped Manning’s list. Unfortunately, James Bryant Conant and Organic Research Unit No. 1 had been struggling with how to produce it. In the freezing labs of McKinley Hall, the men wrestled with a knot of problems that made the substance maddeningly difficult to make on a large scale. Toward the end of 1917, his unit had successfully prepared a batch of mustard using the slow chlorohydrine method that Victor Meyer had described and that the Germans were using. On paper, there was a better way. Mixing ethylene gas with sulfur monochloride yielded dichloroethyl sulfide and sulfur. When cooled, the sulfur then reacted with the dichloroethyl sulfide, yielding liquid mustard. But this process also produced chlorine in the solution, which made the resulting mustard impure and less effective. So much manpower and effort were expended on the king of the war gases that the experiment station gained a new nickname: Mustard Hill.
The problem consumed Conant day in and day out. The British and French had also been working on the mustard problem, so Lieutenant Colonel William H. Walker, the chief of the Chemical Service Section, cabled British counterparts for a description of their process. As Washington braced for more snow, help arrived from Britain in the form of a January 26 cablegram from the U.S. military attaché in London. The cablegram accompanied a report from British chemist Frederick Pope, who had discovered a much-simpler process for synthesizing mustard, using a catalyst to combine ethylene and sulfur dichloride. It was a breakthrough. “The British Authorities request that more than usual precautions be observed in keeping this matter secret,” the attaché warned.
Still, a viable method for synthesizing mustard was only a first step. Mass production was another problem altogether, fraught with technical obstacles. For two grueling weeks in February, Conant tried to replicate the British method for synthesizing the compound. Soon after, a second cable arrived with another breakthrough. The chemists discovered that lowering the temperature during the reaction prevented chlorination and helped keep the mustard pure. Even with the new information from the British, Conant still had difficulties. The British method using ethylene and sulfur monochloride worked but still yielded impure mustard when Conant’s unit attempted it at American University. The resulting material was extremely unstable, breaking up into hydrochloric acid and a thick black oil. In theory the British system was better for industrial-scale production, but it was also difficult and dangerous, tending to clog pipes and machinery. To work out these problems, the Bureau of Mines set up experimental satellite labs in several locations. One was outside New York City, and another was at the Dow Chemical plant in Midland, Michigan. A third was tucked into downtown Cleveland, about two miles from Nela Park, where the industrial wizard Frank Dorsey was in charge.
Gas wasn’t the only Research Division product that proved difficult to put into mass production. After the debacle with the first twenty thousand gas masks, Arno Fieldner and his mask division had worked to fix the flaws that had caused the first masks to fail so miserably, making a series of improvements on the British mask. But while the British mask worked better, it was also extremely uncomfortable and difficult to wear for more than a few minutes, and the Americans began looking closely at adopting the lighter, more comfortable French Tissot mask. Gas mask charcoal and filters had been steadily getting better. The Hero Manufacturing Company in Philadelphia had been making masks at a rate of five thousand a day, but that was nowhere near enough. The army had estimated it would need twenty thousand masks produced every day after January 1 but couldn’t find a company that could produce that volume.
To solve the problem, the War Department had taken another unprecedented step in November 1917. Rather than contract out the manufacture of masks, the Sanitary Corps of the army decided to open its own government gas mask factory. The army had already opened a charcoal-production factory in New York at the Astoria Light and Power Company, where tons of nutshells and coconut husks were shipped to be baked into activated carbon for masks. The army quartermaster found an industrial building for rent on Jackson Avenue in nearby Long Island City.
This was an extraordinary move. The president clearly had the power to go to war, but nowhere did the Constitution say that the president had authority to go into manufacturing. War Department lawyers pored over the legal ramifications of opening a government gas mask factory. The judge advocate general of the War Department had decided the idea was legal, and Secretary of War Newton Baker issued a memo to the surgeon general ordering him to establish the factory in the Jackson Avenue building, which the government began leasing on November 24 for $112,000 a year.
On January 1, 1918, ten sewing-machine operators were hired. Eventually, the Astoria plant and the Long Island City factories would merge into a five-building campus known as the Gas Defense Plant, a million square feet and thousands of women stitching, assembling, and testing gas masks. By the war’s end, 12,350 employees at the plant were turning out forty-two thousand masks per day.
By January of 1918, dozens of officers, scientists, and chemists from every corner of the Research Division now crowded into the w
eekly meetings at the experiment station. The research demands were outgrowing American University, and the limited space there was becoming a liability. Manning had no option: if there was no more room for research on the hill, the bureau would have to create more. Not just a room here and there—an entirely new laboratory. The spot he pinpointed as the best location was between McKinley Hall and Massachusetts Avenue, on an open area of the campus across the quad from the College of History building. He calculated it would need to be about 250 feet long, with an outbuilding for the power and heating plant. This wouldn’t be just any laboratory—it would be a huge research facility intended for three hundred or more chemists, one that Manning intended to be the most advanced and well-outfitted research laboratory anywhere in existence.
The building depended on approval from the engineers at Camp American University. “Our work is growing to such proportions that it is absolutely necessary to put up more buildings to take care of it, and I trust you will grant the necessary authority for the new installation,” Manning wrote to Colonel Mitchell, the commanding officer of the Twentieth Engineers.
But the engineers also coveted open space on the campus, and the colonel responded to Manning’s demand with obvious irritation. There was no space remaining for recruits to train and drill, and now the bureau wanted to lay claim to one of the few areas of level ground anywhere on the hill. He suggested another spot, to the west of McKinley Hall. Manning refused.
The work would go ahead on the original spot he requested, illustrating his leverage as the founding father of the gas research work. But his persistence didn’t help the bureau’s already antagonistic relationship with the engineers.
It would be months before construction could even start on the new building, and the technical men struggled to do their work in the hill’s cold and crowded labs. The urgency of the work and the chemists’ determination required them to tolerate the unfortunate condition of the McKinley Hall labs, with all of the accompanying hazards and inconveniences.
Those very conditions, within the elastic hierarchies of the civilian Bureau of Mines, would have a consequence which would result in one of the most important research discoveries of the war. It began with a winsome recruit named Winford Lee Lewis.
Lewis was another crucial addition to the war gas work. A thirty-nine-year-old chemistry professor from Evanston, Illinois, he had first come to Washington in October of 1917 at the suggestion of Charles Parsons, the chief chemist at the Bureau of Mines and secretary of the American Chemical Society. Lewis was a native Californian, a rarity among the bureau’s pedigreed scientists from Ivy League schools. His ebullient mind crackled with humor, and though he insisted that he was shy at heart, he loved dispensing advice and regaling an audience with a good yarn that reduced them to paroxysms of laughter.
Born in 1878, he had grown up in Gridley, California, a dusty village about sixty miles north of Sacramento. The Lewises were a pioneer family; both parents had come from the same town in Indiana, but his father, George, had gone west with a covered wagon train at age sixteen, joining the streams of prospectors and settlers pursuing dreams of an American El Dorado. He claimed a homestead of several hundred acres, with another 160 acres on an adjoining property, and yet another property he jointly owned with his uncle. Land rich and prosperous, he retraced his steps to Indiana several years after he left and returned to California with Lewis’s mother as his bride.
The Lewis farm sat in a tawny landscape of sepia-toned hills and catkins of live oaks. When Lewis was young, the family moved to the smaller, adjoining farmhouse, which sprouted a room every time a new child was born—there were eight children in all, five by Lewis’s mother, who died when he was about four, and three more from George’s second wife. Lee Lewis was about five when he started school at the one-room schoolhouse in nearby West Liberty. One June, an itinerant lecturer arrived at the school. With the children watching, the woman broke an egg into a saucer of alcohol, and the children watched as the egg turned white, pickled in the alcohol before their eyes. In a second experiment, she lit a match over another dish of alcohol and set the dish aflame as the wide-eyed young Lewis watched in awe. This traveling sorceress, it turned out, was a teetotaler, a temperance advocate trying to frighten the children over the evils of alcohol. Young Lee Lewis remembered it not as an abstinence lesson but as his first chemistry experiment, a demonstration of the combustible union of fire and alcohol.
When Lewis headed off to attend a fledgling new college called Stanford, his father wanted him to study law, but Lewis was drawn instead to chemistry. Mines of all sorts riddled California, and chemists were in constant demand as assayers, to determine the mineral content and value of extracted ores. At Stanford, his chemistry skills weren’t always evident. After he spilled a beaker of nitric acid onto a professor, the teacher suggested that Lewis try a different profession.
Still, Lewis stuck with it. He went to the University of Washington for graduate school, taught there for several years, then moved on to teach at Morningside College, a small Methodist school in Sioux City, Iowa. In the back of one of his classes, a dark-haired beauty named Myrtilla Mae Cook sat taking notes. He ended up marrying her in 1906.
From Sioux City, he went on to get a chemistry Ph.D. at the University of Chicago, with a minor in bacteriology. A stint followed as a chemist for the Department of Agriculture, and he began teaching at Northwestern University in 1910, serving as the city chemist for Evanston, Illinois, at the same time. He specialized in sanitation and chemistry. After a minor typhoid breakout in Evanston, he published papers about using hypochlorite—bleach— to counter the spread of the disease. When he began contemplating how to aid in the war effort in 1917, Lewis assumed he would be most suited for work related to sanitation and food safety. Charles Parsons, a captain in the Sanitary Corps’ food division, sent Lewis a job description and invited him to Washington, apologetically telling him he would have to pay his own way.
Lewis bought a train ticket to Washington in the fall of 1917, where he met with officers from both the Sanitary Corps and the Ordnance Department over several days. He decided against the former because it would almost certainly require him to go abroad; moreover, he didn’t relish the idea of making nutritional surveys of army mess halls, picturing himself pursued by irate camp cooks waving meat cleavers. Instead, he reached an agreement with the Trench Warfare Section of the Ordnance Department. He would finish out his semester teaching at Northwestern and return to Washington in January to work at American University.
Lewis had wired Myrtilla on October 9 to expect him home that Friday at 2:00 p.m. Lewis’s healthy sense of humor and quick wit showed even in the Western Union telegram telling her he had secured a position and a commission in the army, and a pay cut to go with it. “It seems we will have to live on prunes and beans for I am quite in beyond my means as a captain in the army,” Lewis wrote. He called himself “Captain Dad,” a family nickname that stuck with him for the duration of the war.
After Lewis finished teaching the semester at Northwestern, he arrived in Washington. His first job at American University was studying how metal shells corrode from the toxic liquids within. It was a crucial safety issue—new combinations of chemicals could react with the metal shells in unexpected ways, and damaged shells could leak their contents or explode. There was a fairly easy way to test the effect of the chemical agents on the metal: submerge shell shards in the liquid chemicals and observe the results. Lewis, however, took a different route. To mimic the conditions of actual shells, he instead put the liquid chemicals inside shells, heated them up, let them stand, and then reopened them to observe the effect of the chemicals on the interior.
However logical this approach, it had unpleasant side effects. On his second day, one of his assistants was gassed so badly that he was sent to the hospital. When the bureau sent Lewis replacement men, he refused to continue unless he had a safer work environment. It was well known that the chemists had little regard for the
army’s rigid hierarchies, and it was not the first such instance of casual rejection of military strictures. The chemists saw such displays as amusing signs of freewheeling independence. For officers, though, it showed not only insolence but disorganization and lack of discipline. Had the army been in charge of the research work at that point, Lewis might have been court-martialed for disobeying a superior officer or for insubordination. But because the bureau was a civilian-led organization, directed by scientists more concerned with problem-solving than rank, Lewis’s demand got him his way.
To the east of American University, in the city’s Brookland neighborhood, Catholic University of America had offered up its campus for the war effort as well. Unlike American University’s hastily rigged lab, Catholic’s Martin Maloney Chemistry Laboratory had been under construction long before the Bureau of Mines began looking for a laboratory, and it had opened in November of 1917. Catholic’s chemistry department was an enormous granite building trimmed with limestone and marble. The entire first floor of the east wing was dedicated to a state-of-the-art inorganic chemistry lab, while an organic chemistry lab took up the west wing on the opposite site. The main analytical laboratory was big enough to accommodate almost fifty researchers, and as many as sixty more scientists could comfortably work in more labs on the second floor. Luckily for the bureau, Father J. J. Griffin, the priest who headed the lab, had attended Johns Hopkins with Sunny Jim Norris, who was the chief of the offensive section of the Research Division.
After the bureau approved moving some of the research, Norris told Lewis to requisition a truck and a driver to take him across town to Catholic, where his Organic Research Unit No. 3 would resume its work at Father Griffin’s lab. Unsure of what he would need, Lewis selected an assortment of chemicals and acids that he thought would be useful, loaded the jars and vials and beakers into the truck along with a cage full of rats, and the wary driver set out across the city. It was about five miles to Brookland, but roads on Washington’s outskirts were not well maintained. En route, the truck hit a rut and bounced, smashing several bottles of hydrochloric acid and ammonia. When hydrochloric acid and ammonia vapors come into contact, they create a roiling cloud of ammonium chloride. Perhaps that was what caused the nervous driver, well aware of the work on the hill, to bolt from the truck on foot. It took a few minutes for Lewis to coax him back, and they eventually continued on—most likely at a slower pace—toward Catholic University.