Tuxedo Park

by Jennet Conant

  On May 11, London suffered its worst air raid of the war to date, with more than 1,400 killed. Three days later, the great British warship Hood was sunk. On May 22, Bush called on Stimson, and they had a long conference concerning Bush’s desire for a new organization for scientific research for the army and navy:

  He told me that the Navy needed it much more than the Army—they were more backward in it—but that the Army needed it somewhat. He told me that his proposition was that a new Assistant Secretary of War and Secretary of the Navy should be created which had this in charge, and when I asked who he recommended for it in regard to the Army, he said, “Alfred Loomis.”

  Five days later, on May 27, the president gave a vigorous radio speech that, while falling well short of what Stimson had suggested, firmly asserted the doctrine of the freedom of the seas and made it clear America intended to use “all additional measures necessary” to assure the delivery of supplies to Great Britain. Roosevelt also declared an “unlimited national emergency,” giving his administration broader powers in dealing with the crisis.

  On June 5, Loomis and Stimson had lunch at Woodley to discuss the situation and Bush’s proposal in particular. “We hammered out the various ways and methods which he would have to do in his work,” wrote Stimson. “On the whole I think it is a very satisfactory arrangement, or will be one.” Loomis continued to frequent Woodley in June, and predominant in all of his talks with Stimson was his message, which he stressed over and over again, that many of the Rad Lab’s new airplane detectors were ready and should be put into use as soon as possible. A few days later, Stimson noted in his diary that after dinner with Loomis, he and Bundy “talked over Alfred’s particular specialty and what we should do to get the better system of communications and the protection system into the Army.” Bundy, who was a lawyer with no background in science, did his best to maintain good relations with the military while trying to help clear a path for the scientists. Whenever they hit a roadblock, he later recalled: “Bush would needle me and then I would needle the secretary and then the secretary would hit the army over the head.”

  The only problem was that the army knew that Stimson’s sympathies lay with the scientists—with Bush, Compton, and Loomis—and that he, too, was impatient with them for failing to modify their weaponry and methods as soon as the new technical advances became available. Both sides had nothing but harsh opinions of the other. As Bundy put it, “The military don’t like to be needled particularly. And they would have naturally the feeling that these damn scientists weren’t practical men; they were visionaries. . . . And they didn’t want to waste time on something that wasn’t going to win the war.” So back and forth the arguments went, with Loomis making urgent back door appeals to Stimson to do something. A week later, on June 19, Stimson wrote:

  Bundy has come back with word from Loomis and Karl Compton, who have been conducting investigations and experiments up in Boston. He reports them as saying the time has come to freeze the present situation—to waste no more time in experimentation but to go on and build plenty of instruments as we can with the knowledge we now have. They said the developments had gone along far enough so that we could depend on them now. There is always a reluctance of the Department to stop experimenting and I knew we would find it here particularly. . . . However, Loomis and Compton are going to be down here in person next Monday, so we arranged for another meeting with them to clinch the matter as to all the delay.

  No matter how hard Loomis tried to push ahead, now regularly going over Bush’s head straight to Stimson, he could not get the Army Signal Corps to move faster. On June 23, Stimson, after a conference on the delay in constructing airplane radar detectors, vented his own frustration in his diary:

  It has been terribly held up by the finesse of the Signal Corps, who have been fussing over it for years instead of copying the workable arrangement the British have. I was fairly shocked to find how little they had done today. I dined with Bundy, and he had Loomis and Compton there, and also had in McCloy and [Robert] Lovett and we talked the whole thing over in the evening and if the fur doesn’t fly tomorrow I’ll miss my guess. The same old story of the better being the enemy of the good! and our Departments are worse sinners in this respect than anybody I know. They fuss over things trying to better them until the crisis is on us and the troops haven’t got any of the equipment in question.

  By keeping up the pressure, Loomis eventually achieved his end, and that spring Stimson ordered the first radar for the Army Signal Corps into immediate production. In the months to come, the white-haired secretary of war, who at seventy-three was in the position of having to evaluate and approve a whole new generation of advanced weaponry, would lean heavily on his cousin’s technical expertise, as well as the scientific counsel of Bush, Compton, and Conant. Through these “dippings down,” as Stimson called his practice of consulting directly with a trusted adviser on the progress in a specific field, he was able to cut through the bureaucratic double talk of the military and maintain a surprisingly accurate picture of what was really going on within his organization.

  Meanwhile, Bush had been working on a solution to the stalemate. On June 28, Roosevelt, by executive order, created a new, greatly expanded organization called the Office of Scientific Research and Development (OSRD). Directed by Bush, with Conant as his number two, the OSRD would be run by scientists like Compton and Loomis as a flexible, fast-moving, and creative source of new weapons. They would be the first civilians to “push their heads into the generals’ tent”: they would be working toward military objectives, but independent of military control and unburdened by their outdated notions of what was and was not possible. The scientists had prevailed. Finally, substantial federal funds would be poured into university laboratories, not only greatly accelerating the pace of work, but enabling them to move beyond pure research to the production of revolutionary new devices that would make all the difference in the coming contest.

  NOW that there was no longer any question that the laboratory would continue, it began to grow exponentially. By the spring of 1941, the Rad Lab staff had already grown to more than 140: 90 physicists and engineers; 45 mechanics, technicians, guards, and secretaries; and 6 Canadian guest scientists. Over that summer and fall, it would swell to almost 500 people, and more than $19 million would be committed to the secret radar systems they were developing and assembling. The penthouse roof laboratory was so dangerously overloaded that Cambridge authorities worried that it presented a serious fire hazard and urged MIT to relocate the whole operation to Mitchell Field on Long Island. Loomis and Compton dismissed this idea, but a new two-story building was slapped up and promptly filled to overflowing. More space on campus was procured, and almost every week, MIT students would arrive at a classroom only to find it sealed off and teaming with strange men.

  Loomis and Lawrence’s handpicked crew, which had labored with bunkered intensity on the AI radar, unencumbered by bureaucracy or, for that matter, any kind of formal routine, was evolving into a massive research and development organization. “It was a magnificent enterprise—staggering,” recalled Bowles, notwithstanding his frequent complaints about Loomis’ loose management style and indifference to housekeeping chores. “We went up by octaves on our money.” He remembered being in Compton’s office one afternoon when the MIT president was calculating that they had a budget of about $500,000 or so, and seeing where things were headed, he multiplied it by two. “But even then he was well under it,” said Bowles. “There’s nobody that can waste money like a physicist, but I think the result was extraordinary.”

  The sustained chaos of the first year could no longer serve as a management style, so an older, seasoned administrator named F. Wheeler Loomis (no relation), the longtime chairman of the Physics Department at the University of Illinois, was hired to sort out the personnel problems—not all egos adjusted equally well to teamwork—and impose discipline and order. A skilled bureaucrat, he was, as one early recruit observed, exactly wha
t the unruly mob of prima donnas required, “a son of a bitch.” If virtually every request for further funds or equipment had met with the approval of Alfred Loomis, under the day-to-day direction of the easygoing DuBridge, almost nothing got past Wheeler Loomis, who made frequent use of the word no.

  As Loomis and his physicists kept envisioning new devices and setting off in new directions, the Rad Lab kept getting bigger and spawning new projects. The core group in the lab was still concentrated on Project I, perfecting AI equipment for aircraft. Lawson designed a new rugged spark-gap TR box, and in April it was incorporated into the B-18 bomber system, making it possible to pick up ships at a distance of fifteen miles. By late May, one of the rooftop model AI sets was sent at the army’s request to Bell Labs for production, escorted there in the protective custody of two Rad Lab physicists.

  At the same time the Roof Lab was mastering the art of ten-centimeter radar, Rabi, who was head of research, was already pushing on to the three-centimeter model, which would provide even sharper focus and more detailed information. As far as he was concerned, the lab’s mission was “to develop something which could do as much harm as possible to the enemy.” In considering any new tactical device, his standard query was “How many Germans will it kill?” Developing the three-centimeter cavity magnetrons demanded new components and even more challenging techniques, and while the military regarded it as another wasted effort, the policy makers at the Rad Lab were determined to pursue every promising avenue. Loomis made sure the three-centimeter project went forward, and it would succeed beyond all expectations. Because they now needed magnetrons in large quantities for their radar devices, Raytheon was also contracted to manufacture the disks and would supply the first three-centimeter magnetrons that would be used against the Germans.

  Work on Project II, the microwave gun-laying radar, which had begun in January, was also progressing quickly. Loomis, in part because he was one of the few scientists at the lab with a background in astronomy, had suggested early on that they should use a conical scan to give precise azimuth and elevation, an innovation that played a vital part in the system’s success. The other key role was played by Louis Ridenour, a brilliant and caustic physicist from the University of Pennsylvania, who bullied his group into going for broke in trying to build the first fully automatic tracking system. At the time, all naval fighting sets were manual, and automatic tracking was not considered feasible. Ridenour wanted to develop a ten-centimeter microwave radar set that could pick up an enemy plane on the screen, lock in on it, and follow it while continuously feeding the coordinates into a computer, which would point the antiaircraft gun at the target. After working out the theory for Loomis’ conical scanning and borrowing freely from the physicists working on the airborne set, Ridenour’s group was able to get a set to automatically track a plane from the roof of Building 6 on the last day of May. Six months later, an improved system was demonstrated for the Signal Corps at Fort Hancock. Obviously superior to its predecessors, it would become the prototype for the SCR-584 automatic tracking radar, one of the most important radar sets to come out of the Rad Lab, which was used by the army throughout the war. Thousands of SCR-584s would be deployed in battle and would play a crucial role in protecting the ground troops from air attacks.

  THANKS to Loomis’ preoccupation with what had come to be known as his “shower idea,” the Rad Lab was also making great headway on Project III, the need for a long-range system of navigation independent of weather conditions. Back in October 1940, in the thick of the marathon planning sessions for the new radar lab, Loomis had talked to Bowen about Tizard’s conviction that the North American continent was much better suited than war-torn Europe to develop and test a long-range navigation system. Loomis, “who must have been working a 24-hour day,” recalled Bowen in his memoirs, “had fully appreciated this and practically overnight—on the basis of the description I had given him of the British GEE—came up with the suggestion of doing a similar thing. . . .” Pacing back and forth in the library of his New York penthouse, Loomis had excitedly elaborated on his idea to Bowen:

  What about a pulsed hyperbolic system, like GEE, but on long waves which would be reflected from the ionosphere and would, therefore, give a range of one or two thousand miles. Since the two ground stations would themselves be about a thousand miles part, there was a problem of synchronisation. This he proposed solving using his specialty—in this case highly accurate quartz clocks—at each station.

  Bowen had thought it a “marvelous idea,” and from that time on, Project III had proceeded along the specific lines Loomis had suggested, becoming the basis for a new long-range navigation system, originally called LRN for Loomis radio navigation, though after Loomis objected to its being named after him, it was changed to Loran, for long-range navigation. Loomis had proposed a rather ingenious scheme in which pulsed radio waves from fixed shore stations would produce a grid of hyperbolic lines from which planes or ships, equipped with a specially designed pulse receiver, could fix their position. The key to Loran, as Alvarez later wrote, was Loomis’ use of a time-measuring technique—a system of receiving and comparing the time of arrival of pulses—an expertise he had accrued during his long obsession with precision timekeeping:

  The Loran concept of a master station and two slave stations can be traced to the Shortt clocks, which had a master pendulum swinging in a vacant chamber, and a heavy-duty pendulum “slaved” to it, oscillating in the air. To obtain a navigational “fix” with Loran requires the measurement of the time difference in arrival of pulses from two pairs of transmitting stations. Each such time difference places the observer on a particular hyperbola. The observer’s position is fixed by the intersection of two such hyperbolas, each derived from signals originating from a pair of long-wave transmitting stations. . . . The techniques for separating the signals and for measuring their differences in arrival time were “state of the art” at that time, but the problem of synchronizing the transmissions to within a microsecond, at points hundreds of miles apart, was a new one in radio engineering. Loomis proposed the following solution: the central station was to be the master station, and its transmissions were timed from a quartz crystal. The other stations also used quartz crystals, but in addition, monitored the arrival times of the pulses from the master station. When the operators noted that the arrival time of the master pulses was drifting from its correct value, relative to the transmitting time at that particular “slave station,” the phase of the slave’s quartz crystal oscillator was changed to bring the two stations back into proper synchronization. This procedure was able to bridge over periods when the signals at one station “faded out,” and it was also what made Loran a practical system during World War II. . . .

  With so many brilliant physicists pursuing independent lines of research, it certainly did not hurt that Loran was Loomis’ own idea. His proposal was quickly approved by the microwave committee, and a group was set up to order the necessary parts, test equipment, and oversee the installation. A small group headed by Melville Eastham began work on the system early in the summer of 1941, and while waiting for equipment to arrive, they made a series of improvements, including moving to a longer wavelength to allow over-the-horizon operation. The basic system was completed in September, and the first field tests with a system using medium frequencies were conducted over the next three weeks.

  A tunable receiver had been installed at Harvard’s Cruft Laboratory, which had been made available, and another was set up at Lawrence’s room at the MIT Graduate House. They had also obtained two abandoned lifeboat stations from the Coast Guard—one off Montauk Point, at the end of Long Island, and the other at Fenwick Island, off the coast of Delaware. They continued their investigation, running tests between the coastal stations and receiver stations in the Midwest in order to get a general idea about the behavior of sky waves over land. The main receiving station was set up by Donald Kerr in Ann Arbor, Michigan, in the home of the scientist S. A. Goudsmit, and they mad
e control observations with a receiver mounted on a station wagon, stopping at Springfield, Missouri, and Frankfort, Kentucky. The tests strongly supported the possibility of stable sky-wave transmission and were so promising that they decided to abandon the original plan, which called for the ultrahigh frequency. As a result, Loran became the sole Rad Lab product not based on microwaves, an irony that was not lost on any of the Tuxedo Park pioneers.

  Loran proved to be an extremely important new method of navigation, its principle virtues being that it was simple and highly successful. By means of Loran charts, created by the Hydrographic Office, an operator could plot his position accurately in about two minutes. More important, for wartime use, the ship or plane equipped with Loran emitted no signal that might betray its position to the enemy. It also proved relatively impervious to weather, with only severe electrical storms disrupting the system. By day, fixes could be obtained up to 700 miles from the transmitting stations, and by night, up to 1,400 miles. The NDRC immediately ordered that Loran be put into service in the North Atlantic. On September 25, Loomis reported on the Rad Lab’s rapid progress to Stimson, who noted that it was “a very interesting talk. Things here at last seem to be jumping along.”

  While Loomis was congratulated for the dazzling ingenuity of Loran, there were those who found its similarity to the British system—the two schemes turned out to be virtually identical, though at the time the Americans were not permitted to know the details of GEE—too coincidental. Loomis’ loyalists credited him with arriving at the idea independently, granting that the sketchy facts furnished by various members of the Tizard Mission might have helped to “clarify or perhaps crystallize the project.” Bowles, who had always chafed at working under Loomis, was outraged that the financier had somehow managed to usurp their British partners, and he made no secret of it. His efforts to stir up controversy were stymied by Bowen, the mission’s chief technical expert, who fully supported Loomis’ account of his bathtime brainstorm and later testified to the fact when the navy applied for a patent in Loomis’ name.