THE CODEBREAKERS
Page 75
Another antidetection measure was transfer. German agents would write their message in invisible ink on one sheet of paper, then press this tightly against another sheet. Moisture in the air would carry some of the ink to the second sheet without the telltale differential wetting of the fiber papers on which the iodine test relied. This compelled T.O.D. to find the specific reagent required.
Perhaps the most interesting development of the secret-ink war was the German instrument discovered by Shaw, Pierce, and Richter in 1945 and dubbed the “Wurlitzer Organ” because of its resemblance to that musical instrument. They found a burned-out shell of one “organ” in the bombed remnants of the Munich censorship station, and an undamaged one in the censorship station on an upper floor of the Hamburg post office. It examined suspected letters on an assembly-line basis by ingeniously exploiting some principles of physics to make the invisible ink glow. It first exposed the paper to ultraviolet light. This pumped energy into chemicals of the ink, boosting their electrons out of their normal orbits into higher ones. The chemical was then in a metastable state. The heat from a source of infrared then nudged the electrons from their higher orbits back into their regular ones. As they did so, the substance would give up, in the form of visible light, the energy that it had absorbed from the ultraviolet. Since this phenomenon will occur for nearly all substances, even common salt, though some will naturally shine more brightly than others, the Germans had a system that would develop a good many inks.
The chief difficulty with secret inks was their inability to handle the great volume of information that spies had to transmit in a modern war. One way of channeling large amounts was to dot the meaningful letters in a newspaper with a solution of anthracene in alcohol. This was invisible under normal circumstances but glowed when exposed to ultraviolet light. But with newspapers being carried as third-class mail, this was hardly the fastest method of getting information to where it was going.
The Germans then came up with what F.B.I. Director J. Edgar Hoover called “the enemy’s masterpiece of espionage.” This was the microdot, a photograph the size of a printed period that reproduced with perfect clarity a standard-sized typewritten letter. Though microphotographs (of a lesser reduction) had carried messages to beleaguered Paris as far back as 1870, a tip to the F.B.I. in January of 1940 by a double agent, “Watch out for the dots—lots and lots of little dots,” threw the bureau into a near panic. Agents feverishly looked everywhere for some evidence of them, but it was not until August of 1941 that a laboratory technician saw a sudden tiny gleam on the surface of an envelope carried by a suspected German agent—and carefully pried off the first of the microdots, which had been masquerading as a typewritten period.
At first the microdot process involved two steps: A first photograph of an espionage message resulted in an image the size of a postage stamp; the second, made through a reversed microscope, brought it down to less than 0.05 inches in diameter. This negative was developed. Then the spy pressed a hypodermic needle, whose point had been clipped off and its round edge sharpened, into the emulsion like a cookie cutter and lifted out the microdot. Finally the agent inserted it into a cover-text over a period and cemented it there with collodion. Later, one Professor Zapp simplified the process so that most of these operations could be performed mechanically in a cabinet the size of a dispatch case. The microdots, or “pats,” as T.O.D. called them, were photographically fixed but were not developed; consequently, the image on them remained latent and the film itself clear. In this less obtrusive form they were pasted onto the gummed surface of envelopes, whose shininess camouflaged their own. The pats could show such fine detail because the aniline dye used as an emulsion would resolve images at the molecular level, whereas the silver compounds ordinarily used in photography resolve only down to the granular level.
The microdots solved the problem of quantity flow of information for the Nazis. Professor Zapp’s cabinets were shipped to agents in South America, and soon a flood of material was being sent to Germany disguised as hundreds of periods in telegraph blanks, love letters, business communications, family missives, or sometimes as a strip of the tiny film hidden under a stamp. The very first discovered, and the most frightening, was one in which a spy was asked to discover “Where are being made tests with uranium?” at a time when the United States was fighting to keep secret its development of the atom bomb. The “Mexican microdot ring,” which operated from a suburb of Mexico City, microphotographed trade and technical publications that were barred from international channels—a favorite was Iron Age, with statistics on American steel production—and sent them to cover addresses in Europe on a wholesale basis, with as many as twenty pats in a single letter. Technical drawings also went by microdot. Other microdots talked of blowing up seized Axis ships in southern harbors, the deficient condition of one of the Panama Canal locks, and so on. Censorship discovered many of these, now that it knew what to look for, and this enabled the F.B.I.’s wartime Latin American branch to break up one Axis spy ring after another.
With mail and cable routes being screened so closely and subject to unpredictable delays, it was not unlikely that Axis agents would take to the ether to gain speed and avoid censorship. But here, too, the United States was ready for them.
The Radio Intelligence Division of the Federal Communication Commission had the job, in peacetime, of policing the airwaves, which are public property, for violations of federal radio regulations. During the war, its 12 primary and 60 subordinate monitoring posts and about 90 mobile units patrolled the radio spectrum for enemy agent radios. Teletype linked them into a direction-finding net coordinated from Washington. R.I.D. employed the latest radio equipment, including an aperiodic receiver that would give an alarm whenever it picked up a signal on any of a wide range of frequencies, and the “snifter,” a meter that a man could carry in the palm of his hand while inspecting a building to see which apartment a signal came from.
In the routine day-and-night operation of a monitoring station [wrote George E. Sterling, R.I.D.’s chief], the patrolman of the ether would cruise his beat, passing up and down the frequencies of the usable radio spectrum, noting the landmarks of the regular fixed transmission, recognizing the peculiar modulation of a known transmitter or the characteristic fist of a familiar operator, observing an irregularity in operating procedure and pausing long enough to verify the call letters, or finding a strange signal and recording the traffic for close examination, and then sometimes alerting the nation-wide net to obtain a fix on the location of its source. More than 800 such fixes would be made in an average month, requiring the taking of some 6,000 individual bearings.
This sort of efficiency, built up during the long prewar years, so terrified the Japanese that when an agent requested permission to set up a transmitter, he was turned down on the ground that the F.C.C. would nab it as soon as he went on the air. It evidently made its impress on the Nazis as well, for only one bona-fide Axis station was ever heard in the United States. It was the German embassy in Washington which, using the call letters UA, tried to make contact with Berlin in the few days after Pearl Harbor. They never succeeded—nor did any other Nazi agent.
But R.I.D. did not live up to its acronym only within the United States. It insinuated its supersensitive antennae into the furtive Morse whisperings between other continents and thereby made an unexpected contribution to the Allied war effort. This began even before Pearl Harbor, when monitors at Miami heard station UU2 using irregular procedures. R.I.D. antennae swung silently, and the station was soon pinpointed in Lisbon. After a month of listening, monitors in Pittsburgh and Albuquerque finally picked up its correspondent: station CNA in South America. Then Lisbon station BX7 was identified from characteristics of its signal as being UU2 with different call letters, and a week later BX7’s correspondent, NPD, was discovered in Portuguese West Africa. Sterling’s men continued monitoring the little network, and two staff members who had become interested in cryptanalysis, Albert Mcintosh and Abr
aham Checkoway, solved the transposition cipher in which the traffic was enciphered. The decrypted messages disclosed German agents in the neutral colonies and countries of Africa reporting on all manner of things—ship sailings, troop movements, political conditions. When McIntosh and Checkoway solved a message from Lisbon, indiscreetly ordering an agent codenamed ARMANDO in Portuguese West Africa to have one of his assistants “deliver letters personally to Porter Hotel, Duas Hacoes, Victoria Street, for Mr. Merckel,” the fate of the German ring was sealed. Several weeks later, Allied counterintelligence cleaned it out.
Following this demonstration of its ability, R.I.D. was asked early in 1942 by its British counterpart, the Radio Security Service, to cooperate in watching German diplomatic and espionage networks. The two agencies discovered that many clandestine Nazi transmitters changed call-signs every day in a rota that ran for a month and then repeated. They catalogued the characteristics of individual transmitters and operators so that they could be recognized on different circuits. Counterespionage services told them how Nazi radio spies were trained in a school near Hamburg, how each agent’s “fist” was recorded to make radiotelegraph forgery by the Allies that much more difficult, how the spies set up their suitcase-sized transmitter-receivers and connected them to directional antennae that focused maximum signal power on Germany and minimized dispersion. As the war in Europe mounted toward its climax, R.I.D. found itself monitoring 222 frequencies used in clandestine European traffic, breaking most Axis radio-spy codes and ciphers, and reading nearly all the messages of nearly all German networks.
Some of its most spectacular results came in Latin America, where its interceptions of the numerous Axis radios enabled the F.B.I. to help local authorities weed out these infestations. R.I.D. gathered in the transmissions of the three main agent networks in Brazil, centered on transmitters LIR, CEL, and CIT. Solutions of the columnar-transposition cryptograms enabled R.I.D. to feed information to the F.B.I.’s wartime Latin American branch; its agents then gumshoed these leads until it ferreted out the members of the rings.
Such was the story of the CIT ring. In April of 1941, Josef Jacob Johannes Starziczny, an engineer who had been trained at the Nazi spy school, arrived in Rio de Janeiro as Niels Christian Christiansen, smuggling ashore a black leather bag containing a transmitter, four coding books, and microfilmed instructions. A month later CIT went on the air. It and two associated transmitters poured quantities of information into the ether, mostly directed at the Hamburg control station that used call-sign ALD for this operation and the signature Stein in many of its plaintext messages. The information was usually of high quality, and the agent radios reached a frenzy of activity early in March, 1942, when the liner Queen Mary arrived in Rio with 10,000 troops aboard. Sinking her would be a tremendous blow to the Allies, both in terms of the actual loss of the troops and the transport capacity of the vessel and in terms of morale. Because of her speed, she traveled without a convoy, and the agents in Brazil pounded out message after message to Germany in an attempt to enable the U-boat wolf packs to get on her trail.
“Queen Mary arrived here today at 10:00 … she must [go] to cellar,” wirelessed CIT on March 6, 1942. Two days later, CEL flashed: “Queen Mary sailed on March 8, 18 o’clock local time.” The next day, CIT tapped out, “With Queen Mary falls Churchill … Good luck.” Unknown to the operators, however, R.I.D. was eavesdropping. And when, on March 13, LIR sent a slow hand-keyed message on 11,220 kilocycles, an R.I.D. operator at Laredo, Texas, copied it with ease:
From one of LIR’s first messages, R.I.D. had discovered that the group based its calls and its transposition cipher on Axel Munthe’s The Story of San Michele, using an edition excluded from the United States and the British Empire. The agent determined the page to be used by adding his personal key-number to the number of the month and the date. The last line of that page furnished the call letters that LIR was to use that day—the first three letters reversed for the station in Germany and the last three letters reversed for the agent post. From analysis of previous messages, R.I.D. knew that the LIR operator’s keynumber was 56. Added to 3, for the month, and 13, the day, this would total 72. The last word on page 72 was “give,” so EVI was the proper call-sign for the sender. The repeated V stood for von (“from”).
The message had been sent in the early hours of the 13th, but had actually been enciphered on the 12th with the key for that date, which would be found on page 71, from 56 + 3 + 12. The first line on this page began, “I would have known how to master his fear,” and R.I.D., like the agent, assigned numbers to the first nine different letters:
I W O U L D H A V
1 2 3 4 5 6 7 8 9
This key deciphered the first four groups, with other letters acting as nulls:
I W E O F WO N U G I U V B J D L V C P
1 2 3 2 3 4 149 659
which meant “12 March, 2304 hours, 149 letters, 659th message.” The next group, NABRS, identified the agent, and the 149 remaining letters were then inscribed into a columnar transposition block in accordance with a key derived by taking the initial letters of the first twenty lines on page 71 (skipping indented lines) and forming a numerical key from them. The result was:
This single transposition, with x’s as word-dividers, sufficed, and the text can be read directly from the tableau. In English, and without the repetitions: “Message six zero from Vesta to Stein. Queen Mary reported off Recife by Campeiro on eleventh at one eight hours MEZ [Middle European zone time].”
But all the German effort that went into the collection and transmission of this intelligence was frustrated when the Queen Mary, perhaps alerted by these solutions, dashed across the Atlantic, eluding the submarines and making port safely with her 10,000 men. The Queen Mary messages were, in fact, the last the three Nazi radio rings ever sent. Brazilian police, tipped off by the F.B.I, and helped by an R.I.D. agent, swooped down on Starziczny on March 10 and on the other two rings soon thereafter. Two hundred agents and sympathizers were arrested and the spy nets destroyed.
The stupidity of sending a key so that anyone could understand it was not confined to LIR. The best-organized Nazi espionage apparatus in South America, which had transmitted its messages to Germany via stations PYL and PQZ in Chile, was finally smashed in February of 1944, largely through R.I.D. and F.B.I. information. Afterwards, its chief, Major Ludwig von Bohlen, set down his “Experiences Gained from the Valparaiso Process,” and his very first item read: “The cardinal mistake was the insufficiency of the original key, and the transmission of the second key together with the first. If the communication identification word had not been deciphered, the code would not have been broken.”
There were, however, two stations communicating with Hamburg which R.I.D. left undisturbed—even though both were operating on American soil. The first used the call CQ DX V W2—CQ being the general call to anyone to listen, DX meaning that a long-distance contact was desired, V meaning “from,” W2 being the prefix assigned to amateur radio operators’ stations in the second call area, which embraced Long Island. The two or three letters that should follow the W2 prefix were not used. This station began operating at 6 p.m. on May 15, 1940, trying to reach the Hamburg station on a wavelength of 14,300 to 14,400 kilocycles from a small house at Centerport, Long Island. After several tentative contacts, AOR in Hamburg flashed back on May 31 with the first full-length coded message, demanding information on monthly airplane production, how many were exported to all countries, especially England and France, whether they were delivered by ship or air, and whether payment was credit or cash and carry. The messages were enciphered in a transposition cipher like LIR’s, keyed on Rachel Field’s best-selling All This, and Heaven Too. This seems to have been the standard Nazi spy cipher, since it was still being used in 1943 by two German agents caught in Newark. The F.B.I., however, was able to read the Centerport messages without crypt-analyzing them—not because the G-men were clairvoyants or miracle-workers, but because they had made them up in t
he first place.
The ostensible operator of the Centerport set was William G. Sebold, a German-born American citizen secretly working for the F.B.I. When, in the summer of 1939, Sebold returned to his native Mulheim for a visit, the Gestapo had stolen his passport and threatened to harm his Jewish grandfather unless Sebold promised to become a spy for Germany in America. Sebold, who had meanwhile contacted American authorities in Cologne, pretended to agree. After a course at the Hamburg spy school, he returned to the United States on February 8, 1940, contacted the Nazi agents whose names he had been given, and set up the radio station to transmit their information to Germany.
Two F.B.I. agents were in fact manning the Centerport set, enciphering and sending messages whose contents had been carefully screened to include enough true information to seem authentic and enough false data to be misleading. At the same time other F.B.I. agents were using the contact to build up evidence against Frederick J. Duquesne and other Nazi agents. On June 28, 1941, the F.B.I. , in a sudden series of raids, smashed the largest spy ring to be uncovered before Pearl Harbor.
Even more successful in tricking the Nazis was the double-agent ND98. An importer-exporter when the Nazis recruited him for espionage, he sold out to the Americans as soon as he arrived in the Western Hemisphere to carry on his spy activities for Germany. Like Sebold, he established a radio station on Long Island under F.B.I. supervision and began feeding the Nazis with a careful mixture of true and false data. For example, he hinted that the United States planned a major operation against the northern Kurile Islands in the Pacific. In fact this was to be merely a feint for the real assault, on the Marshall Islands. As hoped, Germany relayed the information to Japan, and later the Joint Chiefs of Staff advised the F.B.I. that ND98’s information may well have contributed to the success of the invasion of the Marshalls in February of 1944. The Nazis liked ND98’s information so much that they paid him $55,000 for it and maintained contact with him from February 20, 1942, when the first of hundreds of messages was sent, right up to the very end of the war. They only stopped communicating on May 2, 1945, when the Hamburg station was captured by the advancing British.