13

 

The Coming of the Ultra Americans

 

 

While the British were happy to see the influx of American fighting men, they were not so sure about the advent of U.S. codebreakers. A serious question arose as to whether the Americans and the British could ever reach an agreement to collaborate in signals intelligence. There were convincing reasons why they should unite their cryptanalytic programs, and a good many lives were lost because of their failures to do so. But there were almost equally powerful reasons why they should not. Chief among these were security concerns. No military unit with a hold on some aspect of Sigint trusts any other unit to protect its secrets. So it was then: security gave teeth to interservice rivalries. The U.S. Navy would not entrust the U.S. Army with its methods of cryptanalysis, much less make them available to the British. The British, for their part, saw American intelligence operatives as gabby, loose-lipped security risks who couldn't even keep the U.S. press from blabbing the Midway triumph. Seen from this perspective, both British and American intelligence seemed riddled with small-minded staffers who used the security issue more to guard their own turf than to question the advisability of cooperation.

Slowly both sides realized, however, that they had significant assets that could be shared. The Americans had their Magic; the British had Ultra. Both sides also had intelligence leaders who saw the need for mutual support. As early as August 1940, the head of the U.S. Army planning staff, Brigadier General George V. Strong, led a contingent to England to propose a "periodic exchange of information" between the British and American governments. Hinsley reported that at this meeting, Strong described U.S. progress against Japanese and Italian ciphers—this before the break-in on Purple was completed!

The British reciprocated by sending some of their most respected and talented representatives, including Alan Turing and Alastair Denniston, to discuss cryptologic matters with the likes of William Friedman. But the response was far from complete openness about the doings at Bletchley Park.

In February 1941 the Americans followed with another act of unprecedented generosity. Four junior American officers, including Friedman's disciples Abraham Sinkov and Leo Rosen, arrived in London bearing the gift of a Purple machine or—the records conflict—possibly two, which the British quickly put to work cracking Japanese diplomatic messages in Europe and the Middle East. In reciprocation the Sinkov team received a whirlwind tour of Bletchley Park that included lectures on the general cryptanalytic methods used by the British.

After these promising overtures, however, advances became a matter of two steps forward and one back. The Sinkov mission actually stoked the fires of U.S. opposition. The Americans' exposure to BP had been so superficial—it did not include even a look at a British bombe—that it was seen as a poor trade for the goodwill shown by the U.S. On the British side, surprisingly, Winston Churchill got his back up about yielding any greater depth of Ultra material to the U.S. and declared himself in favor of a "stiffer attitude" toward sharing secret intelligence. The cause was not furthered by that new indication of American security laxness: Colonel Fellers's unwitting revelations to Rommel of British plans in North Africa.

Nor was it helped by the opinion of the U.S. ambassador to Britain, Joseph P. Kennedy, that the British were doomed to defeat and should not be entrusted with U.S. intelligence secrets.

One event that strengthened cooperative attitudes in both camps was the Lend-Lease Act that Franklin Roosevelt negotiated through Congress in March 1941. The British, with their cash reserves depleted, saw this as an imaginative way to "lend" matériel that no one would be expected to return or make postwar payments for, as had been the case in the Great War. Senator Robert Taft compared it to lending chewing gum: you certainly didn't want it back. To avoid having much-needed U.S. lend-lease cargoes loaded onto ships that would then be sunk by U-boats, both navies became more forthcoming in their exchanges of North Atlantic intelligence. The Admiralty, as noted previously, advised the U.S. on setting up a Submarine Tracking Room similar to that of the Royal Navy. Further, an agreement with the Danish ambassador in Washington allowed the U.S. to establish a base in Greenland and later one in Iceland so that the U.S. took responsibility for protecting convoys over an increasingly broad area of the North Atlantic.

Gradually the collaborationists won out over the die-hard resisters, an evolution that accelerated with American entry into the war. In May 1943, this progress culminated in the BRUSA (Britain and USA) Agreement, calling for a comprehensive exchange of intelligence information. One important decision was what Hinsley called "an effective division of labor." The United Kingdom was to concentrate on the ciphers of the European and Mediterranean theaters while the U.S. took the lead responsibility for securing and disseminating intelligence about the forces of Japan.

An interesting outgrowth of BRUSA was an invitation to have American soldiery join in the British codebreaking effort. In fairly short order during 1943 the U.S. Army Signal Corps organized the 2nd Signal "Service Battalion, which was to select and train three outfits to serve in Britain. These were the 6811th, 6812th and 6813th Signal Service Detachments. The original plan called for 485 officers and enlisted men to be assigned, but the number swelled past 500 as more Americans were absorbed into the various detachments. In addition, U.S. Army Intelligence assigned a select group of officers to work at Bletchley Park.

With the integration of these Americans into the Ultra program in late 1943 and early 1944, the promise of effective Anglo-American collaboration could be said, despite a continuance of much bickering and internecine suspicion, to have been, very belatedly, fulfilled.

 

 

6811: Interceptors of Axis Radio Traffic

 

The process of signals intelligence begins with interception. Radio operators must tune in to the rush of Morse dit-dah symbols and meticulously copy them down so that the texts cryptanalysts read are as free as possible of skips and garbles. Even for the most dedicated and experienced operators, the copying of Axis radio traffic in World War II posed a demanding test. Enemy transmitters deliberately used only enough power to reach the farthest station in their network. Allied radio ops trying to listen in from well outside the receiving loop had to strain to catch what they weren't meant to hear. Frequently the stream of Morse they were assigned to copy was only the thinnest thread of sound to be picked out amid a welter of booming senders closer in.

"The intercepts had to be fought for, in a battle of skills, wits and wills," British intercept operator Joan Nicholls has reported in her self-published book about her life at Britain's Beaumanor station. "The Germans were adept at placing obstacles in the way of the eavesdroppers." The obstacles included screening transmissions with "what sounded like military bands, operatic arias, Wagner and speech, all played backwards and at high speed. It sounded like hell let loose in our ears, and underneath it all were the faint sounds of our station." Wavering signals and weather interference could make the job still tougher. A GI operator said it was like going to a noisy cocktail party and trying to listen in on a whispered conversation on the far side of the room.

The Americans had to struggle against the impression of being Johnny-come-latelies. Before the GIs ever arrived, the British network of intercept stations, mostly operated by young women like Joan Nicholls, were doing very well, thank you, without a bunch of Yanks getting into their act. The 7 officers and 195 enlisted men of the 6811th were asked to establish and operate an additional station, which the British labeled Santa Fe. While U.S. help was no doubt useful in covering the widening range of German networks, there was always the feeling among the troops of Santa Fe, of which I was one, that we were added as a courtesy.

To house the station, the Yanks were given an ancient manor, Hall Place, in Bexley, Kent, on the southeastern edge of London. Hall Place was a crumbling relic with some of its walls shored up against collapse. In its cold, drafty interior, whose medieval ceilings seemed to push down on us better-nutritioned Yank residents, the dining hall was converted into a Set Room, with banks of gray-metal Hallicrafters receivers lined up on plank tables. The detachment included one group of radio operators newly trained in the U.S. and a more experienced group transferred from the intercept station they had been manning in Newfoundland.

Santa Fe went into operation on a round-the-clock three-shift basis beginning March 1, 1944. The British assigned it some twenty-five German networks to cover, mostly German air force nets. Many, if not all, of these they were already covering. Redundancy in intercept, we were to learn, is not a waste, since in this game, in which accuracy is needed so acutely, a second operator may correctly copy a signal that the first one partially or wholly missed.

To be assigned a Luftwaffe network meant that the 6811th interceptor had to copy every message sent by all the operators on that net, with the German ops often impatiently waiting their turn. The result was that the Yank was likely to emerge after an eight-hour shift with shaking hands and bloodshot eyes, badly in need of a pint in the nearest pub.

To each shift of radio ops in the Set Room was added a team of eight men rudimentally trained as "cryptographers." Any fantasies the cryptographers harbored of actually breaking significant enemy messages, with generals anxiously awaiting the decrypts, ran up against the reality of our identifying nomenclature: we were the CRR teams, for "Compilation of Reports and Records."

In fact, because of the limitations of the "need to know," the common soldiers at Santa Fe never learned whether the masses of five-letter code transmissions the radio ops received and the cryptographers processed were ever broken. The outfit settled into the routine of handling endless yards of gibberish and making sure that each unit of it was accurately filled in and correctly identified as to time, network and frequency, without ever knowing the content of any of it.

In addition to the radio operators and us cryptographers, each of the shifts also included a small team of Teletype operators who forwarded the intercepts by landline to "Station X"—in actuality, Bletchley Park. On the telephone links to British intercept stations and to X, the fine female British voices oozed condescension toward us late-come Americans.

To do our job, we cryptographers did need to know one important secret. This we summed up in three words: "predicted call signs." In military radio the senders of coded messages try to obstruct interceptors as well as codebreakers. One of the ways they do this is by regularly changing the letters identifying their stations. Unlike commercial radio stations, which want to be recognized as WOR or WLW, military senders try to leave interceptors floundering uselessly in the ether by abruptly switching to a different frequency and employing a whole new set of call signs. What was ABC on 4031 kilocycles suddenly becomes XYZ on 2778 kilocycles.

For their systems the Germans devised fat books of call letters, forty thousand three-letter call signs to a book. At changeover times, usually shortly after midnight, the German ops were armed with a daily numerical shift. The shift told them where to move forward or backward in the call sign books to arrive at their new rows of call letters.

By the time the 6811th set up shop, the British had this whole call sign mechanism completely pinned down. Santa Fe was supplied with the "Elephant" book, which meant that it was the fifth in the series. It hadn't simply been captured somewhere; it had been compiled by infinitely patient analysts. It had blanks in it that were filled in as new call sign lists and placement coordinates came down from X.

The significance of predicted call signs was clear. If a German spy should hear any reference to the prediction, the result could be calamitous. He would know that the German system had been compromised, and if that phase of it was broken, he could arouse suspicions that the whole German encoding superstructure was breached. Many of us had nightmares in which we had one too many glasses of bitters in a pub and blurted out our most closely guarded secret in the presence of a German agent.

The kind of minuscule contributions the Yank cryptographers were able to make is exemplified by one that I recall as probably my singly most important deed of the war. It came on a day in March 1945. At that late date, with the war collapsing around them, the Germans managed to do what they had not done earlier: they made a clean break to a whole new call sign and frequency system. Suddenly everything in the radio spectrum was chaos. The German networks weren't at their usual frequencies, and what was by then the F book was dead without anything to replace it. The CRR could no longer provide predicted call signs. Each radio operator went on search, trying to find his network by detecting the sending idiosyncrasies of one or more of his German transmitters. In preparation for this moment, we and our operators had been boning up for days on any distinguishing quirks of the nets assigned to Santa Fe.

I was leaning over the shelf beyond which Jim Hammet, one of our brighter operators, was twiddling his dials trying to latch on to something recognizable. "That's funny," he said.

"What's funny?" I asked.

"This net leaves a space between the first three letters and the last three letters of the indicator. See this: ARZ space DLY. That's different from my net. They run all six letters together without a pause."

Excitement swelled within me. "We have only one network that does that. Benny Abruzzo's. Give me the frequency and a couple of the call signs and let me have Benny check it out."

I carried my slip of paper to Benny's set. He turned his tuning dial with the delicacy of a safecracker. His face lit up. "That's him! That's one of my guys! I'd know that Heinie meat fist anytime, anywhere!" He grabbed his pad and pencil and began happily copying.

So we three pinned down one frequency and one set of call signs for one day of that one net's operation. It was just such small victories, endlessly accumulated at Station X and fitted like pieces in a giant jigsaw puzzle, that helped the geniuses crack the Germans' systems.

It did seem to me, when I reported our find to Station X, that the British female voice at the other end of the line was, this time, not quite so patronizing.

Other than for moments like this, the work was tedious in the extreme. Even so, everyone in the 6811th went at his tasks with one thought ever in mind: that this message, or the next one, or the one after that, could, when broken, make the difference for some soldier or sailor or airman, perhaps for a whole contingent of soldiers or sailors or airmen, out there on the fighting front. Any examples of surrender to ennui were few and far between.

The prevailing attitude can best be characterized by an incident in September 1944 when Hitler began sending over his V-2 rockets. One hit so close one day that it broke out windows and brought down a ceiling or two in the manse. A radio operator named Arthur Koester, a T/5 corporal, was at work copying his net when the frame of a window above him was ripped out and hurled on top of him. The wood and glass tore a gash in his head, sending blood coursing down over his face. Koester never stopped copying. Sergeant Vernon Pemberton, the Set Room chief, quickly had another op tune to Koester's net and begin copying the same traffic. Then he yelled at Koester, "You can break off now, Art. We've got your net covered." Koester just waved him away. "Come on now, Art," the sergeant pleaded. "Break if off so we can start taking care of you." Koester simply growled and set himself more firmly. Finally the sergeant grabbed Koester's pad and jerked it from under his hands. Koester shot to his feet, shouting at his NCO, "You stupid son of a bitch. Can't you see I'm copying?"

Art was one of two Purple Heart recipients in the 6811th. Our captain was so craven for recognition of the unit that he refused to distinguish between Koester's deserving wound and that of another GI who was hurt when a V-2 hit near the house in which he was shacked up with his British girlfriend.

 

 

6812: Operators of the Bombes

 

Bletchley Park relied on members of the Wrens to operate the bombes used to determine Enigma key settings. As the number of the Germans' Enigma-based ciphers increased, so did the numbers of bombes needed. At the peak, nearly two thousand Wrens were operating the bombes round the clock. For the women it was heavy going. Petty Officer Diana Payne subsequently recalled work on "the monster deciphering machines" as "soul-destroying but vital." After many months of this merciless routine, the strain began to show. As Payne reported, one Wren collapsed, rendered unconscious by overstrain. Another began to have nightmares from which she woke up desperately clutching a phantom drum of the sort she had been loading onto the bombes day after day. A girl who was due to have an operation became frantic with the thought that under the effects of the anesthetic she might give way and talk.

When in early 1944 fresh cadres of Americans began to move into that comer of Britain to help operate the bombes, their arrival was most welcome. These were the men of the 6812th Signal Service Detachment, 5 officers and 149 enlisted men. Whoever was doing the code-naming of the American outfits must have been a railway buff. The name for this unit was Rock Island. The troops were billeted on the northwest outskirts of London until they could get their own camp built, partly by their own labor, in the Middlesex town of Eastcote. Their three-shift work assignment was to operate an array of bombes in a building they called the Factory. They were close enough to London to be on the receiving end of a few V-weapons that overshot their target, but far enough away from Bletchley Park that they never went there and knew of its existence only via phone lines and teleprinter links. Here, too, the "need to know" limited the GIs from learning definitely that the messages on which they worked were ever broken. "We did what we were told," Paul Best of the detachment remembered later, "and had to take it on faith that our work was being effective in helping to solve the German messages."

Their orders came from Bletchley Park in the form of "menus" instructing them what to do. As Best explained,

 

Each drum of a bombe was an analogue of a rotor of the Enigma machine, with 26 letter positions on each drum. BP would send us a menu based on a crib of a German phrase that they hoped would be found in the message. The operator would place the drums on the bombe in the order specified by BP, turn them to the menu positions, plug in the cables at the rear, again according to the menu instructions, and switch on the machine. Essentially we were seeking a yes-or-no answer as to whether the crib was actually a part of the message. If the answer was no, we would keep the bombe running through its fifteen-to-twenty minute cycle and nothing would happen. We would call BP and report, "Sorry, that menu was no good." But if there was a "hit," the bombe would hunt for a voltage that indicated a circuit closing and would stop. We would note all the rotor positions and set up our replica of an Enigma machine in those positions to test whether the hit was confirmed. Then we would send the confirmation back to BP and wait for our next menu. With all that array of bombes in the Factory, we could check out hundreds, thousands, of possible cribs in a single day.

 

As with the men of the 6811th, those of the 6812th were haunted by their secret knowledge. "It was a load to carry," Best recalled. "Here we were, operating replicas of the Germans' encoding machines and, what's more, also operating the still more complex machines that, we felt sure, were being used to conquer the German machines. We dreaded the possibility of one of us going off his rocker and being unable to hold it in any longer. It never happened, but the possibility hung over us like an inescapable cloud."

Here, too, the work was a triumph of determination over boredom. The Americans did something the Wrens hadn't been doing. They kept records of the bombes' performance. Harold Keen, BP's representative at the British Tabulating Machine Company, which produced the bombes, was delighted. He used the GIs' records to make improvements in subsequent bombes.

 

 

At BP, Two Groups of American Codebreakers

 

The first American serviceman to be stationed at Bletchley Park was a young navy lieutenant junior grade. Joe Eachus had been one of the bright lads looking after Alan Turing in his 1942 visit to the U.S. Impressed, Turing had him assigned, in the autumn of that year, to Hut 8. Eachus later recalled those times when denizens of the Hut were struggling with the U-boat codes. "You could tell by people's faces how we were doing," he remembered. "When we were locked out, the faces were grim and discouraged, because we knew what havoc the Germans were creating out there on the seas. But when we were cracking the codes and helping our ships evade the U-boats, there were smiles all around."

Eachus was a special case. Only when the BRUSA Agreement was reached were sizable contingents of Americans invited to join in the code-breaking at Bletchley Park. In the last half of 1943 two separate groups began arriving.

First were the men of the 6813th. Of the three Signal Corps detachments of "Ultra Americans," this one was clearly the elite. To be assigned to it, the men had to have achieved particularly high scores on the Army Classification Tests. They received the more prestigious title of cryptanalysts while the members of the other units were identified as cryptographers, traffic analysts and the like. The 6813th heavily featured Ph.D.s, college academicians, chess masters and accomplished mathematicians. Completing the U.S. railroad cover designations, the unit was labeled Rio Grande.

Second of the U.S. groups at BP was a contingent that reported directly to Military Intelligence at the Pentagon. When it came to staffing U.S. Army Intelligence, Secretary of War Stimson looked to Alfred McCormack, like Stimson a former Wall Street lawyer. As reported in Thomas Parrish's book The Ultra Americans, McCormack slipped directly into the uniform of an army colonel and went at his wartime job with great energy and unflagging intelligence. An important task he took on was to organize and oversee the daily summaries of Magic decrypts that did so much in shaping the decisions that U.S. leaders made about the conduct of the war. With the signing of the BRUSA Agreement, Secretary Stimson also asked him to seek out the best available men for duty at Bletchley Park.

When he visited Bletchley he saw it as a "personnel heaven" because the codebreaking program got first pick of the best brains available. It was not that way in the U.S. Official personnel channels confronted McCormack with the rigidities of the civil service and the stifling rules of the military branches. It was maddening, since he was determined to send only first-class minds to BP, men who would fit into that select group without embarrassing the U.S.

McCormack found ways to get around the barriers. One was to comb the ranks of men already in the service. That way he came up with Lewis Powell, an air force intelligence officer who in later life became a justice of the Supreme Court, and Alfred Friendly, subsequently editor of the Washington Post and a television personality. McCormack also used his contacts at Ivy League schools to enlist men such as Telford Taylor, a Harvard Law School graduate then serving as general counsel of the Federal Communications Commission.

The two groups reported to different commanders and were billeted separately. Heading the 6813th was Captain William Bundy, of a Boston Brahman family and married to the daughter of Dean Acheson, then the undersecretary of state. In charge of McCormack's group, which became known at Bletchley Park as 3-US because they were assigned to Hut 3, was Taylor.

Three members of the 6813th supplied the vanguard of the U.S. troops dispatched to BP: Paul Whitaker, Selmer Norland and Arthur Levenson. Whitaker and Norland, both fluent in German, were assigned as translators in Hut 3; Levenson worked as a cryptanalyst under Gordon Welchman in Hut 6. He was soon joined there by Bundy.

Varied placements would become the rule for the members of the 6813th. Instead of functioning as a cohesive unit, they were integrated piecemeal into the complex British operation, a few here and a few there. The unit's Walter Sharp has recollected, "We came to BP on the same trucks, dispersed to our assigned offices and under the principle of 'need to know' didn't chatter about what we or the other fellows were up to."

Sharp's assignment was in what was called the Machine Room because each of the operators there sat before an Enigma clone. "The head of the shift," he recalled, 

 

was responsible for directing those operating the bombes, located at several remote stations, as to what problems they should attack. The rest of us were there to test the "stops" that the bombe crews telegraphed in. Once we had proved that a particular key for a given army or air force unit was broken for the day, the settings were sent to the Decoding Room, where all the messages for that day were deciphered, then to be passed on to translators, intelligence analysts and disseminators. At the same time, the bombe team working on the now-broken key was pulled off that and given another task.

 

Of his work in Hut 3, Norland remembered, "When a message was broken, the plaintext German was still in five-letter groups. We had to sort that out, try to fill in garbles and omissions and render the message into understandable English. Then we submitted it to the analyst on the watch, who determined its degree of priority and to whom it should be distributed."

A memoir by the late Jim Nielson told of his work in the Quiet Room. Here the operators took on messages that had resisted deciphering and for which there were no verified call signs. The Quiet Room's personnel made a second go against the messages to see whether they could be made to yield.

George Vergine, whom Parrish described as "a brilliant young cryptanalyst," wrote a memoir "for my relatives to tell them what I did during the war." At Bletchley Park, he worked for a time on the Enigma and then transferred to the "Newmanry," where he collaborated with Max Newman on deciphering Fish teleprinter codes. To read his memoir is to marvel at the mathematical complexities young minds could master under the pressure of breaking the codes that revealed "the thinking and planning of the whole German High Command."

As for McCormack's 3-US men, a number of them, after learning the ropes at Bletchley Park, were assigned to Special Liaison Units attached to American operational units. Don Bussey, for example, transferred over to the continent to become an officer in the intelligence section of the U.S. Seventh Army.

By being spread throughout Bletchley Park the Americans learned virtually every phase of the codebreaking program. "If someone had decided to pull all the Americans together and set them up as an operational unit," according to Sharp, "I think we could have done very well. By the end of the war we would have gained from our experience with the British folks most of the arcane skills needed to identify enemy communications, direct their interception, read and translate messages, and have some idea what to do with the information they contained."

The Americans of the 6813th were also the elite in their off-duty hours. At BP they were part of a diverse group that included many civilians and that ignored the niceties of military rank. They all were known to each other by their first names. The GIs drew neither KP nor guard duty; all that was done by people specifically assigned to those jobs. At their billet the Americans had their own very good cooks; "We probably ate as well as any Americans, at home or abroad," said Sharp. In the mess hall, officers and enlisted men did seat themselves separately. "Someone among the officers," Sharp reported, "had the idea of hanging a curtain between the two sections. The first time it was drawn, the racket from the EM side was such that the officers got the message. The curtain was not drawn again."

 

 

For the Ultra Americans, Praise and Honors

 

By war's end, the U.S. soldiers involved in Ultra had pretty well erased the idea that they were mere tokens of Anglo-American cooperation. They had earned their way. Recognition of their efforts came from the European theater's chief signal officer, Major General William S. Rumbough.

To those of the 6813th he wrote, "The extreme value of your work has been recognized." He commended the 6811th for "operating on a par with British intercept stations which had been in action for a period of three years or more and, in numerous instances, actually surpassing them in performance." He also congratulated the Teletype operators for their accuracy and reliability. The 6812th received a similar commendation from the general. The detachment's work was also praised in David Kahn's writings for "producing two to three times as many solutions as a comparable Wren unit, not because it ran the machines faster but because the men changed setups much more quickly."

British appreciation of the Ultra Americans' contributions at BP is indicated by the postwar awards they approved for individual performance. Bundy received the Order of the British Empire, while Norland was recognized as a Member of the British Empire. Robert Carrol, George Vergine, Cecil Porter and Harold Porter of the 6813th were awarded the British Empire Medal.

While the war continued, and for those thirty secret years afterward, the three Signal Corps detachments were almost entirely unaware of each other's existence. This changed when Fred Winterbotham's The Ultra Secret opened the floodgates of information. One member of the 6811th, Robert Fredrickson, gave himself a retirement career of locating the members of the three detachments and bringing them together. Now, more than half a century later, the Ultra Americans have begun to hold the kind of reunions that have become old hat to most World War II servicemen. In addition to posting regular newsletters and organizing the yearly reunions, Fredrickson periodically leads a trip back to the outfits' old haunts. They marvel at Hall Place, handsomely restored and with beautiful gardens replacing what had been GI drill grounds and ball fields, and they enjoy finding their way to Eastcote and the Factory. At Bletchley Park they are pleased to see that the mansion and its huts have been saved from the bulldozers bent on converting the grounds into a housing development during the thirty years when the wartime activities there remained a mystery. They are gratified to see BP made into a cryptology museum and to take part in the tours that have made it an increasingly popular tourist attraction. And in the mansion's library the American visitors take a moment to reread a plaque mounted on the oak-paneled walls that offers these lines of summary from Shakespeare's Henry V, act 2, scene 2:

 

The King hath note of all that they intend,

By interception which they dream not of.