Most Secret War

by R. V. Jones

  There is a twist to the story. In 1952 Churchill asked me to spend a year or so back from my University in the Ministry of Defence, to see what could be done to bring Intelligence back to its wartime efficiency; as a result, I was able to give only a few lectures to my students in that year. These were therefore summary lectures on topics that I thought would be useful, and one was on simple physical principles which I had myself found to be of great use in practical applications. Among these were centres of gravity, my thesis being that although calculations of centres of gravity were a dull process of seemingly little importance, there were many occasions when the exact location of a C.G. was vital, and I told the story of Cotton and his cameras. To add to the colour of the story I mentioned that the last I had heard of Cotton, whom I did not mention by name, was gun-running in the Mediterranean.

  After the lecture one of the students who, as many were, was an ex-Serviceman, came up and asked me, ‘Was your friend’s name Sidney Cotton, sir?’ When I replied that it was, the student said, ‘In that case, sir, I can cap your story! In 1948 Cotton was running guns into Hyderabad, and I was Flying Control Officer on the airfield where his aircraft used to land. On one flight he was flying in a field gun, which must have broken from its fixings, and moved as the aircraft came in to land. The result was that the aircraft became tail heavy and crashed, killing the crew.’ So Cotton’s encounters with centres of gravity did not always have a fortunate ending.

  I was once more to hear of him, and of his Hyderabad venture when I encountered yet another buccaneer who had joined forces with him in order to evacuate the Nyzam of Hyderabad at the time of the Indian take-over in 1948. The idea was that Cotton would provide the air lift, and the second buccaneer would make arrangements for accommodating the Nyzam in the manner to which he had been accustomed. The proposal was that they should rent one of King Farouk’s palaces in Egypt, and the second buccaneer approached the King. I cannot recall all the details, but the buccaneer told me that the King’s terms were that ‘I want 25 percent of all he brings out’ and the buccaneer had replied ‘25 percent of one hundred million pounds, Your Majesty, is a lot!’ The transaction was duly made, but not taken up because the Nyzam insisted on staying to say a last prayer as the Indian Forces were approaching, and the buccaneers had to escape without him.

  But returning to the war, what the Royal Air Force had failed to realize was that photographic reconnaissance was a specialized activity well worth thought and trouble, and not just a backwater for personnel who were not much good for anything else—and the same applied to a greater or lesser degree to the attitude of the Services to all forms of Intelligence activity. Cotton, who had a genius for getting things done—and frequently doing them himself—had seen where the effort needed to be made; fast high flying aircraft of long range, with good cameras and outstanding pilots, and with specialist interpreters for examining the photographs. Before long, his irregular methods were too much for the R.A.F. and he was ‘organized out’ of his leadership of photographic reconnaissance. He had a raw deal, for his contribution was great: and I for one am glad to have known him.

  On 8th September Enigma gave us the pinpoint to the nearest mile of the Cherbourg Knickebein as 49° 40.5’ north 1° 51’ west, and it should surely be visible on the 20-inch focal length photographs that, thanks to Cotton, were then becoming available. Nothing was reported by the interpreters, but I was in a better position to examine the photographs because I had at least some idea of what I was looking for: either a ‘squat tower’ or, more probably, some form of turntable. During a daytime raid on about 17th September, when we had to take cover in the basement, I took a collection of photographs of the Hague peninsula with me to make some use of my small space in the crowded cellar. Ultimately I spotted an object which looked rather like one of the circular filter beds of a sewage farm, with an arm some 30 metres long across it that had rotated between successive photographs just as the sprinkler arm of a sewage farm would do. But this was no sewage farm—it had not been there on photographs taken on 20th June—it must be the turntable of the Knickebein with a girder across it carrying the aerials.

  Fig. 2a Tentative sketch of a Knickebein beam transmitter made from the vertical air reconnaissance photograph at Plate 5a

  Fig. 2b. What this particular transmitter really looked like; but see Plate 5d for a Knickebein transmitter that more closely resembled the upper sketch

  It is shown at Plate 5(a); Figure 2a shows the optimistically detailed sketch that I made from the photograph, while Figure 2b and Plate 5(b) shows what it was really like. Plate 5(d) shows a low oblique photograph of the large Knickebein, 100 metres across, probably the one near Cleves, which bears a distinctly closer resemblance to my sketch. Plate 5(c) shows the entrance to the bunker of a second and later Knickebein near Cherbourg, with its pompous legend ‘ERBAUT UNTER ADOLF HITLER IN KAMPF GEGEN ENGLAND 1941’ (‘Built under Adolf Hitler in war against England 1941’).

  On 18th September I wrote a report describing the identification of Knickebein, thanking the officers who had supplied the photographs. This caused a minor stir at the headquarters of the Photographic Interpretation Unit at Wembley. So far they had not taken my requests as more than a sideshow to the operational demands for photographs of invasion preparation: when they found that I was in a position to write reports going up to the Prime Minister mentioning individuals by name for meritorious work, their attitude changed. The Commanding Officer, Wing Commander Lemnos Hemming, promptly allocated to my work one of his very best interpreters, Claude Wavell (a former surveyor and a distant relative of the Field Marshal, whom he resembled in appearance), who was henceforward to play a major part in my story.

  CHAPTER SEVENTEEN

  The X-Apparatus

  TWO OR three nights before the bombing of London started on 7th September, my sleep was interrupted by an event which made more impression on my memory than any bomb ever did. This was the telephone at my bedside ringing in the small hours of the morning and an excited voice saying, ‘This is Norman at Bletchley. We’ve got something new here. God knows what it is, but I’m sure it’s something for you!’ By this time I had come to know that, although he was a technical tyro, Norman had a magnificent ‘nose’ for sensing that something was important even when he understood nothing of it, so I readily agreed to drive to Bletchley in the morning.

  The cryptographers had broken a new line of Enigma traffic. There were mentions of beams, including one which said that the beam width was eight to ten seconds of arc, or an angle of one in twenty thousand, which would imply that the beam was no wider than twenty yards at two hundred miles. And there was the electrifying word ‘X-GERÄT’ which was being fitted to an aircraft with a call sign 6N+LK, which identified it as belonging to Kampf Gruppe 100. The unit had attempted to attack Birmingham on 13th/14th August; and, as we afterwards found, it was their raid on 23rd August that I had witnessed while at Hoar Cross. I quickly correlated the new beams with those which Scott-Farnie told me had just begun to be heard on frequencies around 70 Megacycles per second from the Cherbourg and Calais areas; and I asked Bletchley to put every possible effort into making further breaks into the new line of traffic. In this, I was in a much better position than if I had been simply in the Air Ministry because of the close and informal terms of my relationship with the Bletchley staff that had arisen through Winterbotham’s agency.

  So the X-Gerät was indeed something distinct from Knickebein, and by 11th September I had circulated my first report. By 24th September we had identified six beams to which the Germans had given the code names of rivers: Weser, Spree, Rhein, Elbe, Isar, and Oder, and we had the exact positions of the first two which were again on the Hague peninsula north-west of Cherbourg. The next three were near Calais, and the last near Brest. Kampf Gruppe 100 seemed to be working in somewhat irregular order through a book of numbered targets, and the chief scientist involved appeared to be a Dr. Kühnhold. We had the actual directions for the beams for 20th Septemb
er: the Germans had specified them to the nearest five seconds, implying an aiming accuracy of about ten yards at two hundred miles—if they had heeded the same lesson as I had learnt in my school physics of not specifying a measurement to a greater accuracy than your practical achievement would justify.

  Could such accuracy be attained with the radio waves of frequencies around 70 Megacycles per second that I had already associated with the X-Gerät (which would have wavelengths around four metres) or would it require still shorter wavelengths of, say, less than a metre? The Germans talked of coarse and fine beams, and the 70 Megacycle beams might be the coarse ones only. Further, there were mentions of centimetres in the Enigma messages. I had already been alarmed by the fact that we had no listening receivers for centimetric waves where, both according to the Oslo Report and to Bainbridge-Bell’s examination of the Graf Spee, German radar might well be, so I used the centimetric beam possibility as a lever to get a special listening watch on these wave-lengths across the Straits of Dover. This watch was undertaken by some young workers from the Telecommunications Research Establishment, including D. J. Garrard and E. G. Ackermann, and it was almost immediately fruitful for they detected radar-type transmissions on a wavelength of 80 centimetres, which appeared to be ranging on our convoys and directing the fire of the German guns.

  There was fairly frequent mention of something called ‘Anna’ and this was usually associated with a number between 10 and 85 which was usually a multiple of 5. By 17th October I had collected the following numbers: 10, 15, 25, 30, 35, 44, 47, 55, 60, 75 and 85. Another set of numbers that I collected at the same time gave the frequencies (such as 8750 kilocycles per second) of crystals that were issued to the beam stations for stabilizing their transmissions. It is normal for the stabilized frequency to be eight times the crystal frequency and if you multiplied the crystal frequencies by 8, you obtained the following series: 66.5, 67.0, 67.5, 68.0, 69.0, 69.5, 70.0, 71.0, 71.1, 71.5, 72.0, 72.5, and 75.0 Megacycles per second (1 Megacycle = 1000 kilocycles, so 8 f#8212; 8750 kilocycles = 70.0 Megacycles). The crystals were very much in the range of frequencies of the new beams that we were hearing, and if, as I had come to suspect, ‘Anna’ represented at least the tuning dial of the aircraft receiver, if not the receiver itself, there ought to be a simple relation between the ‘Anna’ numbers and the actual frequencies. The chance that the one set should nearly always end in 0 or 5, and the other in 0 or ·5, was so small as to be hardly coincidental, and it was simple to deduce that the ‘Anna’ reading had to be divided by 10 and either added to or subtracted from some constant number.

  One evening when the bombing in Richmond was fairly intense, and Vera and I had joined some neighbours with a flat on the ground floor, I left them and worked in the flat above because I thought that by now there ought to be sufficient numbers to resolve the problem unambiguously. Besides the numbers themselves, I knew that a Feldwebel Schumann had signed a return for three crystals for frequencies 69·5, 70·0, and 71·1; I traced him to the station at den Helder, which then had these three crystals only, and which was ordered to transmit on Anna numbers 30 and 35.1 soon found that the constant had to be 66.5 if one tenth of the ‘Anna’ number had to be added to it or 73.0 if it had to be subtracted. And since I knew that crystals for 75 Megacycles per second existed, the second possibility could be dismissed. I obtained further confirmation from the two crystals whose frequencies were not exact or half integers, and the problem was solved. Looking over the figures in retrospect, it was only on 17th October that we had collected just sufficient figures, and my feeling that the problem could now be solved was purely instinctive—if I had tackled it even the night before I could not have found a unique solution.

  A further fact that came out of the ‘Anna’ numbers was that the fine beams, as well as the coarse, were in the same range of frequencies between 66.5 and 75 Megacycles per second, so that wavelengths below one metre were not employed. It turned out that the occasional mention of centimetres referred to the precision with which a monitoring vehicle was to take up its position a kilometre or so in front of the beam station to align the beam in the desired direction.

  A further value of the ‘Anna’ numbers was that if we could obtain the information in time, among the instructions sent out to the beam stations during the afternoon before a raid, we should be able to tell 80 Wing the frequencies on which they should jam. At first, however, my interpretation of the ‘Anna’ numbers was not accepted, because there appeared to be frequencies outside the range that I had determined. These, as it transpired, were due to bad measurements of the frequencies of the German beams on the part of our countermeasures organization, a feature that was to plague us through the whole battle. The fault in this case probably lay not with the observers but with the calibration of our receivers which were not up to the German standards of precision.

  I had recommended on 11th September that similar countermeasures to those which we were employing against Knickebein should be developed against the X-beams, because even if we were only able to jam the coarse beams this might well stop KGr100 from locating the fine beams. We now knew that similar measures would work against both coarse and fine beams, and Robert Cockburn had the development in hand. Corresponding to the ‘Aspirins’ for Knickebein he produced the ‘Bromides’ for the X-system and these began to appear in October. However, they did not seem to have much effect on KGr100 which continued to bomb more or less as it pleased.

  I began to think about other countermeasures, encouraged by the fact that a far miss by a bomb intended by Bomber Command for Cherbourg actually hit the second X-beam station on the Hague peninsula, some 20 kilometres away, destroying the station and, unfortunately perhaps, making the Germans think that we knew a great deal about their activities, and leading them to postulate several theories as to how this remarkable feat was accomplished. They proceeded to tighten their security regulations, examining the past history of all the personnel on the stations. I wondered if we could repeat the feat intentionally, by flying along the X-beam ourselves, which should at least lead us directly over the station, and we attempted to do this but without much luck.

  During October KGr100 began to drop flares over its targets in England, and Lywood at once hailed this as either a success for our countermeasures, or at least as evidence that the X-beams did not work, because, he argued, the pilots were so unsure of the X-beams that they were dropping flares to find out where they were. I pointed out that we had no evidence that KGr100 had been so far disconcerted by our countermeasures, and that the real explanation might well be that they themselves could locate the targets with the X-beams, and were dropping flares in practice for operations where they would act as pathfinders for their de-Knickerbeined comrades in the rest of the Luftwaffe.

  This interpretation did not please Lywood, and added to the unpopularity that I had already acquired in some quarters on the Air Staff shortly after 15th September. Robert Cockburn remembers that when it was claimed by the Air Staff that we had shot down 186 aircraft that day, I stated that we had only found the wreckage of 38 aircraft and that I did not believe that there were more than another 38 in the sea, which was the Air Staff explanation for the discrepancy between claimed certainties and the number that we actually picked up. After the war we found that even my estimate was rather too generous, in that the German losses for the day were no more than 62. The explanation was, of course, that the Air Staff was fooling itself because it was quite unreasonable to expect any pilot who thought that he had destroyed an enemy aircraft to follow it down to the ground and watch it crash, and overclaiming was inevitable. Despite any unpopularity, I survived because war is different from peace: in the latter fallacies can be covered up more or less indefinitely and criticism suppressed, but with the swift action of war the truth comes fairly quickly to light—as Churchill said, ‘In war you don’t have to be polite, you just have to be right!’

  On my advice Lindemann minuted Churchill on 24th October: ‘There is so
me reason to believe that the method adopted is to send a few KGr100 aircraft fitted with special devices to assist in blind bombing on these expeditions in order to start fires on the target which any subsequent machine without special apparatus can use.’ And it turns out that on the other side Milch, the Head of the Luftwaffe, was advising Goering that the current policy of night attacks was useless without special radio-beam devices, like the new X-Gerät. He recommended that KGr100 should receive priority in personnel and aircraft, and that attacks should then be possible even on the darkest nights or through cloud.

  If only we could decode the Enigma messages in time, we could find where and when KGr100 was going to attack, and so to counter them by having fighters waiting and by having our jamming ready on the right frequencies. This would make great demands on the codebreakers, for the orders did not go out to the beam stations until the afternoon, giving only two or three hours to make the break. But for such a prize they strained every resource of human intelligence and endurance; and it was a great day, late in October, when they achieved this fantastic feat for the first time. Thereafter, they were able to repeat it on about one night in three. I was then able (having first worked out the position of the cross-beam stations near Calais) to tell the Duty Air Commodore at Fighter Command the exact place of attack, the time of the first bomb to within ten minutes or so, the expected ground speed of the bombers, their line of approach to within 100 yards, and their height to within two or three hundred metres. Could any air defence system ask for more? Despite this detailed information—and much to our disappointment—our nightfighters repeatedly failed to locate KGr100 aircraft, and I almost began to wonder whether the only use the Duty Air Commodore made of my telephone calls was to take a bet with the rest of the Command as to where the target would be for that night.