given us by the "non-crashing" wheel order rule. Suppose we had got out the 1st and 2nd of the month on W.O. 268 and the 5th and 6th in W.O. 874 - then the "non-crashing" rule told us that for the missing "sandwiched" pair, the 3rd and 4th, neither wheel 2 nor 8 would be used in the 1st position, neither 6 nor 7 in the 2nd and neither 8 nor 4 in the 3rd. Moreover the W.O. always contained a 6, 7 or 8 (as the three latest introduced wheels they had a special sanctity for the keymaker) and these rules combined reduced the number of W.O's to 105 - a lot compared with the 10 or so usually left in by Banburismus but a good deal less than the full 336.
11. Another rather different reason for which we sometimes ran all W.O. cribs was "flaps". Banburismus, needing as it did the whole of the day's traffic, inevitably involved some delay; a very good Banburismus might break within 1[?] hours of the completion of the day, so that if the paired day was quickly got out (as was usually possible) we might be current for a few hours. If there was some important operation as e.g. a Murmansk PQ convoy going North, or movements of the Scharnhorst or Gneisenau or later of the Tirpitz expected, this degree of currency was not good enough and the first good crib to come in was run as an "all wheel order" job without waiting for the results of Banburismus. Such a procedure could not be adopted as regular procedure until considerably later (June 1943) as it hamstrung the efforts of the German Air Section also badly in need of bombes; it says a good ideal for the good sense and cooperation of all concerned that there was never any friction with the Air Section over these sudden descents on the machines.
12. The chief unsolved problem now was that of Offizier. As explained in Chapter I [Para. 52], Offizier messages were doubly encyphered first by the officer - using the same W.O. and ringstellung, but different stecker from that used on the ordinary key - and then by the normal coding clerk. Thus when the day's keys were broken we were left with these special messages still unbroken. Two problems faced us (1) could we find any cribs which would enable us to get into even one
-32-
Back to Cryptographic History of Work on the German Naval Enigma contents