which revealed that April 1st - 10th all had the same W.O. and Ringstellung. The new Seahorse procedure (I do not know when it came into force) was to have 3 W.O's only for the month lasting for 1st - 10th, 11th - 20th, 21st - 31st respectively; this meant a very great saving in bombe time as once one day was broken the nine associated days gave little trouble. Following the discovery the Americans broke most of the keys for 1944 and, as 1945 obligingly repeated 1944 keys, its traffic could also be read.
33. About this time Op. 20 G began a long and unsuccessful attempt to break into Pike, a surface raiders and blockade runners key used outside Home Waters; this attempt never succeeded and is a good example of the difficulty of making a break into a new key if no initial help is given by the indicating system and if there is no real knowledge of the traffic. The attempts were made on possible re-encodements from Sunfish (the test to verify their correctness being the usual one of "non crashing" between cypher and text) but inspite of a number of fairly plausible shots being produced, no success was obtained.
June.
34. In June Porpoise and Grampus abandoned the throw-on system, a belated reformation. This made little immediate difference to Porpoise, on which cribs were good, except to increase its running time; ever since the partial abandonment of the system in February we had been breaking on cribs, using the alphabets merely to reduce the running time by a technical dodge from that of a 4 wheel key to that of a 3 wheel key. This refinement we could not now use and bombe time was about doubled. Grampus, a less important key and one with worse cribs, was more seriously hit by the change. There was a blank period at the beginning of June followed by a temporary revival when a re-encodement from Henno, the Mediterranean reserve hand cypher, was found and from June 17th to August 12th most of the traffic was read; after this the capture of Constanza killed the Henno R.E. and on August 17th the key was officially abandoned since it was clear that the bombe time needed to
-72-
Back to Cryptographic History of Work on the German Naval Enigma contents