As in the course of April and May 1943 the standard of Short Signal menus and the results obtained dropped while the consumption of bombe time rose steadily, a conflict arose in Hut 8 between the Reencoders and the Short Signallers. The Short Signal faction had large numbers of extremely bad cribs and bombes suitable for running them, the Reencodement faction had a number of absolutely certain cribs with now two High Speed machines in action - but in action at this stage did not mean that they worked anything like all the time: the only High Speed job to have come out so far was a Limpet job which had been taken to Letchworth for an experimental run on Feb. 13th. The source of these reencodements, which were now miraculously on the increase and which never again failed us, were the Arctic U Boats which had supplied Doenitz' famous message to his 'U Bootsmaenner' which broke March 14th, 1942: these Arctic U Boats used Dolphin until September 1944 when Narwhal was introduced. It was the belief of the reencodement faction that a better return would be had from using 3 wheel bombes with 4th wheel plugboards and running reencodements slowly, as had been done in March 1942, than was being had from the long series of Short Signal menus, which very, very rarely came out. Eventually a change of policy was agreed upon.
The gradual abandonment of the Short Signal attack makes interesting reading in the Weekly Reports of the time and they show clearly how our attitude to the policy gradually changed:
June 4th. We continued to have a very sticky time on Shark, only 28th and 29th May being broken despite great expenditure of bombe time.
June 18th. We have again got stuck on Shark, no day after June 12th having come out despite heavy expenditure of bombe time. One day had a number of apparent dummy messages in spite of which we failed to get it out and most of the other days looked fairly hopeless from the beginning. It is difficult to say how far it is really worth while spending the very large amount of bombe time that we do on running extremely seedy cribs.....We used 5,505 bombe hours or between 55% and 60% of the total - a new high record.
June 25th. The 4 wheelers had their first success when Trinidad (Trinidad was a name given to a bombe, the reference to Washington is to an early bombe at Op-2O-G) got out June 16th after which we were able to break 15th, 17th, 18th and Washington got up May 31st...... We had a meeting with Block A and Hut 6 to consider again the policy of running weak Short Signal Cribs. Analysis of the results of running 'long shots' seems to show that the average menu of this kind has about a 3% chance of coming out. We have moreover now got a number of