GREENBERG | 91
received). IJTV USYX was the indicator (repeated at the end of the message). The final line,
added by the British interceptors, gave the time of interception, 1231, and the transmission
frequency, 7640 kilocycles (kHz).
The message, from the admiral commanding U-boats to all U-boats in the Bay of Biscay, said:
BISMARCK MUST NOW BE ASSUMED TO HAVE SUNK. U-BOATS TO SEARCH FOR
SURVIVORS IN SQUARE BE6150 AND TO NORTH WEST OF THIS POSITION.
The number of intercepted Naval Enigma messages increased steadily throughout the war,
from about 300 messages a day in 1940 to 1500–2000 messages a day during 1944–45. The high-
est number of intercepted naval messages recorded on a single day was 2133, on 12 March 1945.
The daily key
The daily key (or daily setting) told the operator how to set up his machine for the day. The naval
daily key had four components:^8
• ‘Plugboard pairs’: usually ten letters were connected to ten other letters at the plug-
board. The remaining six unplugged letters were known as ‘self-steckered’, meaning
‘self-plugged’. Each of the self-plugged letters remained unchanged when current
passed through the plugboard.
• ‘Wheel order’: three wheels selected from eight were placed in the machine in a specific
order. There were 8 × 7 × 6 (i.e. 336) possible permutations of the wheel order (more in
the M4).
• ‘Ground setting’: this was a group of three letters that specified the position of the
wheels the operator was to use at the start of encrypting the ‘message setting’. The mes-
sage setting, described in what follows, is itself a trio of letters, such as BDK, denoting
the position of the wheels at the start of encrypting the actual message text. In either
case, the operator turns the wheels by hand and when the three prescribed letters show
in the viewing windows, the wheels are in their starting position.
• ‘Ring setting’: this was the position of the rotatable alphabetic ring around each wheel.
With a separate setting for each wheel’s ring, there were 26 × 26 × 26 (i.e. 17,576) possi-
ble configurations of the rings. The purpose of the ring setting was to disguise the start-
ing position of the wheels. Even if one knew the message setting, it was not possible to
infer the actual starting position of the wheels from it unless the ring setting was also
known. This is because the letters making up the message setting were not fixed relative
to the wheels but inscribed on the movable ring.
The complete daily key was not in fact changed every day. On adjacent days, called ‘paired
days’, the wheel order and ring setting would remain unchanged; in a 30-day period, the wheel
order and ring setting would generally change only fifteen times. In 31-day months, there was
normally one 3-day period with the same wheel order and ring setting. On some Naval Enigma
ciphers, a complete change of daily key was made as infrequently as every ten days.
The German ‘key makers’ restricted themselves quite unnecessarily in a number of ways.
Hut 6 discovered the rules used by the compilers of army and air force keys, calling these ‘rules
of keys’.^9 The naval key makers had unnecessary ‘wheel order rules’: for example, the wheel
order generally contained a wheel VI, VII, or VIII. The wheel order rules could on occasion