The History of Hut Eight

at A and proceed as before, coming out at N. N will not, however, light up on the lampboard, but rather T, because N has been steckered to T. This steckering process affects 20 letters, the remaining 6 are referred to as self-steckered and, when they are involved, the current proceeds directly to or from the wheels.


In order to decode a message one has then to know wheel order, clips, starting position of message, and stecker. Any 3 of the 8 wheels may be chosen - 336 possibilities. There are 17000 possible clip combinations and 17000 possible starting positions - in the 4 wheeled machine half a million. The number of possible Stecker combinations is in the region of half a billion. In fact the number of ways the machine may be set up is astronomical and it is out of the question to attempt to get messages out by a process of trial and error. I mention this as it is a hypothetical solution to the problem often put forward by the uninitiated, when in fact all the coolies in China could experiment for months without reading a single message. On most keys the wheel order and clips were changed every two days - this is the so-called "innere Einstellung" which was supposed to be changed only by officers and which was printed on a separate sheet of paper. The Stecker and Grundstellung (of which more later) normally changed every 24 hours. In a 31 day month the odd day was coped with by having 3 days on 1 wheel order - a triplet. This sometimes came at the end of the month and sometimes in the middle. Triplets happened also in 30 day months with the result that the last day was a "singleton" with a wheel order of its own. These rules were not obeyed by all keys but we shall meet the exceptions as we proceed with the historical survey. The later 4 wheeled keys had a choice of 2 reflectors and 2 reflector wheels; these were only changed once a month.


The next essential is to understand the indicating system. Various indicating systems were used simultaneously, but if we examine the most complicated fairly carefully the others will be simple to explain.

Dolphin, Plaice, Shark, Narwhal, and Sucker all built up their indicators with the help of bigram tables and K-book.


One half of the K book consists of a Spaltenliste containing all 17576 existing trigrams, divided into 733 numbered columns of 24 trigrams chosen at random. The second half consists of a Gruppenliste where the trigrams are sorted into alphabetical


< previous

next >