1. Stabilisation and synchronisation are carried through. We have enough Enigmas.

Plain catalogue is complete and punched. At moment under revision. Will be duplicated for French. This is the only work on hand at present.

3. CYCLOMETER.

Gadget (a). An old (P.O.) cyclometer too elaborate for use, since original purpose does not now obtain.

Gadget (b). A small hand sex-cyclometer which was a mere draft, cannot be worked at a remunerative speed and is more often in course of repair than working.

4. PUNCHES.

We have two, and two more punches on order. Probably two more punch machines will be required.

URGENTLY NEEDED.

6. See Appendix I.

7. See Appendix II.

DOUBTFULLY NEEDED.

8. Machine of silo order to work on cyclometer results. No good results can be obtained from applying hand methods either.

Owing to the recent increase of the High Frequency Enigma Air Traffic on the Main Discriminant, there is now an extremely good chance of solving these messages on a smaller machine than that now under construction.

For this purpose, it is suggested that a "4-Bombe" (8 Enigmas), similar to the original Polish Bombe, should be built as soon as possible.

The time taken to solve a day's messages on the "4-Bombe" would not be unreasonable, although any precise estimate cannot be made at present.

The solution of Naval Enigma will divide itself into two parts, that of solving one message of a day, and that of solving further messages.

The first problem is to be tackled by:

(a). Analytical methods, using Jeffrey's statistics (virtually hopeless).

(b). By the machine now being made at Letchworth, resembling, but far larger than the Bombe of the Poles (superbombe machine).

For the second problem; i.e. solving further messages, we may either:

(i) Guess three or four letters of the message.

(ii) Make use of another machine, the "rack", which operates by so setting the messages that the decode contains sufficiently many letters E.

We have at present no information which will be of use for Method (i), although when a number of messages have been solved it may be applicable. Without a "rack" we shall, therefore, not be able to get any further if, for instance, position Stecker were captured from a submarine.

With the "rack" we shall, in such cases, almost certainly be able to solve 40% of the messages, and probably 70%. If by that time we are able to apply Method (i) as well, we may be able to solve as many as 200 messages on that day. If this ever happens it will be possible to solve the indicating system; i.e. to obtain the bigram list. This will enable us to solve all further messages for that day at once, and, on later days while the bigram list lasts, to solve all the messages as soon as a single message has been solved for that day.

We feel that no unnecessary time should be lost in experimenting with and constructing such a machine.