find the keys for other days at the beginning of May and they actually found the Stecker for the 2nd, 3rd, 4th, 5th and 8th, and read about 100 messages. The indicators and window positions of four (selected) messages for the 5th were
| Indicator | Window start |
| KFJX EWTW | P C V |
| SYLG EWUF | B Z V |
| JMHO UVQG | M E M |
| JMFE FEVC | M Y K |
The repetition of the EW combined with the repetition of V suggests that the fifth and sixth letters describe the third letter of the window position, and similarly one is led to believe that the first two letters of the indicator represent the first letter of the window position, and that the third and fourth represent the second. Presumably this effect is somehow produced by means of a table of bigramme equivalents of letters, but it cannot be done simply by replacing the letters of the window position with one of their bigramme equivalents, and then putting in a dummy bigramme, for in this case the window position coressponding to JMFE FEVC would have to be say MYY instead of MYK. Probably some encipherment is involved somewhere. The two most natural alternatives are i) The letters of the window position are replaced by some bigramme equivalents and then the whole enciphered at some 'Grundstellung', or ii) The window position is enciphered at the Grundstellung, and the resulting letters replaced by bigramme equivalents. The second of these alternatives was made far more probable by the following indicators occurring on the 2nd May
| EXDP IVHO | V C P |
| XXEX JXJY | V U E |
| RCXX JLWA | N U M |
With this second alternative we can deduce from the
-136-
Back to Turing's Treatise on the Enigma. Chapter VII.