Obviously, the rotors had to be marked in some way on the outside so that the different positions could be identified.
However, here entered yet another element of complexity.
Each rotor was encircled by a ring bearing the 26 letters, so that with the ring fixed in position, each letter would label a rotor position.
(In fact, the letter would show through a window at the top of the machine.)
However, the position of the ring, relative to the wirings, would be changed each day.
The wirings might be thought of as labelled by numbers from 1 to 26, and the position of the ring by the letters A to Z appearing in the window.
So a ring-setting would determine where the ring was to sit on the rotor, with perhaps the letter G on position 1, H on position 2, and so forth.
It would be part of the task of the cipher clerk to make the ring-settings, and thereafter he would use the letters on the ring to define the rotor-settings.
From the cryptanalyst's point of view, this meant that even if it were openly announced that rotor-setting 'K' was being used, this would not give away what at Bletchley they would call the core-position—the actual physical position of the wiring.
This could only be deduced if the ring-setting were also known.
However, the analyst might know the relative core-positions; thus settings K and M would necessarily correspond to core-positions two places apart.
So it was known that if K were at position 9, then M would be at position 11.