[Next] [Up] [Previous] [Index]
Main : Index : Rotor Machine Ciphers : Early Machine Ciphers : Reihenschieber


This device really belongs among the paper-and-pencil systems. But it is placed here in consideration of the fact that, briefly, from 1958 to 1960, it was considered adequate to safeguard even top secret material.

The Reihenschieber resembled a slide rule in shape. Ten square sticks, taken from a set of 26, were placed in it, and a grille slid along the device.

The 26 sticks that came with the device were each labelled with a letter of the alphabet. Which ten sticks to use in the device was determined by the daily key, a sequence of ten different letters. Each stick had ten dimples on each of its four sides, which allowed it to be placed between the Reihenschieber's rails in ten positions. These were determined by another daily key, this time of ten digits. The ten digits (with the last one repeated to make an eleventh digit) were used to encipher a five letter area indicator, which was sent with the message, repeated twice using the same method that the digits the Reihenscheiber generates are used to encipher messages: each digit indicates the column in a table containing the enciphering alphabet, and in addition the following digit, depending on whether it is low (0-4) or high (5-9), selects one of two tables.

The sticks were printed with random sequences of digits, interspersed with the odd dot. The dots were ignored when encountered, being present to make the sequence of digits produced by the device irregular. To the left of these digits, there was an expanse with some letters and blank spaces; this expanse contained each of the 26 letters once in an area 10 characters long on the four sides of the stick. The ten letters, generated from the ten digit key called the "message key", were placed in a line, and thus these 10 letters specified both the horizontal placement and the side which was to be uppermost, for each stick.

A frame that held grilles on the front and back (the strips were also printed with digits on the back) slid along the device like the cursor on a slide rule. With each month's supply of key material, three sets of grilles were supplied, each for use for a period of 10 or 11 days. As the strips ran along the device, going from left to right, the numbers the device produced were those read out from the grille vertically, first from top to bottom, and then from the leftmost column to the rightmost column. After reading out the digits shown at one slide position, the slide was flipped over to read the numbers showing on the back as well before the slide was advanced along the ruler for the next group of numbers. The grilles had 10 columns; the article which is my source claims they had 25 holes, but gave an example with 26.

The grille could be slid through 52 positions, labelled first with the lowercase alphabet and then the uppercase alphabet. The first position used was the one with the lowercase letter corresponding to the last letter of the enciphered area indicator as used to slide and rotate the ten sticks.

Therefore, the Reihenschieber looked something like this:

                  ____________                             ______________
           ______|            |___________________________|              
          /      |            |                           |_             
         |       |            |           abcdefghijklmnopq|  ----------
  ---------------|            |---------------------------|  |  -      -|
 |X    u  ae l pr|            |      1701.5433.626.90512.3|  | |7|    |4|
  ---------------|            |---------------------------|  |- -   -  -|
     |E      i qj|            |ooo       849.216.55.1034.2|  |6|   |.|  |
    -------------|            |---------------------------|  |--   -- - |
   |G    m uwi  o|            |o       65.20473.12.9.14516|  ||4| |.||3||
    -------------|            |---------------------------|  | - - - -- |
         |P     a|            |ooooooo       9.250.3961490|  |  |0| |7| |
     ------------|            |---------------------------|  |-  - - - -|
    |V    oi u xm|            |oo       131.6150492.747318|  |9|  |2| |5|
   --------------|            |---------------------------|  |--   -  --|
  |A     py x q r|            |       24619.43.529.6565.34|  ||.| |5||.||
   --------------|            |---------------------------|  | --  - -- |
      |W    hwm z|            |oooo       5.1693.40453.112|  | |.|  |3| |
     ------------|            |---------------------------|  |- -  - -  |
    |C     i eqrl|            |oo       87.21295.306172.91|  |4|  |3|   |
     ------------|            |---------------------------|  |-  - --  -|
          |T    c|            |oooooooo       .40541.331.6|  |  |1||.||8|
      -----------|            |---------------------------|  | - -- - --|
     |J    u mtnw|            |oooo       259.10129.6542.3|  ||8||4| |1||
      -----------|            |---------------------------|_ | -  -   - |
         |       |            |           abcdefghijklmnopq|  ---------- 
          \______|            |___________________________|              
                 |____________|                           |______________

and in the position shown, the encipherer would read off the digits 6944870142537331458 for use in enciphering. The slide continues on like this:

     |                                     \
--   |IJKLMNOPQRSTUVWXYZ                    |
 -|  |----------------------------------------
|4|  |802.1221607.435549.2076.24391462      EF|
 -|  |--------------------------------------------
  |  |.4335.20623.49.17029.33406.415.3307.      EF|
- |  |--------------------------------------------

Also, the four sides of one of its sticks would have looked something like this:

  |A     py x q r    oooooooooo       24619.43.529.6565.3471.52.403569.19382.1
  |A    u mvjo zt    oooooooooo       .95190.317.45602.12995.138.440.27061.630
  |A    ek lbd c s   oooooooooo       437.1512.44691.703977.69.405631.29331.95
  |A     nh gfi aw   oooooooooo       2026.19.5113.4072.6198.20541.35622.9271.

The A at the left end indicated which square stick it was; on the other end, a code of two capital letters indicated which set of sticks a stick belonged to. There are 52 usable positions for the grille; as the grille is 10 columns wide, and there are 10 possible horizontal positions for the slide, the area of numbers and dots is 52 + 9 + 9 or 70 characters long.

Thus, the Reihenschieber did not encipher text directly. It simply was used to generate a stream of digits, which was used as a key.

The digits were used to encipher text using a table of twenty alphabets, organized as two tables of ten alphabets. One digit was used per letter to be enciphered; the digit picked one of the alphabets from a table; which table was decided by whether the next digit (which will later be used to encipher another letter) was high (5-9) or low (0-4). A new set of tables was provided each month.

My source for the information in this section is an article from the April 1996 issue of Cryptologia. The author, Michael van der Meulen, had also done a few previous articles on German paper-and-pencil systems.

Comments on the Reihenschieber

The Reihenschieber as described here is certainly an interesting and unusual type of device. It is possible that the old cipher device known as Ducros' Scotograph was somewhat similar to this, rather than being a variant of the Bazeries cylinder as usually thought. And a device of this type has the potential to be quite secure.

But the Reihenschieber as described here seems to have some flaws that reduce its potential security.

If one is ambitious, one might even advocate using a monthly table with 100 alphabets, and using two digits to encipher every letter, rather than using 20 alphabets (a number less than 26, and therefore also dangerous).

[Next] [Up] [Previous] [Index]

Chapter Start
Table of Contents
Home Page