128. By ELIA, JR. (Variant, Autokeyed).
O O U J V J M K N C B U Q L P F U L A S A Z F T G M P B V A Y V S Q J
L F A W S P C H A E I U N R S M F V W S S O O H M E B E A M K F A A X
R H K Z R J Q A O I A V M E I B T O P D J G P R J N F R X T I I G X F
K D H X A F T H J Q H L A R K T G D L P S B M V Y E E V A O A C S M U
V U W C V C T S K S M W L O N P A O O H M W W P Y P O H I L G A Z Q B
Q U Z B Q P K M B O V K W J H P J A G D C H X G W Q B K O G Y A K S I
W N W E X Q N U S U C V O E Y H Y J J C B T B V J Q M N S P A R V P X
O A G T A V L V C Z B D I X N F M W U E Z L N N N W B M O X G T C P K.
129. By ELIA, JR. (Beaufort, Autokeyed).
O A N C Q R O Q N Y Z G K P L V G P A I A B V H U O L Z F A C F M A V
Z J H T I L L X V B C Z M M T O W Z W A O Q V P M M Q D L Q K O H K F
G O B T L R A U X Y T Y S F N O C Y G R P M U U H T H E W P O O S R G
R Z Z S L Y G K I A N K M M T O W Z W A O Q V A B E U X W T C T J I O
G L P H T E F U F B R X M U Z V L D B P K N S Y A Q B I V M O H P V L
G Z Y F C C W C O M C A W N A A A E V A W M P E B Q X O D O V P X A T
E M A J A T P M J E A Z Z M D S B B N A A A F G L I D N X A M K H K P
D B B P Q Y.
130. By ELIA, JR. (St. Cyr, Autokeyed).
T B F N Q X E F D G F W E A F X S Q U N I G A H E U N B B J L O B Q P
H F A K A S N X G B P E E J W W L Z J O M L L A P R V Y T N M X H Y V
O S E S Q V O A Q M O G V P A J K P Y I U Z F Q G Y J Y T L D F E L Q
Z L W Y Y U Y Z N E P P F W B R W M E E F R W X J W E P R V Y B U M P
Z Z M T S B U K K B A L K Z I L Q A L Z K K F S X Z U S T G J T H A R
G S B X I W V L Z B Z M P I K Y I U R H R V W C V A U F V L W F Q Z U
D I G F W H T Z M S F B K T Z U T R K I V F Z X W L C A U J P A N V S
E O Z U X G I X D S X M G Q E L Q T V B L E I D I A L L A I N O E N L
V J I O I S W Q T D E C T M.

CHAPTER XVII
Some Periodic Number-Ciphers

The use of numbers in a periodic cipher does not, in itself, create a problem essentially different from that of the letter-ciphers. Numbers may, in themselves, cause weakness; we saw such a case in the last of our autokey examples, where a complete disarrangement in their order did nothing to conceal their size. But oftener than not, the weakness lies in the construction of the cipher or in the manner of its application, and while this is fully as true of letter ciphers, the numbers, for some reason, appear to be more inviting for certain kinds of misuse.

In order to observe a weakness which need not have existed, let us consider the slide partially shown in Fig. 131. The use of two-digit numbers will furnish a hundred substitutes; but a strip of that length is awkward to handle, and the constructor of the present slide has confined its length to forty numbers. Then, since

Figure 131
A Slide Carrying a NUMBER-Alphabet(and Keys) ( Plaintext Alphabet - Stationary (
( A B C D E F G H I J K L M ..... ( 10 20 30 40 50 60 70 80 90 00 11 21 31 41 51 61 71 .....
Sliding Cipher-Alphabet: Key-letters may be added: A B C D E F G H I J K L M N O)
( * P Q R • S T U * V W X Y Z *)
The addition of key-letters makes it possible to employ a keyword. For the present
forty-number slide, it was necessary to double them up as in Porta. A slide having
fifty-one numbers would have accommodated all twenty-six of the key-letters.

he has only fifteen different cipher alphabets, and wishes to make use of word-keys, he has adjusted his 26 key-letters to fit the number of alphabets. Now if the alphabet of the slide (that is, the cipher alphabet, or series of numbers) is written in straight 1-2-3 order, and if it is considered that letters may have two or more values, so that A has the values 1, 27, 53, etc., B the values 2, 28, etc., C the values 3, 29, and so on, a slide of this kind is exactly the equivalent of the Saint-Cyr slide, since any cryptogram accomplished with it could be promptly converted to a Vigenère cryptogram by substituting letters for numbers. The keys, of course, might differ. The constructor of the slide has desired something more difficult. But instead of carrying his forty numbers through a transposition block, and really mixing them, he has been content to group them, in regular order, by their tens. We shall see in a moment what happens to his cryptograms. But he neglects also an opportunity: Presuming that his circumstances are such as to make the use of numbers practical at all, why waste the opportunity to use the full one hundred substitutes? The remaining sixty numbers might have been placed on the next two rows, and thus, in every position of the slide, he could have had two or three optional substitutes for every letter — a much more difficult case than the simple periodic.

The cryptogram of Fig. 132 was enciphered with the slide of the preceding figure, using the key-word CABLO (equivalent to the numerical key 30-10-20-21-51), and its period, 5, can be determined in the usual way. However, we have already seen the Kasiski method; suppose, here, we look at another, originated by Ohaver; and, since Ohaver himself, explaining his method in connection with a number-cipher of much the kind we have here, illustrated with single numbers instead of with sequences, it seems fitting that it be illustrated again in the same way.