Another point, however, must not be overlooked: the long repeated sequences HVXXU, ZDYFZJX, DRKVT, GSRRVT. Repeated sequences of these lengths will usually come from repeated whole words, making it possible, to some extent, to attack the cryptogram by word-division methods. It is, in fact, the repetition of sequences, these and many others, which, in the beginning, has led us to assume that both cryptograms are using the same key. As to the recovery of this key, we need not wait until solution is complete. Even in simple substitution, it is well, during the identification of substitutes, to have before us a sort of skeleton key, in which the plaintext alphabet has been written out in normal order, so that the substitutes, as fast as their identities are discovered, can be placed below their originals.
Thus, having identified as many as twelve letters in our present cryptogram, this skeleton key, or framework, might begin to assume the appearance which is indicated in the upper tabulation of Fig. 70. Here, we are able to note a reciprocal encipherment between A and S, F and N, R and T, and U and Y, suggesting that the whole encipherment may have been reciprocal; if so, we have the identities of four additional substitutes: O, I, H, E, representing d, j, k, v, respectively. If they are present in the cryptogram, these four substitutions may be tried; but with or without their presence in the cryptogram, they can be added to the skeleton key, as in the lower tabulation of the figure. Notice that when this has been done, the cipher alphabet is beginning to show alphabetical sequences (reversed). We find H I J K, and, just before this, D F, which is an alphabetical sequence if the letter E has been taken out for use in a key-word. Between DF and HIJK of the cipher alphabet, we need only G to fill out the sequence; therefore either l or m must belong to the key-word; comparing this with what is found at the other end of the sequence, we find that either L or M would be the substitute for g. Between NO, we find V,
| Figure 70 Supposing 12 substitutes to have been identified: Plaintext alphabet: 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 CIPHER ALPHABET: S V N K J F D T A R Y U Assuming reciprocal substitution: Plaintext alphabet: 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 CIPHER ALPHABET: S O V N K J I H F D T A R Y E U Q?P? * L? G? C?B?* * T? M? |
evidently misplaced; and, following O and preceding S, we find two positions which may be occupied by two of the letters PQR, of which R has already been placed (under t). That is, where the encipherer has used a key-word-mixed alphabet without troubling to carry it through a transposition process of any kind, we are often able to build it up again, and make it help us in the solution. This is especially true if he has used reciprocal encipherment; with the substitutes which may actually be found in our foregoing cryptograms, a little rearrangement is all that is needed in order to discover exactly what the original key was. When the cipher alphabet has been carried through a transposition block, it is not so easy to recover during the actual process of solution; afterward, however, it is not usually difficult to treat it by one of the transposition processes, just as if it were a transposition cryptogram of 26 letters. In the examples which follow, the key-word-mixed alphabets were used as they stood, though we believe that none of the encipherment was reciprocal. In one case, however, the plaintext and cipher alphabets were both mixed, according to different key-words, so that the recovery of this key may prove troublesome.
45. By PICCOLA.
S C Y J T O P N R M J T U E A W S R O R O A E P Q R J C R O A R M P H
Q K J Q S R S J H A X P F K E A Q R M Y S R P Q P M P S E C A H G A W
S R O P E E E S H A Q O P V S H I R O A Q P F A E A H I R O P H N P Q
R J H T F U A M C J M R Y R O M A A W A E E B T Q R W M S R A S R J H
A I M J T K U A E J W P H J R O A M P H N Q A A W O P R Y J T Q A A L.
46. By PICCOLA. (Plaintext and Cipher Alphabets have each a key-word).
J C W E H S N D F S B N J I V T E A G V D H O C Q Q I Q F R P H F K Q
E A R F Q A R F A H F Q E J C B N J N H B E O C B N L N O V H B L F Q
J B N A B L F V H C A J I V B N W N S T B L E A G V A J N S R F W N S
Y R V S S C A E H V A Q F C J E A G J N A W N S O V B V C Q Y D C S P
H E H O C S P E A G B E O N A F R L C A G N E A K C S O N S H A C B E
F A C Q X.
47. By PALOMITA. (No key-word).
B O Y B A N K I L L A P K R I Y A P Y Y U P B L Y E R P B P L G Y G M
H L A B O Y K J A K L P Y L H H J A C R P O R C Q U Y N B H L A B O Y
G N A Z N Y L H B O Y K N A N P R B R W O J C B R C Q D N P K.
48. By PICCOLA. (Of these two, one has normal word divisions; the other has not).
W T E I C H E P P C A E P T J W P O Y D Q P R M E L U E I N D E P Q T C
Q D Q D P C P D H K G E P U O P Q D Q U Q D J I C. I S Y E Q T C P V E M Y R
E W M E K E C Q E S P E L U E I N E ? P D Q H U P Q C G P J T C V !
E O E E I Q M C I, P K J P E S X Q T E Q M C I P K J P D Q D J I D P U P U
C G G Y J R Q T E V E M Y P D H K G E P Q F D I S.
49. By PICCOLA.
P B K L A B E I C D J D B I L Y P K L D O I X L Y I P K V Y A L ?
A G F Y A M I L K L Y I K I D C A G G L D O I X V D J R K L Y I C P B R P B N
X D Q A J I ? Q K J I S P B R K L Y A L A B R M X Q F P L F P E O L D
I B R V Y P E Y O B D V X D Q P C E G I A J F I J C I E L G X P K
S I A B P B N P L K. A J P K L D E J A L A B B D L P K L Y P K B D !