book cover
From FLYNN's November 19, 1927

SOLVING CIPHER SECRETS

Edited by M. E. Ohaver
DID YOU KNOW THE ROPES WHEN YOU TACKLED CIPHER No. 88 OF OCTOBER 29?
HERE THEY ARE, WITH SOME OTHER BRAIN WRINKLERS FOR ALL CIPHERITES

IF you knew the ropes you should have had no difficulty in solving Cipher No. 88 (Edward H. Thimme), reprinted here from the October 29 issue. For even in a short message the system there used usually leaves easily recognizable clews leading to its speedy solution.

    5     10     15     20
THCSF  JGRTU  KGNLU  TLPHZ

   25     30     35     40
KMOKO  WJFXR  QHTMH  HUKCO

   45     50
GEBEK  OHVEU  CXHT

Thus the fewness of vowels, and the frequencies of such letters as K, X, and Z, should assure the solver almost at a glance that the cipher is not of the transposition class. If a more accurate test on this point is desired, the vowels may be counted. About forty per cent of the letters in a transposition cipher should be vowels. Here there are but eleven vowels, only about twenty per cent of the fifty-two letters in the cryptogram.

Assuming a substitution cipher, the next step is to find the number of alphabets. One test for this purpose depends on factoring the intervals between recurrent groups. Here the following groups recur at the places and intervals given:

Groups Places Intervals Factors
UK 37 — 10 = 27; I, 3, 9- 27
HT S3—3 2 = 21; I, 3, 7, 21

The largest factor to the most intervals in such a table, as 5 in this case, ordinarily indicates the period, or number of alphabets. Hence, the cryptogram may be grouped by threes, and frequency tables made for the first, second, and third letters of each group.

SymbolsAlphabets
 No. 1No. 2No. 3
Az  x  y 
Bay z 
Cb z aⅠⅠⅠ
Dc a b 
Edbc
Fecd 
Gfde
HgeⅠⅠⅠⅠⅠf
Ih f g 
Ji g hⅠⅠ
Kj hⅠⅠiⅠⅠⅠ
Lk iⅠⅠj 
MlⅠⅠj k 
Nmk l 
OnⅠⅠⅠlm 
Po m n
Qpn o 
Rq op
Srp q 
TsⅠⅠq rⅠⅠⅠ
UtⅠⅠrs
Vu s t
wv tu 
Xwuv 
Yx v w 
Zy wx 

Next, some of the symbols in these alphabets must be identified. Perhaps the best place to begin is in alphabet No. 2, where H, occurring five times, is probably the substitute for e. Taking H=e, and writing the alphabet in regular order opposite the cipher alphabet in the first column, we find that no count occurs on j, k, q, x, and z. This is a favorable indication of a shifted normal alphabet.

Now fifty-four letters of average text only contain about six E's. Hence, it may be assumed that, if H=e in alphabet No. 2, the most used symbols in the other alphabets probably signify other letters than e. In alphabet No. 3, for example, the eight-letter interval between C and K—used three times each—suggests that they stand for a and i, respectively.

Alphabet No. 1 may now be determined in a similar manner, or even easier by context after translating the symbols in the other two. For example, the third and fourth groups, "-or -he," will become "for the," as illustrated below, if a forward shift of one place is used for this alphabet. Deciphered in full the message is: Search jar the missing will in the upper left-hand cabinet drawer.

No doubt many of our readers have already recognized our correspondent's cipher as the famous Gronsfeld, with the key 132. In this system a number selected as key is repeated above the message as many times as required to supply each letter with a key figure. The substitute for any letter is then found by counting forward in the alphabet the number of places indicated by its key figure, thus:

Key:      1 3 2  1 3 2  1 3 2  1 3 2  1 3 2 ...
Message:  s e a  r c h  f o r  t h e  m i s ...
Cipher:   T H C  S F J  G R T  U K G  N L U ...
Grouped:  THCSF      JGRTU       KGNLU      ...

Last week's answers:

Straight substitution Cipher No. 92 conveyed the message: WE RAID TUESDAY AT MIDNIGHT. HAVE THE BOYS READY, AND SHOOT TO KILL. CARTER, CHIEF.

No. 93 used the subjoined alphabet in expressing the message: " The guilt of enforced crimes lies on those who impose them."

   S  E  N  C  A
8  B  D  F  G  H
3  IJ K  L  M  O
6  P  Q  R  T  U
4  V  W  X  Y  Z
   2  1  9  5  7

Mr. Walker described his No. 94 as "the easiest possible sort of substitute-transposition." . In enciphering, the letters of the message (a), "Seeing his days are determined...," were first replaced by those next following them in the alphabet (b). These pairs were then merely reversed (c), and grouped by fives (d) for transmission.

(a) SE EI NG HI SD AY SA ...
(b) TF FJ OH IJ TE BZ TB ...
(c) FT JF HO JI ET ZD BT ...
(d) FTJFH  OJIET   ZDBT—

This week's "Q. and A." cipher uses two straight substitution alphabets, cne for the question, and the other for the answer. You'll be sure to like it. As for No. 96, a full description of the system is given in the message. So it can't be so very difficult. But can you get it?

Without any hints you would no doubt find No. 97 quite baffling. But the method is simple, as only a working knowledge of the typewriter keyboard and the continental telegraph code are necessary. Mr. Cotton writes us that this cipher is in active use to-day in a certain organization, and that it has proved singularly effective in composition and solving.

He also advises us: "If you feel that it will add a new wrinkle to the cerebrums of the other cipherites, submit it to them for their approval." So here it is, fansi You know what that means!

CIPHER No. 95.

Question: OYTA  JM  T  ZYXWVWUXTS?

Answer: LX  YXDICYJWYMX  IMX-
UVTYXQ  L  ELWV.  WKHD  WKV
CMALX  XHAVCLOD  YX  "OMCE KLUV
AVCIYV  UJMX  UD"  VPJCVDD  WKV
TVLC  AEIOUUUY.

CIPHER No. 96 (Perry Philips, Lincoln, Nebraska).

HK  WQA  FNWQY  TGJRKEGE  UKJV
GICUUGO  VPWQE  MECD  QYV  UGECNR
PK  GJV  VGDCJRNC  TQH  JECG  TGV-
VGN  FPC  FCGT  JECG  RWQTI  UFTCY-
MECD.  R-2B.

CIPHER No. 97 (D. W. Cotton, Hood River, Oregon).

Mor e wa rm5  pose on at two  ba ra bian
s were to you  as p ret tyo rm ade $ .983
par or.6¾by 1 and 1%ad mac won one your
u ra tay a out  wer em a we out tyde on
mone d up ma no  it me po 1 out  you hav
emor eon a it jam by  in me bet  a bag
1%ad mac won one  hour one on it by  in
no man a trow I eke  wag one or by gob
they 9