From FLYNN's August 6, 1927

IN WHICH IS EXPLAINED THE SYSTEM EMPLOYED IN CIPHER No 44, OF

JULY 16—ALSO, SOME DANDY NEW CIPHERS AND LAST WEEK'S ANSWERS

JULY 16—ALSO, SOME DANDY NEW CIPHERS AND LAST WEEK'S ANSWERS

I N cipher No. 44 (J. McA. T.), published in the July 16 issue, the reader was given a cryptogram and its full translation, and asked to discover the cipher system.

A complete description of this system follows. But the special key is still withheld, so that the reader, now having the cryptogram, translation, and full knowledge of the system, will not be denied the fascinating problem of working it out for himself.

This cipher is one of the polygram systems using multifid alphabets set forth by
F. Delastelle in his *Traité
Elémentaire de Cryptographie*, Paris, 1902. In these systems,
groups of message letters are converted into corresponding groups of cipher letters,
or *vice versa*, by transposing and regrouping the units of their symbols.

The present cipher employs a *bifid*, or two-unit, alphabet. Delastelle suggests
several variations to this cipher; but here we shall have to limit ourselves to
that used by our correspondent.

The key to this cipher consists of the order of letters in the cipher alphabet, and the length of the groups into which the message is to be divided. Delastelle makes no provisions for arranging this key, and it is assumed that our correspondent's key is arbitrary. To facilitate the use of the cipher, however, these features are controlled in the present description by a keyword, which must consist of an odd number of letters.

To construct the alphabet upon the keyword PHALANX, for example, first cross out any repeated letters after the first time used. Then, combining I-J—Delastelle omits W from his alphabet—write the remaining letters of the alphabet in lines of the same length as the keyword, and in regular order beneath it, thus:

P H A L — N X B C D E F G I K M O Q R S T U V W Y Z

Taking the columns so formed by verticals, downward and from left to right, number the resulting series of letters in regular order, using the twenty-five combinations of the digits 1 to 5, as follows:

11 P 21 C 31 O 41 Y 51 G 12 B 22 M 32 W 42 F 52 S 13 K 23 V 33 L 43 R 53 X 14 V 24 A 34 E 44 Z 54 IJ 15 H 25 D 35 Q 45 N 55 T

To encipher a message, first divide it into groups of the prescribed length. Delastelle
specifies that these groups consist of an *odd* number of letters. The illustrative
message, AWAIT FURTHER ORDERS, has here been grouped by *sevens (a)* as determined
by the length of the keyword.

Next, write the numerical substitute for each letter *vertically* below it,
as shown in line *(b)*. The figures in each group are now pointed off *horizontally*
by twos. In this way the numbers 24-32-24-54-55-42-14 of the first group, for example,
are converted into the new series, 23.25.54.14.24.45.24.

Encipherment is completed by replacing these last numbers by their corresponding
letters in the cipher alphabet, as in line *(c)*. If desired, the cryptogram
may be regrouped by fives *(d)* for transmission.

(a)A W A I T F U R T H E R O R D E R S(b)2 3.2 5.5 4.1 4 5.1 3.4 3.4 2 3.4 5. 4.2 4.4 5.2 4 3.5 5.4 3.1 3 5 4.3 2 (c) V D I U A N A N K R R T R K V N I W (d) VDIUA NANKR RTRKV NTW

When the final group is shorter than the prescribed length, it is enciphered just as it stands. DERS, for example, here becomes VNIW in cipher. Decipherment is merely the reverse of encipherment. This is one of the most interesting problems ever published in this department. Enough of No. 44 and its translation is given below to permit reconstruction of the key. Can you get it?

Last week's ciphers are now due for attention. No. 48 conveyed the message: "Hero-worship exists, has existed, and will forever exist, universally among Mankind." The following alphabet of coupled pairs— in which O= Z, Z=O, C=Y, Y=C, and so on—based on the keyword OCEAN, was used.

O C E A N B D F G H I J K Z Y X W V U T S R Q P M L

This cipher offered some dandy clews in the terminations of the groups, XEPFDF, XEPFDXT, and XEPFD, where the endings STS, STED, and ST, at once.suggest themselves. After which all is over but the shouting.

No. 49, the " box of candy" cipher, like No. 44, also used a two-unit alphabet, but with letters—representing the flavors—instead of figures, thus:

V A O L P (Vanilla) V A B C D E (Almond) A F G H I K (Orange) O L M N O P (Lemon) L Q R S T U (Peppermint) P V W X Y Z

To decipher, the layers could be taken in regular order, but with rows in reverse order, 3, 2, 1, and the candies in each row from right to left. This would be equivalent to giving the box a half turn, and reading in regular order, thus:

VP-OA-VA-VV-LO-LO-PL ... E M B A S S Y ...

*Message*: Embassy safe; three left turns 10 ninety; two right to fifty;
one left to ten.

No. 49 was thus a simple substitution cipher, except for the minor element of transposition introduced by the uncertainty as to te order in which the cryptogram was to be "tasted." Four of the symbols were divided between rows. And one symbol, that for T in LEFT, was divided between the first and second layers.

The answer to No. 50 will not be published for two more weeks.

This week's ciphers are next on the program. No. 51 is a simple substitution cryptogram, calculated to delight the hearts of "crypt" fans. Look at the pronounceable words in No. 52; are they going to fool you? No. 44 is the Delastelle cipher.

CIPHER No. 51.

IQW HRNUWD JX BZFQWD CECIWNC ZC FDMBIZBMAAE PZIQJRI AZNZI MHV NMHE JX IQWN MDW BMFMUAW JX MH MAMJCI ZHXZHZIW BQMHKW UE NWMHC JX YMDZMUAW LWEC

CIPHER No. 52 (J. A. Dockham, Oakland, California).

TUSTIMK PIA LOOX CYDUT OALZ ZUJWIEPA HA AZTYE RIEV YOFADB AY LO KIRMVIJVU UR NEZBERQ DOM WUL

CIPHER No. 44 (J. McA. T.).

IWOND ERWHE THERT HESIM VZPOD BWEZL WBTLO ACXPC PLEAN DDIRE CTMET HODOF PRHGE GBVXN QMMMD ZYXTF ENCIP HERME NTHER EUSED RCGKS AVIYE OAEXY ICGHD ISREA DILYD ISCOV ERABL IBHUK KLDLO IUICT BALAE EITHI NKNOT BWEQS NZFET

Solutions to all three code messages in the April 23 issue were received from Lieutenant Commander E. H. Barber, United States Naval Station, Cavite, Philippine Islands, but too late for inclusion in the list of solvers published three weeks ago.

The translations and full explanations to ciphers Nos. 51 and 52 will be published next week. The key to No. 44 will be published in three weeks.