The Complete Works of Edgar Allan Poe - Edgar Allan Poe [1391]
Were two individuals, totally unpractised in cryptography, desirous of holding by letter a correspondence which should be unintelligible to all but themselves, it is most probable that they would at once think of a peculiar alphabet, to which each should have a key. At first it would, perhaps, be arranged that a should stand for z, b for y, c for x, d for w, &c. &c.; that is to say, the order of the letters would be reversed. Upon second thoughts, this arrangement appearing too obvious, a more complex mode would be adopted. The first thirteen letters might be written beneath the last thirteen, thus:
n o p q r s t u v w x y z
a b c d e f g h i j k I m; and, so placed, a might stand for n and n for a, o for b and b for a, &c. &c. This, again, having an air of regularity which might be fathomed, the key alphabet might be constructed absolutely at random.
Thus, a might stand for p
b " " " x
c " " " u
d " " " o, &c.
The correspondents, unless convinced of their error by the solution of their cipher, would no doubt be willing to rest in this latter arrangement, as affording full security. But if not, they would be likely to hit upon the plan of arbitrary marks used in place of the usual characters. For example, ( might be employed for a
. " " " b
: " " " c
; " " " d
) " " " e &c.
A letter composed of such characters would have an intricate appearance unquestionably. If, still, however, it did not give full satisfaction, the idea of a perpetually shifting alphabet might be conceived, and thus effected. Let two circular pieces of pasteboard be prepared, one about half an inch in diameter less than the other. Let the centre of the smaller be placed upon the centre of the larger, and secured for a moment from slipping; while radii are drawn from the common centre to the circumference of the smaller circle, and thus extended to the circumference of the greater. Let there be twenty-six of these radii, forming on each pasteboard twenty-six spaces. In each of these spaces on the under circle write one of the letters of the alphabet, so that the whole alphabet be written — if at random so much the better. Do the same with the upper circle. Now run a pin through the common centre, and let the upper circle revolve, while the under one is held fast. Now stop the revolution of the upper circle, and, while both lie still, write the epistle required; using for a that letter in the smaller circle which tallies with a in the larger, for b that letter in the smaller circle which tallies with b in the larger &c. &c. In order that an epistle thus written may be read by the person for whom it is intended, it is only necessary that he should have in his possession circles constructed as those just described, and that he should know any two of the characters (one in the under and one in the upper circle) which were In juxta-position when his correspondent wrote the cipher. Upon this latter point he is informed by looking at the two initial letters of the document, which serve as a key. Thus, if he sees a m at the beginning, he concludes that, by turning his circles so as to put these characters in conjunction, he will arrive at the alphabet employed.
At a cursory glance, these various modes of constructing a cipher seem to have about them an air of inscrutable secrecy. It appears almost an impossibility to unriddle what has been put together by so complex a method. And to some persons the difficulty might be great; but to others — to those skilled in deciphering — such enigmas are very simple indeed. The reader should bear in mind that the basis of the whole art