other characters to represent the divisions between words, the various stops, or even blanks. Although Davies Gilbert, I believe, and myself, both arrived at it from our own efforts, I have reason to think that it is of very much older date. I am not sure that it may not be found in the "Steganographia" of Schott, or even of Trithemius.
One great aid in deciphering is, a complete analysis of the language in which the cipher is written. For this purpose I took a good English dictionary, and had it copied out into a series of twenty-four other dictionaries. They comprised all words of
One letter,
Two letters,
Three letters,
&c. &c. :
Twenty-six letters.
Each dictionary was then carefully examined, and all the modifications of each word, as, for instance, the plurals of substantives, the comparatives and superlatives of adjectives, the tenses and participles of verbs, &c., were carefully indicated. A second edition of these twenty-six dictionaries was then made, including these new derivatives.
Each of these dictionaries was then examined, and every word which contained any two or more letters of the same kind was carefully marked. Thus, against the word tell the numbers 3 and 4 were placed to indicate that the third and fourth letters are identical. Similarly, the word better was followed by the numbers 25, 34. Each of these dictionaries was then re-arranged thus:—In the first or original one each word was arranged according to the alphabetical order of its initial letter.
In the next the words were arranged alphabetically accord-
