Augustus of Brunswick-Lüneburg (Wikipedia), also known as Gustavus Selenus, the author of Cryptomenytices et Cryptographiae (1624), used an interesting simple substitution cipher in which the 24 letters of the alphabet forms eight 3-letter groups and each letter is replaced by the other two letters in the group. The scheme is called a triplet cipher herein (not to be confused with a "trigraphic" cipher, a "trifid" cipher, or a "triliteral" cipher).
Augustus of Brunswick-Lüneburg (1579-1666) spent his life in literary pursuit, until he succeeded as Duke of Brunswick-Lüneburg in 1635 after the death of the reigning duke without heirs. He even wrote books under the pseudonym Gustavus Selenus, one of which was Cryptomenytices et Cryptographiae (1624).
Authoring an extensive book on cryptology is one thing but actually using secure ciphers is another. In a letter of 1 May 1624 (the very year of the publication of the book) to Dr. Theodor Bussius, he proposed using a simple monoalphabetic cipher (Strasser (2015) p.49).
A letter (1635) from Friedrich von Mata, the Duke's correspondent in Vienna, appears to use a simple cipher representing each letter by a letter pair (Strasser (2015) Fig.4).
a | b | c | d | e | f | g | h | i | k | l | m | n | o | p | q | r | s | t | u | w | x | y | z |
Hx | in | mq | [sp] | fu | ue | ty | ax | bn | [yt] | [oz] | [qc] | ib | zl | ds | [cm] | gW | pd | ky | ef | rg | aH | tk | ol |
The symbols in brackets [ ] are conjectures, derived from the cipher structure below. |
This substitution table can be derived from a keyword, which I conjecture to be "imperato." (If any one finds a more plausible keyword, please contact the present author. While Strasser says the duke used the double name of his third wife (Sophie Elisabeth) as a keyword about this time, it seems to refer to some other letter.) By writing the letters of the alphabet not used in this keyword in alphabetical order under the keyword, the following table can be obtained.
i | m | p | e | r | a | t | o |
b | c | d | f | g | h | k | l |
n | q | s | u | w | x | y | z |
The first column has "i-b-n" and these three letters are enciphered with each other. That is, "i" is enciphered as "bn", "b" as "in", and "n" as "ib." Similarly for the other triplets.
The same scheme was used in a letter (1641) of Augustus to Hilmers von Oberg, in which the keyword "HILOBERG" was used (Strasser (2015) p.50).
h | i | l | o | b | e | r | g |
a | c | d | f | k | m | n | p |
q | s | t | u | w | x | y | z |
The Duke's note of this cipher key with a keyword "FERDINAT" written in 1663 has been found inserted in his copy of Athanasius Kircher's Polygraphia nova published that year (Strasser (2017) p.14, Strasser (1983) p.211).
f | e | r | d | i | n | a | t |
b | c | g | h | k | l | m | o |
p | q | s | u | w | x | y | z |
It is yet to be seen whether this scheme is described in the Duke's earlier work Cryptomenytices et Cryptographiae or Athanasius Kircher's Polygraphia nova.
Gerhard F. Strasser, "Die Wissenschaft der Alphabete. Universalsprachen vom 16. bis zum frühen Neuzeit -- Methoden, Probleme, Forschungsbedarf" in Geheime Post (2015)
Gerhard F. Strasser, "Wolfenbüttel, a Minor German Duchy but a Major Center of Cryptology in the Early Modern Period" (2017) (researchgate)
Gerhard F. Strasser, "Duke August the Younger of Brunswick-Luneburg (Gustavus Selenus) and His Cryptological Activities" (1983), Cryptologia