Summary Insights

Sullivan’s Key: Insights on the VMS Encryption
I have been periodically attacking the Voynich cypher since the 1970s with little success until late. In the last year I have made good progress at understanding the key structure of MS408. Since I am not an academic and I live and work in the real world, I simply do not have the time nor equipment to do what needs to be done.
I have tried to solve the cypher by creating a cypher that would produce the same characteristics of the VMS.
I had a breakthrough when I analyzed the middle column key sequence on page 66. This is a sequence of 34 characters with some repetitions. I noted that the letter 9 in Revised Currier occurs at intervals equivalent to the position of vowels in the Latin alphabet. The end of the sequence though, did not fit, but then I realized that at O, the next letter is z and it runs to the end of the sequence in reverse. This was my first solid evidence that the underlying language is Latin.
The most common characters that occur at the beginning of paragraphs are B, V. and P and these should reflect forms of hic and qui, which they do. This key sequence, excepting for duplications, appears to be solid. There is an underlying system which I have not figured out but I think it involves three columns with multiple sequences of equivalent letter groups.
By using the single letter key, I have been trying to assign letter values to repeating groups of letters. There are about thirty five groups that neatly correspond to the F66 Key Sequence.
At this point it appears numbers are expressed by X words. The suffix seems to be a arabic number and the prefix is probably a multiple of ten. One is 9, and the other numbers are O, OE, 2, OS, AJ, OR, AD, AM, 89. This same structure led be to look at the Q, Y, W words, and combined with my decipherments, I think this is going to get complicated.
The first five lines reads in revised Currier:
F1R1 haec est a (ap5) (ZO1) (zo18) (q314) E (FO8) (ZO10)
F1R2 RUE 1 2 6 SS (ZP6) (z11) (Q1) (Q1)
F1R3 REM (ZC14) D EST (ZO6) (Q112) (Q8) SATRE
F1R4 (?o82om) (OP14) (OP43) DE (OP1) (Q1) ET (OF6) E (OF5)
F1R5 CLENA (Q5) (W1) (Y4)
I am definitely getting Latin. But the Latin is only being used to tie together Code. The VMS is encyphered code text.
I am assigning numbers by there position on the key sequence and it generally makes sense.
Presently, I am working on refining the key so that the code can be separated out. Since the key sequence was part of the book, it is very possible that the code key may be hidden in the manuscript. I am very suspicious of the last section that is a long series of short paragraphs.
If this project interests anyone, I would welcome any suggestions. Try it yourself and see what you get. If anyone could use a computer to automatically replace letter groups with letters it could help greatly.
The code groups need to be pulled out and it actually appears that there is a lot of context to reconstruct the code book. But this doses not appear to be anything that is going to happen quickly.