Another Morse Code cipher but a little different than the Pollux cipher.  This is another cipher method that fulfills all of what I would consider to be requirements of a K4 ciphering system.  Too bad I couldn’t get it to work out to some plaintext.*

G. Worley gives a pretty good description:

```After encoding the plaintext in Morse Code, the encoded text is broken into
n-graphs, usually trigraphs because there are 26 Morse Code trigraphs (you never
have three dividers in a row, so you don't count that one).  Here's a sample
trigraph mapping (using a keyword):

Keyword:  LOOKINGGLASS

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

So, breaking our message into trigraphs we get:

. x . - x . . - - - . x - .
. . . - . - x . x - . x . -
x - x - . x x - - x - . - x
~~~~~~~~~~~~~~~~~~~~~~~~~~~
Ct:  K T K H R G B D P J O Y D G

Cracking this is much harder because the dividers are obfuscated.  The only way
to attack this is to do a statistical analysis for common Morse Code trigraphs.
You can try brute force, but the keyspace is a little large:

26!

Of course, this is still in shooting range, so you might want to use an n-graph
Fractionated Morse cipher.  This has a much larger keyspace:

(3^n - sigma(i = 3, n, (-1)^i * 3^(n-i)) )!```

Returning to K4:

OBKR
UOXOGHULBSOLIFBBWFLRVQQPRNGKSSO
TWTQSJQSSEKZZWATJKLUDIAWINFBNYP
VTTMZFPKWGDKZXTJCDIGKUHUAUEKCAR

And Cipher Clerk.

M     E M   T M E I T E N U N E   A Q N T     K M M A   I
A M W T T I   T   T 7   E L   S G     A E N A   K T G   T T
L   E E E T E S E V

Then Kryptos-keyed:

A E   X     U 0   K   E   E U T   T   E     T Q   U M E   E
I U W T W E E A T E E I R S   T   T T E =     I T T T E
L R I   T   T   2 Y   D T C X

Hydra-keyed:

T K K T   A   W   A A   T E   M I U   A W T E   T   E   I P
I A R   T   T T W E N B T Y I   A T 3     M E   E M     N
T   G T 6 G   T   E E N E S 6 D 8

Medusa-keyed:

T N J . T T T E 2 O N E   E I E G   M   A N I U     W T E
M E     T   T Q   E N A   T I   E A E   A     N E   E S   N
T   O O   T T   E Z D N K   A

This is not entirely unlikely for a possible K4 translation but I think we’d need the specific keyword.  For now it just remains an interesting cipher method.

*Cipher Exchange lists this one for 110-150 plaintext letters