Okay, I think I've figured out the problem with my code that back propagates cipher wheels to the beginning of the code, and ran it on Exhibit 1 again. It actually matches 603 characters of input, then stalls, after searching 1.8 billion cipher encoding possibilities. I am now trying to figure out if it is a typo in the original, or some slim bug that causes it to go awry.
Because others might be pursuing their own solution of this issue, I'll hold off on publishing the key that I've recovered for a bit longer. If anyone needs it, let me know, and I'll send you the spoiler.
Addendum: There is a typo in the version of the cipher text that I downloaded from the Chaoscipher clearing house. I'll write up more when I've learned more.
Addendum2: More than one transcription error. More later.