Just found some good and bad info in the shop. The good is, I clamped a precision straight edge across the face of my chuck, and mounted an indicator on the new MLA-11 cross slide, and swept it the entire travel of the feed screw. Got less than .001" deviation, and even better the deviation ran progressively towards the chuck, as it should. Very happy with that. Also virtually zero play on the cross slide when I tried to move it left/right and fore/aft( I locked the carriage for this test).
Now the bad. Disassembled the compound for a quick inspection on the surface plate. Holy cow, no wonder this thing has given me fits! The lower casting 704-016 isn't bad. The top dovetail is parallel to the swivel base base to within .001"-.002", and the dovetails are parallel to each other to a worst case of .003". The top casting 704-017 is another matter. Must have been made on Friday afternoon or Monday morning. The parallelism of the dovetails to the top surface( tool post mount ) is out by .008". The parallelism of the dovetails to each otheris off by .030"! That's not a misprint. They both converge towards one another at one end. No wonder I could never tweak the gib adjustments to a satisfactory level. I'll start a separate thread when I begin it's rehab.
Update:
Well, there will be no compound rehab thread, as the top casting is pretty much shot. Both dovetails angled in toward each other pretty severely. Looked like something shifted when they were machining it. Anyway, by the time I got them parallel, the gib would have to be damn near triple in thickness to snug it up. Thinking my best option is to simply make a new one out of Durabar, and mate it to the factory lower casting, which is dimensionally pretty good. Once I have my rigid tool post mount made, I can make this new compound of a thickness that will stack up to the exact height of the rigid mount, and will allow the Aloris toolholders to remain set.