>>Now comes the problem.  The Mac will not boot off of the internal HD with
>>this partition set-up.  If I make one big Mac partition, it will then boot
>>properly, so I think it is the partitioning not the disk.  I don't have my
>>IIcx handy, so I can't check the partitioning on it.
>
>It must be Silverlining, not the partitioning.

I think this problem exists in a couple of different formatters. What 
seems to be happening is the Partition Map table is mapped to cover N-1 
disk partition entries so the last one on the disk gets lost.  I first 
saw this with Apple's "HD SC Setup" from my System 7.5.3 distribution.  
Other disk formatters I've tried recognized the disk as being damaged and 
offered to fix it, and the fix was updating the Partition Map entry that 
defines the Partition Map itself. Putting a minimal sized dummy partition 
at the end of the disk is another good work-around.

>3. When you partition the Free Space, create mac partitions at the end
>   of the disk (higher numbered blocks).  Create Un*x partitions at the
>   start of the disk (lower numbered blocks).  Why?  Because if you have
>   several Mac HFS partitions and several Un*x partitions, and you put
>   the Mac ones first, you won't be able to get to all the Un*x partitions
>   because only the first 8 are available in Mkfs and Installer.
>
I believe with the lastest version of Mkfs it will scan as many as 20 
Disk Partitions looking for valid entries.  The Kernel does the same 
thing, and the Installer may as well.  Historically the A/UX partitions 
were created before the MacOS ones.  That may have been because a user 
who wanted to load A/UX on a Mac would probably be running more A/UX than 
MacOS.  Or it may have been that early on the A/UX Utilities, like the 
early NetBSD utilities, could only access the first 1Gig of disk space on 
the physical volume.

-bob