Ah - that makes sense.  I never programmed the Honeywell in anything but Dartmouth BASIC (mostly) and any early FORTRAN (very little) and the whole idea of storage size was somewhat oblivious to me at the point as I was a youngster when I did that.  Any idea did the Honeywell treat chars as 36-bit entities also?  Steve, maybe you remember?

Also, please remember that the standard does not yet exist for a good 10 years ;-)  At this point, the 'standard' was the Ritchie Compiler for the PDP-11.

At the time, we to run wanted the program on all of the UNIX/v6 systems and CMU's version of TOPS-10 and later TOPS-20 as an interchange format.  Thus, I have memories of having to use the  "c =& 0177" idiom in the backup/dumper program in a number of places [remember tar does not yet exist, and tp/stp was a binary program].  Beyond that, I don't remember much about the running C on the 10s.  I think we started trying to move Harvard's stp to TOPS-10, but ran into an issue [maybe the directory size] and stopped.  Since backup (dumper) was heavily used, we were trying to get IUS and SUS to be able to be backed up and handled the same way the operators did the backup for the 10's.  In my own case, I had learned SAIL (and BLISS) on the 10s before C on the PDP-11, plus this was an early C program for me, maybe my second or third non-trivial one after I worked with Ted on fsck, so coming from the PDP-10/SAIL/BLISS et al world, 7-bit chars certainly seemed normal.   I also remember having an early 'ah-ha' moment, when the difference between a 7-bit and 8-bit char started to become important.

Clem