I've lost track of Holmgreen who would be your best best.   That said, when I asked him about it a few years ago, Chesson told me that he had it in his archives.  @ the time, I thought I had a copy in my CMU archives, but when I managed to restore the tape I believed to contain it, the "arpanet" directory was empty.  We must have had it on a separate RK05 image which I failed to make a copy of when I left.  Dan Klein might have kept a copy and don't know what Lisa did with Ted Kowalski's archives when he passed two years ago (I'll try to ask her).

Thinking about it some more, it might be in one of the early USENIX tapes.   But you'd have to do some digging.  Note if so, remember that the format of those old tapes will be the original tp program as well as the original ar, so pulling them apart will take a little work.

FYI:  Assuming I can still read it, I do have a 9-track tape of the MIT Chaos distribution tape circa '82 from the Steve ward's RTS folks, which I was under the impression was based on that the UofI kernel hacks for the NCP; although Noel probably has more complete MIT archives than I.