I don't like making up numbers for cars that didn't have numbers. This has happened in F1 research and has later caused problems. The F1R Register invented P numbers for Parnell-built Lotus 24s but made it completely clear they were inventions and that worked well. The numbers Thomson invented for the LDSs worked a lot less well as it wasn't so clear they were invented and owners started putting them on false chassis plates. Then Lawrence came up with a different set of numbers and live got much worse. People now argue about who has LDS 06 and who has LDS 07, completely forgetting that these numbers were a 1970s invention.
I know LM 77 and LM 78 weren't official but the magazine coverage makes it quite clear than one car was built new for 1977 and another car was new for 1978 so all I'm doing is adding the year on the end so we know what we're talking about. My Indy records use this notation extensively (Eagle 68, Huffaker 64, Gerhardt 65, etc) as it allows a model of car to be identified without having to invent anything. The Eagle T1F and T1G notation used widely in F1 history books was an invention. Eagle never referred to a model but probably regarded the car as the Mk 1. The 1967 Indy Eagle, for example, was numbered as if it was the Mk 3 but never called the Mk 3 so I just call it the Eagle 67.
If we use LM2, LM3 and LM4, we are making the assumption that there was no car in between the 77 and the 78. What if one day we find that there was? Maybe due to an accident to the original 77 car? Would we call it LM3-and-a-half? If these numbers were used by the team, officially or even informally, then they'd be fine to use. But I don't see that they were.
Anyway, sorry for the long lecture. I'm watching the Superbowl and it's not really holding my attention.
Allen