Zoran, thanks for this. It's much what I was thinking, except that we'd have several sprockets fixed to a tube around one of the axle shafts. The problem in a higher-performance application would be torque steer, i.e. the long axle shaft winds up under power and unwinds when power is released, resulting in more power first to one wheel and then to the other. It is why the short driveshaft on a transverse front-drive arrangement usually has a narrow waist somewhere along its length, to make it as springy in torsion as the long driveshaft. So one would have to incorporate some sort of torsional spring in the hub etc. to achieve that. But I'm investigating other differential options. Edit: as long as the diff is open, wind-up shouldn't be a problem, as it would be equalized in the differential's action. So, Stu's arrangement would work perfectly, with the necessary adjustments. Things might get a bit hairy with a limited-slip differential, though. The same applies to the principle of a Detroit Locker, for instance, which isn't really a differential at all but an automatically-unlocking spool. That could in principle be widened into a laterally- sliding splined axle with clutches at the wheels, but the haphazard locking and unlocking wouldn't suit this sort of car. The consensus in road-racing seems to be moving in the direction of, the grippier the tyres you have, the more sense a plain open diff makes. I think the old trials practice of separate left and right handbrakes acting on the respective rear wheels would rather suit the character of a GN/FN, despite being wholly redundant on the originals.