Right. The notes from the Scottish Register come right out of the STA notes on this tartan, and they do reflect the earliest known date of this particular MacDonald tartan: 1845.

Now, if we could find any evidence of this tartan being used at all in the region in question in the eighteenth century, then your theory might be plausible. But so far as we know, the Seaforth tartan pre-dates the MacDonald of the Isles Hunting tartan by a good 70 years. So to suggest that the MacDonald tartan (whatever it may have been called) not only predates the Seaforth/MacKenzie tartan, but was in fact the basis of it, goes contrary to the known evidence that is available to us.

I don't mean to sound like I'm being dismissive of your theory. I'm just saying that it's going to need some evidence to back it up.