David,

I think a lot of what we are seeing has to do with the fact that tartan is not an exact science today, and was even less so in the nineteenth century. Categories such as "dress" and "hunting" etc. arose well after many tartans were already in use, and no doubt many of them were shoved into categories after the fact because of certain characteristics.

For example, the Menzies "dress" tartan you mention was certified by the chief of the clan in 1816 in the Highland Society of London collection as the clan tartan. Period. No "dress." I suspect it was labelled as "dress" erroneously at some point later on simply because it contained a lot of white. The "hunting" version of this tartan, which is red and green (or red and black) dates to 1893. I suspect that it was likely after alternate tartans were introduced for the clan (such as the hunting) that the white version came to be called "dress."

As for the dress tartans that are obviously variations of recognized clan tartans with more white added, I did a cursory search of the ITI for tartans with "dress" in the name, and the earliest of these I found seemed to come from the 1880s (Clans Originaux), which means they were in production at that time. Then there seems to be another influx of dress designs dating from 1930-50 (The MacGregor-Hastie collection), and many more introduced in the 60s and 80s. I suspect as time went on, there was a "Hey, why don't we have a dress tartan?" syndrome going on. :-)

It is interesting, also, to note that a lot of tartan designs may date to an earlier period, but the name "dress" being associated with them would seem more recent. For example, the Wallace dress tartan is documented as an unnamed sett in the Highland Society Collection (1816), but in the MacGregor-Hastie collection (from between 1930 and 1950, as I have said) it is labelled "Wallace Dress."