Why not horns?

Macrojack,
Those TC-50s are notorious for showing up in any discussion of time coherence. They've become practically an iconic item in the discussion.
Because they used a first order filter on the mid/woofer driver and had that distinctive slanted baffle it is often assumed that they were time coherent.
Stereophile measured the speaker before they obtained the MLSSA system so some of the measurements related to time were not available.
However, it has since been shown that they are not time coherent. It is estimated that they sounded so damn good because the primary driver handling the midrange was using a first order filter and therefor had a very nice impulse response, which the old stereophile measurements DO show.

Cheers!

Unsound, ALL time coherent designers think/thought this. Vandersteen, Thiel, Johnson, etc.
It is the single measurement that most distinctly shows the output signal as it relates to time. And timing was the paramount issue for all of these guys.

"Another thing that makes the Ohm speakers unique (then and now) is that they are almost a 1-way speaker. The CLS driver handles frequencies from the bass through about 10kHz, at which time they (finally) hand off to a (metal) dome tweeter. In doing so, they remain completely free of all known deleterious effects of crossovers usually located in the all-important midrange. Coherence is the natural by-product of one driver doing most of the talking - or singing as it were. So are phase coherence and time alignment."

This quote from a six moons review of one of the OHM Walsh designs pretty much states the case for the OHM CLS driver achieving a high degree of coherency in that a single driver handles all but the uppermost frequencies. I think this is an established fact regarding the OHM CLSs in that it has been reported in many reviews over the years and never brought into question or challenged at least in any reputable publication I know of.

Also, Macrojack, John Dunlavy was known for claiming that his later Dunlavy Audio Labs speakers were superior to his earlier Duntech designs precisely because the "new" (at the time) MLSSA system gave him the opportunity to measure far more precisely and implement his theories to a greater degree.
This was late '93, early '94.

Anyway, back to horns!!! (didnt want to hijack this thread, haha.)

Mapman, the original Ohms Walsh driver was not a one way speaker, there were different segments. Though it might not have used an electrical cross-over, there were indeed mechanical cross-overs. Still a brilliant idea, that I think might has been improved upon by the German Physiks DDD, which appears to be a single driver. I think we're getting off topic now.