The W.E. 596A and 597A (differing only in style of connectors, I believe) were an important milestone in loudspeaker development; they were literally the world's first tweeters.

In L.G. Bostwick's article "A Loudspeaker Good to Twelve Thousand Cycles" (Bell Laboratories Record, May 1931) he describes his driver as having a response from 3,000 Hz. to 12,000Hz. with an overall efficiency of about twenty percent. He states: "Twelve thousand cycles is within the highest octave that can normally be perceived by the ear, but yet it has been found that certain musical instruments and voices, and many common sounds such as jingling of keys or coins, have overtones or harmonics that make such a loud speaker necessary for perfect reproduction." It is hard to imagine now, but he was writing this at a time when people had never heard such sounds reproduced properly before.

As Marty pointed out, these drivers were way ahead of their time, reproducing mostly noise in the early Vitaphone sound sources. They were installed in the W.E. "Wide Range" theatre sound systems of the early to mid 1930s, which augmented the earlier "Voice In Action" full range snail horn systems by adding woofers and tweeters to a flat baffle mounted under the large horns.

Bostwick was well aware of the problems of such extended frequency response. He concludes his article with these thoughts: "...a loud speaker efficient at high frequencies introduces other difficulties that would not be encountered if the high frequencies were suppressed. For example, amplifier overloading becomes much more strident, and noise may increase to an objectionable extent. A loud speaker of the type described, therefore, cannot be used to full advantage in systems where these factors are not favorable."

While I'm sure that others would have developed tweeters eventually, Mr. Bostwick did it in about 1930! Here is another example of so many good things, speaker wise, tracing their origins back to Bell Telephone Laboratories.

I have listened to a 596A. I sounded very sweet, perhaps not as extended as some more modern tweeters, but not the least bit raspy either. Although its 180 milligram moving mass should allow response higher than 12k, the single slit exit path probably introduces a phase cancellation that knocks out higher frequencies.

The high prices paid for these are mostly fueled by their rarity and value as status symbols, but this would not be the case if they did not sound very good to begin with. I'll bet that Mr. Bostwick would enjoy watching these auctions if he was still with us.