I read somewhere that some very early Morse code transmitters used on aeroplanes were fed raw AC from a wind driven alternator. If the AC frequency was high enough it would have transmitted a tone, not just steady carrier.
The Alexanderson alternator was the first, high power, CW generator for, what were at the time, high frequencies (~100kHz).
Although I don't know how practical that would've been onboard aircraft; maybe not much, but then maybe just fine with a tow line antenna.
Tim
If its a biplane then a loop antenna would have been fairly easy - run just under the skin of the upper surface of the top wing, then to the bottom wing at the outer strut then back under the skin of the bottom surface if the bottom wing and continuing the same on the other side.
A trailing end-fed long-wire antenna isn't going to be particularly effective without a metal fuselage as a counterpoise. Metal skinned airplanes became increasingly common from the mid 1920s onwards, as did amplifying vacuum tubes, so the odds that anyone used a trailing long wire antennae with an airborne Alexanderson alternator other than purely experimentally are quite low - by the time metal skinned planes that could form an effective counterpoise were on the market, the Alexanderson alternator was very much on the way out. Also an Alexanderson alternator is a heavy brute - there's no way of avoiding it at the speed it spins using 1920's materials and metallurgy, so it would have taken up a serious chunk of the payload capacity of early planes.
There were WW1 era artillery observer aircraft that had an on-board transmitter and a 150' trailing wire antenna, but they used spark transmitters. A 150' wire is a quarterwave at about 1.6MHz - well above the maximum practical frequency for a small Alexanderson alternator of a few hundred KHz.