My copy of that magazine is in storage and I don't currently have hold of it. It can probably be found in a searchable data base associated with the Library of Congress, or maybe somewhere else on the Internet. But the tablature was simple enough. It was the treble cleff, with a black circle in the first three spaces, up from the bottom, one above the other. Each circle had a slanted line going up to the right, and the ends of these lines were connected by a vertical line. Normally, this would indicate three eighth notes played simultaneously. Still, the proof is in the puffin' (as they used to say on the Grand ole Opry). It is possible to play three strings at the same time on a violin. You can even play all four strings at once by taking the hair off the bow, turning the bow upside down under the strings, then replacing the hair, over the strings. You can also play one note, then another, while continuing to hold the first one. I don't know how this would be indicated on tablatur, but If you hear the riff in person, the notes sound sequential, with the overtones coming together to form a spherical surface which is a few times larger than the violin. If you're within a few feet, at least some of the sound seems to come from the surface in the air, rather than the instrument itself.
The magazine didn't say how Keely played the notes, only that he "played three notes on a violin" and this launched the cannon ball. Keely's tablature drawing showed what the three notes were - F, A, and C. What wasn't shown was the order the notes were played in. The reason I know is because I once heard an old time fiddler play the riff, and observed the effect. (1951). Back in those days, tricks like that were passed around and handed down. Somewhere, some fiddler must have heard what Keely did, and shared it.
For what it's worth, it only takes two notes to cause a longitudinal pulse to condense. A longer, faster baseline overtaking a shorter, slower one. Or you can use four or five frequencies. Pulse length is always equal to or greater than four pi times pulse width. Keely may have found that he needed three notes for the pulse to be long enough to effectively launch the mass of the bowling ball. And that's why you might not be able to use just the 42.8 kHz, even if it were the active harmonic component of the overall pulse. When Puharich did his hole punching experiment, through the ceiling and the roof, his test tube setup was producing the key frequency. But he also had a couple other frequencies mixed in, as a standing wave, trying to produce sedimentation.
