My theory is that the different tube somehow was able to do some damage, but it is not obvious how upon looking up the two different tubes.
Atmasphere,
I was wondering the same thing as neither heater of the tubes would have been energized. But what if the designer of the amp connected one lead of the filament secondary transformer winding to chassis to save wire or trace and at each of the tube sockets connected one heater socket pin terminal to chassis? If that were the case I can see a completed circuit where plate voltage would have been directed through one section of a triode through the heater to power supply B- chassis. Not sure what that load/s could do to the power supply. Just a thought.
On another note, as a kid I remember if a coupling/blocking capacitor was bad it would not block the high plate DC voltage from a preceding tube stage from entering the grid of the next tube along with the signal. Just going from memory the music was heard through the speaker only in very short segments of on and off repeated sequences repeatedly. Back then that was also called motor boating.
Jim