Barton -- If I may, something is a little screwy here. You've measured the DC resistance for each half of the primary winding in your output transformers, as well as the voltage drop across each half, and determined your plate current. No problem with procedure, but here's what should be investigated:
1. Collectively, you're indicating that in plate current alone, your output tubes are drawing 169.4 mA -- which is awfully high given the rather low plate (and by reasoning the screen) voltages your tubes are running at.
2. If you add in a collective additional 16 ma to account for screen current, that means your output stages are collectively drawing 185.4 mA -- again, very high for a set of four 6BQ5 tubes.
3. That much current equates to 18.4 volts across a 99.5Ω cathode bias resistor, yet in your post #27, you state there is only 10.8 volts at the cathode, for a collective current of 108.5 mA through all four tubes, which is a much more typical Magnavox operating level.
Clearly there is no agreement here between the calculated plate and screen current, versus measured cathode current. I would bet that #3 is much more closer to the truth, but certainly worth investigating.
Dave
.