Having now reviewed my above thinking, etc. I can now see that that thinking was clearly in error. I really shouldn't do things like this when its very late at night and most of my brain cells have gone to sleep! (Especially with voltages of these magnitudes.
) But enthusiasm being what it is . . .
However, with a bit of a re-calc., etc., things are now working as they should.
350 v. in; 1k5 series resistor; 150 kΩ resistor from stabilized rail to anode of 2nd. valve; similarly, 220 kΩ resistor to anode of 3rd. valve. That puts about 1 mA start-up current into each of those two valves to initiate the striking voltage. The voltage across each valve is then as it should be. Current through the valves is approx. 20 mA.
On test, Vstab is 322 v. no-load; 320 v. at 20 mA. That is more than adequate for what I have in mind.
Al.