Thanks for the circuit Fred! I like the way you have coloured coded it - makes it easier to understand the various paths
You mentioned in your circuit that the primary impedance can be 5 to 8K. I don't know what the primary is on the OP tranny I have & going by the valves originally used it could be 10K. Is there anyway I can measure the impedance? Does it even matter?

I had toyed with the idea of using 807's as I have a new pair & they do look very impressive. But since this is not my amp it is probably prudent that I don't use them for the obvious reasons!

And thank you Josh for the valve heaven link.

Yes, don't sweat about the output transformer too much, just so long as it is intact to begin with so the voltages feed to the valves and you can connect a load to it.
About the impedance. To keep it simple, the load speaker you are going to use has an 'impedance' expressed in ohms like 2,4,8,16 ohms for a normal radio/amplifier. PA types talk more in terms of volts as each PA speaker has a high impedance via a coupling transformer so the load may be a '110 volt line'.
The output valves exhibit an internal impedance usually in the thousands and if you stick a 4 ohm speaker straight across them you virtually short them out and get nothing. Hence the use of a transformer. By virtue of the ratio of the turns, low turns on the secondary / high turns on the primary the low ohms of your 4 ohm speaker get "multiplied" to a high ohms to suit the valves. Thus the transformer is an "impedance changer".
Your OPT may have 110v turns on the secondary and may have a low turn tapping to suit a radio type speaker. You can sometimes pick a line or low ohm winding by the ohms reading. A normal radio OPT will have a primary with hundreds of ohms (lots of turns, small gauge) and a secondary with a couple of ohms (low windings turns thick wire). A Line OPT will have hundreds of ohms on each side but the secondary may also have a tap with much lower ohms to feed a radio speaker.
On first power up, if the OPT looks like a line type then you can use a high wattage couple of hundred ohm resistor as a load sink (heater element/light globes/bunch of 10 watt resistors) or even use a cheap line to 8 ohm transformer from jaycar/altronics and a radio speaker or a low ohm dummy load (bunch of 10 watt resisitors).
Of course to be correct the load should match the valves for maximum power transfer theory but that comes at the end of the job, first get the thing going!
Also don't worry too much about the actual value of components to make it work. This amp was built in the days of plus/minus 10% resistors, capacitors. Ditto with the electros in the power supply, all that was available back then were 8 or 16 μF axial and that will be fine in all the positions. Because of you 450 volt PT I would be more concerned with the voltage rating, Hence recommending using the indirect heat 5V4 to keep the peak voltage down. If you used a direct heat tube like the 5Y3 you get 600 volt plus within seconds and it takes another 30 seconds for the output valves to come on line and load the HT rail down. The 5V4 will keep pace with the 6V6 or 6CA7 and the surge will be minimised. Still I would prefer to have 600v electros next to the rectifier.
Leave the 807's alone that top cap is a killer and needs a proper all covering ceramic cap as you have DC 500 and AC 1000 volt up there for unwary fingers! Lovely valve but a killer in domestic equipment.
Fred.

Looking at that diagram raises some immediate issues..

Firstly the power transformer is completely shorted out, as you have the centre-tap connected directly to B+.
Then, there's that 10μF cap and 1H filter coil which are doing nothing useful. At least get the power supply sorted out, as that is elementary stuff. The 200K 2W unit would only serve the purpose of a bleeder resistor... can't see the point of it myself.

The next bone of contention is the anode resistor of the 79 pin 2 is not going to B+, it is going to ground via the cathode resistor of the 42. Can't see that working too well.

Then there's 2 paths of HT getting to the tone control circuit, one via the 42's anode, and one via the 79 pin 5 anode. It looks odd, but at least it won't blow up.

The rest looks ok at first glance.

So, to summarise -
1. Rewire the power supply so that the filter components actually do something, and the power transformer won't catch fire.
2. The 250K resistor that is going to the lower 42's cathode, join it to B+ instead.

After that, with the correct valves (assuming your guesses are correct) it should do something.

Yes I noticed the power tranny shorted through a resistor, I'm guessing the centre tap negative should cross the HT + rather than join it. I love the circuit though, when you get it working please let me know because I've been searching for a really primitive amp using 42's to use as a guitar amp, sort of an electronically super retro 1930's thing to try and get a really crazy sound! That would be perfect, I've got a few old 42's laying around and NOS ones do come up every now and then on flea bay, I bought 2 brand new matched ones, Philco brand last year and popped them in two of my consoles and still have the old ones as spares and a couple of unknown condition ones, beyond that they work and one rattly one I wouldn't chance using! I just love 42's good loud valves!

Back-bias via the choke was quite a common technique back then - be mindful of how that is meant to work to set up your output stage bias. The lamp across the choke would have been a service operating indicator.

Any photos?

I n that original circuit. I do wonder at the 200K from B+ To ground: That's the sort of thing I would expect for a screen divider. The 250K to the cathode of the lower 42 is going to see the phase splitter plate with no plate voltage.

The 250k should go to a B+, as recommended earlier, to get that paraphase PI circuit ok.

I think I misread the choke before - it doesn't appear to be a back-bias, but rather just placing the choke in the CT to 0V line - that would reduce voltage stress on the choke insulation. That would make it a choke input filter, as was common, and the 200k would be a small pre-load. They may have put a bulb across the choke. The 42 cathodes appear to have a common bias resistance.