Moving back to the gain control section, V9 looks as if it is part of a feedback volume control using VR4. Except that the feedback loop is broken by switch S1/3 when it is in the “gram” position. Then VR1 appears to act as a current-dividing control. It certainly looks odd.
The selectivity switch wafer connected to the anode of V9 switches in additional load resistors (in AC parallel with the main V9 anode load). It would appear to adjust the AF gain according to selectivity position, quite reasonable given that IF gain is likely to be higher in the narrower bandwidth positions.
My best guess as to the purpose of V10 is that it is the centre of the volume expansion circuit. Its output is in parallel with that of V9, in that its HT pathway from its anode is ultimately via V9’s anode load resistor. Its input is fed back from its output via an R-C network that involves the V10 diodes (setting the grid bias according to signal level) and probably sets the time constants and frequency response of the expansion process. Its immediate anode load, R46, appears to be bypassed in some positions of the selectivity switch, suggesting that the amount of expansion varies with the bandwidth setting.
Cheers,