Greetings Willy,
Yes, the Seafire III has identical tankage to the Spitfire V.
The RAF Spitfire V pilot's handling notes are available on line;
http://books.google.co.uk/books?id=...ed&dq=Supermarine+Spitfire+V+operating+manual
Almost all of the content applies equally to the Seafire III. Make sure you paste that URL as one line with no spaces. Use the page forward and backward arrows at top left to navigate the POH. You cannot simply download it.
In short the 48 Imperial gallon tank is on top of the 37 IG tank and both are immediately ahead of the cockpit firewall bulkhead. CG moves aft as fuel is consumed with logical trim and pitch inertia consequences.
You are however entirely missing the point. What you need to know is the (calorific) mass of the fuel, not its volume. The real tank volume is only relevant to the necessary calulation of the mass.
Recent versions of MSFS incorrectly assume all AVGAS is modern US formulation low lead AVGAS with a density of only 6.0 pounds per USG. The FD author must therefore research the real (calorific) mass of earlier high lead fuels and always encode the false volume of modern low lead fuel that equates to the same (calorific) mass. The Merlin 55 required British formulation high TEL 100/130 Octane AVGAS whose density was 7.7 pounds per Imperial Gallon. Thus the larger upper tank must be encoded to have;
48 x 7.7lbs of fuel = 361.9lbs of fuel
and must be encoded as a
361.9 / 6 = 60.32 USG tank
so that MSFS has access to the real world (calorific) mass of fuel. The same process must then be applied to the lower tank.
If the FD author is experienced and understands how MS(C)FS works, and the many ways in which MS(C)FS flight dynamics are differently broken in each successive version of MS(C)FS, the volume encoded for vintage era aircraft will rarely match the real tank volume because it must instead provide the real world (calorific) mass for four reasons;
1) so that the current weight of the entire aeroplane can be calculated in real time, as mass x G = weight, as we move the joystick to vary aircraft weight at will to vary our radial or straight line acceleration at constant power.
2) so that CG is realistic versus time
3) so that inertia in each axis of rotation is realistic versus time
4) so that the endurance, range, and combat radius of the given aeroplane, at any given pound per hour (PPH) of fuel flow, is realistic.
Remember current inertia in each axis is highly dependent on fuel mass (remaining) and placement. Remember achievable roll rate (which depends on span) is irrelevant for all practical purposes. What matters in real life is roll acceleration which depends instead on roll inertia, which depends on fuel mass (remaining) and location. Consequently in an air superiority fighter (jaeger) the fuel (and guns) should never be in the wings. In an interceptor (zerstorer) they can be, but it is always the mass that matters, not the volume. High lead, high density fuel imposes a small agility penalty which offsets the small combat radius gain for offensive sweeps, or defensive CAP endurance.
So, maybe you only 'think' the encoded volume is 'wrong' because you do not understand how the differently broken flight dynamics within different recent versions of MS(C)FS work, and how the errors and omissions in different recent versions (including P3D) must be allowed for and corrected by the (experienced) third party FD author at (version specific) design time?
Only very old versions of MS(C)FS allowed the FD author to encode the real fuel volume and then encode the variable density of the ancient fuel in the air file to correct the (calorific) mass. Thus for jet aircraft exactly the same problem exists for AVTAG versus AVTUR. Recent versions of MSFS incorrectly assume modern AVTUR and if the fuel in the jet during its original deployment was really AVTAG, then false volume must again be encoded to deliver true mass, and location of mass.
There is much more to creating realistic flight dynamics than copying numbers from a book. A huge part of the task is always allowing for and correcting all of the design time errors and omissions in different versions of MS(C)FS.
Of course this post assumes that the intention is to simulate operation of the Seafire III during WW2 and not modern era (virtual) 2013 airshow circuit use while filled with cheap low quality modern era AVGAS.
FSAviator.