I have NO idea of how to even guess at how many Farads these *condenser-buildings *were or, how many watt-seconds they could dump into a spark gap?

Total capacitance is easy: If you put your numbers for plate area and separation into this calculator (which accepts US units),

http://www.calctool.org/CALC/eng/electronics/parallel_plateand 1 for the dielectric constant of air (strictly 1.00059 at STP), you get 6.477nF per plate pair. There were 288 plates at Cape Breton, so 287 6" gaps between pairs of plates (its a fencepost problem), so multiply by that and you get 1.86uF.

Energy stored is fractionally harder - you'd have to know the working voltage, then apply

**E = (1/2) C V**^{2}. Of course, that's stored energy and is likely to be be significantly greater than the discharge energy in the spark gap due to circuit losses.

To calculate that you'd need a good model for the spark gap and the rate the capacitor discharges into the gap.

Rate of discharge is incredibly more difficult - you'd have to have details of where the connections were and the external circuit sufficient to give you decent figures for the circuit inductance and resistance (including dynamic resistance of the gap) , and also treat the pairs of plates as transmission lines. Without those circuit details all you can do is set an upper bound for it based on most favorable conditions i.e. zero plate resistance and discharge into matched resistance via external circuit of negligible length connected at the center of one long side of the plates, but even so, the calculation is going to be a stone bitch, best solved by modelling by Finite Element Analysis on a supercomputer.

*Edit: Aagh! I forgot the parallel plate capacitance calculator link. *