Being now convinced that the oxide layer on the anode is the dielectric and the water electrolyte simply an extension of the cathode, I thought it might be a good idea to try to calculate the capacitance of Bob's 100 cell electrolyser, or at least get a 'ball park' figure.
Working on Bob's plates being 150mm square, I have calculated the following capacitance based on a 10 micron layer of chromuim oxide.
Chromium oxide has a dielectric constant of around 13.
Capacitance = Area of plates/distance between plates x static permittivity x dielectric constant.
I calculate this to give each plate a capacitance of around 258nF (assuming the oxide layer to be 10 microns thick)
He has 100 plates in series which reduces the overall capacitance by 100x, hence giving an overall capacitance of 2.58nF
It should however be noted that I have simply made a judgement on the thickness of the oxide layer. If it were 100 microns thick then the capacitance of each place would decrease by a factor of 10.