It needs to be an indirect toner so that it only affects the remaining silver halides after the first development. The bleach following, therefore, should only remove the silver reduced by the first development, leaving us with a (toned) positive image.
As for the hydrolysis theory, I'm curious than why potassium ferricyanide seems to remain stable in an alkaline environment. Would you say that there is a rate of oxidization?
Ferric Ammonium EDTA loses oxidation potential as pH becomes more alkaline than about 6.7. It stops bleaching. As you continue to raise the pH, the Ferric ion begins to precipitate out as Ferric Hydroxide and you lose all oxidation power completely. Basically, you end up with a slurry of glop that stinks like ammonia.
The basic 5 steps of reversal processing are
1. First developer
2. Bleach (non-halogenating)
3. Re-exposure *
3. Second developer (may be a fogging developer)
* Re-exposure is not required if a fogging developer is used.
There are also washes and a clearing bath but these are the basic steps.
Note that a rehalogenizing bleach is not used as it defeats the purpose. You do not want the image procuced by the first developer when bleached to be sensitive to light.
For specific information go to the following site. Be sure to click on the last line "REVERSAL PROCESSING" for full information. http://www.ilfordphoto.com/applications/page.asp?n=90
Just curious which indirect toning technique you plan on using. Unless you use very special dyes you shouldn't have an issue with acid, and if you thoroughly tone with Sulfide neither acid nor ferricyanide should be a problem. I suspect you want to red or blue tone, and some of the resulting compounds will be attacked by ferricyanide. If that is indeed the case, you should be able to work with Ammonium Ferric EDTA based BLIX regardless of pH.
If, as Rudeofus mentions, you wish to use a dye toner it is better to do this after the reversal process.
The indirect toner I'm using could be any true toner (i.e. one that converts the silver grains formed in the first developer to a different metallic grain such as selenium, iron or silver-sulfide).
A dye toner (or tinter) does not actually convert the silver to a new metal; instead, the bleach (or mordant in this case) simply allows the dyes to bound to the grains which would otherwise have no affinity to them. Therefore, using a tinter would not be as viable since the silver grains that represent our positive image would still be susceptible to the blix. The resulting image, therefore, would be solely formed of the dyes which simply do not have enough density on there to be pictorially pleasant.
Your right though that the pH isn't an issue. That was never really an issue. In fact, I've already processed some 5,000 feet with the process using an acid-based BLIX. Here's a sample to give you an idea: http://vimeo.com/51802644.
However, despite how the process works, I wanted to see if I could go further with it and maintain the entire process at a specific alkalinity, which according to PE, isn't possible - or at least with the bleach agent at hand...