Doug,
I think that the FeO/FeO2 is one posible cause of the colour change.
More likely still it is due to the presence of anthocyanin (blue) and flavonol (yellow), two plant dyes that act much the same as litmus paper.
The effect of these dyes can be seen in the living plant - a hydrangea growing in an acid soil will produce pink flowers, but if the soil is made more alkaline the flowers will be blue.
The dye responsible in this example is anthocyanin.
Another example of natural indicator dyes can be seen in cherry juice, which may be red (pH = 2.5), orange (pH = 4.5), brown (pH = 7), or green (pH = 10). Dyes found in red cabbage react in a similar manner. Yet another is the dye found in the dreaded turmeric, which is commonly seen as the bright yellow of curry stain, but can be turned pink by the addition of a base solution.
With all these the relevant factor is whether the prevailing conditions are acid or base.
Clearly wines are complex compounds, so the simple dye colour change/ pH relationship does not follow the same rules as with the raw dye, but my experience tends to support the theory.
These colour changes can be seen without the involvement of micro-splitters, but it may be that other actions associated with the splitting of soils make these changes more apparent.
I hope that this also goes some way to answering both of Pauls' questions