This suggestion is not meant to insult you but:

Were the prints 100% dry before mounted? If you have a split selenium tone, it is far more apparrent in the wet print and can largely disappear when dry. Therefore if yor images were seemingly dry, but not 100%, there may have been further drydown to swallow up the purple hues. A print that is dry to the touch and no longer floppy (but perhaps not as bone dry as it could be) will dry mount perfectly well as it dries out in seconds. Pure speculation, but a thought, as I to cannot imagine that heat would cause the problems you mention. I use a sealmount without adjustable temperature (laminate or dry mount settings only - I use laminate as dry mount is never hot enough for my tissues) and simply use for minimum dwell time required. Occassionally I have had the odd corner refusing to adhere so it has been in and out a few times. Never have I had what you report, weird! To change colour I would have thought that the chemical change would have to be reversed or at least the grain structure seriously altered (as I thought the colour was a result of grain structure and reflectance....), but my lack of of chemistry PhD might render this total rubbish.

on the subject of temperature...tho one may use supposedly higher temperatures than other, the temperature the print is subject to may not be higher? One may use a thicker print sandwich and lesss time. If you subject tissue to direct heat ie with a tacking iron it pretty well instantly melts or bonds. I therefore hypothesise that if you use the minimum time to get a perfect bond, then the print will have been subject to the same temperature regardless of what the dry mount press dial reads....about the operating temperature of your particular tissue. For me 45s = perfect bond, 35 s= no bond to partial bond (vacuum press). I would have thought that where long dwell times and lower temperatures are used, it perhaps takes that bit longer for the heat to conduct through and reach full temperature?
Just guessing...