The mysty product website works fine for me, albeit a bit slow to load.http://the-flower.biz/?p=37
They use the same sort of process as discussed throughout this thread.
You soak in one solution to dehydrate and remove color and then you soak in another solution to dye and to replace the solvent with another that won't evaporate.
I decided to do a little testing of my own.
I was interested whether it would be possible to find a solvent which preserved the color while displacing the water.
So, I bought a red rose to test the effect of various solvents on petal clippings.
Methanol was very effective at removing color.
My subjective impression of their relative strengths:
The DMSO acted much slower than the methanol, but over a couple days, it has rendered portions of some of the petal clippings transparent.
The alcohols only achieved translucency.
A drop of HCl into the acids immediately turns the solution a bright red, demonstrating the pH-dependence of the rose anthocyanidin pigment.
Even the isopropyl removed a great deal of color and I doubt that it would be easy to preserve a red rose with the original color intact by this sort of procedure.
However, if you want to dye your flowers anyway then methanol seems like a good initial solvent.