Thanks. That was helpful. Here is what I have figured out so far.
Here is the general scheme:
1.) rock (containing Mg 2+) + HF (able to dissolve rocks)---->MgF2 (and if Ca, Sr, and Ba are in the rock they will precipitate as well)
2.) To be sure that the above mentioned ions are not part of the MgF2 precipitate
(MgF2+ precipitate) + Sulfate ---->MgSulfate(aq) + possible Ca, Sr, Ba,sulfate precipitate
3.) I would then discard the precipitate and use the aq MgSulfate sln to determine the initial Mg content in the rock.
Does this seem plausible to you? Also, do you have any suggestions as to how I can determine the Mg content from the MgSulfate aqueous solution?