It may be a problem to get a precise age, since many gabbros and basalts do not contain minerals with high Sm/Nd, which is what you need to get a well-defined isochron. If there is garnet around, you're in luck. Otherwise, it may be a futile exercise, depending on the expected age of the rocks. If you really have no idea of their age, and are happy to know whether they are ca. 500 or 1000 Ma old, take cpx (and/or hbl) and WR; plagioclase contains very little of anything. If you have an alkalu-gabbro, you could use apatite. If you want to know whether the rocks are 400 or 450 Ma, you may have a problem to get an age that is precise enough to be meaningful.

Dear Marlina Elburg, thank you very much for your such soon and kind reply. My sample is a gabbro containing apatite. Its age may be between 300 Ma and 200 Ma.  There seems no garnet around the sample. According to you reply, did you mean that I may not have  a meaningful Sm-Nd  isochron age of this sample.?

I think it will be problematic to obtain an age that will be precise enough to be meaningful. Perhaps you can do some literature research whether anyone else managed anything similar.

If your gabbro does not contain zircons or titanites for U-Pb, I think your best bet may either be Hornblende Ar/Ar if your rock is not altered and if it is, perhaps U-Pb on apatite .

Both of these methods have closure temperatures around ~450-550 degrees so if your rock is metamorphosed, you will get the age of the metamorphic overprint.

regards

This was a great question I recently searched, as I see no mafic rock ages directly measured for Idaho Proterozoic rock. Thanks for the discussion.