Strange, yes.

My first guess was another suite of evaporite minerals. Wrong!

I've seen light colored mafic rocks here and there, "komatiites" are the most mafic I can think of, but they are not white, more gray. They are light enough to confuse one for a more felsic rock, though.

I agree with PsiPhi about the limitations of the data, applying possible mineral compositions to curves, etc. They all add up to new puzzles.

PsiPhi's syenite idea is interesting, this is an interesting class o' rocks. Great names, too: the extrusive varieties are called "trachytes". But they typically have alot of K-spar and are generally considered felsic rocks (quartz poor granites). However, "shonkinite" has lots of pyroxene and some olivine. They are sometimes associated with carbonotites, really strange, pretty rare, typically ancient, limestone volcanos. Yeah, you heard me: carbonate volcanics. They are thought to come from carbonates in the mantle. Some of the rocks that come from carbonotites are really strange. But would they send back a basalt spectra? I doubt it.

Here's one: Anorthosite. Composed of mostly of plagioclase and some mafic minerals, anorthosites are associated with layered mafic complexes or within gabbros (intrusive basalts). They do not occur as volcanics. The more intermediate varieties can be lighter in color. But again, would their spectra be similar to basalt?

Could be very weathered basalt, or cooked as REB supposes. But would the elemental composition of the surfaces (RAT'd or not) be different?

I'll keep looking at the graphs, scratching my head, waiting for Nick's 2 cents, and waving my arms.

BTW, extended explanations of the figures can be found by clicking on the thumbnails here.
(This page updates all the time, if you are reading this a few days after it was posted, you'll not see the graphs here, you'll have to go back to older pages. The slide numbers are PIA05756-PIA05758, you can use the search box to find them quickly.)