First of all, what a cool job, being in the "department of scientific research" of the Metropolitan Museum of Art at New York, scraping tiny little bits off great paintings and analysing where the artists bought their colours :)
But seriously, I think this modified Raman method the author has optimised for the application to artworks looks promising, as it allows to identify organic pigments with only a microscopically small sample, which was previously nearly impossible.
While his previous efforts on detecting dyes in solution or in fabric fibres were published in the Journal of Raman Spectroscopy, what got this one into PNAS was obviously the fact that the author not only managed to adapt the technique to paintings and glazes on sculptures, but that he also pushed back the record for the earliest records of humans preparing organic dyes from plants or animals by a cool 7 centuries by studying 4000 year old sample from the Middle Kingdom.
Microanalysis of organic pigments and glazes in polychrome works of art by surface-enhanced resonance Raman scattering
Published online before print August 10, 2009, doi: 10.1073/pnas.0906995106