Over the last decades, archaeology has been undergoing a revolution of sorts driven by advances in the archaeological sciences and the ability to extract data from sediments. These new methods work at the micro- or even molecular scale and thus require extremely high levels of precision for the context of the samples. While there have been significant advances in proveniencing techniques, by and large, basic excavation methods have hardly changed in half a century or more. We document excavations better, we save more things more systematically, and we move slower, but excavations are still primarily focused on the recovery of objects. In our experience, sampling for the archaeological sciences is shoehorned into the existing methodology. Here we describe an excavation methodology - vial-based excavation - that is instead designed from the start for the archaeological sciences and, in particular, to systematically collect the invisible components of the archaeological record. To do this, we remove intact blocks of sediment to a laboratory, employ a vacuum system for complete recovery of sediments, collect sediments as a large number of very small samples in glass vials, use micromorphology extensively to track micro-contexts, and use digital systems to document contexts efficiently. We applied this methodology to the investigation of fire residues in Layer 8 of the Middle Paleolithic site of Pech de l’Azé IV, France. Our purpose here is to share this methodology as a potentially useful companion to more traditional excavation methodologies.
