Dark Earth deposits immediately overlie late Roman urban stratigraphy across northwest Europe, representing the crucial but poorly documented collapse of provincial Roman urbanism. Deficient in artefacts, they have proved recalcitrant to traditional methods of archaeological interpretation. Here, they are used as a vehicle to promote a more integrated, holistic approach to scientific archaeology. This recognises the great potential value of applying a wide range of geoanalytical techniques to the finer-grained matrices that enclose (or, more precisely, grade into) artefacts in archaeological sequences. As this multifaceted approach is time-consuming, deposits chosen for analysis should contain potential answers to profound historical questions. Comparative studies are necessary, in which samples of known age, provenance, environment and mode of deposition outnumber those of a more equivocal nature. Considerable knowledge is required to select the optimal range of complementary techniques for application to a particular suite of materials; this case study outlines the relative merits of analyses for fabric orientation, particle size, micromorphology, bulk geochemistry (ICP), particle geochemistry (microprobe), heavy mineralogy, plant remains (pollen, phytoliths, wood), animal remains, macroscopic artefacts, and radiometric dating. The resulting large bodies of data are best summarised by multivariate analyses (notably ordination algorithms), together with semivariograms for spatial data. Interpretations should take full account of the range of anthropogenic processes and products inherent in archaeological deposits.