Richly decorated enamelled glass objects and fragments of different provenance and epoch have been analysed using mobile and fixed Raman instruments: some fragments of the outstanding Begram treasure (Musée des arts asiatiques – Guimet, Paris) dated to the 1^st century AD, mosque lamps and bottles of Syrian/Egyptian provenance dated to the 13^th/14^th century (collections of Musée du Louvre and of Musée des arts décoratifs, Paris). The techniques are compared using the data obtained from the study of a group of similar objects and fragments discovered in Melfi Castle in the South of Italy in an archaeological context dated to the last quarter of the 13^th century. The glass body was difficult to analyse requiring the use of high-energy high-power laser beams and/or sampling that allowed determining the soda-lime type precisely. In contrast, a variety of colouring agents was identified: lapis lazuli and/or cobalt for blue, antimonate pyrochlore solid solution for yellow, with the addition of cobalt/lapis lazuli for green, hematite for red and calcium phosphate/cassiterite/calcium antimonate for white. Where present, gilding was found applied on a rough and matt red enamel base probably in order to guarantee the physical adherence of the gold leaves. The comparison between the above mentioned groups of objects and between them and data existing in the literature about Roman enamelled glass allowed us to follow the evolution of the technology of this class of precious artefacts and to discuss the potential of the mobile Raman analysis.

The present overview answers the need of assessing the current state of the art concerning the application of principal components analysis (PCA) to Raman spectroscopy investigations of cultural heritage and related materials. An increment of the employment of this multivariate statistic technique to Raman results in the mentioned field began between 15 and 10 years ago, after a very slow start at the turn of the millennium. A delay of about a decade was observed with respect to PCA applied to elemental quantitative data of archaeometric analyses, likely a consequence of the required spectral pre‐treatment and to results of complex interpretation. Therefore, it is by now the time to summarize this evolution in a comprehensive, yet very specific way. In this overview, painting constituents were considered, both colouring materials and binders, in addition to natural and synthetic glasses, and biogenic and mineral gemmological materials. A marked unbalance between the studies pertaining to the different sections has been noticed, revealing a concentration of the work mainly on painting materials, including the study of ageing and alteration. The different aims of PCA application to Raman spectra, the various approaches and the achievable results, with the possible arising problems, were underlined, too. Special attention was given to the pre‐treatment of the spectra, which was observed to be essential to overcome the influence of several issues concerning bands intensity, spectral noise, background, fluorescence and so on.