This study is concerned with the compositional analysis of Roman and early post-Roman glass from both domestic assemblages and the remains derived from glass working and producing sites in Britain, using Inductively Coupled Plasma Spectrometry (ICPS). Samples analysed were from glassworking waste from Mancetter (midsecond century), Leicester (third century) and Worcester (first to third centuries), glass production debris, probably manufactured from the raw materials, in conjuncton with a domestic assemblage, from Coppergate (first to fourth centuries, or possibly later), and a domestic assemblage from Fishergate (spanning both the Roman and immediate post- Roman periods). All the glass analysed was shown to be of a typical and uniform soda-limesilica composition, except for a small number of vessel fragments from York which were higher in calcium. Any compositional differences between blue-green glasses typologically dated either to the Roman or Saxon periods, were found not to be consistent. Analysis of the colourless glass showed that the majority appeared to be actively decolorized using antimony, in conjunction with apparent differences in the compositions of the raw materials, when compared to glass of the same date in other colours. Compositional differences between melted waste from Mancetter, Leicester and Worcester, were apparent, but not to an extent which allowed characterization to be successful. Analysis of glassmelting pots from Coppergate showed some high temperature glassworking (and possibly glassmaking) could have occurred. Other debris, thought to be indicative of glassmaking was also analysed and compared to the composition of the Roman domestic assemblage.

This study is concerned with the compositional analysis of Roman and early post-Roman glass from both domestic assemblages and the remains derived from glass working and producing sites in Britain, using Inductively Coupled Plasma Spectrometry (ICPS). Samples analysed were from glassworking waste from Mancetter (midsecond century), Leicester (third century) and Worcester (first to third centuries), glass production debris, probably manufactured from the raw materials, in conjuncton with a domestic assemblage, from Coppergate (first to fourth centuries, or possibly later), and a domestic assemblage from Fishergate (spanning both the Roman and immediate post- Roman periods). All the glass analysed was shown to be of a typical and uniform soda-limesilica composition, except for a small number of vessel fragments from York which were higher in calcium. Any compositional differences between blue-green glasses typologically dated either to the Roman or Saxon periods, were found not to be consistent. Analysis of the colourless glass showed that the majority appeared to be actively decolorized using antimony, in conjunction with apparent differences In the compositions of the raw materials, when compared to glass of the same date in other 0 colours. Compositional differences between melted waste from Mancetter, Leicester and Worcester, were apparent, but not to an extent which allowed characterization to be successful. Analysis of glassmelting pots from Coppergate showed some high temperature glassworking (and possibly glassmaking) could have occurred. Other debris, thought to be indicative of glassmaking was also analysed and compared to the composition of the Roman domestic assemblage.