The large number of paintings of the fifteenth and sixteenth centuries in which colourless powdered glass has been found as a paint additive demonstrates how widely it was used. In the northern European paintings studied here it was present in translucent red lake containing paints and in mordants for gilding. In the Italian paintings in this study it was found not only in red lake glazes, and in several mordants, but also in paint of other colours, generally of grey, brown or orange hues. It was, in addition, a rather common component of the primings.
Aside from the references to powdered glass as an aid to the grinding of orpiment, the earliest documentary sources found so far in which it is mentioned are from the second half of the sixteenth century, but they do not indicate its purpose. It is not until the very last years of the sixteenth century in English documentary sources that it is suggested as a drier. In some sources there is some hint that it is being used specifically for colours that lack body. Experiments carried out to investigate its effect on the properties of oil paint have indeed confirmed that it does give a thicker paint with good handling properties. These reconstructions have not proved conclusive on the question of whether it has any influence on drying time, and further experiments that replicate the particle size and composition of the glass found in paintings, as well as following instructions in the sources, such as initial grinding in water, would be valuable. Even if the manganese in the glass is not sufficiently available to act on the oil, it may be that its alkalinity has an effect, since it is known that basic pigments can interact with acidic oxidation products which can otherwise soften a dried film note 80). Artists may have believed that it could act as a drier, even if in fact it was not capable of performing this function. It could also have been effective as a transparent filler or extender, appropriate for the relatively expensive red lake pigments. Translucency might be desirable in a priming layer since it would allow underdrawing beneath it to remain easily visible, but this cannot have been its purpose where it is used in mordant layers hidden beneath metal leaf. It is also unlikely to have been added as an extender where it has been found mixed with cheap earth pigments. The reasons for adding glass to the paint may have varied, but in Italian paintings, at least, the consistent manner and extent of its use seem to suggest that it was not dependent on the idiosyncrasies of individual artists but was instead an established general practice.
Artists appear to have used whatever colourless glass was available in the vicinity, whether it was locally made or imported, as the composition shows the same general geographical distribution established from studies of vessel and window glass. In the Italian paintings it is almost exclusively soda ash glass, with only one exception. In the German and Netherlandish paintings soda ash glass, wood ash and wood ash–lime glass are present, in common with what has been found in archaeological sites from these areas. In the results from Northern paintings presented here, the locally produced glass types are still predominant, although the number of analyses is rather more limited than those from Italian paintings, and more are needed before general conclusions can be reached. The quantitative analysis of the glass in Italian paintings has also made it evident that there are some differences in composition that can be related to the raw materials, particularly in the use of either sand or river pebbles as the silica source, and that this does seem to have some relationship to the location in which the painting was made. Again, while this does identify some possible directions for future investigation, no firm trends can be established until a more geographically diverse group of occurrences in paintings has been analysed, especially those in Northern Italian works, combined with a deeper comparison with the literature on glass-making practices and raw materials used in this period in this area.
The earliest paintings in which glass has been found are Netherlandish and German works from around 1430. It does seem to be associated with an oil medium and it is therefore not surprising that the earliest occurrence in a painting produced in Italy is from the 1470s, a time when Italian artists were beginning to use oil more widely but were still predominantly working in egg tempera. In Northern Europe it has a far longer history as a paint binder, and far earlier works in the Netherlands, Germany, England and Norway have been confirmed through analysis to have an oil medium. Further work to investigate the origins of this practice should therefore be concentrated on paintings from these areas (note 81). Nevertheless, it is already interesting that artists added a material to their paint that did not function as a pigment but would have modified the working and handling properties, and perhaps also the drying properties, and that it can be traced back as far as the early Netherlandish artists such as van Eyck and van der Weyden, who were famous for their ability to manipulate oil paint to achieve the remarkable detail and impressive effects that can be seen in their paintings.