Page:Archaeological Journal, Volume 7.djvu/486

352 mixed, thus making very irregular forms; so that I became convinced that the verdigris—for such in reality it is—was the result of some chemical change, and on scraping off a portion from the surface of the tessellæ this proved to be the case; as when the green coating was removed it was found to conceal a beautifully coloured ruby glass. Here, then, the nature of the unsatisfactory colouring was made apparent. On making a new tracing, and putting the bright red—though colour but badly imitates its richness and transparency—the head-dress, a chaplet of flowers, as also the branch held in the hand, were clearly shown to be meant for "ruby gems," and as such became intelligible, and at the same time struck us as being exceedingly appropriate.

This, perhaps, may serve to show us that in many instances Mosaic pavements may have undergone great changes in colour from decomposition; so that, whenever we see any of these decorations faulty or unsatisfactory m chromatic arrangement, we should in all instances examine the matter more closely, for it may happen, as in the example before us, that the colouring, as first observed, would render the subject confused, when on copying from amended observations all would become plain and harmonious.

But as these changes depend on chemical action, it behoves us to inquire deeper, and with this conviction I requested my friend and coadjutor, Dr. Augustus Voelcker, Professor of Chemistry to the Royal Agricultural College, to undertake a chemical analysis of the only fragment of the glass that could be spared for that purpose; the result of that inquiry I shall lay briefly before the Society.

The analyses—which were only qualitative, as we had not materials sufficient to determine quantities—resolved themselves into two subjects; namely, an examination of the green powder on the surface of the glass, and an inquiry into the constituents of the glass itself, which gave the following results:—