But it is neither from Italy nor from the countries bordering the Eastern Mediterranean that the most important supply of Roman glass has been obtained. Putting aside objects of quite local _provenance_, it will be found that in the museums of England, France, and Germany, by far the larger part of the glass exhibited—and this is above all the case with the blown glass—has been found within the limits of the ancient Gallia. Spain, contrary to what we might have expected, has yielded little Roman glass of any artistic merit, partly perhaps for want of systematic search. But there are few districts in France or in the west of Germany where the exploration of Roman cemeteries has not yielded a plentiful crop. If we travel northward from the estuary of the Rhone by way of Arles and Nismes to Avignon, Valence, and Lyons, then across by the country on either side of the Jura to the valley of the Rhine, and follow that river by Strassburg to Cologne, we pass for the whole way through a district especially rich in Roman glass. And this is what might well be looked for. The third and fourth centuries—a little earlier or a little later, according to locality—are above all the great centuries for the prevalent use of glass, and it was during this period that the central tract of country that included the two great metropolitan cities of Arles and Trèves began to take the prominent place that it maintained throughout the early Middle Ages. Even our English glass of this time, so much of which comes from districts to the north and the south of the estuary of the Thames, may be brought commercially at least into connection with the wealthy provinces of Northern and Eastern Gaul. It was from these provinces that glass was first imported, and from them, no doubt, the glass-workers passed over to Britain. In the case of the rich collection of Roman glass in the British Museum, the backbone, as it were, is formed by the specimens excavated from tombs in the neighbourhood of the lower Rhone valley—from Vaison, near Vaucluse (the Comarmond collection), from Apt, and from Alais. At Arles, in that district of tombs, the Aliscamps, which furnished Dante with a well-known image, beneath the Christian sarcophagi (in these, too, not a little glass has been found), the earlier Roman tombs lie on the bed-rock. From these tombs numberless urns of glass, in cases of lead or stone, have been taken, as well as many examples of glass of rare and exceptional shapes—among others what is apparently an alembic for use in distillation. Some of these vessels contain a red liquid which may _represent_ at least the wine with which they were originally filled (Froehner, p. 109). In this town of Arles, too, in the suburb of Trinquetailles, there were probably extensive glass-works, as we may infer from the quantity of vitrified paste there found (Quicherat, _Revue Archéologique_, xxviii.).