sources of the production of red wares. The earlier factories appear to have been situated in the country of the Ruteni at La Graufesenque, in the modern department of Aveyron, at Montans in Tarn, and at Banassac in Lozere—all in South-Western France. Somewhat later a second set of red-ware potteries sprang up in the country of the Arverni, the modern Auvergne,—at Lezoux near Clermont-Ferrand, and neighbouring localities in the valley of the Allier. Of these potteries our literary sources tell us nothing. Their history, as far as we know it, has been worked out purely by archaeological methods. In the excavations of the sites alluded to, many remains of potters' kilns have been met with, and large collections of fragments of vases and moulds have been obtained. In this way, the types of vessels made, the names of the potters, and the designs characteristic of each pottery have been ascertained, and from the manner in which they occur on sites which can be dated with some approximation to accuracy, Déchelette has deduced certain limits within which the activity of the various potteries may be placed.

La Graufesenque (with which we may include the neighbouring potteries of Banassac and Montans) appears to have begun its exportation early in the first century A.D. Its products, however, do not appear at Haltern. On the other hand, they are plentiful on the Lower Rhine at Vechten, at Xanten, and at Novaesium. They have been found in the cemetery of Andernach along with a series of coins ending with Nero, from which Déchelette concludes that the beginning of the exportation must be placed between the year A.D. 16 and the rise of the Flavian dynasty. Its close is, however, of more importance so far as Newstead is concerned. At Pompeii, destroyed in A.D. 79, nineteen bowls of Rutenian origin have been noted, while the pottery of Lezoux is entirely absent. Again, in the forts of the German Limes that date from the end of the first or the early years of the second century A.D., such as Waldmössingen, Heidenheim, and Okarben, the decorated products of the Rutenian potteries hold the field unchallenged. But with the end of the first century or the beginning of the second, the prosperity of the workshops of La Graufesenque would seem to have ended, and the activity of the Rutenian potters to have given place to that of those of Lezoux.

Déchelette dates the beginnings of Lezoux as early as the year A.D. 40. He places its first period between that year and A.D. 75, and considers it to have been a period of limited production. A second period of greater activity, marked by the use of many new designs, is placed between the years A.D. 75 and 110. It is during this period, and probably towards the end of it, that the

193