ROMANO-BRITISH DERBYSHIRE (5) Not a few coins have been found. A copper coin of Domitian was discovered in 1841 outside the fort near a stone coffin beside Glossop Brook, and another or, it may be, the same otherwise described is said to have been discovered in the fort at some date previous to I85O. 1 More recently the excavations have revealed three denarii, one each of Galba, Domitian, and Trajan ; a ' second brass ' of Hadrian, a ' first ' and a ' second brass,' each illegible, a copper coin of Gallienus, and (strange discovery) a coin minted by the Jewish insurgent leader Barcochab in A.D. 134. (6) The pottery includes embossed Samian of the first and second centuries, belonging to the types numbered by DragendorfF as 29, 30, and 37 ; a bit of Castor or New Forest ware ; a pelvis lip stamped FECIT ; an amphora handle stamped SGA, and a curious little earthenware ornament, 4$ inches long, representing a saddled horse and perhaps used as a toy. 3 Lastly, the roads. The well-known Doctor Gate, as Watson observed, led eastwards across the moors to Brough and ultimately to Templeborough. Another road, stated by Watson to be visible in 1771 and well gravelled, led to Stockport and Manchester. A less well-attested road perhaps led direct to Buxton, and Watson thought he detected a fourth road, connecting with Staley Street and the route from Manchester to York. Here, as at Brough, we must for the present refrain from any attempt to write the history of the fort. We may think that we see in the fashion of the ramparts and the head-quarters building and in the coins, some indications that the fort was established at a comparatively early period, perhaps during the last third of the first century, in the reigns of Vespasian or Domitian, and an inscrip- tion shows that at some time or other the Cohors i Frisiavonum was employed here. For the rest, we must wait, as elsewhere, for more light. The Romano-British name of the fort is unknown. Watkin, noting in the Ravennas the series : Navione (= Anavione, Brough), Aquis (Buxton), Arnemeza, Zerdotalia, conjectured for the fourth name Zedrotalia, connected it with the river Etherow or Edrow, and pro- duced thus a title for the fort. But it is obviously a wild and uncritical guess, and one must regret that it has found some local credence. The name Melandra itself is equally obscure. It cannot be traced back beyond 1771 ; it is not susceptible of philological explanation, and its curiously Greek look suggests that its form is not entirely free from learned influence. 3 1 Hamnett, Derb. Arch. Jour. xxi. 15, 1 6 ; xxiii. 102. Beaumont, Brit. Arch. Assoc. Journ. vii. (1851) 18. Two copper coins, one of Postumus and one of Carausius, are said to have been found here also, but more probably in the vicinity of Melandra than in the fort. The coins mentioned by Watkin, Derb. Arch. Journ. vii. 89 (Marcus to Severus Alexander), as found at Melandra were really found at Hooleywood in 1838 : see Padfield in the alphabetical index. 8 Personal inspection, by the kindness of Mr. Hamnett and Prof. R. S. Conway. 3 Watkin, Derb. Arch. Journ. vii. 86 ; information from Mr. W. H. Stevenson. 215
