having been found at Bettenberge. It is made of bronze, and has been overlaid with silver. A head-piece, which has evidently undergone alteration in barbarian hands, was discovered at Thorsbjerg, in Schleswig. Of the masks, some fifteen in all are known. Hitherto none of the finds have been made in circumstances which rendered it possible to fix the exact period to which this type of helmet belongs. Dr. Donner von Richter,[1] whose treatise on the subject is one of the most recent, places the Ribchester and Nikopolis helmets at the beginning of the third century. Now, however, the unmistakable association with first-century pottery has furnished a new and valuable clue. The Newstead helmets must have been deposited in their resting-places prior to 100 A.D. The probability is that both the Ribchester and the Nikopolis helmets belong to the same period. At Ribchester, certainly, the pottery shows that the fort is old enough for a first century helmet to be lost in it.

Their use

The exact purpose for which the visor-helmets were intended is somewhat uncertain. It seems clear that in battle they would be cumbrous and of little service. Benndorf, who made an elaborate study of the group, came to the conclusion that they were not part of the Roman soldier's equipment, but were really parade armour, destined to figure in certain ceremonies, and especially to be used as death-masks, laid upon the face of the body when it was deposited in the grave. It is an undoubted fact that such masks have occurred as grave furniture. A case in point is the recent discovery of a mask of hammered iron at Chassenard, in France.[2] At the same time, it is impossible to suppose that the Newstead helmets, with their woollen padding, and with their visor-masks showing signs of repair and of alteration, had no other purpose than to be laid on a dead soldier's face. A passage from Arrian's Τεχνη Τακτικη, cited by Benndorf, probably furnishes us with the true explanation, although Benndorf himself seems hardly to have appreciated the full importance of his quotation. The Τεχνη Τακτικη was written in the twentieth year of the reign of Hadrian,—that is, in 136 A.D. The last ten or twelve chapters are devoted to a description of the sports or exercises indulged in by the Roman cavalry. The author apologises at the outset for the difficulty he feels in making the necessary explanations. In many cases, he says, there is not even a Latin equivalent for the technical terms, some of which are Iberian, and some

1 Donner von Richter, Op. cit. pp. 40 and 41.

2 Déchelette, 'La Sepulture de Chassenard,' Revue Archéologique, 1903, tome i. p. 235.