On an Italian oblong bronze coin or _quincussis_, 6⅝ inches by 3½ inches, and weighing about 3½ lbs., is the representation of a leaf-shaped sword with a raised rib along the centre of the blade, and in general character much like Fig. 345. A specimen of this coin is in the British Museum,[1049] and bears upon the reverse the figure of a scabbard with parallel sides, and a nearly circular chape. Another coin of the same type, engraved by Carelli,[1050] has a nearly similar scabbard on the reverse, but the sword on the obverse is either represented as being in its scabbard or is not at all leaf-shaped, the sides of the blade being parallel. The hilt is also curved, and there is a cross-guard. In fact, upon the one coin, the weapon has the appearance of a Roman sword of iron, and on the other that of a leaf-shaped sword of bronze. These pieces were no doubt cast in Umbria, probably in the third century B.C., but their attribution to Ariminum is at best doubtful. From the two varieties of sword appearing on coins of the same type, the inference may be drawn either that at the time when they were cast, bronze swords were in Umbria being superseded by those of iron; or that the type originally referred to some sacred weapon of bronze such as is represented on the coin in the British Museum, but was subsequently made more conventional so as to represent the sword in ordinary use at the period. [Illustration: Fig. 346. Tiverton. ¼] [Illustration: Fig. 347. Kingston. ¼] The sword with a central rib was sometimes attached to the hilt in a different manner from any of the blades hitherto described, as will be seen by Fig. 346, copied from the _Archæological Association Journal_.[1051] This sword was found at Tiverton, near Bath, and it is provided with four rivets, a pair on each side of the continuation of the central rib along the hilt-plate. Human remains and stag’s-horns are said to have been found near it. In the British Museum is a blade of the same kind (19⅝ inches), with semicircular notches for the four rivets. It was found in the Thames at Kingston. Another from the Thames (21 inches) has the two upper holes perfect. Leaf-shaped swords of the ordinary type also occasionally had their hilts attached in the same manner. Fig. 347 shows a blade from the Thames,[1052] near Kingston (16⅛ inches) with the rivet-holes thus arranged. I have another, from the Hugo Collection (18 inches), found in the Thames about a mile west from Barking Creek,[1053] which has had four rivet-holes arranged in the same manner, though the margins are now broken away, so that only traces of the holes remain. Another apparently of this type was found in Lincolnshire.[1054] In Canon Greenwell’s Collection is a leaf-shaped blade of the same character (15¾ inches), which, however, has only two rivet-holes, one on each side of the hilt-plate. It was found at Sandford,[1055] near Oxford, together with a rapier-shaped blade.