151 riJE-UISTORlC ARCIIJIOLOGY OF EAST DEVON. Wiltshire barrows described by Sir R. C. Iloaro. A few inches below this was a finely patinated bronze dcXgger, finished with graceful symnietr}, and in somewhat imperfect preservation (fig. 1). This weapon or implement measnres nearly 2 inches in length, and the leaf-shaped blade at the widest part l;,' inches in breadth. Towards the lower part of the blade there are two small oblong perforations, which appear to have been made at the time of the casting of the instrument, and not to have been subsequently punched. hen discovered it retained one of the bronze rivets by which it had been attached to the haft, thereby showing that the handle had been formed of wood a!id not of metal, as in many of the swords and daggers of the same relative age that occur in Denmark. In reference to the weapons of this type. Dr. Wilson observes that " the metal is too brittle to resist violent contact with any hard body ; but if the edge of a bronze weapon is hammered till it begins to crack, and then ground, it acquires a hardness, and takes an edge not greatly inferior to the ordinary kinds of steel." From the discovery of this weapon we arc led to infer that one of the last honours paid to the buried wai'rior was to lay the well-proved weapon by the side of" the ashes of the heio who had wielded it, before that his companions in arms piled over the remains the barrow or memorial cairn. The custom still obtains ; the soldier's favourite sword is laid on his bier when his comrades bear him to his last resting-place. Proceeding with the careful removal of this accumulation of burnt bones and charcoal, that I have mentioned as oc- curring within the cist of b.-iked clay, we found another example of the skill and ingenuity of native manufacture, consisting of a .sepulchral vessel of small size, and of the typo kncnvn as tiie " incense-cu|»" (fig. 4). As we cannot doubt that the' burietl records of the progress attained by the ancient occupants ol' these ishuuls have been as yet only paitially disclu.sed. so also we may hope that as other e.- aiii|.l(S of native lietile ware are from time to time exhumed horn the store-hou.se in which th'-y lie safely garnered, and a.s th(3 Arelneologi.st is thereby stimulated to greater precision in the study of ancient interments and urn-hurials, we shall collect an amount of aecurately-ohscrved r-icls which may •Miable us to classit'v into distinct pi rio.ls the pottery found
