The next class of socketed celts which has to be noticed consists of those in which the loop is absent. No doubt, in some cases, this absence arises either from defective casting, or from the loop having been accidentally broken off, and all traces of it removed; but in many instances it is evident that the tools were cast purposely without a loop. It seems probable that many of them were intended for use as chisels, and not like the looped kinds as axes or hatchets. The similarity between the looped and the loopless varieties is so great that I have thought it best to describe some of the instruments which may be regarded as undoubtedly chisels in this place rather than in the chapter devoted to chisels, in which, however, such of the socketed kinds as are narrow at the edge, and do not expand like the common forms of celt, will be found described. [Illustration: Fig. 159. Reach Fen. ½] [Illustration: Fig. 160. Carlton Rode. ½] The small tool shown in Fig. 159 may safely be regarded as a chisel. It does not show the slightest trace of ever having been intended to have a loop, and is indeed too light for a hatchet. It was found with a tanged chisel, a hammer, numerous socketed celts, and other articles, in the hoard from Reach Fen, Cambridge, already mentioned at p. 112. I have seen another, 2⅛ inches long, with a somewhat oval socket and no loop, which was found in Mildenhall Fen, and was in the collection of the Rev. S. Banks, of Cottenham. A longer celt of the same character is engraved by Dr. Plot.[480] It was sent to him by Charles Cotton, Esq., and according to Plot “seems to have been the head of a Roman rest used to support the lituus, the trombe-torte, crooked trumpet, or horne pipe used in the Roman armies.” Another of nearly the same form was found on Meon Hill,[481] near Camden, Gloucestershire. A celt or chisel of this character found at Düren, in North Brabant, is in the museum at Leyden. Another was found at Zaborowo,[482] in Posen, in a sepulchral urn. A celt of the octagonal form of section and without a loop is shown in Fig. 160. It formed part of the great hoard found at Carlton Rode, near Attleborough, Norfolk, of which some particulars have already been given. The joint marks of the moulds are still very distinct upon the sides. This specimen is in the Norwich Museum, and was kindly lent by the trustees for me to have it engraved. A nearly similar Scottish celt is shown in Fig. 165. A celt from the hoard of Cumberlow, near Baldock,[483] has been figured as having no loop, but I believe that this has arisen from an error of the engraver, as in a drawing which I have seen the loop is present. One of hexagonal section and socket from a hoard found on Earsley Common,[484] Yorkshire, in 1735, is engraved as having no loop. Celts without loops are not uncommon in France, and are often found of small size in Denmark.[485]