Another noteworthy palstave, found at Miltown, Co. Dublin, is shown in Fig. 105. In this the side wings are not hammered over, and the stop is supported by a conical bracket. The shoulders, instead of being nearly square to the midrib, are inclined upwards at an angle of nearly 45°, so as to form receptacles in which the wedge-shaped ends of the split handle would be held tight against the blade. These inclined stops have been observed in other palstaves of different forms, and Sir W. Wilde[379] has called attention to them in connection with a palstave much like that now under consideration, but without any projection or loop on the side. The most remarkable feature in the Miltown example is a projecting, slightly curved spike or neb placed near the top of the blade rather above the position usually occupied by the loop. At first sight it looks like an imperfect loop, but, on examination, it is evident that the casting is perfect; and, on consideration, it seems clear that this projection would serve quite as well as a loop for receiving a cord to hold the blade back upon its haft, while for the actual tying it would be more convenient, as the cord would have merely to be passed over a hook, and not to be threaded through a loop. In a somewhat similar palstave (3⅞ inches) in the Museum of the Royal Irish Academy[380] there is also a projecting neb, but more semicircular in outline. I am not sure that it was intended for the same purpose. A looped palstave of this type, but with the bottom of the side socket more circular, is engraved by Vallancey.[381] [Illustration: Fig. 105.—Miltown. ½] Some of the socketed celts from the Bologna hoard have curved nebs on each side instead of rings. Instruments of the same character, also from Italy, have been engraved by De Bonstetten,[382] Schreiber,[383] and Caylus.[384] Double-looped palstaves, with a loop on either side, and in character like Fig. 86, are almost or quite as rare in Ireland as in England. The only specimen engraved by Wilde[385] is in the collection of Lord Talbot de Malahide. It is 6¼ inches long, with the loops not quite symmetrical. It was supposed to be unique. I have, however, another specimen of this type (6⅜ inches) found at Ballincollig,[386] Co. Cork, in 1854, which was formerly in the collection of the Rev. Thomas Hugo, F.S.A. It so closely resembles Fig. 86 that it is not worth while to engrave it.