324 ARCHAEOLOGICAL INTELLIGENCE. the South Downs, and amongst them from Mr. William Holt, a silver- smith of Eastbourne, informing me that, in 1805, or about that time, four large gold armillaj, of the same pattern as those found at Holjingbury Hill, were discovered near Southbourne, having been brought to light by a fall of the cliff, owing to the inroads of the sea. They were sent to Sir Joseph Banks, who found a purchaser of these ornaments, and JNIr. Holt stated that they were exhibited to the Society of Antiquaries, in 1807. In 1818, another armilla of gold was found in the same neighbourhood, and obtained by Mr. Holt. It was quite plain, and bent to the shape of the wrist. It was sold for twelve pounds to some collector of antiquities in Sussex ®." The armillse of gold, to which Mr. Holt's observation refers, do not ap- pear, however, to have been similar in form to those in Dr. Mantell's pos- session, as from some erroneous impression he had stated in his letter. They were exhibited to the Society of Antiquaries on March 19, 1807, and a representation of one of them, given in the Archseologia, shews that they were simple rings, with dilated extremities, resembling certain gold orna- ments found repeatedly [in Ireland. It is remarkable that celts, a portion of a broken sword of bronze, and lumps of pure copper, were found with these armillfe. HoUingbury castle is an earth-work, consisting of a square entrenchment, enclosing an area of about five acres ; it crowns the summit of a conspicuous conical hill, scarcely two miles distant from Brighton. The objects here represented were found under a low mound of earth, stated to have been within the vallum of the fortress, and at a slight depth beneath the turf. They lay, in the regtdar arrangement, placed as here shewn, in a slight excavation on the face of the chalk i-ock. The tore, of the simple funicular type, twisted like a cord and tapering towards the ends, measures 19 inches in length, and had been broken asunder in the middle, apparently by design. On each extremity of this tore was found a ring of bronze spirally twisted, considerably larger than the tore, and possibly intended for some adjustT ment in fastening the garments. Within the circle of this tore lay a large bronze celt, of the type formed with a central stop-ridge, according to Mr. Du Noyer's useful classification of these implements, and without any lateral ear or loop for fastening it to. the haft. This celt was also designedly broken into two pieces ; at the four corners around the tore and celt, were placed four massive bronze ar- millse of simple construction, formed of a fillet bent into a loop at one end, with a hook at the other, which was adjusted to the loop and formed a fast- ening. These armlets vary a little in size and thickness of the metal bars of which they were formed : the largest weighs nearly half a pound. The extremities of the tore found at HoUingbury castle were not re- curved, as most usual in objects of this kind, but straight, and bluntly pointed. The use of the spiral rings, through which when found, the ends of the tore were passed, is not obvious : a gold tore found near Boyton in « Arcli.xologia, vol. xvi. pi. 68, p. 363. exhibited at the Norwich Meeting of the Four other gold arniillae have been lately Institute. They are in the possession of found near the same locality, and were Col. Paine.
