^{page 13 note 1} Fig. 1. National Grid Reference SJ229075.

^{page 13 note 2} Mr. Shepherd believes that all the articles found were recovered by him from the workmen. Having spoken to the men concerned, I am inclined to agree.

^{page 14 note 1} I had the welcome assistance of Mr. Shepherd and also of Air-Commodore Douglas Iron, C.B.E., and there were two labourers. The ready help of Alderman W. Richards and of the Borough Surveyor, Mr. F. A. Lodge, are also to be recorded.

^{page 14 note 2} Information from Mr. Lodge.

^{page 14 note 3} See Lloyd, J. E., A History of Wales to the Edwardian Conquest (3rd ed., 1939), i, 248.Google Scholar Properly, Trallwng Llywelyn, Llywelyn's Bog.

^{page 14 note 4} Cited in Wales, R.C.A.H.M. & Monm. I, Montgomeryshire (1911), 182—mon. 911Google Scholar.

^{page 14 note 5} See Nash-Williams, V. E., The Roman Frontier in Wales (1954), 48–56Google Scholar, for brief account and refs.

^{page 16 note 1} In Roman barrows, burial was commonly at or about ground-level: see Dunning, G. C. and Jessup, R., ‘Roman Barrows’, Antiquity, x (1936), 38Google Scholar. The small nail can scarcely have been used to join the planks of a cist, although wooden cists are quite common (Dunning and Jessup, op. cit., 38 and gazetteer).

^{page 16 note 2} Mr. Shepherd was aware that traces of a cremation might be found, and since he had had experience in helping to excavate a Saxon cremationcemetery near Hull, knew the appearance of cremated bone. None was found either in 1959 or 1960, so far as is known.

^{page 16 note 3} Not uncommon in the highland zone.

^{page 16 note 4} Jessup, R. F.et al., ‘A Roman Barrow at Holborough, Snodland’, Arch. Cant, lxviii (1954), 5, 9, 11, 15–19 passimGoogle Scholar.

^{page 16 note 5} Ibid. 22. Magnetite (‘mill-scale’), Fe_aO₄, a product of the heating of iron, was identified at Holborough as a distinct inner layer. Superficially, the Welshpool specimens appeared the same, but Dr. R. F. Tylecote of King's College, Newcastle upon Tyne, was unable to establish this.

^{page 17 note 1} Kindly identified by Mr. H. A. Hyde, Keeper of Botany in the National Museum of Wales.

^{page 17 note 2} This was kindly verified by Mr. Hyde.

^{page 17 note 3} A parallel to the careful packing of bronzes is at the Langley (Kent) cemetery: Jessup, R. F., ‘Bar-rows and Walled Cemeteries in Roman Britain’, Journ. Brit. Archaeol. Assn. ser. 3, xxii (1959), 27Google Scholar.

^{page 17 note 4} Ibid. and Dunning and Jessup, op. cit., passim; and the continental material cited.

^{page 17 note 5} Allen, J. Romilly, ‘Two Kelto-Roman Finds in Wales’, Arch. Camb. ser.6, i (1901), 31–33, 44 nGoogle Scholar.

^{page 17 note 6} Including direct, two-piece casts from paterahandles (Grimes, W. F., ‘Holt, Denbs.: the Works-depot of the Twentieth Legion’, Y Cymmrodor, xli (1930), 174Google Scholar, fig. 60, 4; and personal inspection. Copies of bronze vessels, more or less true to the prototypes, are common in legionary ware both at Holt, Caerleon, and, for example, Vindonissa (for the last two see R. E. M., and Wheeler, T. V., Archaeologia, lxviii, fig. 21Google Scholar; and Ettlinger, E., Journ. Rom. Stud, xli, 105Google Scholar).

^{page 17 note 7} Cf. Frazer, J. G., Adonis Attis Osiris (2nd ed. 1907), 73–76Google Scholar; Cumont, F., Recherches sur le symbolisme funéraire des romains (1942), 392, 395–6Google Scholar; Drioux, G., Cultes indigoènes des Lingons (1934), 105Google Scholar.

^{page 18 note 1} Conveniently, Hawkes, C. F. C. and Dunning, G. C., ‘The Belgae of Gaul and Britain’, Arch. Journ. lxxxvii (1930), 304–5Google Scholar and notes.

^{page 18 note 2} Barrow II. Gage, J. in Archaeologia, xxv (1834), p. 8Google Scholar, pi. iii, 9.

^{page 18 note 3} Conveniently, Smith, R. A. ‘On Late-Celtic Antiquities discovered at Welwyn, Herts.’, Archaeologia, lxiii (1911–12), 5 ffGoogle Scholar.

^{page 18 note 4} Note in Antiq. Journ. xxxviii (1958), 89 f.

^{page 18 note 5} See Piggott, 's table, ‘Fire-dogs again’, Antiquity, xxii (1948), 27Google Scholar.

^{page 18 note 6} See Appendix.

^{page 18 note 7} Type, for example, Brit. Mus. Guide to the Antiq. of Rom. Brit. (1951), 45, fig. 20, 5.

^{page 18 note 8} For example, Richter, G. M., Greek Etruscan and Roman bronzes (Metro. Mus. of Art, New York, 1915)Google Scholar, Tomb Group III, 180–1, of the fourth-third century B.C., has three candelabra and four firedogs.

^{page 18 note 9} For example, Tienen, Mertens, J., L'Antiquité classique, xxi (1952), fig. 9 and pi. iii, 2Google Scholar; citing analogous pieces.

^{page 18 note 10} Dunning and Jessup, op. cit. 40.

^{page 18 note 11} The date of the individual Welshpool pieces is considered in the Appendix.

^{page 18 note 1} See R. A. Smith, op. cit. 8–12, and C. R. Smith, Collectanea Antigua, ii (1852), 25 ff. for more details of Mt. Bures.

^{page 19 note 2} See the articles in the Appendix.

^{page 19 note 3} Tacitus, Annales, xii, 33.

^{page 19 note 4} In whose territory much of the fine metalwork found in Wales seems to have been made: Fox, C., Pattern and Purpose (1958), 75, 144.Google Scholar

^{page 19 note 5} Montgom. Coll. xlii (1931–2), 85, 104.

^{page 19 note 6} Found at the secondary modern school and presented to the National Museum of Wales by Mr. P. Spurgeon.

^{page 19 note 7} Nash-Williams, op. cit. 46.

^{page 20 note 1} Llanfrynach, : Bull. Board of Celt. Stud, xiii (1950), 105–8Google Scholar; Usk valley. Rea valley: O.S. Map of Roman Britain (3rd ed. 1956) with information kindly sent by Mr. C. W. Philips, F.S.A.

^{page 20 note 2} S. Piggott in ch. 1 of Richmond, I. A. (ed.), Roman and Native in North Britain (1958)Google Scholar; Wheeler, R. E. M., The Stanwick Fortifications (Res. Rept. Soc. Antiq. Lond. xvii, 1954), 27–30Google Scholar.

^{page 20 note 3} Nash-Williams, op. cit. 46, 57 f., 48.

^{page 20 note 4} Ibid. 52.