page 369 note 1 ‘The Llynfawr and other Hoards of the Bronze Age’, by O. G. S. Crawford and R. E. M. Wheeler, Archaeologia, lxxi, 133 ff.

page 371 note 1 Mr. Stow recollected that the sword was part of the find when reminded of its existence.

page 372 note 1 Archaeologia, lxxi, pl. XII, 1.

page 372 note 2 Archaeologia, lxxx, 32.

page 372 note 3 Loc. cit., p. 20.

page 372 note 4 The Mabinogion, trans. Elliott and Lloyd, i, 53–4. Miss Chitty drew my attention to this apposite quotation. Mr. I. C. Peate tells me that the Welsh word for cauldron in this text [Pair=peir in 12th century =coire in Irish) has no Roman connexions and is undoubtedly pre-Roman. The translators’ comment on this episode is: ‘We are here introduced to the great “cauldron”, later called the “cauldron of regeneration”, which plays an important part in Celtic myth. It is sometimes referred to as the “cauldron of inspiration”, the origin of all poetry and arts. It is generally associated with deep waters, e.g. Llyn Tegid (Bala Lake).’

page 373 note 1 Cauldron I is distorted and the test cannot be applied.

page 373 note 2 Nos. 18 and 22 of Leeds's schedule in the National Museum at Dublin show the same planes of wear, but not nos. 26 and 29.

page 375 note 1 See W. Parker Brewis, ‘The Bronze Sword in Great Britain’, Archaeologia, lxxiii, pls. XLI and XLII, figs. 25, and Peake, The Bronze Age and the Celtic World, pl. xiv, 5–7.

page 375 note 2 As von Sacken, Das Grabfeld von Hallstatt, Taf. v, fig. 2; and example in the British Museum.

page 375 note 3 Peake, The Bronze Age and the Celtic World, pl. xi, 1.

page 375 note 4 See my fig. 1; and Wheeler (loc. cit., pp. 134–5 and pl. ix) for a description of the Llyn Fawr chapes.

page 375 note 5 Déchelette, ii, 728 and map; and Henry, F., Les Tumulus du Département de la Côte d'Or (1933), 45 ff.Google Scholar, where 31 burials with such swords are discussed.

page 375 note 6 See Mile Henry's chronological analysis, loc. cit., pp. 76–8.

page 375 note 7 See Map, fig. 5, Dunlop, , ‘L'Age du Bronze en France’, in L'Anthropologie, 1939, p. 471Google Scholar, and cf. this author's fig. 16, p. 497.

page 376 note 1 This type of razor ‘à tranchant semkirculaire et à l'anneau’ is commonly associated with the long iron swords in the Côte d'Or tumuli; five examples are recorded by Mile Henry, loc. cit., fig. 16, and table opposite p. 56.

page 376 note 2 George, J. and Chauvet, G., Cachette d'objets en bronze, découverte à Vénat, Charente, 1895, pl. XVI, 168Google Scholar. I owe this reference to Dr. H. N. Savory.

page 376 note 3 Mr. W. F. Grimes, M.A., F.S.A., late Assistant Keeper of the Department of Archaeology, was active in seeking information on the Llyn Fawr problems.

page 378 note 1 Since the cauldron was not found with the hoard, Leeds (op. cit.) did not feel able to regard the hoard as providing any help towards the dating of cauldrons in Britain, and omitted it from consideration.

page 378 note 2 Dr. Wheeler records roots of trees with axe-marks on them. One would not construct a crannog with such. In Prehistoric and Roman Wales, p. 202, he mentions ‘hewn timbers’, which may be more correct. I have seen a photograph taken when the trees referred to by Dr. Wheeler had become talked about. One or two roughly cut logs lie in front of peat dumps; they are unfortunately not in situ. They show no mortise and tenon construction (frequent in crannog structure), and no shaped ends.

page 379 note 1 Evidence suggesting that the two cauldrons came from one household has already been presented.

page 379 note 2 Two were in the Cardiff hoard, which included two Hallstatt razors, and so is strictly comparable in date to the Llyn Fawr hoard. See Nash-Williams, V. E. in Antiq. Journ. (1933, p. 299)CrossRefGoogle Scholar and fig. 7, this paper. The third comes from the Llantwit Major hoard, also in the National Museum.

page 379 note 3 Fox, Aileen, Arch. Camb., 1936, pp. 100–17Google Scholar.

page 380 note 1 Fox, Aileen, Arch. Camb., 1936, map 11, opp. p. 109Google Scholar.

page 380 note 2 The list has been published by Aileen Fox in Bull. B.C.S. viii, 280–4.

page 382 note 1 The Ely was undoubtedly the estuary chiefly favoured in Late Bronze Age times. It is a sluggish stream, up which Channel-traffic could pass at least as far as Fairwater by Llandaff; by dug-out canoe one could travel many miles farther.

page 382 note 2 Dr. A. Mahr has plotted the cauldrons on a map in his paper on a La Tène cauldron of wood in Proc. R.I.A., 1934, fig. 6, p. 26Google Scholar, but he did not comment on the British distribution.

page 383 note 1 As in fig. 47 b, Bronze Age Guide, Brit. Mus., 2nd ed., p. 57.

page 383 note 2 This was owned by a man who also possessed a rapier and palstave. Babington, Camb. Antiq. Soc. Communications, vol. iii, pp. 361–2.

page 384 note 1 Ancient Bronze Implements, fig. 236.

page 384 note 2 This essential feature of the evolutionary process was noted by Dr. Wheeler in 1920; see Arch. lxxi, 71, p. 135. For the Cardiff hoard see above, p. 379, and Guide to Pre-historic Collections, National Museum of Wales, p. 169 and fig. 66.

page 384 note 3 For illustrations of such see Coffey, Bronze Age in Ireland, fig. 76, 3; and Evans, Bronze, fig. 237. There is one sickle of this class in the English lowlands, from Icklingham, Suffolk. I think it is of Irish derivation.

page 384 note 4 A survey of the sickles of the British Isles, from which the above notes are extracted and on which the map described below is based, will shortly be published. It will include an annotated list of the 63 known specimens, and acknowledgements of my indebtedness to many helpers, particularly to Mr. E. Estyn Evans, F.S.A.

page 385 note 1 The wide extension possible in a culture diffused by sea-trade such as this is illustrated by the occurrence in the bed of the Seine at Paris of a sickle of Llyn Fawr type. It is in the Evans Collection, Ashmolean Museum.

page 385 note 2 See Evans, Bronze Implements, figs. 133–5, and compare examples from East Anglia in British Museum (e.g. Quy Fen, Fen Ditton) and Cambridge Museum (e.g. Hockwold and Reepham). There are, however, in addition to Llyn Fawr, six of these axes in the Bristol Channel region: (a) Cardiff hoard, one; (b) R. Avon, Bristol Docks, two; (c) Over, near Gloucester, one; (d) Crumlin, Mon., one. (a) and (d) in National Museum of Wales; (b) and (c) see Cat. Br. Imp., in Brit. Mus. It should be added that examples comparable to the larger specimens in our series are known from Brittany; but the differences, though slight, confirm these axes as of insular origin. A survey of this axe-type, by Miss L. F. Chitty, F.S.A., and myself is in preparation.

page 386 note 1 See Dr. Wheeler's description (loc. cit., p. 136, and fig. 2), and cf. E. C. Curwen's comment, Proc. Prehist. Soc., 1938, p. 41Google Scholar. I have no observations to make on the iron spear, save that a comparable example from London has been studied since Dr. Wheeler wrote his paper, Antiq. Journ., 1929, pp. 376–7Google Scholar.

page 387 note 1 The limits suggested by the author are 700–400 B.C. Arch. Journ. lxxxv, p. 173.

page 387 note 2 Cf. Special Reports on the Mineral Resources of Great Britain, vol. x. Iron ores—the haematites of the Forest of Dean and South Wales. Geological Survey, 1927. Dr. North (to whom I am much indebted) regards the South Wales orefield (Glamorgan, the Llanharran-Miskin area) as less obvious than that of the Forest of Dean; it is also less extensive. Leeds's comments, Arch. lxxvi (1927), 237Google Scholar, on the probable importance of the Forest of Dean orefield in the third century B.C. are of interest in this connexion. We have also Professor Gowland's opinion (Proc. Soc. Ant. xx, 1903–5, p. 194Google Scholar) that a currency bar submitted to him was probably made of Forest of Dean ore. Both orefields were known to the Romans.

page 389 note 1 Archaeology in England and Wales, T. D. Kendrick and C. F. C. Hawkes, p. 138.

page 389 note 2 Arch. Camb., 1931, pp. 354–5Google Scholar.

page 389 note 3 H. O'Neill Hencken describes the Irish gold ornaments of late type found in Cornwall, and mentions (Cornwall and Scilly, p. 165) a bronze vessel found in a tin stream in Luxulyan parish in 1792, which he considers to have been a cauldron of Irish type. This illustrates the pull which Cornwall may have exerted on the Irish trade.

page 390 note 1 See Cyril Fox, ‘On two Bronze Age Cairns in South Wales’, Arch. lxxxvii, p. 150, and ‘Sleds, Carts and Waggons’, Antiquity, 1931, p. 192Google Scholar.

page 390 note 2 Evans, Bronze, p. 119 and fig. 126 (from Mynydd-y-glew, Glam.). Dr. R. E. M. Wheeler, Prehistoric and Roman Wales, p. 156, discussed the type and its distribution. There is a large number of illustrations of the type in the Guide to the Prehistoric Collections of the National Museum. Of 51 axes from South Wales in the museum 35 are ‘Welsh’. H. O'Neill Hencken (Cornwall and Scilly, p. 92) follows Evans in describing Cornish examples as the ‘Welsh’ type.

page 390 note 3 On the periphery of the area of distribution, links with other types are met with—the Stogursey and Bourton-on-the-Water hoards provide instances of this.

page 390 note 4 Jacquetta Hawkes, Archaeology of the Channel Islands, vol. ii, Jersey, pl. VIII, second from left (Mainlands hoard).

page 392 note 1 Dr. North and the writer were enabled to visit and examine this site by the courtesy of the Glamorgan County Council.

page 393 note 1 I am indebted to the Glamorgan County Surveyor for information embodied in this map.

page 394 note 1 I wish to record with gratitude the assistance which I received throughout the investigation from Mr. J. W. Davies, who accompanied me in the field and helped with the borings, carried out the routine preparation of the peat samples for microscopic examination, and prepared the various diagrams.

page 394 note 2 The degree of humification is expressed in terms of Von Post's scale, in which H. 1 stands for entirely fresh and H. 10 for completely humified peat. L. Von Post, Das genetische System der organogenen Bildungen Schwedens, Comité internat. de Pédologie, Commission IV, no. 22, 1924.

page 397 note 1 Godwin, H., and Mitchell, G. F. (1938), ‘Stratigraphy and development of two raised bogs near Tregaron, Cardiganshire’, New Phytologist, xxxvii, 425CrossRefGoogle Scholar.

page 399 note 1 Pearsall, W. H. (1931), ‘The Aquatic Vegetation of the English Lakes’, Journal of Ecology, viii, 163Google Scholar.

page 401 note 1 Granlund, E. (1932), ‘De svenske Högmossernas Geologi’, Sveriges Geolog. Undersök. Årsbok, xxvi (German summary)Google Scholar.

page 401 note 2 Jessen, K. (1934), ‘Archaeological datings in the history of North Jutland's vegetation’, Acta Archaeologica, v (København)Google Scholar.

page 401 note 3 Jessen, K. (1937), ‘Preliminary Scheme of Post Glacial Development in Ireland’, in Dr. Mahr's Presidential Address, Proc. Prehist. Soc., N.S. iiiGoogle Scholar.

page 403 note 1 Prepared by my wife and myself. Dr. Wheeler's list has been drawn upon (Prehist. and Rom. Wales, p. 156). Miss L. F. Chitty, F.S.A., has kindly supplied the following additional records (not on the map): Brecknock: Crickadarn, and the Walton, Brecon, both Brecon Mus. Bedfordshire: Eaton Ford, V.C.H. Hunts, i, pl. 11, 9.