^{page 71 note 1} Variously spelt as Parc le Breos or Parc-le-Breos and pronounced locally as ‘Bruce’. Older spellings give Parke le Bruce, Parke Price and Parke Brewis; the original form was Braose. William de Braose was granted the Lordship of Gower in the thirteenth century.

^{page 71 note 2} Ordnance Survey, Southampton, 1936.

^{page 71 note 3} Proc. Preh. Soc., 1936, pp. 106 ffGoogle Scholar.

^{page 73 note 1} Journal of the Ethnological Society, New Series, Vol. 11, no. 4, Jan., 1871, p. 416 ffGoogle Scholar. (Abbreviated below to J.E.S., 1871).

^{page 73 note 2} Arch. Comb., 1871, 168–172 Google Scholar.

^{page 73 note 3} id., 1887, 192–201.

^{page 73 note 4} Brief accounts of these visits are printed in id., 1886, 344; id., 1920, 296–299.

^{page 73 note 5} id., 1936, 321 ff.; ibid., Western Mail, March 12, 1937, p. 8.

^{page 73 note 6} e.g. by Fergusson, , Rude Stone Monuments, 164 Google Scholar. Redrawn plans are given by Wheeler, , Prehistoric and Roman Wales, 71 Google Scholar; and by Piggott, , Proc. Preh. Soc., 1935, 118 Google Scholar.

^{page 73 note 7} Dr Arbour Stephens gives the azimuth of the central gallery as s.23°w. ( Arch. Camb., 1920, 297)Google Scholar.

^{page 73 note 8} The term ‘gallery-grave’ is here used to describe long rectangular chambers in preference to such other English equivalents of allée couverte as long cist, covered gallery and entrance grave.

^{page 73 note 9} A thick deposit of leaf mould and earth within the chambers makes the recognition of low features such as foot-courses and sills very difficult at present.

^{page 75 note 1} Arch. Camb., 1886, 344 Google Scholar.

^{page 75 note 2} id., 1887, 197.

^{page 75 note 3} J.E.S., 1871, 417.

^{page 75 note 4} Arch. Camb., 1887, 197 Google Scholar.

^{page 76 note 1} Wheeler, (Prehistoric and Roman Wales, 76)Google Scholar describes these as ‘the fragmentary remains of at least four individuals.’

^{page 76 note 2} Arch. Camb., 1887, 198 Google Scholar.

^{page 76 note 3} Note that my side-chamber A is Avebury's side-chamber 1 and Morton Douglas's case no. 1; my B-Avebury's 4, Douglas's 5; my C, Avebury's 2, Douglas's 2; my D, Avebury's 3, Douglas's 4; while Douglas's case no. 3 is the central gallery.

^{page 76 note 4} Swansea, writes (Arch. Camb., 1887, 198)Google Scholar, ‘I caused the remainder of the bones, after taking such portions as were required for scientific purposes, to be re-buried, each set in their former resting place, enclosed in the fire-clay retorts which we use for the manufacture of spelter.’! The bones that were not re-buried have recently been re-discovered by Rix in the University Museum at Oxford.

^{page 76 note 5} The Ashmolean Reference number is 1486–1886. My thanks are due to Mr Piggott for drawing my attention to these sherds, to Mr Harden for his kindness in showing them to me, and to the Curator of the Ashmolean for his courtesy in allowing me to publish them. They have recently been figured by Rix, in Arch. Camb., 1936, 322 Google Scholar.

^{page 76 note 6} Arch. Camb., 1886, 344 Google Scholar.

^{page 76 note 7} loc. cit. supra.

^{page 77 note 1} Sernander, V., Geol. Fören. Stockh. Förk., XXX, 1908, 465 ffCrossRefGoogle Scholar; von Post, , ‘Die postarktische Geschichte der Europäischen Walder’, Congr. Intern. Inst. des Recherches Forest., Stockholm, 1929 Google Scholar.

^{page 77 note 2} Grave-mounds and their contents, 2.

^{page 77 note 3} Antiquity, 1930, 89 Google Scholar.

^{page 77 note 4} Arch. Camb., 1916, 101 ffGoogle Scholar.

^{page 77 note 5} Proc. Prehist. Soc., 1936, 5 Google Scholar.

^{page 77 note 6} Ordnance Survey Professional Papers, New Series, no. 6, 3.

^{page 78 note 1} On lowland long barrows Crawford, v., Antiquity, 1930, 357–8Google Scholar; on lowland round barrows v. ib., id., 1933, 290 ff; Leeds, , Ant. J., 1934, 414–6Google Scholar; Phillips, C. W., Proc. Preh. Soc., 156–7Google Scholar; J. G. D. Clark, id., 1936, 2–5. In South Wales three burial chambers—Carreg Coitan Arthur at Newport (Pembs.); Broomhill Burrows, Angle (Pembs); Heston Brake, Portskewett (Monmouth)—as well as Parc le Breos Cwm, are below 100 feet. Clark has rightly stressed ( Proc. Preh. Soc., 1936, 5)Google Scholar the part played by air-photography in the recognition of these lowland sites.

^{page 78 note 2} Antiquity, 1930, 92 Google Scholar.

^{page 78 note 3} Antiquity, 1930, 357 Google Scholar.

^{page 78 note 4} Proc. Preh. Soc., 1936, 5 Google Scholar.

^{page 78 note 5} Antiquity, 1930, 92 Google Scholar.

^{page 78 note 6} Proc. Preh. Soc., 1936, 114 Google Scholar. See also his introduction (p. 6) to the Map of South Wales showing the distribution of long barrows and megaliths.

^{page 78 note 7} Arch. Camb., 1936, 259 ffGoogle Scholar.

^{page 78 note 8} id., 267.

^{page 79 note 1} See the admirable article by Grimes, W. F. on ‘Two Bronze Age Burials from Wales’, Arch. Comb., 1936, 293 ffGoogle Scholar.

^{page 79 note 2} Antiquity, 1935, 429 Google Scholar. He also stresses the need for the objectivity of the environmental base-maps.

^{page 79 note 3} And too often, alas, as the spread of megalithic civilisation.

^{page 79 note 4} Excepting of course the parallelism produced by convergent development. For a discussion of these points see Childe, , Trans. Glasgow Arch. Soc., New Series, Vol. 8, III, 121 Google Scholar; id., Prehistory of Scotland, 24.

^{page 79 note 5} ‘The “Dolmens” of Southern Britain’, (Antiquity, June, 1937)Google Scholar, passim.

^{page 79 note 6} For details of the Scilly and Penwith group see the work of Hencken; The Bronze and Early Iron Ages in Cornwall and Devon (MS 1929), and The Archaeology of Cornwall and Scilly, 1932. The work of Thurnam, Crawford and Mrs Cunnington on the Cotswold-Severn group is well known. On the Medway group see Crawford, Ordnance Survey Prof. Paper, New Series, no. 8, and Jessup, Archaeology of Kent. On the South Welsh material see Wheeler, , Prehistoric and Roman Wales and Grimes, Proc Preh. Soc., 1936, 106 ffGoogle Scholar. Grimes's important recognition of two groups among the chambers tomb of South Wales is first made in P.P.S.E.A., 1932, 89 ffGoogle Scholar. I have argued elsewhere ( Proc. Preh. Soc., 1936, 259 Google Scholar; Antiquity, June, 1937, op. cit.) the dual nature of his Western group.

^{page 80 note 1} An ‘open’ chamber as opposed to a closed one is one in which access may be obtained to the chamber without disturbing the primary construction of the barrow.

^{page 80 note 2} Archaeology of Cornwall and Scilly, 20 ff.

^{page 80 note 3} The term passage-grave is used throughout this discussion for burial chambers morphologically like lie Longue or Cunha Baixa or Bryn Celli Ddu.

^{page 80 note 4} Prehistoric and Roman Wales, 86.

^{page 81 note 1} Although Crawford, using the word passage-grave in a wider sense than that in which it is here used, has described as passage-graves tombs such as Nymphsfield and Uley which, as has been pointed out ( Handbook to the Prehistoric Archaeology of Britain, 1932, 25)Google Scholar, are more properly described as gallery graves with paired sidechambers.

^{page 81 note 2} J.E.S., 1871, 418.

^{page 81 note 3} British Barrows, 450–1.

^{page 81 note 4} P.P.S.E.A., 1932, 90 Google Scholar.

^{page 81 note 5} id.

^{page 81 note 6} Proc. Preh. Soc., 1936, 128 Google Scholar.

^{page 81 note 7} id.

^{page 81 note 8} And perhaps, orthostatically walled forecourts also. There is, I think, a trace of one of these at Trefignath.

^{page 82 note 1} And perhaps Maen Pebyll also, although it seems to me unlikely that this ever had a chamber. Grimes, (Proc. Preh. Soc., 1936, 106 ffGoogle Scholar, passim) would connect Capel Garmon and Maen Pebyll with the second sub-group of the Cotswold-Severn Group, but this seems hardly justified.

^{page 82 note 2} But again the North Irish analogies of the semi-circular forecourt might be used to justify the postulation of long barrows for some of the Dyfed chambers. This postulate would however be difficult to defend; it would involve the assumption of a most peculiar differential destruction of barrows—one which left long barrows in the Cotswolds but none in Pembrokeshire, while it left some round barrows covering chambers in the latter area.

^{page 82 note 3} Proc. Preh. Soc., 1935, 115 ffGoogle Scholar.

^{page 82 note 4} It should however be said in fairness that they are all based on Avebury's plan.

^{page 82 note 5} There is at present no northern chamber at Penmaen Burrows.

^{page 82 note 6} Rev. Thomas Bere's account ( Gents. Mag., 1789, LIX, 392, 602 Google Scholar) suggests there were three pairs of chambers, but Scarth's plan ( Proc. Som. Arch. Soc., VIII, ii Google Scholar, plate III) shows seven pairs as well as two entrances. I think little reliance can be placed on this plan.

^{page 83 note 1} Parc le Breos Cwm itself is 90 feet by 55 feet, Uley is 120 feet by 85 feet, Stoney Littleton 100 feet by 50 feet. Contrast Lanhill, 160 by 40; East Kennet, 350 by 100; and West Kennet, 340 by 75.

^{page 83 note 2} Mr Grimes has suggested to me that Ty Illtyd (Llanhamlach) might be the remains of another. Capel Garmon might be claimed as one although it is to me more like Carrickard.

^{page 85 note 1} Both these tombs, as pointed out supra, reveal specialised features, and are probably typologically late.

^{page 85 note 2} The following account owes much to my discussion of the Irish material with Mr T. G. E. Powell and of the Morbihan material with Miss V. C. C. Collum. I take this opportunity of expressing my indebtedness to them both.

^{page 85 note 3} Archaeological Review, II, 1889, 314 Google Scholar.

^{page 85 note 4} Ant.J., 1921, 107 Google Scholar. Fergusson, (Rude Stone Monuments, 182)Google Scholar gives a plan of this site. A more reliable plan is given by Wood Martin, Rude Stone Monuments of Ireland, Fig. 44.

^{page 85 note 5} Wood Martin, op. cit., Fig. 135.

^{page 85 note 6} Trans. Glasgow Arch. Soc., New Series, VIII, iii, 135.

^{page 85 note 7} Arch. J., 88, 55.

^{page 85 note 8} C.R. of the Congress of Prehist. and Protohist. Sciences, London, 1932, p. 115 Google Scholar.

^{page 85 note 9} For plans see Bull. Soc. Polym. du Morb., 1866 (sketch plans by de Closmadeuc).

^{page 85 note 10} Figured by Lukis, , J.B.A.A., 1866 Google Scholar, pl. 17, fig. 12.

^{page 86 note 1} See Forde, , The Megalithic Gallery in Brittany, Man, 1929 Google Scholar, no. 80.

^{page 86 note 2} Goward for instance is 115 feet by 50 feet; Ballyalton, 115 by 65; Dunloy, 76 by 48.