^{page 170 note 1} The Transepted Gallery Graves of Western France, P.P.S., 1939, p. 143 fGoogle Scholar.

^{page 170 note 2} P.P.S., 1937, p. 86.

^{page 171 note 1} Plan in le Rouzic, , l'Anth., 43, 1933, p. 240, fig. 11Google Scholar.

^{page 171 note 2} e.g., La Varde, Guernsey and Les Monts Grantez, Jersey (Kendrick, and Hawkes, , Arch, of Channel Islands, vol. 1, p. 189Google Scholar, fig. 86 and 2, p. 208, fig. 54), and in Ireland, Dowth; (Powell, , P.P.S., 4, 1938, p. 291, fig. 2, 18Google Scholar), and in Caithness, Kenny, Cairn, Childe, , Prehistory of Scotland, p. 37, fig. 8Google Scholar.

^{page 171 note 3} In Jersey e.g., Hougue, La Bie (Arch. Channel Islands, vol. 2, p. 197, fig. 51)Google Scholar, in Ireland Carrowkeel Carn K and Lochcrew Cam T (Powell, loc. cit.).

^{page 171 note 4} e.g., in Iberia the Dolmen de Soto, Huelva (Obermaier, , B.S. Esp. Exc., 32, p. 1Google Scholar), Cueva de Menga, Antequera (Hemp, , Ant. J., 14, 1934, p. 405Google Scholar), Los Millares, Tomb 8 (Siret, , Rev. Quest. Sci., 34, 1893, p. 522 f.Google Scholar) and Cova d'en Daina, Catalonia (Pericot, , Civ. Megalitica Catalonia, Barcelona, 1925, p. 24Google Scholar); in S. France St. Eugenie, Aude (Sicard, , B.S.P.F., 1930, p. 536Google Scholar); in Brittany Gavr'inis and Mané Kerioned 1 and 111; in the Channel Islands le Trépied and La Varde, in Guernsey (Kendrick, , Arch. Chann. Is., vol. 1, pp. 189 and 104Google Scholar) the latter equipped with one secondary chamber lateral to the main chamber; in Jersey Mont Ubé, Les Monts Grantez, with one side chamber, and La Hougue Bie, with 3 subsidiary chambers arranged in a cruciform manner (Hawkes, , Arch. Chann. Is., vol. 2, pp. 214, 208 and 197Google Scholar).

^{page 171 note 5} Proc. First Int. Cong. Prehist. Sciences, 1932, p. 115Google Scholar.

^{page 171 note 6} There is a tomb intermediate between the normal ‘passage grave’ and the ‘angled gallery’ at Kergonfals, Bignan, in which a square chamber(2.6 m.) with a sill stone at the entrance is approached by an angled passage (6.15 m. long).

^{page 172 note 1} l'Anth., 43, 1933, p. 242Google Scholar.

^{page 172 note 2} P.P.S., 1939, p. 143, note 2.

^{page 172 note 3} For recent plans on a uniform scale see Powell, , P.P.S., 4, 1938, p. 241, fig. 2, 19Google Scholar.

^{page 172 note 4} The plans of the central and western tombs in La Haute Folie are those of monuments too ruinous to determine their original form satisfactorily, but there is some indication that the western tomb was also of this form. The plan of the eastern tomb in Les Mousseaux is apparently a bisection of that of the western whereby the end chamber extends asymmetrically only to the east of the passage and the one lateral chamber off the passage lies on the east side.

^{page 172 note 5} Mat., 3e ser. 3, 1886, 278–9 (my italics)Google Scholar.

^{page 173 note 1} Plans of two undifferentiated passage graves by de Wistnes are reproduced by Dr Daniel and the tomb described by de Lisle in 1891 is indicated to be a passage grave probably with a lateral chamber. Bull. Arch., 1891, p. 36 fGoogle Scholar.

^{page 173 note 2} Viz., Dolmen de la Barbière, 1 km. from Crossac, arrondissement of St. Nazaire, described as a ruined gallery with a square chamber to the side. The gallery oriented E—W was 12 m. long and the chamber 2 m. 70 × 2 m. 10; Donges, arrondissement of St. Nazaire, ‘allée couverte’ with lateral chambers (de Lisle, P., Bull. Soc. Arch. Nantes, 1882, XXI, pp. 123–4 and 137 f.Google Scholar)

^{page 173 note 3} He records finding in the north-east chamber ‘fragments d'une urne en forme de calice en terre rouge’ (Mat., 1886, 3e ser. 3, p. 284). This is the term he also uses to distinguish beakers from what we now know as western neolithic ware in a later passage of the same paper quoted by DrDaniel, , (P.P.S., 5, 1939, p. 159)Google Scholar where he seeks to contrast the wares found on the opposite sides of the river Loire.

^{page 173 note 4} He notes (Bull. Arch., 1891, 37–8Google Scholar) that it ‘differs in the irreproachable perfection of its contour and the solidity of the ware, being so regularly made that it seems to have been made with the aid of a wheel.’ These qualities would well apply to the finest Atlantic beakers. The scattered gold beads and the callais bead (not a necklace) from this tomb are also types found in Morbihan megalithic tombs. The long-tanged flint arrowhead and the bronze arrowhead with a quadrangular tang also found suggest that burials were still being made in this tomb in the Breton Bronze Age.

^{page 173 note 5} See especially, Childe, , ‘The Continental Affinities of British Neolithic Pottery,’ Arch. J., 83, 1932Google Scholar. Hawkes, J., Arch. Chann. Is., 2, pp. 70–87Google Scholar, and eadem, ‘The Significance of Channelled Ware,’ Arch. J., 95, 1939, p. 126 fGoogle Scholar.

^{page 173 note 6} P.P.S., 1939, p. 159.

^{page 174 note 1} Although there is difference of opinion between different observers in the case of Er Ro'h the one tomb that conforms to the ‘transepted gallery’ definition.

^{page 174 note 2} The Paris basin long cists from which the latter are probably derived in many instances preserve the passage derived from their own passage grave ancestry in the form of a short and slightly narrower ‘vestibule,’ cf. e.g., the well known La Justice tomb and Preslés, , Rev. Arch., 28, 1928, p. 1 fGoogle Scholar. For evidence in support of this derivation see Forde, , ‘The Megalithic Gallery in Brittany,’ Man, 29, 1929, p. 80CrossRefGoogle Scholar.

^{page 174 note 3} Daniel, , P.P.S., 1939, p. 161Google Scholar; it should be noted that for the parenthetically alleged spread of the gallery grave from the Loire to the Paris Basin no evidence at all is presented.

^{page 174 note 4} The fact that the adjacent mutilated Tomb 6, now at least, shows no evidence of a passage is not likely to be taken as evidence that it belongs to a distinct class of ‘gallery grave.’

^{page 175 note 1} The evidence for any barrowlike superstructure associated with the Mallorcan rock cut tombs is still slight. We have only remnants of what may have been a sub-rectangular reveted mound in the case of one tomb. Concerning the morphologically closely similar group of tombs on the mainland near Aries, on the other hand, there is evidence, more definite and abundant, of a circular superstructure, although whether this was a barrow or a stone ringed platform remains uncertain. This is probably yet-another instance of the marked instability of barrow form associated with collective tombs which Dr Daniel fails to take sufficiently into account. This instability is abundantly illustrated in Britain by the segmented cists of S.W. Scotland and the passage graves of N. Scotland where long and round, horned and unhorned cairns cover tombs of dated forms. The Severn–Cotswold tombs are indeed probably exceptional in the stability of the barrow form.

^{page 175 note 2} Even de Fondouce's inadequate diagram showed that in addition to the rock cut ramp there is a covered passage 18 feet 6 inches long markedly lower and narrower than the chamber to which it leads. The lateral embayments form part of the forecourt at the foot of the ramp, and not of the passage or chamber and may not have been roofed. If, however, as is consistent with the view referred to above, the ramp is here also regarded as equivalent to a passage we have a duplex approach element with lateral chambers off the outer section.

^{page 175 note 3} Salmon, P., Dolmen avec tumulus et Cromlech à Kerlescan, Paris, 1887, pp. 8–10Google Scholar and figs. 1 and 3 and le Rouzic, , l'Anth., 43, 1933, p. 245, fig. 15Google Scholar.

^{page 175 note 4} Pericot, , La Civilization Megalitica Catalana, Barcelona, 1925, p. 26Google Scholar.

^{page 175 note 5} Hemp, , Ant. J., 14, 1934, p. 277 f.CrossRefGoogle Scholar: Goby, , Cong. Cannes-Grasse, 1929Google Scholar; Moreover the thorough investigation of the superstructures of Iberian tombs has yet to be undertaken. The elongation of the barrow covering some of the large megalithic tombs in N.W. France does not, however, necessarily imply external stimulus particularly if the long, reveted cist mounds in Morbihan to which Piggott has called attention are accepted as starting earlier than the period of the large collective tombs. In this connection it should be recalled that the circular structure in the Mané Groh barrow, which is compared with Notgrove, has its own likely predecessors in the Crucuny cist barrows (Piggott, , Ant., II, 1937, 441 f. and fig. 3Google Scholar). But competent excavation is needed to investigate the character of constructional features such as revetment before the genuinely elongated structure of the chambered barrows in question can be regarded as well established. In their present denuded condition their shapes are difficult to determine and may not closely reflect the original form.

^{page 175 note 6} This degeneration process is also demonstrable in the Carrowmore and Inverness-shire groups of chambered tombs.

^{page 176 note 1} I would agree here with DrMahr, (P.P.S., 1937, 352)Google Scholar, considering that the balance of evidence favours the view that such resemblances as exist between the Ulster–Scottish long cists, the Sardinian Giants' Graves and the segmented long cists on both sides of the Pyrenees are due to parallel degeneration whereby the terminal chamber or its distinction from the passage has been lost, while the other Iberian features, segmentation and the forecourt, have been retained in some groups. The segmentation so striking in the British long cists and at La Halliade is, it must be remembered, a well established and widespread feature of passage graves occurring in such tombs in Almeria, Algarve, Estremadura, Ireland, the Hebrides and Denmark.