^{page 352 note 1} Grateful thanks to Messrs J. Drayson and K. Hoy, to the Henry Harbin School and Poole Grammar School Archaeological Societies, whose efforts were so freely given in a race against time; and to the Whiffen family, tenants of Lake Farm, for their co-operation.

^{page 352 note 2} Cf. Dr Dimbleby's note on the charcoal from the site, below.

^{page 352 note 3} Proc. Dorset Nat. Hist. & Arch. Soc., vol. 60 (1938), pp. 73–4Google Scholar.

^{page 354 note 1} The Roman discoveries, necessarily shown in the illustrations, will be reported in the Proceedings of the Dorset Natural History and Archaeological Society.

^{page 355 note 1} Comparison might be made with pits sunk through irregular shallow depressions at Hazard Hill in Devon (Devon Arch. Explor. Soc. Trans., vol. 21 (1963), p. 12Google Scholar).

^{page 358 note 1} Neolithic Cultures of the British Isles (1954), p. 67Google Scholar.

^{page 358 note 2} Count based on published illustrations in Proc. Devon Arch. Explor. Soc., 1930, 1931, 1932, 1935, and in Wheeler, ‘Maiden Castle, Dorset’ (Reports of Research Committee, Soc. Ant. Lond., XII (1943))Google Scholar.

^{page 358 note 3} As pointed out in the original report.

^{page 358 note 4} Windmill Hill and Avebury: Excavations by Alexander Keiller, 1925–1939, ed. Smith, I. F. (1964)Google Scholar, Chap. IV. The pre-enclosure pottery and much of that from the ditches now seems best classed as Hembury ware.

^{page 358 note 5} 3230 ± 150 B.C. (BM–134); Antiquity, XXXVIII (1964), pp. 139–40Google Scholar.

^{page 358 note 6} Three determinations consistently place the Neolithic occupation at Hembury before 3000 B.C. and probably within the period 3180–3300 B.C. Antiquity, XXXVII (1963), pp. 228–9Google Scholar.

^{page 359 note 1} Windmill Hill and Avebury, fig. 37.

^{page 359 note 2} Oxoniensia, XVII/XVIII (1952–1953), p. 10Google Scholar.

^{page 359 note 3} For full details, see Windmill Hill and Avebury, pp. 89–91 and fig. 38.

^{page 360 note 1} See also p. 364 for data on a similar Late Neolithic industry from Puddlehill, Dunstable, Beds.

^{page 363 note 1} Cf. Piggott, , The Neolithic Cultures of the British Isles (1954), fig. 57Google Scholar; PPS, II (1936), p. 195 and pl. xlGoogle Scholar: 13, 14.

^{page 363 note 2} For comparison, see Windmill Hill and Avebury, fig. 41, and PPS, XXVI (1960), p. 219, fig. 12Google Scholar.

^{page 363 note 3} Palaeohistoria, V (1956), pp. 1–38Google Scholar.

^{page 363 note 4} Windmill Hill and Avebury, pp. 95, 241 and fig. 82.

^{page 365 note 1} See Ewbank, Phillipson and Whitehouse with note by E. S. Higgs in the current volume, p. 423.

^{page 366 note 1} Proc. Camb. Ant. Soc., LIV (1961), p. 9Google Scholar.

^{page 366 note 2} Clark, J. G. D., Excavations at Star Carr (1954). p. 89Google Scholar.

^{page 366 note 3} Probably occupied in the 7th century A.D.

^{page 367 note 1} PPS, XXVI (1960), pp. 202–45Google Scholar.

^{page 367 note 2} ibid., p. 211.

^{page 367 note 3} No attempt is made to discuss pits in Ireland, for though they are fairly numerous the context of many is far from clear. Straightforward pits of the type under discussion occurred amongst the Neolithic houses at Lough Gur (Proc. Roy. Irish Acad., vol. 56 (1954), pp. 297 ff.Google Scholar), but others are associated in an ambiguous way with ring-barrows (for summary, see PPS, XXVII (1961), p. 207Google Scholar) or with other sites of possibly ritual character (ibid., p. 202; Ulster Journ. Arch., vol. 26 (1963), pp. 43 ffGoogle Scholar.).

^{page 369 note 1} Only a few reached a depth of 18 inches; only 35 per cent were over 2 feet in diameter and none exceeded 4 feet. PPS, XXVI (1960), pp. 206–7Google Scholar.

^{page 369 note 2} e.g. at Hazard Hill and Hembury Fort in Devon, on lava and shale and on clay-with-flints over greensand respectively. See register for details of dimensions.

^{page 369 note 3} Nearly a third of the sites listed, including some of those with numerous pits, are inadequately documented

^{page 369 note 4} Except in the case of depressions described as irregular or saucer-shaped and shallow, or of obvious post-sockets.

^{page 369 note 5} Devon Arch. Explor. Soc. Trans., vol. 21 (1963), pp. 16–18Google Scholar.

^{page 370 note 1} But certain pits which the late Dr Stone thought to have been used for the disposal of ‘ritual’ rubbish (Wilts. Arch. Mag., LII (1948), pp. 287–306Google Scholar) have not. When the evidence is reviewed as a whole these are not found to possess sufficiently distinctive characteristics to warrant exclusion.

^{page 370 note 2} e.g. at Woodbury, Little (PPS, VI (1940), pp. 52–4Google Scholar).

^{page 370 note 3} Figs. 1 and 2 in Fox's, The Personality of Britain (1947)Google Scholar are still the most convenient for purposes of comparison. More recent compilations indicate, in round numbers, over 100 chambered tombs in Cornwall and Wales and in Scotland about 350 in the north and east alone; Beaker graves and finds in the area of eastern Scotland to the south of the Moray Firth alone number over 200.

^{page 370 note 4} Cf. the Monamore cairn on Arran, Antiquity, XXXVIII (1964), p. 52Google Scholar.

^{page 370 note 5} Fox, loc. cit., Map B.

^{page 370 note 6} Information by kindness of Mr I. H. Longworth.

^{page 370 note 7} Scotstarvit, Fife (Proc. Soc. Ant. Scot., LXXXII (1947–1948), p. 225)Google Scholar; Clegyr Boia, Pemb. (Arch. Camb., CII (1953), pp. 20–47)Google Scholar; Nottage, Glam. (Savory, Exc. of a Neolithic Dwelling … at Mount Pleasant Farm, Nottage' Glam., Nat. Mus. Wales, 1955Google Scholar); and Ronaldsway, I.o.M. (PPS, XIII (1947), p. 143Google Scholar), where a pit against the house wall is fairly convincing, but may equally well have been a large post-socket.

^{page 371 note 1} Muirkirk, Ayr. (Proc. Soc. Ant. Scot., XLVIII (1913–1914), pp. 373 ff.Google Scholar; LIV (1919–20), pp. 210 ff.; LXI (1926–27), pp. 269 ff.) and Castell Bryn-Gwyn, Anglesey (Arch. Camb. (1962), pp. 25 ffGoogle Scholar.), where the nature of the sites is open to more than one interpretation; on the latter see also comments by Grimes, in Culture and Environment, ed. Foster, and Alcock, (1963), p. 147Google Scholar.

^{page 371 note 2} The radiocarbon dates for Hembury indicate ‘Western’ Neolithic settlement there before 3000 B.C. (Antiquity, XXXVII (1963), pp. 228–9Google Scholar); some of the Late Neolithic and Beaker sites shown on the map are likely to represent occupation around 1600 B.C.

^{page 371 note 3} ‘Western’ Neolithic sites, 24; Peterborough, 16; Rinyo-Clacton, 18; Beaker, 18.

^{page 371 note 4} The Personality of Britain, p. 53.

^{page 371 note 5} In Roman and Native in North Britain, ed. Richmond, I. A. (1958), Map 2Google Scholar.

^{page 371 note 6} As demonstrated, for example, by grain impressions in pottery from Unstan and Eday in the Orkneys (Jessen, and Helbaek, , ‘Cereals in Gt. Britain and Ireland in Prehistoric and Early Historic Times’, Kong. Danske Videns. Selskab, Biol. Skrift., III (1944), p. 18Google Scholar.

^{page 372 note 1} Judging from the paucity of information given by Piggott, in The Neolithic Cultures of the British Isles (1954)Google Scholar, where only one find is noted.

^{page 372 note 2} Cf. Grimes, , The Prehistory of Wales (Nat. Mus. Wales, 1951), p. 44Google Scholar.

^{page 372 note 3} PPS, XVIII (1952), pp. 224–6Google Scholar; XXVI (1960), p. 213.

^{page 372 note 4} e.g. on ‘Western’ Neolithic sites at Hazard Hill and Hembury, Devon; Maiden Castle, Dorset; Wingham, Kent; Hurst Fen, Suffolk; New Barn Down, Sussex; Windmill Hill, Wilts.; and on Beaker sites at Swarkeston, Derbyshire, and Itford Hill, Sussex. See register for references.

^{page 372 note 5} Where sites are mentioned without reference to publications, full details will be found in the register.

^{page 372 note 6} Devon. Arch. Explor. Soc. Trans., vol. 21 (1963), pp. 15–16Google Scholar.

^{page 372 note 7} At the rate of one bushel to 1848 cubic inches. The largest pit had a mean diameter of 4½ feet and a depth in the subsoil of 1 foot 9 inches.

^{page 372 note 8} In the past these large pits were sometimes interpreted as ‘pit-dwellings’ or ‘hut-basements’.

^{page 372 note 9} PPS, XX (1954), p. 112, note 8Google Scholar.

^{page 372 note 10} Chippenham, Cambs.; Hembury, Devon; Dovercourt, Essex; Heathrow, Middlesex; also a giant beaker from a pit at Peterborough, Northants.

^{page 373 note 1} Antiquity, X (1936), pp. 25–36Google Scholar.

^{page 373 note 2} PPS, XVIII (1952), pp. 207–8Google Scholar.

^{page 373 note 3} Lady (Aileen) Fox, who was consulted on this occasion, did in fact emphasize this and has since assured the writer that there can be no question of Iron Age contamination.

^{page 373 note 4} PPS, XVIII (1952), pp. 209, 233Google Scholar.

^{page 373 note 5} Only emmer and barley were in fact identified amongst the impressions on sherds from Hembury. Some of the charred grains from Pit Q, now preserved separately in the museum at Exeter, were found adhering to the inner surface of a large Neolithic sherd.

^{page 374 note 1} PPS, XXVI (1960), pp. 17–22, 34Google Scholar.

^{page 374 note 2} Only those proved by excavation are shown on the map. They include the recently discovered site at Staines in Middlesex (Archaeol. News Letter, vol. 7 (1962), pp. 131–4Google Scholar) and the earlier of the two earthworks at Rybury in Wiltshire, where a recent test excavation by Mr D. J. Bonney seems to have produced the requisite evidence (Wilts. Arch. Mag., vol. 59 (1964)—forthcoming)Google Scholar.

^{page 374 note 3} Two pits near the enclosure on Knap Hill, Wilts., have accidentally been omitted from the map and register. Both contained ‘Western’ pottery. Wilts. Arch. Mag., XXXVII (1911), p. 53Google Scholar.

^{page 374 note 4} With the possible exception of a likely-looking impression on a Peterborough bowl from the Thames in Reading Museum.

^{page 374 note 5} Peterborough ware is known from a number of sites in Wales and as far north as southern Scotland; Rinyo-Clacton ware from sites in western and northern Scotland.

^{page 374 note 6} Grain impressions occur so infrequently in Early Iron Age pottery in the south of England that if it were necessary to rely entirely on this kind of evidence quite erroneous inferences could be drawn. Cf. PPS, XVIII (1952), p. 207Google Scholar.

^{page 374 note 7} Antiquity, XXXVII (1963), pp. 274–81Google Scholar.

^{page 375 note 1} By Mr Emery and Dr D. R. Brothwell, both of whom are engaged in studies of this question, and with whom the subject of this note has been discussed.