The main source of the Lea is at Leagrave Marsh, about 3 miles N.W. of Luton, and 376 feet above Ordnance datum. On the surface near this place, Mr. Smith[1] found a flat ovate implement, in form much like those from Warren Hill or that from near Dunstable, Fig. 17. He says that it may be neolithic, but that he has found palæolithic flakes, both ochreous and grey, in situ in gravel at Leagrave. At Houghton Regis,[2] 112 miles north of Dunstable, Mr. Smith found a fragment of an ovate implement on the surface. Another implement, found so long ago as 1830 by Mr. William Gutteridge, at Dallow,[3] or Dollar farm, 34 of a mile west of Luton, is distinctly palæolithic in form.
The most interesting of Mr. Worthington Smith's discoveries have, however, been made on or near the summit of a hill, a good 2 miles from the Lea, and somewhat nearer the Ver. At and around the village of Caddington there are several brickfields, some of them no longer worked. The original surface of the ground in some of these is as much as 550[4] to 595 feet above the Ordnance Datum. The brick-earth is of great thickness, in places fully 50 feet, and overlies the Chalk. The upper portion of the beds is much contorted, and has in it occasional seams of flint gravel or tenacious clay, in which cream-coloured or brownish palæolithic implements occur. In the gravel, brown, ochreous, slightly abraded implements and flakes are found, and at the base in many cases is the old land-surface or "Palæolithic floor" resting on and surmounted by brick-earth. In one pit were three heaps of flints brought by hand in Palæolithic times from flint-bearing beds either above or in the Chalk. On the Palæolithic floor were numerous sharp-edged flakes, which had hardly been moved from the original place at which they were struck off. Mr. Smith has replaced more than 500 flakes either on to other flakes or on to implements and cores from the same floor.
One old land-surface was full of narrow vertical fissures, due perhaps to the heat of a burning summer sun. While they were still open 18 inches of watery brick-earth, perhaps brought down by a heavy storm of rain, filled up the fissures, covered up the old surface and formed a new surface at a higher level. The upper deposits often resemble contorted masses of half-frozen mud and stone pushed over an old water-laid and perhaps frozen surface of brick-earth. Mr. Smith's view is that Palæolithic
