Retreat of the Late Pliocene and Lower Pleistocene Crag sea
clearly indicate transport by river, with subsequent reworking by currents and longshore drift. Prior
to deposition of the Westleton Beds, pebbles are a very minor constituent of the Crag and possibly
relate only to major flood events.
Within the Red Crag Formation, stone counts include quartz, reaching 3.0-7.0% between Elsenham
and Hemel Hempstead, suggesting a small but significant input from west or northwest of the London
Basin, a composition which "foreshadows that of the proto-Thames" (Green & McGregor 1999). In
the Norwich Crag Formation, the figure always reaches 3.0%, is in the region of 20% at Great
Blakenham and in the Potters Bar area (Well End and Well Wood) and reaches a maximum of 58%
at Stoke Row. near Henley-on-Thames. The ratio of quartz to other far-travelled clasts is high,
indicating the quartz is a more important external input than the other lithologies. This external input
is consistent and particularly high in the west, strongly suggesting the existence of an early Thames.
Quartz gravels are poorly represented at Easton Bavents. Possibly this indicates a contribution from
the Bytham River, which is associated more with a quartzite input. Easton Bavents lies north of the
ridge that continues the high land separating the Sudbury-Stradbroke Basin from the Ipswich Basin.
This would support the view that the early Thames was utilising the Ipswich Basin and the Bytham
River the Stradbroke -Sudbury Basin. This conclusion is equivocal as there is Greensand chert, from
the Weald, at Easton Bavents, suggesting a Thames/Medway influence.
Sinclair (pers, comm.) gives an indication of the difficulty of interpreting the stone count data. At
times of low sea-level, in the cold stages, the rivers would have extended over the sea-floor as the
sea retreated. Thus, in the harsher climates of the cold stages, gravels could have been deposited far
out into the into the North Sea basin and then reworked into younger deposits when sea-level rose.
This could account for the Lower Greensand chert at Easton Bavents.
However, in the overlying Westleton Beds, stone counts again show fluvial inputs, but now from
three possible sources (Green & McGregor 1999), from the north {Rhaxella chert, spicular flint),
from the west (quartz, quartzite, Carboniferous chert) and the south (Lower Greensand chert). Thus
the early Thames and Bytham River were in existence and there was a northern input by another
river, the Ancaster River (Rose et al., 1999; Rose 2000; Rose et al. 2000,2001), or by longshore drift
(Sinclair 1990) (fig. 3a).
The Early Thames in East Anglia
Associated with the switch from the sand-dominated Red and Norwich Crags to the gravel-dominated
Wroxham Formation marks a very significant change in river behaviour. Prior to 2.5 million years
ago, the scale of climatic change was low. The climate in the cold stages was not severe, consequently
rivers were low energy and moved only small proportions of coarse bcdload. After that time, the
cold stages became far more severe and coarser bedloads were generated, though they were worked
downstream slowly and it was not until just after 2.0 million years ago that coarse sediment was
reaching East Anglia regularly. With the establishment of vigorous river systems, their catchments
can be determined from the lithologies involved (Table 3).
In Norfolk, in the members of the Wroxham Formation, stone counts from members of the Wroxham
Formation illustrate the changing influence of the Thames (Tabic 3). The Dobbs Member is
characterised by 80-90% flint and about 10% far-travelled lithologies, mainly white or colourless
quartzose rocks, Carboniferous and Rhaxella chert and traces of Greensand chert, in the How Hill
Member, the far-travelled component dramatically increases to 30-40% quartzose rocks, now
Essex Naturalist (New Series) 18 (2001)
17